diff --git a/docs/advanced.rst b/docs/advanced.rst index 6f8567ba..9c30f1c5 100644 --- a/docs/advanced.rst +++ b/docs/advanced.rst @@ -148,7 +148,8 @@ In addition to tesseract, OCRmyPDF uses the following external binaries: - ``gs`` (Ghostscript) - ``unpaper`` -- ``qpdf`` +- ``pngquant`` +- ``jbig2`` In each case OCRmyPDF will search the ``PATH`` environment variable to locate the binaries. diff --git a/docs/api.rst b/docs/api.rst index c38bb400..ce6583fc 100644 --- a/docs/api.rst +++ b/docs/api.rst @@ -56,8 +56,8 @@ OCRmyPDF does not. On Windows, the script that calls ``ocrmypdf.ocr()`` must be protected by an "ifmain" guard (``if __name__ == '__main__'``) or you must use ``ocrmypdf.ocr(...use_threads=True)``. If you do not take at least one - of these steps, Windows fork semantics will prevent OCRmyPDF from working - correct. + of these steps, Windows process semantics will prevent OCRmyPDF from working + correctly. Logging ------- @@ -105,8 +105,8 @@ Reference :members: :undoc-members: -.. autoclass:: ocrmypdf.ExitCode +.. autofunction:: ocrmypdf.configure_logging + +.. automodule:: ocrmypdf.exceptions :members: :undoc-members: - -.. autofunction:: ocrmypdf.configure_logging diff --git a/docs/errors.rst b/docs/errors.rst index 825cd656..bf0b53d0 100644 --- a/docs/errors.rst +++ b/docs/errors.rst @@ -22,14 +22,20 @@ As the error message suggests, your options are: - ``ocrmypdf --skip-text`` to skip OCR and other processing on any pages that contain text. Text pages will be copied into the output PDF without modification. +- ``ocrmypdf --redo-ocr`` to scan the file for any existing OCR + (non-printing text), remove it, and do OCR again. This is one way + to take advantage of improvements in OCR accuracy. Printable vector + text is excluded from OCR, so this can be used on files that contain + a mix of digital and scanned files. + Input file 'filename' is not a valid PDF ======================================== -OCRmyPDF passes files through qpdf, a program that fixes errors in PDFs, -before it tries to work on them. In most cases this happens because the -PDF is corrupt and truncated (incomplete file copying) and not much can -be done. +OCRmyPDF checks files with pikepdf, a library that in turn uses libqpdf to fixes +errors in PDFs, before it tries to work on them. In most cases this happens +because the PDF is corrupt and truncated (incomplete file copying) and not much +can be done. You can try rewriting the file with Ghostscript: diff --git a/docs/index.rst b/docs/index.rst index bf042ae1..2591b1b9 100644 --- a/docs/index.rst +++ b/docs/index.rst @@ -26,8 +26,8 @@ image processing and OCR to existing PDFs. docker advanced batch - security performance + pdfsecurity errors .. toctree:: @@ -35,6 +35,7 @@ image processing and OCR to existing PDFs. :maxdepth: 2 api + plugins contributing Indices and tables diff --git a/docs/security.rst b/docs/pdfsecurity.rst similarity index 95% rename from docs/security.rst rename to docs/pdfsecurity.rst index bcc69e8e..36246960 100644 --- a/docs/security.rst +++ b/docs/pdfsecurity.rst @@ -68,7 +68,7 @@ license, OCRmyPDF's GPL license, and any other licenses. Setting aside these concerns, a side effect of OCRmyPDF is it may incidentally sanitize PDFs that contain certain types of malware. It -runs ``qpdf`` to repair the PDF, which could correct malformed PDF +repairs the PDF with pikepdf/libqpdf, which could correct malformed PDF structures that are part of an attack. When PDF/A output is selected (the default), the input PDF is partially reconstructed by Ghostscript. When ``--force-ocr`` is used, all pages are rasterized and reconverted @@ -144,10 +144,9 @@ set, the document cannot be viewed without the password. Either way, OCRmyPDF does not remove passwords from PDFs and exits with an error on encountering them. -``qpdf``, one of OCRmyPDF's dependencies, can remove passwords. If the -owner and user password are set, a password is required for ``qpdf``. If -only the owner password is set, then the password can be stripped, even -if one does not have the owner password. +``qpdf`` can remove passwords. If the owner and user password are set, a +password is required for ``qpdf``. If only the owner password is set, then the +password can be stripped, even if one does not have the owner password. After OCR is applied, password protection is not permitted on PDF/A documents but the file can be converted to regular PDF. diff --git a/docs/plugins.rst b/docs/plugins.rst index e22cea36..f2ac0b94 100644 --- a/docs/plugins.rst +++ b/docs/plugins.rst @@ -2,23 +2,79 @@ Plugins ======= -You can use plugins to customize the behavior of OCRmyPDF at certain -points of interest. +You can use plugins to customize the behavior of OCRmyPDF at certain points of +interest. -Currently, it is possible to: - override the decision for whether or not -to perform OCR on a particular file - modify the image is about to be -sent for OCR +Currently, it is possible to: + +- add new command line arguments +- override the decision for whether or not to perform OCR on a particular file +- modify the image is about to be sent for OCR +- modify the page image before it is converted to PDF + +OCRmyPDF plugins are based on the Python ``pluggy`` package and conform to its +conventions. Note that: plugins installed with as setuptools entrypoints are +not checked currently, because OCRmyPDF assumes you may not want to enable +plugins for all files. Also, plugins must be functions, not classes. How plugins are imported ======================== -Plugins are imported on demand, by the OCRmyPDF worker process that -needs to use them. As such, plugins cannot share state with each other, -and will be imported many times, once for each worker process. +Plugins are imported on demand, by the OCRmyPDF worker process that needs to use +them. As such, plugins cannot share state with other plugins, cannot rely on +their module's or the interpreter's global state, and should expect asynchronous +copies of themselves to be running. Plugins can write intermediate files to the +folder specified in ``options.work_folder``. -Plugins currently cannot override the same hook. +Plugins should work whether executed in threads or processes. -How plugins are invoked -======================= +Script plugins +============== -Plugins may be called from the command line: +Script plugins may be called from the command line, by specifying the name of a file. + +.. code-block:: bash + + ocrmypdf --plugin example_plugin.py input.pdf output.pdf + +Multiple plugins may be called by issuing the ``--plugin`` argument multiple times. + +Packaged plugins +================ + +Installed plugins may be installed into the same virtual environment as OCRmyPDF +is installed into. They may be invoked using Python standard module naming. + +.. code-block:: bash + + ocrmypdf --plugin ocrmypdf_fancypants.pockets.contents input.pdf output.pdf + +OCRmyPDF does not automatically import plugins, because the assumption is that +plugins affect different files differently and you may not want them activated +all the time. The command line or ``ocrmypdf.ocr(plugin='...')`` must call +for them. + +Third parties that wish to distribute packages for ocrmypdf should package them +as packaged plugins, and these modules should begin with the name ``ocrmypdf_`` +similar to ``pytest`` packages such as ``pytest-cov`` (the package) and +``pytest_cov`` (the module). + +Plugin hooks +============ + +A plugin may provide the following hooks. Hooks should be decorated with +``ocrmypdf.hookimpl``, for example: + +.. code-block:: python + + from ocrmpydf import hookimpl + + @hookimpl + def prepare(options): + pass + +The following is a complete list of hooks that may be installed and when +they are called. + +.. automodule:: ocrmypdf.pluginspec + :members: diff --git a/docs/release_notes.rst b/docs/release_notes.rst index 9a6eeb3b..f494063f 100644 --- a/docs/release_notes.rst +++ b/docs/release_notes.rst @@ -13,6 +13,33 @@ Note that it is licensed under GPLv3, so scripts that ``import ocrmypdf`` and are released publicly should probably also be licensed under GPLv3. +v10.0.0 (not yet released) +========================== + +**Breaking changes** + +- Support for pdfminer.six version 20181108 has been dropped, along with a + monkeypatch that made this version work. +- Ghostscript is no longer used for finding the location of text in PDFs, and + APIs related to this feature have been removed. +- Output messages are now displayed in color (when supported by the terminal) + and prefixes describing the severity of the message are removed. As such + programs that parse OCRmyPDF's log message will need to be revised. (Please + consider using OCRmyPDF as a library instead.) +- Code describing the resolution in DPI of images was refactored into a + ``ocrmypdf.helpers.Resolution`` class. +- A deprecated parameter in ``ocrmypdf.exec.ghostscript.generate_pdfa`` was + removed. +- The deprecated module ``ocrmypdf.exec.qpdf`` was removed. +- The ``ocrmypdf.hocrtransform`` module has been updated to follow PEP8 naming + conventions. + +**New features** + +- PDF page scanning is now parallelized across CPUs, speeding up the "Scan" + phase for files with a high page count. +- Colored log messages. + v9.8.2 ====== diff --git a/misc/example_plugin.py b/misc/example_plugin.py new file mode 100644 index 00000000..d6c93363 --- /dev/null +++ b/misc/example_plugin.py @@ -0,0 +1,53 @@ +# © 2020 James R Barlow: https://github.com/jbarlow83 +# +# This program is free software: you can redistribute it and/or modify +# it under the terms of the GNU General Public License as published by +# the Free Software Foundation, either version 3 of the License, or +# (at your option) any later version. +# +# This program is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with this program. If not, see . + +import logging + +from PIL import Image + +from ocrmypdf import hookimpl + +log = logging.getLogger(__name__) + + +@hookimpl +def add_options(parser): + parser.add_argument('--grayscale-ocr', action='store_true') + + +@hookimpl +def prepare(options): + pass + + +@hookimpl +def validate(pdfinfo, options): + pass + + +@hookimpl +def filter_ocr_image(page, image): + if page.options.grayscale_ocr: + log.info("graying") + return image.convert('L') + return image + + +@hookimpl +def filter_page_image(page, image_filename): + output = image_filename.with_suffix('.jpg') + with Image.open(image_filename) as im: + im.save(output) + return output diff --git a/misc/watcher.py b/misc/watcher.py index c1a3d963..d2381050 100644 --- a/misc/watcher.py +++ b/misc/watcher.py @@ -33,11 +33,11 @@ import ocrmypdf INPUT_DIRECTORY = os.getenv('OCR_INPUT_DIRECTORY', '/input') OUTPUT_DIRECTORY = os.getenv('OCR_OUTPUT_DIRECTORY', '/output') -OUTPUT_DIRECTORY_YEAR_MONTH = bool(os.getenv('OCR_OUTPUT_DIRECTORY_YEAR_MONTH', False)) -ON_SUCCESS_DELETE = bool(os.getenv('OCR_ON_SUCCESS_DELETE', False)) -DESKEW = bool(os.getenv('OCR_DESKEW', False)) +OUTPUT_DIRECTORY_YEAR_MONTH = bool(os.getenv('OCR_OUTPUT_DIRECTORY_YEAR_MONTH', '')) +ON_SUCCESS_DELETE = bool(os.getenv('OCR_ON_SUCCESS_DELETE', '')) +DESKEW = bool(os.getenv('OCR_DESKEW', '')) OCR_JSON_SETTINGS = json.loads(os.getenv('OCR_JSON_SETTINGS', '{}')) -POLL_NEW_FILE_SECONDS = os.getenv('OCR_POLL_NEW_FILE_SECONDS', 1) +POLL_NEW_FILE_SECONDS = int(os.getenv('OCR_POLL_NEW_FILE_SECONDS', '1')) USE_POLLING = bool(os.getenv('OCR_USE_POLLING', False)) LOGLEVEL = os.getenv('OCR_LOGLEVEL', 'INFO').upper() PATTERNS = ['*.pdf'] diff --git a/requirements/main.txt b/requirements/main.txt index a9bc5ee8..7dc37803 100644 --- a/requirements/main.txt +++ b/requirements/main.txt @@ -2,8 +2,9 @@ # setup.py lists a separate set of requirements that are looser to simplify # installation cffi == 1.14.0 +coloredlogs == 14.0 # technically optional img2pdf == 0.3.4 -pdfminer.six == 20200402 +pdfminer.six == 20200517 pikepdf == 1.11.1 Pillow == 7.1.1 reportlab == 3.5.34 diff --git a/setup.cfg b/setup.cfg index f307a2e5..487ed30d 100644 --- a/setup.cfg +++ b/setup.cfg @@ -23,7 +23,7 @@ force_grid_wrap=0 use_parentheses=True line_length=88 known_first_party = ocrmypdf -known_third_party = PIL,_cffi_backend,cffi,flask,gs,img2pdf,pdfminer,pikepdf,pkg_resources,pytest,reportlab,setuptools,sphinx_rtd_theme,tqdm,watchdog,werkzeug +known_third_party = PIL,_cffi_backend,cffi,flask,img2pdf,pdfminer,pikepdf,pkg_resources,pluggy,pytest,reportlab,setuptools,sphinx_rtd_theme,tqdm,watchdog,werkzeug [metadata] license_file = LICENSE diff --git a/setup.py b/setup.py index 419a0c97..9dd07aff 100644 --- a/setup.py +++ b/setup.py @@ -27,22 +27,6 @@ if sys.version_info < (3, 6): print("Python 3.6 or newer is required", file=sys.stderr) sys.exit(1) - -# pylint: disable=w0613 - - -command = next((arg for arg in sys.argv[1:] if not arg.startswith('-')), '') -if command.startswith('install') or command in [ - 'check', - 'test', - 'nosetests', - 'easy_install', -]: - forced = '--force' in sys.argv - if forced: - print("The argument --force is deprecated. Please discontinue use.") - - if 'upload' in sys.argv[1:]: print('Use twine to upload the package - setup.py upload is insecure') sys.exit(1) @@ -95,10 +79,10 @@ setup( use_scm_version={'version_scheme': 'post-release'}, cffi_modules=['src/ocrmypdf/lib/compile_leptonica.py:ffibuilder'], install_requires=[ - 'chardet >= 3.0.4, < 4', # unlisted requirement of pdfminer.six 20181108 'cffi >= 1.9.1', # must be a setup and install requirement + 'coloredlogs >= 14.0', # strictly optional 'img2pdf >= 0.3.0, < 0.4', # pure Python, so track HEAD closely - 'pdfminer.six >= 20181108, <= 20200517', + 'pdfminer.six >= 20191110, <= 20200517', 'pikepdf >= 1.8.1, < 2', 'Pillow >= 6.2.0', 'reportlab >= 3.3.0', # oldest released version with sane image handling diff --git a/src/ocrmypdf/__init__.py b/src/ocrmypdf/__init__.py index 2f76bf4f..d64253b8 100644 --- a/src/ocrmypdf/__init__.py +++ b/src/ocrmypdf/__init__.py @@ -15,10 +15,13 @@ # You should have received a copy of the GNU General Public License # along with OCRmyPDF. If not, see . -from . import helpers, hocrtransform, leptonica, pdfa, pdfinfo -from ._version import PROGRAM_NAME, __version__ -from .api import Verbosity, configure_logging, ocr -from .exceptions import ( + +from pluggy import HookimplMarker as _HookimplMarker + +from ocrmypdf import helpers, hocrtransform, leptonica, pdfa, pdfinfo +from ocrmypdf._version import PROGRAM_NAME, __version__ +from ocrmypdf.api import Verbosity, configure_logging, ocr +from ocrmypdf.exceptions import ( BadArgsError, DpiError, EncryptedPdfError, @@ -33,3 +36,6 @@ from .exceptions import ( TesseractConfigError, UnsupportedImageFormatError, ) +from ocrmypdf.pluginspec import OcrEngine, OrientationConfidence + +hookimpl = _HookimplMarker('ocrmypdf') diff --git a/src/ocrmypdf/__main__.py b/src/ocrmypdf/__main__.py index bcfd6528..69d68db4 100755 --- a/src/ocrmypdf/__main__.py +++ b/src/ocrmypdf/__main__.py @@ -19,18 +19,21 @@ import logging import os import sys +from multiprocessing import set_start_method -from . import __version__ -from ._jobcontext import make_logger -from ._sync import run_pipeline -from ._validation import check_closed_streams, check_options -from .api import Verbosity, configure_logging -from .cli import parser -from .exceptions import BadArgsError, ExitCode, MissingDependencyError +from ocrmypdf import __version__ +from ocrmypdf._plugin_manager import get_parser_options_plugins +from ocrmypdf._sync import run_pipeline +from ocrmypdf._validation import check_closed_streams, check_options +from ocrmypdf.api import Verbosity, configure_logging +from ocrmypdf.cli import get_parser, plugins_only_parser +from ocrmypdf.exceptions import BadArgsError, ExitCode, MissingDependencyError + +log = logging.getLogger('ocrmypdf') def run(args=None): - options = parser.parse_args(args=args) + parser, options, plugin_manager = get_parser_options_plugins(args=args) if not check_closed_streams(options): return ExitCode.bad_args @@ -47,10 +50,9 @@ def run(args=None): configure_logging( verbosity, progress_bar_friendly=options.progress_bar, manage_root_logger=True ) - log = make_logger('ocrmypdf') - log.debug('ocrmypdf ' + __version__) + log.debug('ocrmypdf %s', __version__) try: - check_options(options) + check_options(options, plugin_manager) except ValueError as e: log.error(e) return ExitCode.bad_args @@ -61,9 +63,11 @@ def run(args=None): log.error(e) return ExitCode.missing_dependency - result = run_pipeline(options=options) + result = run_pipeline(options=options, plugin_manager=plugin_manager) return result if __name__ == '__main__': + if sys.platform == 'darwin' and sys.version_info < (3, 8): + set_start_method('spawn') # see python bpo-33725 sys.exit(run()) diff --git a/src/ocrmypdf/_concurrent.py b/src/ocrmypdf/_concurrent.py new file mode 100644 index 00000000..6d608eb6 --- /dev/null +++ b/src/ocrmypdf/_concurrent.py @@ -0,0 +1,135 @@ +# © 2020 James R. Barlow: github.com/jbarlow83 +# +# This file is part of OCRmyPDF. +# +# OCRmyPDF is free software: you can redistribute it and/or modify +# it under the terms of the GNU General Public License as published by +# the Free Software Foundation, either version 3 of the License, or +# (at your option) any later version. +# +# OCRmyPDF is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with OCRmyPDF. If not, see . + +import logging +import logging.handlers +import multiprocessing +import os +import signal +import sys +import threading +from multiprocessing import Pool as ProcessPool +from multiprocessing.dummy import Pool as ThreadPool +from typing import Callable, Iterable, Optional + +from tqdm import tqdm + + +def log_listener(queue): + """Listen to the worker processes and forward the messages to logging + + For simplicity this is a thread rather than a process. Only one process + should actually write to sys.stderr or whatever we're using, so if this is + made into a process the main application needs to be directed to it. + + See https://docs.python.org/3/howto/logging-cookbook.html#logging-to-a-single-file-from-multiple-processes + """ + + while True: + try: + record = queue.get() + if record is None: + break + logger = logging.getLogger(record.name) + logger.handle(record) + except Exception: # pylint: disable=broad-except + import traceback # pylint: disable=import-outside-toplevel + + print("Logging problem", file=sys.stderr) + traceback.print_exc(file=sys.stderr) + + +def process_init(queue, user_init): + """Initialize a process pool worker""" + + # Ignore SIGINT (our parent process will kill us gracefully) + signal.signal(signal.SIGINT, signal.SIG_IGN) + + # Reconfigure the root logger for this process to send all messages to a queue + h = logging.handlers.QueueHandler(queue) + root = logging.getLogger() + root.handlers = [] + root.addHandler(h) + + if user_init: + user_init() + + +def thread_init(_queue, user_init): + if user_init: + user_init() + + +def exec_progress_pool( + *, + use_threads: bool, + max_workers: int, + tqdm_kwargs: dict, + task_initializer: Optional[Callable] = None, + task: Optional[Callable] = None, + task_arguments: Optional[Iterable] = None, + task_finished: Optional[Callable] = None, +): + log_queue = multiprocessing.Queue(-1) + listener = threading.Thread(target=log_listener, args=(log_queue,)) + + if use_threads: + pool_class = ThreadPool + initializer = thread_init + else: + pool_class = ProcessPool + initializer = process_init + listener.start() + + with tqdm(**tqdm_kwargs) as pbar: + pool = pool_class( + processes=max_workers, + initializer=initializer, + initargs=(log_queue, task_initializer), + ) + try: + results = pool.imap_unordered(task, task_arguments) + while True: + try: + result = results.next() + if task_finished: + task_finished(result, pbar) + else: + pbar.update() + except StopIteration: + break + except KeyboardInterrupt: + # Terminate pool so we exit instantly + pool.terminate() + # Don't try listener.join() here, will deadlock + raise + except Exception: + if not os.environ.get("PYTEST_CURRENT_TEST", ""): + # Unless inside pytest, exit immediately because no one wants + # to wait for child processes to finalize results that will be + # thrown away. Inside pytest, we want child processes to exit + # cleanly so that they output an error messages or coverage data + # we need from them. + pool.terminate() + raise + finally: + # Terminate log listener + log_queue.put_nowait(None) + pool.close() + pool.join() + + listener.join() diff --git a/src/ocrmypdf/_exec/__init__.py b/src/ocrmypdf/_exec/__init__.py new file mode 100644 index 00000000..8c6d0bb3 --- /dev/null +++ b/src/ocrmypdf/_exec/__init__.py @@ -0,0 +1,18 @@ +# © 2020 James R. Barlow: github.com/jbarlow83 +# +# This file is part of OCRmyPDF. +# +# OCRmyPDF is free software: you can redistribute it and/or modify +# it under the terms of the GNU General Public License as published by +# the Free Software Foundation, either version 3 of the License, or +# (at your option) any later version. +# +# OCRmyPDF is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with OCRmyPDF. If not, see . + +"""Manage third party executables""" diff --git a/src/ocrmypdf/exec/ghostscript.py b/src/ocrmypdf/_exec/ghostscript.py similarity index 64% rename from src/ocrmypdf/exec/ghostscript.py rename to src/ocrmypdf/_exec/ghostscript.py index 856bc0c1..0fb65b1b 100644 --- a/src/ocrmypdf/exec/ghostscript.py +++ b/src/ocrmypdf/_exec/ghostscript.py @@ -20,9 +20,6 @@ import logging import os import re -import warnings -from contextlib import suppress -from functools import lru_cache from io import BytesIO from os import fspath from pathlib import Path @@ -31,10 +28,11 @@ from subprocess import PIPE, CalledProcessError from PIL import Image -from ..exceptions import MissingDependencyError, SubprocessOutputError -from . import get_version, run +from ocrmypdf.exceptions import MissingDependencyError, SubprocessOutputError +from ocrmypdf.helpers import Resolution +from ocrmypdf.subprocess import get_version, run -gslog = logging.getLogger() +log = logging.getLogger(__name__) GS = 'gs' if os.name == 'nt': @@ -57,12 +55,11 @@ if os.name == 'nt': GS = Path(GS).stem -@lru_cache(maxsize=1) def version(): return get_version(GS) -def jpeg_passthrough_available(): +def jpeg_passthrough_available() -> bool: """Returns True if the installed version of Ghostscript supports JPEG passthru Prior to 9.23, Ghostscript decode and re-encoded JPEGs internally. In 9.23 @@ -79,94 +76,25 @@ def jpeg_passthrough_available(): return version() >= '9.24' -def _gs_error_reported(stream): +def _gs_error_reported(stream) -> bool: return re.search(r'error', stream, flags=re.IGNORECASE) -def extract_text(input_file, pageno=1): - """Use the txtwrite device to get text layout information out - - For details on options of -dTextFormat see - https://www.ghostscript.com/doc/current/VectorDevices.htm#TXT - - Format is like - - - - - - :param pageno: number of page to extract, or all pages if None - :return: XML-ish text representation in bytes - """ - - if pageno is not None: - pages = ['-dFirstPage=%i' % pageno, '-dLastPage=%i' % pageno] - else: - pages = [] - - # Note due to bug https://bugs.ghostscript.com/show_bug.cgi?id=701971 - # Ghostscript <= 9.50 will truncate output unless we write to stdout, so - # don't write to a file. - args_gs = ( - [ - GS, - '-dQUIET', - '-dSAFER', - '-dBATCH', - '-dNOPAUSE', - '-sDEVICE=txtwrite', - '-dTextFormat=0', - ] - + pages - + ['-o', '-', fspath(input_file), "-sstdout=%stderr"] - ) - - try: - p = run(args_gs, stdout=PIPE, stderr=PIPE, check=True) - except CalledProcessError as e: - raise SubprocessOutputError( - 'Ghostscript text extraction failed\n%s\n%s' - % (input_file, e.stderr.decode(errors='replace')) - ) - - return p.stdout - - def rasterize_pdf( - input_file, - output_file, - xres, - yres, - raster_device, - log, - pageno=1, - page_dpi=None, - rotation=None, - filter_vector=False, + input_file: os.PathLike, + output_file: os.PathLike, + *, + raster_device: str, + raster_dpi: Resolution, + pageno: int = 1, + page_dpi: Resolution = None, + rotation: int = None, + filter_vector: bool = False, ): - """Rasterize one page of a PDF at resolution (xres, yres) in canvas units. - - The image is sized to match the integer pixels dimensions implied by - (xres, yres) even if those numbers are noninteger. The image's DPI will - be overridden with the values in page_dpi. - - :param input_file: pathlike - :param output_file: pathlike - :param xres: resolution at which to rasterize page - :param yres: - :param raster_device: - :param log: - :param pageno: page number to rasterize (beginning at page 1) - :param page_dpi: resolution tuple (x, y) overriding output image DPI - :param rotation: 0, 90, 180, 270: clockwise angle to rotate page - :param filter_vector: if True, remove vector graphics objects - :return: - """ - res = round(xres, 6), round(yres, 6) + """Rasterize one page of a PDF at resolution raster_dpi in canvas units.""" + raster_dpi = raster_dpi.round(6) if not page_dpi: - page_dpi = res - if not log: - log = gslog + page_dpi = raster_dpi args_gs = ( [ @@ -178,7 +106,7 @@ def rasterize_pdf( f'-sDEVICE={raster_device}', f'-dFirstPage={pageno}', f'-dLastPage={pageno}', - f'-r{res[0]:f}x{res[1]:f}', + f'-r{raster_dpi.x:f}x{raster_dpi.y:f}', ] + (['-dFILTERVECTOR'] if filter_vector else []) + [ @@ -191,7 +119,6 @@ def rasterize_pdf( ] ) - log.debug(args_gs) try: p = run(args_gs, stdout=PIPE, stderr=PIPE, check=True) except CalledProcessError as e: @@ -216,43 +143,17 @@ def rasterize_pdf( elif rotation == 270: im = im.transpose(Image.ROTATE_270) if rotation % 180 == 90: - page_dpi = page_dpi[1], page_dpi[0] + page_dpi = page_dpi.flip_axis() im.save(fspath(output_file), dpi=page_dpi) def generate_pdfa( pdf_pages, - output_file, - compression, - log, - threads=None, # deprecated parameter - pdf_version='1.5', - pdfa_part='2', + output_file: os.PathLike, + compression: str, + pdf_version: str = '1.5', + pdfa_part: str = '2', ): - """Generate a PDF/A. - - The pdf_pages, a list files, will be merged into output_file. One or more - PDF files may be merged. One of the files in this list must be a pdfmark - file that provides Ghostscript with details on how to perform the PDF/A - conversion. By default with we pick PDF/A-2b, but this works for 1 or 3. - - compression can be 'jpeg', 'lossless', or an empty string. In 'jpeg', - Ghostscript is instructed to convert color and grayscale images to DCT - (JPEG encoding). In 'lossless' Ghostscript is told to convert images to - Flate (lossless/PNG). If the parameter is omitted Ghostscript is left to - make its own decisions about how to encode images; it appears to use a - heuristic to decide how to encode images. As of Ghostscript 9.25, we - support passthrough JPEG which allows Ghostscript to avoid transcoding - images entirely. (The feature was added in 9.23 but broken, and the 9.24 - release of Ghostscript had regressions, so we don't support it until 9.25.) - """ - if not log: - log = gslog - if threads is not None: - warnings.warn( - "use of deprecated parameter 'threads'", category=DeprecationWarning - ) - compression_args = [] if compression == 'jpeg': compression_args = [ diff --git a/src/ocrmypdf/exec/jbig2enc.py b/src/ocrmypdf/_exec/jbig2enc.py similarity index 92% rename from src/ocrmypdf/exec/jbig2enc.py rename to src/ocrmypdf/_exec/jbig2enc.py index 5218edbd..deced89a 100644 --- a/src/ocrmypdf/exec/jbig2enc.py +++ b/src/ocrmypdf/_exec/jbig2enc.py @@ -17,14 +17,12 @@ """Interface to jbig2 executable""" -from functools import lru_cache from subprocess import PIPE -from ..exceptions import MissingDependencyError -from . import get_version, run +from ocrmypdf.exceptions import MissingDependencyError +from ocrmypdf.subprocess import get_version, run -@lru_cache(maxsize=1) def version(): return get_version('jbig2', regex=r'jbig2enc (\d+(\.\d+)*).*') diff --git a/src/ocrmypdf/exec/pngquant.py b/src/ocrmypdf/_exec/pngquant.py similarity index 92% rename from src/ocrmypdf/exec/pngquant.py rename to src/ocrmypdf/_exec/pngquant.py index 17721065..61f197fe 100644 --- a/src/ocrmypdf/exec/pngquant.py +++ b/src/ocrmypdf/_exec/pngquant.py @@ -17,17 +17,14 @@ """Interface to pngquant executable""" -from functools import lru_cache -from subprocess import run from tempfile import NamedTemporaryFile from PIL import Image -from ..exceptions import MissingDependencyError -from . import get_version +from ocrmypdf.exceptions import MissingDependencyError +from ocrmypdf.subprocess import get_version, run -@lru_cache(maxsize=1) def version(): return get_version('pngquant', regex=r'(\d+(\.\d+)*).*') diff --git a/src/ocrmypdf/exec/tesseract.py b/src/ocrmypdf/_exec/tesseract.py similarity index 60% rename from src/ocrmypdf/exec/tesseract.py rename to src/ocrmypdf/_exec/tesseract.py index c8a16f42..a253db74 100644 --- a/src/ocrmypdf/exec/tesseract.py +++ b/src/ocrmypdf/_exec/tesseract.py @@ -23,15 +23,21 @@ import shutil from collections import namedtuple from contextlib import suppress from os import fspath +from pathlib import Path from subprocess import PIPE, STDOUT, CalledProcessError, TimeoutExpired +from typing import List, Optional -from ..exceptions import ( +from PIL import Image + +from ocrmypdf.exceptions import ( MissingDependencyError, SubprocessOutputError, TesseractConfigError, ) -from ..helpers import page_number, safe_symlink -from . import get_version, run +from ocrmypdf.helpers import safe_symlink +from ocrmypdf.subprocess import get_version, run + +log = logging.getLogger(__name__) OrientationConfidence = namedtuple('OrientationConfidence', ('angle', 'confidence')) @@ -59,16 +65,11 @@ class TesseractLoggerAdapter(logging.LoggerAdapter): return '[tesseract] %s' % (msg), kwargs -def version(tesseract_env=None): - return get_version('tesseract', regex=r'tesseract\s(.+)', env=tesseract_env) +def version(): + return get_version('tesseract', regex=r'tesseract\s(.+)') -def v4(tesseract_env=None): - "Is this Tesseract v4.0?" - return version(tesseract_env) >= '4' - - -def has_textonly_pdf(tesseract_env=None, langs=None): +def has_textonly_pdf(langs=None): """Does Tesseract have textonly_pdf capability? Available in v4.00.00alpha since January 2017. Best to @@ -77,28 +78,28 @@ def has_textonly_pdf(tesseract_env=None, langs=None): args_tess = tess_base_args(langs, engine_mode=None) + ['--print-parameters', 'pdf'] params = '' try: - # print-parameters can return non-UTF8 if the parameters are so initialized - proc = run(args_tess, check=True, stdout=PIPE, stderr=STDOUT, env=tesseract_env) + proc = run(args_tess, check=True, stdout=PIPE, stderr=STDOUT) params = proc.stdout except CalledProcessError as e: raise MissingDependencyError( - "Could not --print-parameters from tesseract" + "Could not --print-parameters from tesseract. This can happen if the " + "TESSDATA_PREFIX environment is not set to a valid tessdata folder. " ) from e if b'textonly_pdf' in params: return True return False -def has_user_words(tesseract_env=None): +def has_user_words(): """Does Tesseract have --user-words capability? Not available in 4.0, but available in 4.1. Also available in 3.x, but we no longer support 3.x. """ - return version(tesseract_env) >= '4.1' + return version() >= '4.1' -def languages(tesseract_env=None): +def get_languages(): def lang_error(output): msg = ( "Tesseract failed to report available languages.\n" @@ -111,12 +112,7 @@ def languages(tesseract_env=None): args_tess = ['tesseract', '--list-langs'] try: proc = run( - args_tess, - universal_newlines=True, - stdout=PIPE, - stderr=STDOUT, - check=True, - env=tesseract_env, + args_tess, universal_newlines=True, stdout=PIPE, stderr=STDOUT, check=True ) output = proc.stdout except CalledProcessError as e: @@ -125,11 +121,11 @@ def languages(tesseract_env=None): for line in output.splitlines(): if line.startswith('Error'): raise MissingDependencyError(lang_error(output)) - header, *rest = output.splitlines() + _header, *rest = output.splitlines() return set(lang.strip() for lang in rest) -def tess_base_args(langs, engine_mode): +def tess_base_args(langs: List[str], engine_mode) -> List[str]: args = ['tesseract'] if langs: args.extend(['-l', '+'.join(langs)]) @@ -138,7 +134,7 @@ def tess_base_args(langs, engine_mode): return args -def get_orientation(input_file, engine_mode, timeout: float, log, tesseract_env=None): +def get_orientation(input_file: Path, engine_mode, timeout: float): args_tesseract = tess_base_args(['osd'], engine_mode) + [ '--psm', '0', @@ -147,19 +143,13 @@ def get_orientation(input_file, engine_mode, timeout: float, log, tesseract_env= ] try: - p = run( - args_tesseract, - stdout=PIPE, - stderr=STDOUT, - timeout=timeout, - check=True, - env=tesseract_env, - ) + p = run(args_tesseract, stdout=PIPE, stderr=STDOUT, timeout=timeout, check=True) stdout = p.stdout except TimeoutExpired: return OrientationConfidence(angle=0, confidence=0.0) except CalledProcessError as e: - tesseract_log_output(log, e.output, input_file) + tesseract_log_output(e.stdout) + tesseract_log_output(e.stderr) if ( b'Too few characters. Skipping this page' in e.output or b'Image too large' in e.output @@ -181,15 +171,17 @@ def get_orientation(input_file, engine_mode, timeout: float, log, tesseract_env= return oc -def tesseract_log_output(mainlog, stdout, input_file): - log = TesseractLoggerAdapter( - mainlog, extra=mainlog.extra if hasattr(mainlog, 'extra') else None +def tesseract_log_output(stream): + tlog = TesseractLoggerAdapter( + log, extra=log.extra if hasattr(log, 'extra') else None ) + if not stream: + return try: - text = stdout.decode() + text = stream.decode() except UnicodeDecodeError: - text = stdout.decode('utf-8', 'ignore') + text = stream.decode('utf-8', 'ignore') lines = text.splitlines() for line in lines: @@ -198,67 +190,58 @@ def tesseract_log_output(mainlog, stdout, input_file): elif line.startswith("Warning in pixReadMem"): continue elif 'diacritics' in line: - log.warning("lots of diacritics - possibly poor OCR") + tlog.warning("lots of diacritics - possibly poor OCR") elif line.startswith('OSD: Weak margin'): - log.warning("unsure about page orientation") + tlog.warning("unsure about page orientation") elif 'Error in pixScanForForeground' in line: pass # Appears to be spurious/problem with nonwhite borders elif 'Error in boxClipToRectangle' in line: pass # Always appears with pixScanForForeground message elif 'parameter not found: ' in line.lower(): - log.error(line.strip()) + tlog.error(line.strip()) problem = line.split('found: ')[1] raise TesseractConfigError(problem) elif 'error' in line.lower() or 'exception' in line.lower(): - log.error(line.strip()) + tlog.error(line.strip()) elif 'warning' in line.lower(): - log.warning(line.strip()) + tlog.warning(line.strip()) elif 'read_params_file' in line.lower(): - log.error(line.strip()) + tlog.error(line.strip()) else: - log.info(line.strip()) + tlog.info(line.strip()) -def page_timedout(log, input_file, timeout): +def page_timedout(timeout): if timeout == 0: return - prefix = f"{(page_number(input_file)):4d}: [tesseract] " - log.warning(prefix + " took too long to OCR - skipping") + log.warning("[tesseract] took too long to OCR - skipping") -def _generate_null_hocr(output_hocr, output_sidecar, image): +def _generate_null_hocr(output_hocr, output_text, image): """Produce a .hocr file that reports no text detected on a page that is the same size as the input image.""" - from PIL import Image - with Image.open(image) as im: w, h = im.size - with open(output_hocr, 'w', encoding="utf-8") as f: - f.write(HOCR_TEMPLATE.format(w, h)) - with open(output_sidecar, 'w', encoding='utf-8') as f: - f.write('[skipped page]') + output_hocr.write_text(HOCR_TEMPLATE.format(w, h), encoding='utf-8') + output_text.write_text('[skipped page]', encoding='utf-8') def generate_hocr( - input_file, - output_files, - language: list, + input_file: Path, + output_hocr: Path, + output_text: Path, + languages: list, engine_mode, tessconfig: list, timeout: float, pagesegmode: int, user_words, user_patterns, - tesseract_env, - log, ): + prefix = output_hocr.with_suffix('') - output_hocr = next(o for o in output_files if fspath(o).endswith('.hocr')) - output_sidecar = next(o for o in output_files if fspath(o).endswith('.txt')) - prefix = os.path.splitext(output_hocr)[0] - - args_tesseract = tess_base_args(language, engine_mode) + args_tesseract = tess_base_args(languages, engine_mode) if pagesegmode is not None: args_tesseract.extend(['--psm', str(pagesegmode)]) @@ -269,94 +252,70 @@ def generate_hocr( if user_patterns: args_tesseract.extend(['--user-patterns', user_patterns]) - # Reminder: test suite tesseract spoofers will break after any changes + # Reminder: test suite tesseract test plugins will break after any changes # to the number of order parameters here args_tesseract.extend([input_file, prefix, 'hocr', 'txt'] + tessconfig) try: - p = run( - args_tesseract, - stdout=PIPE, - stderr=STDOUT, - timeout=timeout, - check=True, - env=tesseract_env, - ) + p = run(args_tesseract, stdout=PIPE, stderr=STDOUT, timeout=timeout, check=True) stdout = p.stdout except TimeoutExpired: # Generate a HOCR file with no recognized text if tesseract times out # Temporary workaround to hocrTransform not being able to function if # it does not have a valid hOCR file. - page_timedout(log, input_file, timeout) - _generate_null_hocr(output_hocr, output_sidecar, input_file) + page_timedout(timeout) + _generate_null_hocr(output_hocr, output_text, input_file) except CalledProcessError as e: - tesseract_log_output(log, e.output, input_file) + tesseract_log_output(e.output) if b'Image too large' in e.output: - _generate_null_hocr(output_hocr, output_sidecar, input_file) + _generate_null_hocr(output_hocr, output_text, input_file) return raise SubprocessOutputError() from e else: - tesseract_log_output(log, stdout, input_file) + tesseract_log_output(stdout) # The sidecar text file will get the suffix .txt; rename it to # whatever caller wants it named - if os.path.exists(prefix + '.txt'): - shutil.move(prefix + '.txt', output_sidecar) + if prefix.with_suffix('.txt').exists(): + shutil.move(prefix.with_suffix('.txt'), output_text) -def use_skip_page(text_only, skip_pdf, output_pdf, output_text): - with open(output_text, 'w') as f: - f.write('[skipped page]') +def use_skip_page(output_pdf, output_text): + output_text.write_text('[skipped page]', encoding='utf-8') - if skip_pdf and not text_only: - # Substitute a "skipped page" - with suppress(FileNotFoundError): - os.remove(output_pdf) # In case it was partially created - safe_symlink(skip_pdf, output_pdf) - return - - # Or normally, just write a 0 byte file to the output to indicate a skip - with open(output_pdf, 'wb') as out: - out.write(b'') + # A 0 byte file to the output to indicate a skip + output_pdf.write_bytes(b'') def generate_pdf( *, - input_image, - skip_pdf=None, - output_pdf, - output_text, - language: list, + input_file: Path, + output_pdf: Path, + output_text: Path, + languages: List[str], engine_mode, - text_only: bool, - tessconfig: list, + tessconfig: List[str], timeout: float, pagesegmode: int, user_words, user_patterns, - tesseract_env, - log, ): """Use Tesseract to render a PDF. - input_image -- image to analyze - skip_pdf -- if we time out, use this file as output + input_file -- image to analyze output_pdf -- file to generate output_text -- OCR text file - language -- list of languages to consider + languages -- list of languages to consider engine_mode -- engine mode argument for tess v4 - text_only -- enable tesseract text only mode? tessconfig -- tesseract configuration timeout -- timeout (seconds) - log -- logger object """ - args_tesseract = tess_base_args(language, engine_mode) + args_tesseract = tess_base_args(languages, engine_mode) if pagesegmode is not None: args_tesseract.extend(['--psm', str(pagesegmode)]) - if text_only and has_textonly_pdf(tesseract_env, language): - args_tesseract.extend(['-c', 'textonly_pdf=1']) + args_tesseract.extend(['-c', 'textonly_pdf=1']) if user_words: args_tesseract.extend(['--user-words', user_words]) @@ -366,30 +325,23 @@ def generate_pdf( prefix = os.path.splitext(output_pdf)[0] # Tesseract appends suffixes - # Reminder: test suite tesseract spoofers might break after any changes + # Reminder: test suite tesseract test plugins might break after any changes # to the number of order parameters here - args_tesseract.extend([input_image, prefix, 'pdf', 'txt'] + tessconfig) + args_tesseract.extend([input_file, prefix, 'pdf', 'txt'] + tessconfig) try: - p = run( - args_tesseract, - stdout=PIPE, - stderr=STDOUT, - timeout=timeout, - check=True, - env=tesseract_env, - ) + p = run(args_tesseract, stdout=PIPE, stderr=STDOUT, timeout=timeout, check=True) stdout = p.stdout if os.path.exists(prefix + '.txt'): shutil.move(prefix + '.txt', output_text) except TimeoutExpired: - page_timedout(log, input_image, timeout) - use_skip_page(text_only, skip_pdf, output_pdf, output_text) + page_timedout(timeout) + use_skip_page(output_pdf, output_text) except CalledProcessError as e: - tesseract_log_output(log, e.output, input_image) + tesseract_log_output(e.output) if b'Image too large' in e.output: - use_skip_page(text_only, skip_pdf, output_pdf, output_text) + use_skip_page(output_pdf, output_text) return raise SubprocessOutputError() from e else: - tesseract_log_output(log, stdout, input_image) + tesseract_log_output(stdout) diff --git a/src/ocrmypdf/exec/unpaper.py b/src/ocrmypdf/_exec/unpaper.py similarity index 87% rename from src/ocrmypdf/exec/unpaper.py rename to src/ocrmypdf/_exec/unpaper.py index 2984b455..e1a58746 100644 --- a/src/ocrmypdf/exec/unpaper.py +++ b/src/ocrmypdf/_exec/unpaper.py @@ -20,25 +20,27 @@ """Interface to unpaper executable""" +import logging import os import shlex -from functools import lru_cache +from pathlib import Path from subprocess import PIPE, STDOUT, CalledProcessError from tempfile import TemporaryDirectory from PIL import Image -from ..exceptions import MissingDependencyError, SubprocessOutputError -from . import get_version -from . import run as external_run +from ocrmypdf.exceptions import MissingDependencyError, SubprocessOutputError +from ocrmypdf.subprocess import get_version +from ocrmypdf.subprocess import run as external_run + +log = logging.getLogger(__name__) -@lru_cache(maxsize=1) def version(): return get_version('unpaper') -def run(input_file, output_file, dpi, log, mode_args): +def run(input_file, output_file, dpi, mode_args): args_unpaper = ['unpaper', '-v', '--dpi', str(dpi)] + mode_args SUFFIXES = {'1': '.pbm', 'L': '.pgm', 'RGB': '.ppm'} @@ -64,8 +66,8 @@ def run(input_file, output_file, dpi, log, mode_args): "Failed to convert image to a supported format." ) from e - input_pnm = os.path.join(tmpdir, f'input{suffix}') - output_pnm = os.path.join(tmpdir, f'output{suffix}') + input_pnm = Path(tmpdir) / f'input{suffix}' + output_pnm = Path(tmpdir) / f'output{suffix}' im.save(input_pnm, format='PPM') # To prevent any shenanigans from accepting arbitrary parameters in @@ -75,7 +77,7 @@ def run(input_file, output_file, dpi, log, mode_args): # 3) append absolute paths for the input and output file # This should ensure that a user cannot clobber some other file with # their unpaper arguments (whether intentionally or otherwise) - args_unpaper.extend([input_pnm, output_pnm]) + args_unpaper.extend([os.fspath(input_pnm), os.fspath(output_pnm)]) try: proc = external_run( args_unpaper, @@ -110,7 +112,7 @@ def validate_custom_args(args: str): return unpaper_args -def clean(input_file, output_file, dpi, log, unpaper_args=None): +def clean(input_file, output_file, dpi, unpaper_args=None): default_args = [ '--layout', 'none', @@ -124,4 +126,4 @@ def clean(input_file, output_file, dpi, log, unpaper_args=None): ] if not unpaper_args: unpaper_args = default_args - run(input_file, output_file, dpi, log, unpaper_args) + run(input_file, output_file, dpi, unpaper_args) diff --git a/src/ocrmypdf/_graft.py b/src/ocrmypdf/_graft.py index a535d492..de9ebda0 100644 --- a/src/ocrmypdf/_graft.py +++ b/src/ocrmypdf/_graft.py @@ -15,12 +15,13 @@ # You should have received a copy of the GNU General Public License # along with OCRmyPDF. If not, see . -import os +import logging from contextlib import suppress from pathlib import Path import pikepdf +log = logging.getLogger(__name__) MAX_REPLACE_PAGES = 100 @@ -88,99 +89,10 @@ def strip_invisible_text(pdf, page): page.Contents = pikepdf.Stream(pdf, content_stream) -def _graft_text_layer( - *, pdf_base, page_num, text, font, font_key, procset, rotation, strip_old_text, log -): - """Insert the text layer from text page 0 on to pdf_base at page_num""" - - log.debug("Grafting") - if Path(text).stat().st_size == 0: - return - - # This is a pointer indicating a specific page in the base file - pdf_text = pikepdf.open(text) - pdf_text_contents = pdf_text.pages[0].Contents.read_bytes() - - base_page = pdf_base.pages.p(page_num) - - # The text page always will be oriented up by this stage but the original - # content may have a rotation applied. Wrap the text stream with a rotation - # so it will be oriented the same way as the rest of the page content. - # (Previous versions OCRmyPDF rotated the content layer to match the text.) - mediabox = [float(pdf_text.pages[0].MediaBox[v]) for v in range(4)] - wt, ht = mediabox[2] - mediabox[0], mediabox[3] - mediabox[1] - - mediabox = [float(base_page.MediaBox[v]) for v in range(4)] - wp, hp = mediabox[2] - mediabox[0], mediabox[3] - mediabox[1] - - translate = pikepdf.PdfMatrix().translated(-wt / 2, -ht / 2) - untranslate = pikepdf.PdfMatrix().translated(wp / 2, hp / 2) - corner = pikepdf.PdfMatrix().translated(mediabox[0], mediabox[1]) - # -rotation because the input is a clockwise angle and this formula - # uses CCW - rotation = -rotation % 360 - rotate = pikepdf.PdfMatrix().rotated(rotation) - - # Because of rounding of DPI, we might get a text layer that is not - # identically sized to the target page. Scale to adjust. Normally this - # is within 0.998. - if rotation in (90, 270): - wt, ht = ht, wt - scale_x = wp / wt - scale_y = hp / ht - - # log.debug('%r', scale_x, scale_y) - scale = pikepdf.PdfMatrix().scaled(scale_x, scale_y) - - # Translate the text so it is centered at (0, 0), rotate it there, adjust - # for a size different between initial and text PDF, then untranslate, and - # finally move the lower left corner to match the mediabox - ctm = translate @ rotate @ scale @ untranslate @ corner - - pdf_text_contents = b'q %s cm\n' % ctm.encode() + pdf_text_contents + b'\nQ\n' - - new_text_layer = pikepdf.Stream(pdf_base, pdf_text_contents) - - if strip_old_text: - strip_invisible_text(pdf_base, base_page) - - base_page.page_contents_add(new_text_layer, prepend=True) - - _update_page_resources( - page=base_page, font=font, font_key=font_key, procset=procset - ) - pdf_text.close() - - -def _find_font(text, pdf_base): - """Copy a font from the filename text into pdf_base""" - - font, font_key = None, None - possible_font_names = ('/f-0-0', '/F1') - try: - with pikepdf.open(text) as pdf_text: - try: - pdf_text_fonts = pdf_text.pages[0].Resources.get('/Font', {}) - except (AttributeError, IndexError, KeyError): - return None, None - for f in possible_font_names: - pdf_text_font = pdf_text_fonts.get(f, None) - if pdf_text_font is not None: - font_key = f - break - if pdf_text_font: - font = pdf_base.copy_foreign(pdf_text_font) - return font, font_key - except (FileNotFoundError, pikepdf.PdfError): - # PdfError occurs if a 0-length file is written e.g. due to OCR timeout - return None, None - - class OcrGrafter: def __init__(self, context): self.context = context - self.log = context.log - self.path_base = Path(context.origin).resolve() + self.path_base = context.origin self.pdf_base = pikepdf.open(self.path_base) self.font, self.font_key = None, None @@ -195,10 +107,11 @@ class OcrGrafter: self.emplacements = 1 self.interim_count = 0 - def graft_page(self, page_result): - pageno, image, text, _sidecar, autorotate_correction = page_result - if text and not self.font: - self.font, self.font_key = _find_font(text, self.pdf_base) + def graft_page( + self, *, pageno: int, image: Path, textpdf: Path, autorotate_correction: int + ): + if textpdf and not self.font: + self.font, self.font_key = self._find_font(textpdf) emplaced_page = False content_rotation = self.pdfinfo[pageno].rotation @@ -206,7 +119,7 @@ class OcrGrafter: if path_image is not None and path_image != self.path_base: # We are updating the old page with a rasterized PDF of the new # page (without changing objgen, to preserve references) - self.log.debug("Emplacement update") + log.debug("Emplacement update") with pikepdf.open(image) as pdf_image: self.emplacements += 1 foreign_image_page = pdf_image.pages[0] @@ -220,25 +133,23 @@ class OcrGrafter: content_rotation = autorotate_correction text_rotation = autorotate_correction text_misaligned = (text_rotation - content_rotation) % 360 - self.log.debug( + log.debug( f"Rotations for page {pageno}: [text, auto, misalign, content] = " f"{text_rotation}, {autorotate_correction}, " f"{text_misaligned}, {content_rotation}" ) - if text and self.font: + if textpdf and self.font: # Graft the text layer onto this page, whether new or old strip_old = self.context.options.redo_ocr - _graft_text_layer( - pdf_base=self.pdf_base, + self._graft_text_layer( page_num=pageno + 1, - text=text, + textpdf=textpdf, font=self.font, font_key=self.font_key, rotation=text_misaligned, procset=self.procset, strip_old_text=strip_old, - log=self.log, ) # Correct the rotation if applicable @@ -250,10 +161,13 @@ class OcrGrafter: self.save_and_reload() def save_and_reload(self): - # Periodically save and reload the Pdf object. This will keep a - # lid on our memory usage for very large files. Attach the font to - # page 1 even if page 1 doesn't use it, so we have a way to get it - # back. + """Save and reload the Pdf. + + This will keep a lid on our memory usage for very large files. Attach + the font to page 1 even if page 1 doesn't use it, so we have a way to get it + back. + """ + page0 = self.pdf_base.pages[0] _update_page_resources( page=page0, font=self.font, font_key=self.font_key, procset=self.procset @@ -264,12 +178,14 @@ class OcrGrafter: # {interim_count} is the opened file we were updateing # {interim_count - 1} can be deleted # {interim_count + 1} is the new file will produce and open - old_file = self.output_file + f'_working{self.interim_count - 1}.pdf' + old_file = self.output_file.with_suffix(f'.working{self.interim_count - 1}.pdf') if not self.context.options.keep_temporary_files: with suppress(FileNotFoundError): - os.unlink(old_file) + old_file.unlink() - next_file = self.output_file + f'_working{self.interim_count + 1}.pdf' + next_file = self.output_file.with_suffix( + f'.working{self.interim_count + 1}.pdf' + ) self.pdf_base.save(next_file) self.pdf_base.close() @@ -282,3 +198,98 @@ class OcrGrafter: self.pdf_base.save(self.output_file) self.pdf_base.close() return self.output_file + + def _find_font(self, text): + """Copy a font from the filename text into pdf_base""" + + font, font_key = None, None + possible_font_names = ('/f-0-0', '/F1') + try: + with pikepdf.open(text) as pdf_text: + try: + pdf_text_fonts = pdf_text.pages[0].Resources.get('/Font', {}) + except (AttributeError, IndexError, KeyError): + return None, None + for f in possible_font_names: + pdf_text_font = pdf_text_fonts.get(f, None) + if pdf_text_font is not None: + font_key = f + break + if pdf_text_font: + font = self.pdf_base.copy_foreign(pdf_text_font) + return font, font_key + except (FileNotFoundError, pikepdf.PdfError): + # PdfError occurs if a 0-length file is written e.g. due to OCR timeout + return None, None + + def _graft_text_layer( + self, + *, + page_num: int, + textpdf: Path, + font: pikepdf.Object, + font_key: pikepdf.Object, + procset: pikepdf.Object, + rotation: int, + strip_old_text: bool, + ): + """Insert the text layer from text page 0 on to pdf_base at page_num""" + + log.debug("Grafting") + if Path(textpdf).stat().st_size == 0: + return + + # This is a pointer indicating a specific page in the base file + with pikepdf.open(textpdf) as pdf_text: + pdf_text_contents = pdf_text.pages[0].Contents.read_bytes() + + base_page = self.pdf_base.pages.p(page_num) + + # The text page always will be oriented up by this stage but the original + # content may have a rotation applied. Wrap the text stream with a rotation + # so it will be oriented the same way as the rest of the page content. + # (Previous versions OCRmyPDF rotated the content layer to match the text.) + mediabox = [float(pdf_text.pages[0].MediaBox[v]) for v in range(4)] + wt, ht = mediabox[2] - mediabox[0], mediabox[3] - mediabox[1] + + mediabox = [float(base_page.MediaBox[v]) for v in range(4)] + wp, hp = mediabox[2] - mediabox[0], mediabox[3] - mediabox[1] + + translate = pikepdf.PdfMatrix().translated(-wt / 2, -ht / 2) + untranslate = pikepdf.PdfMatrix().translated(wp / 2, hp / 2) + corner = pikepdf.PdfMatrix().translated(mediabox[0], mediabox[1]) + # -rotation because the input is a clockwise angle and this formula + # uses CCW + rotation = -rotation % 360 + rotate = pikepdf.PdfMatrix().rotated(rotation) + + # Because of rounding of DPI, we might get a text layer that is not + # identically sized to the target page. Scale to adjust. Normally this + # is within 0.998. + if rotation in (90, 270): + wt, ht = ht, wt + scale_x = wp / wt + scale_y = hp / ht + + # log.debug('%r', scale_x, scale_y) + scale = pikepdf.PdfMatrix().scaled(scale_x, scale_y) + + # Translate the text so it is centered at (0, 0), rotate it there, adjust + # for a size different between initial and text PDF, then untranslate, and + # finally move the lower left corner to match the mediabox + ctm = translate @ rotate @ scale @ untranslate @ corner + + pdf_text_contents = ( + b'q %s cm\n' % ctm.encode() + pdf_text_contents + b'\nQ\n' + ) + + new_text_layer = pikepdf.Stream(self.pdf_base, pdf_text_contents) + + if strip_old_text: + strip_invisible_text(self.pdf_base, base_page) + + base_page.page_contents_add(new_text_layer, prepend=True) + + _update_page_resources( + page=base_page, font=font, font_key=font_key, procset=procset + ) diff --git a/src/ocrmypdf/_jobcontext.py b/src/ocrmypdf/_jobcontext.py index 42c96282..3091fbc8 100644 --- a/src/ocrmypdf/_jobcontext.py +++ b/src/ocrmypdf/_jobcontext.py @@ -15,40 +15,26 @@ # You should have received a copy of the GNU General Public License # along with OCRmyPDF. If not, see . -import logging import os import shutil import sys +from functools import partial +from pathlib import Path + +from ocrmypdf._plugin_manager import get_plugin_manager -class PicklableLoggerMixin: - def __init__(self): - self._log = None - - @property - def log(self): - if not self._log: - self._log = self.get_logger() - return self._log - - def __getstate__(self): - # Python 3.6 is incapable of pickling a logger and marshalling it to another - # process (threading._RLock error), so we disconnect it before pickling, - # and create a new logger in the worker process. - state = self.__dict__.copy() - state['_log'] = None - return state - - -class PDFContext(PicklableLoggerMixin): +class PdfContext: """Holds our context for a particular run of the pipeline""" - def __init__(self, options, work_folder, origin, pdfinfo): - PicklableLoggerMixin.__init__(self) + def __init__( + self, options, work_folder: Path, origin: Path, pdfinfo, plugin_manager + ): self.options = options - self.work_folder = work_folder - self.origin = origin + self.work_folder = Path(work_folder) + self.origin = Path(origin) self.pdfinfo = pdfinfo + self.plugin_manager = plugin_manager if options: self.name = os.path.basename(options.input_file) else: @@ -56,11 +42,8 @@ class PDFContext(PicklableLoggerMixin): if self.name == '-': self.name = 'stdin' - def get_logger(self): - return make_logger(self.options, filename=self.name) - - def get_path(self, name): - return os.path.join(self.work_folder, name) + def get_path(self, name: str) -> Path: + return self.work_folder / name def get_page_contexts(self): npages = len(self.pdfinfo) @@ -68,27 +51,40 @@ class PDFContext(PicklableLoggerMixin): yield PageContext(self, n) -class PageContext(PicklableLoggerMixin): +class PageContext: """Holds our context for a page Must be pickable, so only store intrinsic/simple data elements """ - def __init__(self, pdf_context, pageno): - PicklableLoggerMixin.__init__(self) + def __init__(self, pdf_context: PdfContext, pageno): self.work_folder = pdf_context.work_folder self.origin = pdf_context.origin self.options = pdf_context.options self.name = pdf_context.name self.pageno = pageno self.pageinfo = pdf_context.pdfinfo[pageno] - self._log = None + self.plugin_manager = pdf_context.plugin_manager - def get_logger(self): - return make_logger(self.options, filename=self.name, page=self.pageno + 1) + def get_path(self, name: str) -> Path: + return self.work_folder / ("%06d_%s" % (self.pageno + 1, name)) - def get_path(self, name): - return os.path.join(self.work_folder, "%06d_%s" % (self.pageno + 1, name)) + def __getstate__(self): + state = self.__dict__.copy() + if state['plugin_manager'] is not None: + del state['plugin_manager'] + state['construct_plugin_manager'] = partial( + get_plugin_manager, self.options.plugins + ) + return state + + def __setstate__(self, state): + self.__dict__.update(state) + if 'construct_plugin_manager' in state: + self.plugin_manager = state['construct_plugin_manager']() + else: + self.plugin_manager = None + del self.__dict__['construct_plugin_manager'] def cleanup_working_files(work_folder, options): @@ -96,29 +92,3 @@ def cleanup_working_files(work_folder, options): print(f"Temporary working files retained at:\n{work_folder}", file=sys.stderr) else: shutil.rmtree(work_folder, ignore_errors=True) - - -class LogNameAdapter(logging.LoggerAdapter): - def process(self, msg, kwargs): - # return '[%s] %s' % (self.extra['input_filename'], msg), kwargs - return '%s' % (msg,), kwargs - - -class LogNamePageAdapter(logging.LoggerAdapter): - def process(self, msg, kwargs): - return ( - #'[%s:%05u] %s' % (self.extra['input_filename'], self.extra['page'], msg), - '%4u: %s' % (self.extra['page'], msg), - kwargs, - ) - - -def make_logger(options=None, prefix='ocrmypdf', filename=None, page=None): - log = logging.getLogger(prefix) - if filename and page: - adapter = LogNamePageAdapter(log, dict(input_filename=filename, page=page)) - elif filename: - adapter = LogNameAdapter(log, dict(input_filename=filename)) - else: - adapter = log - return adapter diff --git a/src/ocrmypdf/_logging.py b/src/ocrmypdf/_logging.py new file mode 100644 index 00000000..5126d97c --- /dev/null +++ b/src/ocrmypdf/_logging.py @@ -0,0 +1,60 @@ +# © 2020 James R. Barlow: github.com/jbarlow83 +# +# This file is part of OCRmyPDF. +# +# OCRmyPDF is free software: you can redistribute it and/or modify +# it under the terms of the GNU General Public License as published by +# the Free Software Foundation, either version 3 of the License, or +# (at your option) any later version. +# +# OCRmyPDF is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with OCRmyPDF. If not, see . + +import logging +import sys +from contextlib import suppress + +from tqdm import tqdm + + +class PageNumberFilter(logging.Filter): + def filter(self, record): + pageno = getattr(record, 'pageno', None) + if pageno is not None: + record.pageno = f'{pageno:5d} ' + else: + record.pageno = '' + return True + + +class TqdmConsole: + """Wrapper to log messages in a way that is compatible with tqdm progress bar + + This routes log messages through tqdm so that it can print them above the + progress bar, and then refresh the progress bar, rather than overwriting + it which looks messy. + + For some reason Python 3.6 prints extra empty messages from time to time, + so we suppress those. + """ + + def __init__(self, file): + self.file = file + self.py36 = sys.version_info[0:2] == (3, 6) + + def write(self, msg): + # When no progress bar is active, tqdm.write() routes to print() + if self.py36: + if msg.strip() != '': + tqdm.write(msg.rstrip(), end='\n', file=self.file) + else: + tqdm.write(msg.rstrip(), end='\n', file=self.file) + + def flush(self): + with suppress(AttributeError): + self.file.flush() diff --git a/src/ocrmypdf/_pipeline.py b/src/ocrmypdf/_pipeline.py index 58ac2495..20bc3dfb 100644 --- a/src/ocrmypdf/_pipeline.py +++ b/src/ocrmypdf/_pipeline.py @@ -15,53 +15,55 @@ # You should have received a copy of the GNU General Public License # along with OCRmyPDF. If not, see . +import logging import os import re import sys from datetime import datetime, timezone -from pathlib import Path from shutil import copyfileobj import img2pdf import pikepdf from pikepdf.models.metadata import encode_pdf_date -from PIL import Image +from PIL import Image, ImageColor, ImageDraw -from . import leptonica -from ._version import PROGRAM_NAME -from ._version import __version__ as VERSION -from .exceptions import ( +from ocrmypdf import leptonica +from ocrmypdf._exec import ghostscript, unpaper +from ocrmypdf._version import PROGRAM_NAME +from ocrmypdf._version import __version__ as VERSION +from ocrmypdf.exceptions import ( DpiError, EncryptedPdfError, InputFileError, PriorOcrFoundError, UnsupportedImageFormatError, ) -from .exec import ghostscript, tesseract -from .helpers import safe_symlink -from .hocrtransform import HocrTransform -from .optimize import optimize -from .pdfa import generate_pdfa_ps -from .pdfinfo import Colorspace, Encoding, PdfInfo +from ocrmypdf.helpers import Resolution, safe_symlink +from ocrmypdf.hocrtransform import HocrTransform +from ocrmypdf.optimize import optimize +from ocrmypdf.pdfa import generate_pdfa_ps +from ocrmypdf.pdfinfo import Colorspace, Encoding, PdfInfo + +log = logging.getLogger(__name__) VECTOR_PAGE_DPI = 400 -def triage_image_file(input_file, output_file, options, log): +def triage_image_file(input_file, output_file, options): log.info("Input file is not a PDF, checking if it is an image...") try: im = Image.open(input_file) except EnvironmentError as e: # Recover the original filename - log.error(str(e).replace(input_file, options.input_file)) + log.error(str(e).replace(str(input_file), str(options.input_file))) raise UnsupportedImageFormatError() from e with im: log.info("Input file is an image") if 'dpi' in im.info: if im.info['dpi'] <= (96, 96) and not options.image_dpi: - log.info("Image size: (%d, %d)" % im.size) - log.info("Image resolution: (%d, %d)" % im.info['dpi']) + log.info("Image size: (%d, %d)", *im.size) + log.info("Image resolution: (%d, %d)", *im.info['dpi']) log.error( "Input file is an image, but the resolution (DPI) is " "not credible. Estimate the resolution at which the " @@ -69,7 +71,7 @@ def triage_image_file(input_file, output_file, options, log): ) raise DpiError() elif not options.image_dpi: - log.info("Image size: (%d, %d)" % im.size) + log.info("Image size: (%d, %d)", *im.size) log.error( "Input file is an image, but has no resolution (DPI) " "in its metadata. Estimate the resolution at which " @@ -96,11 +98,14 @@ def triage_image_file(input_file, output_file, options, log): layout_fun = img2pdf.default_layout_fun if options.image_dpi: layout_fun = img2pdf.get_fixed_dpi_layout_fun( - (options.image_dpi, options.image_dpi) + Resolution(options.image_dpi, options.image_dpi) ) with open(output_file, 'wb') as outf: img2pdf.convert( - input_file, layout_fun=layout_fun, with_pdfrw=False, outputstream=outf + os.fspath(input_file), + layout_fun=layout_fun, + with_pdfrw=False, + outputstream=outf, ) log.info("Successfully converted to PDF, processing...") except img2pdf.ImageOpenError as e: @@ -124,7 +129,7 @@ def _pdf_guess_version(input_file, search_window=1024): return '' -def triage(original_filename, input_file, output_file, options, log): +def triage(original_filename, input_file, output_file, options): try: if _pdf_guess_version(input_file): if options.image_dpi: @@ -137,18 +142,16 @@ def triage(original_filename, input_file, output_file, options, log): return output_file except EnvironmentError as e: log.debug(f"Temporary file was at: {input_file}") - msg = str(e).replace(input_file, original_filename) + msg = str(e).replace(str(input_file), original_filename) raise InputFileError(msg) from e - triage_image_file(input_file, output_file, options, log) + triage_image_file(input_file, output_file, options) return output_file -def get_pdfinfo(input_file, detailed_page_analysis=False, progbar=False): +def get_pdfinfo(input_file, progbar=False, max_workers=None): try: - return PdfInfo( - input_file, detailed_page_analysis=detailed_page_analysis, progbar=progbar - ) + return PdfInfo(input_file, progbar=progbar, max_workers=max_workers) except pikepdf.PasswordError: raise EncryptedPdfError() except pikepdf.PdfError: @@ -156,7 +159,6 @@ def get_pdfinfo(input_file, detailed_page_analysis=False, progbar=False): def validate_pdfinfo_options(context): - log = context.log pdfinfo = context.pdfinfo options = context.options @@ -194,54 +196,56 @@ def validate_pdfinfo_options(context): "form and all filled form fields. The output PDF will be " "'flattened' and will no longer be fillable." ) + context.plugin_manager.hook.validate(pdfinfo=pdfinfo, options=options) def get_page_dpi(pageinfo, options): "Get the DPI when nonsquare DPI is tolerable" xres = max( - pageinfo.xres or VECTOR_PAGE_DPI, - options.oversample or 0, - VECTOR_PAGE_DPI if pageinfo.has_vector else 0, + pageinfo.dpi.x or VECTOR_PAGE_DPI, + options.oversample or 0.0, + VECTOR_PAGE_DPI if pageinfo.has_vector else 0.0, ) yres = max( - pageinfo.yres or VECTOR_PAGE_DPI, + pageinfo.dpi.y or VECTOR_PAGE_DPI, options.oversample or 0, - VECTOR_PAGE_DPI if pageinfo.has_vector else 0, + VECTOR_PAGE_DPI if pageinfo.has_vector else 0.0, ) - return (float(xres), float(yres)) + return Resolution(float(xres), float(yres)) -def get_page_square_dpi(pageinfo, options): +def get_page_square_dpi(pageinfo, options) -> Resolution: "Get the DPI when we require xres == yres, scaled to physical units" - xres = pageinfo.xres or 0 - yres = pageinfo.yres or 0 - userunit = pageinfo.userunit or 1 - return float( + xres = pageinfo.dpi.x or 0.0 + yres = pageinfo.dpi.y or 0.0 + userunit = float(pageinfo.userunit) or 1.0 + units = float( max( (xres * userunit) or VECTOR_PAGE_DPI, (yres * userunit) or VECTOR_PAGE_DPI, - VECTOR_PAGE_DPI if pageinfo.has_vector else 0, - options.oversample or 0, + VECTOR_PAGE_DPI if pageinfo.has_vector else 0.0, + options.oversample or 0.0, ) ) + return Resolution(units, units) -def get_canvas_square_dpi(pageinfo, options): +def get_canvas_square_dpi(pageinfo, options) -> Resolution: """Get the DPI when we require xres == yres, in Postscript units""" - return float( + units = float( max( - (pageinfo.xres) or VECTOR_PAGE_DPI, - (pageinfo.yres) or VECTOR_PAGE_DPI, - VECTOR_PAGE_DPI if pageinfo.has_vector else 0, - options.oversample or 0, + (pageinfo.dpi.x) or VECTOR_PAGE_DPI, + (pageinfo.dpi.y) or VECTOR_PAGE_DPI, + VECTOR_PAGE_DPI if pageinfo.has_vector else 0.0, + options.oversample or 0.0, ) ) + return Resolution(units, units) def is_ocr_required(page_context): pageinfo = page_context.pageinfo options = page_context.options - log = page_context.log ocr_required = True @@ -251,14 +255,15 @@ def is_ocr_required(page_context): elif pageinfo.has_text: if not options.force_ocr and not (options.skip_text or options.redo_ocr): raise PriorOcrFoundError( - "page already has text! - aborting (use --force-ocr to force OCR)" + "page already has text! - aborting (use --force-ocr to force OCR; " + " see also help for the arguments --skip-text and --redo-ocr" ) elif options.force_ocr: log.info("page already has text! - rasterizing text and running OCR anyway") ocr_required = True elif options.redo_ocr: if pageinfo.has_corrupt_text: - log.warn( + log.warning( "some text on this page cannot be mapped to characters: " "consider using --force-ocr instead" ) @@ -285,7 +290,7 @@ def is_ocr_required(page_context): ) elif options.force_ocr: # Warn the user they might not want to do this - log.warn( + log.warning( "page has no images - " "all vector content will be " f"rasterized at {VECTOR_PAGE_DPI} DPI, losing some resolution and likely " @@ -305,7 +310,7 @@ def is_ocr_required(page_context): pixel_count = pageinfo.width_pixels * pageinfo.height_pixels if pixel_count > (options.skip_big * 1_000_000): ocr_required = False - log.warn( + log.warning( "page too big, skipping OCR " f"({(pixel_count / 1_000_000):.1f} MPixels > {options.skip_big:.1f} MPixels --skip-big)" ) @@ -316,14 +321,12 @@ def rasterize_preview(input_file, page_context): output_file = page_context.get_path('rasterize_preview.jpg') canvas_dpi = get_canvas_square_dpi(page_context.pageinfo, page_context.options) page_dpi = get_page_square_dpi(page_context.pageinfo, page_context.options) - ghostscript.rasterize_pdf( - input_file, - output_file, - xres=canvas_dpi, - yres=canvas_dpi, + page_context.plugin_manager.hook.rasterize_pdf_page( + input_file=input_file, + output_file=output_file, raster_device='jpeggray', - log=page_context.log, - page_dpi=(page_dpi, page_dpi), + raster_dpi=canvas_dpi, + page_dpi=page_dpi, pageno=page_context.pageinfo.pageno + 1, ) return output_file @@ -359,33 +362,27 @@ def describe_rotation(page_context, orient_conf, correction): def get_orientation_correction(preview, page_context): - """ - Work out orientation correct for each page. + """Work out orientation correct for each page. We ask Ghostscript to draw a preview page, which will rasterize with the - current /Rotate applied, and then ask Tesseract which way the page is + current /Rotate applied, and then ask OCR which way the page is oriented. If the value of /Rotate is correct (e.g., a user already - manually fixed rotation), then Tesseract will say the page is pointing + manually fixed rotation), then OCR will say the page is pointing up and the correction is zero. Otherwise, the orientation found by - Tesseract represents the clockwise rotation, or the counterclockwise + OCR represents the clockwise rotation, or the counterclockwise correction to rotation. When we draw the real page for OCR, we rotate it by the CCW correction, which points it (hopefully) upright. _graft.py takes care of the orienting the image and text layers. - """ - orient_conf = tesseract.get_orientation( - preview, - engine_mode=page_context.options.tesseract_oem, - timeout=page_context.options.tesseract_timeout, - log=page_context.log, - tesseract_env=page_context.options.tesseract_env, + orient_conf = page_context.plugin_manager.hook.get_ocr_engine().get_orientation( + preview, page_context.options ) correction = orient_conf.angle % 360 - page_context.log.info(describe_rotation(page_context, orient_conf, correction)) + log.info(describe_rotation(page_context, orient_conf, correction)) if ( orient_conf.confidence >= page_context.options.rotate_pages_threshold and correction != 0 @@ -426,21 +423,19 @@ def rasterize( device = colorspaces[device_idx] - page_context.log.debug(f"Rasterize with {device}") + log.debug(f"Rasterize with {device}") # Produce the page image with square resolution or else deskew and OCR # will not work properly. canvas_dpi = get_canvas_square_dpi(pageinfo, page_context.options) page_dpi = get_page_square_dpi(pageinfo, page_context.options) - ghostscript.rasterize_pdf( - input_file, - output_file, - xres=canvas_dpi, - yres=canvas_dpi, + page_context.plugin_manager.hook.rasterize_pdf_page( + input_file=input_file, + output_file=output_file, raster_device=device, - log=page_context.log, - page_dpi=(page_dpi, page_dpi), + raster_dpi=canvas_dpi, + page_dpi=page_dpi, pageno=pageinfo.pageno + 1, rotation=correction, filter_vector=remove_vectors, @@ -454,29 +449,21 @@ def preprocess_remove_background(input_file, page_context): leptonica.remove_background(input_file, output_file) return output_file else: - page_context.log.info("background removal skipped on mono page") + log.info("background removal skipped on mono page") return input_file def preprocess_deskew(input_file, page_context): output_file = page_context.get_path('pp_deskew.png') dpi = get_page_square_dpi(page_context.pageinfo, page_context.options) - leptonica.deskew(input_file, output_file, dpi) + leptonica.deskew(input_file, output_file, dpi.x) return output_file def preprocess_clean(input_file, page_context): - from .exec import unpaper - output_file = page_context.get_path('pp_clean.png') dpi = get_page_square_dpi(page_context.pageinfo, page_context.options) - unpaper.clean( - input_file, - output_file, - dpi, - page_context.log, - page_context.options.unpaper_args, - ) + unpaper.clean(input_file, output_file, dpi.x, page_context.options.unpaper_args) return output_file @@ -488,15 +475,11 @@ def create_ocr_image(image, page_context): output_file = page_context.get_path('ocr.png') options = page_context.options with Image.open(image) as im: - from PIL import ImageColor - from PIL import ImageDraw - white = ImageColor.getcolor('#ffffff', im.mode) # pink = ImageColor.getcolor('#ff0080', im.mode) draw = ImageDraw.ImageDraw(im) - xres, yres = im.info['dpi'] - page_context.log.debug('resolution %r %r' % (xres, yres)) + log.debug('resolution %r', im.info['dpi']) if not options.force_ocr: # Do not mask text areas when forcing OCR, because we need to OCR @@ -512,15 +495,15 @@ def create_ocr_image(image, page_context): # without regard whatever resolution is in pageinfo (may differ or # be None) bbox = [float(v) for v in textarea] - xscale, yscale = float(xres) / 72.0, float(yres) / 72.0 + xyscale = tuple(float(coord) / 72.0 for coord in im.info['dpi']) pixcoords = [ - bbox[0] * xscale, - im.height - bbox[3] * yscale, - bbox[2] * xscale, - im.height - bbox[1] * yscale, + bbox[0] * xyscale[0], + im.height - bbox[3] * xyscale[1], + bbox[2] * xyscale[0], + im.height - bbox[1] * xyscale[1], ] pixcoords = [int(round(c)) for c in pixcoords] - page_context.log.debug('blanking %r', pixcoords) + log.debug('blanking %r', pixcoords) draw.rectangle(pixcoords, fill=white) # draw.rectangle(pixcoords, outline=pink) @@ -530,28 +513,30 @@ def create_ocr_image(image, page_context): im = pix.topil() del draw + + filter_im = page_context.plugin_manager.hook.filter_ocr_image( + page=page_context, image=im + ) + if filter_im is not None: + im = filter_im + # Pillow requires integer DPI - dpi = round(xres), round(yres) + dpi = tuple(round(coord) for coord in im.info['dpi']) im.save(output_file, dpi=dpi) return output_file -def ocr_tesseract_hocr(input_file, page_context): +def ocr_engine_hocr(input_file, page_context): hocr_out = page_context.get_path('ocr_hocr.hocr') hocr_text_out = page_context.get_path('ocr_hocr.txt') options = page_context.options - tesseract.generate_hocr( + + ocr_engine = page_context.plugin_manager.hook.get_ocr_engine() + ocr_engine.generate_hocr( input_file=input_file, - output_files=[hocr_out, hocr_text_out], - language=options.language, - engine_mode=options.tesseract_oem, - tessconfig=options.tesseract_config, - timeout=options.tesseract_timeout, - pagesegmode=options.tesseract_pagesegmode, - user_words=options.user_words, - user_patterns=options.user_patterns, - tesseract_env=options.tesseract_env, - log=page_context.log, + output_hocr=hocr_out, + output_text=hocr_text_out, + options=options, ) return (hocr_out, hocr_text_out) @@ -568,13 +553,15 @@ def create_visible_page_jpg(image, page_context): # might have removed the DPI information. In this case, fall back to # square DPI used to rasterize. When the preview image was # rasterized, it was also converted to square resolution, which is - # what we want to give tesseract, so keep it square. - fallback_dpi = get_page_square_dpi(page_context.pageinfo, page_context.options) - dpi = im.info.get('dpi', (fallback_dpi, fallback_dpi)) + # what we want to give to the OCR engine, so keep it square. + if 'dpi' in im.info: + dpi = Resolution(*im.info['dpi']) + else: + # Fallback to page-implied DPI + dpi = get_page_square_dpi(page_context.pageinfo, page_context.options) # Pillow requires integer DPI - dpi = round(dpi[0]), round(dpi[1]) - im.save(output_file, format='JPEG', dpi=dpi) + im.save(output_file, format='JPEG', dpi=dpi.to_int()) return output_file @@ -587,56 +574,50 @@ def create_pdf_page_from_image(image, page_context): # sandwich renderer would be fine. output_file = page_context.get_path('visible.pdf') dpi = get_page_square_dpi(page_context.pageinfo, page_context.options) - layout_fun = img2pdf.get_fixed_dpi_layout_fun((dpi, dpi)) + layout_fun = img2pdf.get_fixed_dpi_layout_fun(dpi) # This create a single page PDF with open(image, 'rb') as imfile, open(output_file, 'wb') as pdf: - page_context.log.debug('convert') + log.debug('convert') img2pdf.convert( imfile, with_pdfrw=False, layout_fun=layout_fun, outputstream=pdf ) - page_context.log.debug('convert done') + log.debug('convert done') return output_file def render_hocr_page(hocr, page_context): output_file = page_context.get_path('ocr_hocr.pdf') dpi = get_page_square_dpi(page_context.pageinfo, page_context.options) - hocrtransform = HocrTransform(hocr, dpi) + hocrtransform = HocrTransform(hocr, dpi.x) # square hocrtransform.to_pdf( output_file, - imageFileName=None, - showBoundingboxes=False, - invisibleText=True, - interwordSpaces=True, + image_filename=None, + show_bounding_boxes=False, + invisible_text=True, + interword_spaces=True, ) return output_file -def ocr_tesseract_textonly_pdf(input_image, page_context): +def ocr_engine_textonly_pdf(input_image, page_context): output_pdf = page_context.get_path('ocr_tess.pdf') output_text = page_context.get_path('ocr_tess.txt') options = page_context.options - tesseract.generate_pdf( - input_image=input_image, - skip_pdf=None, + + ocr_engine = page_context.plugin_manager.hook.get_ocr_engine() + ocr_engine.generate_pdf( + input_file=input_image, output_pdf=output_pdf, output_text=output_text, - language=options.language, - engine_mode=options.tesseract_oem, - text_only=True, - tessconfig=options.tesseract_config, - timeout=options.tesseract_timeout, - pagesegmode=options.tesseract_pagesegmode, - user_words=options.user_words, - user_patterns=options.user_patterns, - tesseract_env=options.tesseract_env, - log=page_context.log, + options=options, ) return (output_pdf, output_text) -def get_docinfo(base_pdf, options): +def get_docinfo(base_pdf, context): + options = context.options + def from_document_info(key): try: s = base_pdf.docinfo[key] @@ -648,7 +629,6 @@ def get_docinfo(base_pdf, options): k: from_document_info(k) for k in ('/Title', '/Author', '/Keywords', '/Subject', '/CreationDate') } - renderer_tag = 'OCR' if options is not None: if options.title: pdfmark['/Title'] = options.title @@ -659,12 +639,9 @@ def get_docinfo(base_pdf, options): if options.subject: pdfmark['/Subject'] = options.subject - if options.pdf_renderer == 'sandwich': - renderer_tag = 'OCR-PDF' + creator_tag = context.plugin_manager.hook.get_ocr_engine().creator_tag(options) - pdfmark['/Creator'] = ( - f'{PROGRAM_NAME} {VERSION} / ' f'Tesseract {renderer_tag} {tesseract.version()}' - ) + pdfmark['/Creator'] = f'{PROGRAM_NAME} {VERSION} / {creator_tag}' pdfmark['/Producer'] = f'pikepdf {pikepdf.__version__}' if 'OCRMYPDF_CREATOR' in os.environ: pdfmark['/Creator'] = os.environ['OCRMYPDF_CREATOR'] @@ -697,7 +674,7 @@ def convert_to_pdfa(input_pdf, input_ps_stub, context): try: len(pdf_file.docinfo) except TypeError: - context.log.error( + log.error( "File contains a malformed DocumentInfo block - continuing anyway" ) else: @@ -711,12 +688,12 @@ def convert_to_pdfa(input_pdf, input_ps_stub, context): else: safe_symlink(input_pdf, fix_docinfo_file) - ghostscript.generate_pdfa( + context.plugin_manager.hook.generate_pdfa( pdf_version=input_pdfinfo.min_version, - pdf_pages=[fix_docinfo_file, input_ps_stub], + pdf_pages=[fix_docinfo_file], + pdfmark=input_ps_stub, output_file=output_file, compression=options.pdfa_image_compression, - log=context.log, pdfa_part=options.output_type[-1], # is pdfa-1, pdfa-2, or pdfa-3 ) @@ -738,25 +715,21 @@ def metadata_fixup(working_file, context): if not missing: return if options.output_type.startswith('pdfa'): - context.log.warning( + log.warning( "Some input metadata could not be copied because it is not " "permitted in PDF/A. You may wish to examine the output " "PDF's XMP metadata." ) - context.log.debug( - "The following metadata fields were not copied: %r", missing - ) + log.debug("The following metadata fields were not copied: %r", missing) else: - context.log.error( + log.error( "Some input metadata could not be copied." "You may wish to examine the output PDF's XMP metadata." ) - context.log.info( - "The following metadata fields were not copied: %r", missing - ) + log.info("The following metadata fields were not copied: %r", missing) with pikepdf.open(context.origin) as original, pikepdf.open(working_file) as pdf: - docinfo = get_docinfo(original, options) + docinfo = get_docinfo(original, context) with pdf.open_metadata() as meta: meta.load_from_docinfo(docinfo, delete_missing=False, raise_failure=False) # If xmp:CreateDate is missing, set it to the modify date to @@ -804,11 +777,9 @@ def merge_sidecars(txt_files, context): if txt_file: with open(txt_file, 'r', encoding="utf-8") as in_: txt = in_.read() - # Tesseract v4 alpha started adding form feeds in - # commit aa6eb6b - # No obvious way to detect what binaries will do this, so - # for consistency just ignore its form feeds and insert our - # own + # Some OCR engines (e.g. Tesseract v4 alpha) add form feeds + # between pages, and some do not. For consistency, we ignore + # any added by the OCR engine and them on our own. if txt.endswith('\f'): stream.write(txt[:-1]) else: @@ -818,8 +789,8 @@ def merge_sidecars(txt_files, context): return output_file -def copy_final(input_file, output_file, context): - context.log.debug('%s -> %s', input_file, output_file) +def copy_final(input_file, output_file, _context): + log.debug('%s -> %s', input_file, output_file) with open(input_file, 'rb') as input_stream: if output_file == '-': copyfileobj(input_stream, sys.stdout.buffer) diff --git a/src/ocrmypdf/_plugin_manager.py b/src/ocrmypdf/_plugin_manager.py new file mode 100644 index 00000000..9f328c69 --- /dev/null +++ b/src/ocrmypdf/_plugin_manager.py @@ -0,0 +1,67 @@ +# © 2020 James R. Barlow: github.com/jbarlow83 +# +# This file is part of OCRmyPDF. +# +# OCRmyPDF is free software: you can redistribute it and/or modify +# it under the terms of the GNU General Public License as published by +# the Free Software Foundation, either version 3 of the License, or +# (at your option) any later version. +# +# OCRmyPDF is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with OCRmyPDF. If not, see . + +import argparse +import importlib +import importlib.util +import sys +from pathlib import Path +from typing import List + +import pluggy + +from ocrmypdf import pluginspec +from ocrmypdf.cli import get_parser, plugins_only_parser + + +def get_plugin_manager(plugins: List[str], builtins=True): + pm = pluggy.PluginManager('ocrmypdf') + pm.add_hookspecs(pluginspec) + + if builtins: + all_plugins = [ + 'ocrmypdf.builtin_plugins.ghostscript', + 'ocrmypdf.builtin_plugins.tesseract_ocr', + ] + plugins + else: + all_plugins = plugins + for name in all_plugins: + if name.endswith('.py'): + # Import by filename + module_name = Path(name).stem + spec = importlib.util.spec_from_file_location(module_name, name) + module = importlib.util.module_from_spec(spec) + sys.modules[module_name] = module + spec.loader.exec_module(module) + else: + # Import by dotted module name + module = importlib.import_module(name) + pm.register(module) + return pm + + +def get_parser_options_plugins( + args, +) -> (argparse.ArgumentParser, argparse.Namespace, pluggy.PluginManager): + pre_options, _unused = plugins_only_parser.parse_known_args(args=args) + plugin_manager = get_plugin_manager(pre_options.plugins) + + parser = get_parser() + plugin_manager.hook.add_options(parser=parser) + + options = parser.parse_args(args=args) + return parser, options, plugin_manager diff --git a/src/ocrmypdf/_sync.py b/src/ocrmypdf/_sync.py index 07c355ab..60eed5de 100644 --- a/src/ocrmypdf/_sync.py +++ b/src/ocrmypdf/_sync.py @@ -17,21 +17,20 @@ import logging import logging.handlers -import multiprocessing import os -import signal import sys import threading from collections import namedtuple +from functools import partial from pathlib import Path from tempfile import mkdtemp import PIL -from tqdm import tqdm -from ._graft import OcrGrafter -from ._jobcontext import PDFContext, cleanup_working_files, make_logger -from ._pipeline import ( +from ocrmypdf._concurrent import exec_progress_pool +from ocrmypdf._graft import OcrGrafter +from ocrmypdf._jobcontext import PdfContext, cleanup_working_files +from ocrmypdf._pipeline import ( convert_to_pdfa, copy_final, create_ocr_image, @@ -43,8 +42,8 @@ from ._pipeline import ( is_ocr_required, merge_sidecars, metadata_fixup, - ocr_tesseract_hocr, - ocr_tesseract_textonly_pdf, + ocr_engine_hocr, + ocr_engine_textonly_pdf, optimize_pdf, preprocess_clean, preprocess_deskew, @@ -56,20 +55,25 @@ from ._pipeline import ( triage, validate_pdfinfo_options, ) -from ._validation import ( +from ocrmypdf._plugin_manager import get_plugin_manager +from ocrmypdf._validation import ( check_requested_output_file, create_input_file, report_output_file_size, ) -from .exceptions import ExitCode, ExitCodeException -from .exec import qpdf -from .helpers import available_cpu_count -from .pdfa import file_claims_pdfa +from ocrmypdf.exceptions import ExitCode, ExitCodeException +from ocrmypdf.helpers import available_cpu_count, check_pdf, samefile +from ocrmypdf.pdfa import file_claims_pdfa + +log = logging.getLogger(__name__) PageResult = namedtuple( 'PageResult', 'pageno, pdf_page_from_image, ocr, text, orientation_correction' ) +tls = threading.local() +tls.pageno = None + def preprocess(page_context, image, remove_background, deskew, clean): if remove_background: @@ -81,8 +85,23 @@ def preprocess(page_context, image, remove_background, deskew, clean): return image +old_factory = logging.getLogRecordFactory() + + +def record_factory(*args, **kwargs): + record = old_factory(*args, **kwargs) + if hasattr(tls, 'pageno'): + record.pageno = tls.pageno + return record + + +logging.setLogRecordFactory(record_factory) + + def exec_page_sync(page_context): options = page_context.options + tls.pageno = page_context.pageno + 1 + orientation_correction = 0 pdf_page_from_image_out = None ocr_out = None @@ -146,18 +165,22 @@ def exec_page_sync(page_context): visible_image_out = create_visible_page_jpg( visible_image_out, page_context ) + visible_image_out = ( + page_context.plugin_manager.hook.filter_page_image( + page=page_context, image_filename=Path(visible_image_out) + ) + or visible_image_out + ) pdf_page_from_image_out = create_pdf_page_from_image( visible_image_out, page_context ) if options.pdf_renderer == 'hocr': - (hocr_out, text_out) = ocr_tesseract_hocr(ocr_image_out, page_context) + (hocr_out, text_out) = ocr_engine_hocr(ocr_image_out, page_context) ocr_out = render_hocr_page(hocr_out, page_context) if options.pdf_renderer == 'sandwich': - (ocr_out, text_out) = ocr_tesseract_textonly_pdf( - ocr_image_out, page_context - ) + (ocr_out, text_out) = ocr_engine_textonly_pdf(ocr_image_out, page_context) return PageResult( pageno=page_context.pageno, @@ -178,138 +201,50 @@ def post_process(pdf_file, context): return optimize_pdf(pdf_out, context) -def worker_init(queue, max_pixels): - """Initialize a process pool worker""" - - # Ignore SIGINT (our parent process will kill us gracefully) - signal.signal(signal.SIGINT, signal.SIG_IGN) - - # Reconfigure the root logger for this process to send all messages to a queue - h = logging.handlers.QueueHandler(queue) - root = logging.getLogger() - root.handlers = [] - root.addHandler(h) - +def worker_init(max_pixels): # In Windows, child process will not inherit our change to this value in - # the parent process, so ensure workers get it set + # the parent process, so ensure workers get it set. Not needed when running + # threaded, but harmless to set again. PIL.Image.MAX_IMAGE_PIXELS = max_pixels -def worker_thread_init(_queue, max_pixels): - # This is probably not needed since threads should all see the same memory, - # but done for consistency. - PIL.Image.MAX_IMAGE_PIXELS = max_pixels - - -def log_listener(queue): - """Listen to the worker processes and forward the messages to logging - - For simplicity this is a thread rather than a process. Only one process - should actually write to sys.stderr or whatever we're using, so if this is - made into a process the main application needs to be directed to it. - - See https://docs.python.org/3/howto/logging-cookbook.html#logging-to-a-single-file-from-multiple-processes - """ - - while True: - try: - record = queue.get() - if record is None: - break - logger = logging.getLogger(record.name) - logger.handle(record) - except Exception: - import traceback - - print("Logging problem", file=sys.stderr) - traceback.print_exc(file=sys.stderr) - - def exec_concurrent(context): """Execute the pipeline concurrently""" # Run exec_page_sync on every page context max_workers = min(len(context.pdfinfo), context.options.jobs) if max_workers > 1: - context.log.info("Start processing %d pages concurrently", max_workers) - - # Tesseract 4.x can be multithreaded, and we also run multiple workers. We want - # to manage how many threads it uses to avoid creating total threads than cores. - # Performance testing shows we're better off - # parallelizing ocrmypdf and forcing Tesseract to be single threaded, which we - # get by setting the envvar OMP_THREAD_LIMIT to 1. But if the page count of the - # input file is small, then we allow Tesseract to use threads, subject to the - # constraint: (ocrmypdf workers) * (tesseract threads) <= max_workers. - # As of Tesseract 4.1, 3 threads is the most effective on a 4 core/8 thread system. - tess_threads = min(3, context.options.jobs // max_workers) - if context.options.tesseract_env is None: - context.options.tesseract_env = os.environ.copy() - context.options.tesseract_env.setdefault('OMP_THREAD_LIMIT', str(tess_threads)) - try: - tess_threads = int(context.options.tesseract_env['OMP_THREAD_LIMIT']) - except ValueError: # OMP_THREAD_LIMIT initialized to non-numeric - context.log.error("Environment variable OMP_THREAD_LIMIT is not numeric") - if tess_threads > 1: - context.log.info("Using Tesseract OpenMP thread limit %d", tess_threads) - - if context.options.use_threads: - from multiprocessing.dummy import Pool - - initializer = worker_thread_init - else: - Pool = multiprocessing.Pool - initializer = worker_init + log.info("Start processing %d pages concurrently", max_workers) sidecars = [None] * len(context.pdfinfo) ocrgraft = OcrGrafter(context) - log_queue = multiprocessing.Queue(-1) - listener = threading.Thread(target=log_listener, args=(log_queue,)) - listener.start() - with tqdm( - total=(2 * len(context.pdfinfo)), - desc='OCR', - unit='page', - unit_scale=0.5, - disable=not context.options.progress_bar, - ) as pbar: - pool = Pool( - processes=max_workers, - initializer=initializer, - initargs=(log_queue, PIL.Image.MAX_IMAGE_PIXELS), + def update_page(result, pbar): + sidecars[result.pageno] = result.text + pbar.update() + ocrgraft.graft_page( + pageno=result.pageno, + image=result.pdf_page_from_image, + textpdf=result.ocr, + autorotate_correction=result.orientation_correction, ) - try: - results = pool.imap_unordered(exec_page_sync, context.get_page_contexts()) - while True: - try: - page_result = results.next() - sidecars[page_result.pageno] = page_result.text - pbar.update() - ocrgraft.graft_page(page_result) - pbar.update() - except StopIteration: - break - except KeyboardInterrupt: - # Terminate pool so we exit instantly - pool.terminate() - # Don't try listener.join() here, will deadlock - raise - except Exception: - if not os.environ.get("PYTEST_CURRENT_TEST", ""): - # Unless inside pytest, exit immediately because no one wants - # to wait for child processes to finalize results that will be - # thrown away. Inside pytest, we want child processes to exit - # cleanly so that they output an error messages or coverage data - # we need from them. - pool.terminate() - raise - finally: - # Terminate log listener - log_queue.put_nowait(None) - pool.close() - pool.join() + pbar.update() - listener.join() + exec_progress_pool( + use_threads=context.options.use_threads, + max_workers=max_workers, + tqdm_kwargs=dict( + total=(2 * len(context.pdfinfo)), + desc='OCR', + unit='page', + unit_scale=0.5, + disable=not context.options.progress_bar, + ), + task_initializer=partial(worker_init, PIL.Image.MAX_IMAGE_PIXELS), + task=exec_page_sync, + task_arguments=context.get_page_contexts(), + task_finished=update_page, + ) # Output sidecar text if context.options.sidecar: @@ -333,34 +268,27 @@ class NeverRaise(Exception): pass # pylint: disable=unnecessary-pass -def samefile(f1, f2): - if os.name == 'nt': - return f1 == f2 - else: - return os.path.samefile(f1, f2) - - def configure_debug_logging(log_filename, prefix=''): log_file_handler = logging.FileHandler(log_filename, delay=True) log_file_handler.setLevel(logging.DEBUG) formatter = logging.Formatter( - '[%(asctime)s] - %(name)s - %(levelname)7s - %(message)s' + '[%(asctime)s] - %(name)s - %(levelname)7s -%(pageno)s %(message)s' ) log_file_handler.setFormatter(formatter) logging.getLogger(prefix).addHandler(log_file_handler) return log_file_handler -def run_pipeline(options, api=False): - log = make_logger(options, __name__) - +def run_pipeline(options, *, plugin_manager, api=False): # Any changes to options will not take effect for options that are already # bound to function parameters in the pipeline. (For example # options.input_file, options.pdf_renderer are already bound.) if not options.jobs: options.jobs = available_cpu_count() + if not plugin_manager: + plugin_manager = get_plugin_manager(options.plugins) - work_folder = mkdtemp(prefix="com.github.ocrmypdf.") + work_folder = Path(mkdtemp(prefix="com.github.ocrmypdf.")) debug_log_handler = None if (options.keep_temporary_files or options.verbose >= 1) and not os.environ.get( 'PYTEST_CURRENT_TEST', '' @@ -373,21 +301,17 @@ def run_pipeline(options, api=False): # Triage image or pdf origin_pdf = triage( - original_filename, - start_input_file, - os.path.join(work_folder, 'origin.pdf'), - options, - log, + original_filename, start_input_file, work_folder / 'origin.pdf', options ) # Gather pdfinfo and create context pdfinfo = get_pdfinfo( origin_pdf, - detailed_page_analysis=options.redo_ocr, progbar=options.progress_bar, + max_workers=options.jobs if not options.use_threads else 1, # To help debug ) - context = PDFContext(options, work_folder, origin_pdf, pdfinfo) + context = PdfContext(options, work_folder, origin_pdf, pdfinfo, plugin_manager) # Validate options are okay for this pdf validate_pdfinfo_options(context) @@ -412,7 +336,7 @@ def run_pipeline(options, api=False): pdfa_info['conformance'], ) return ExitCode.pdfa_conversion_failed - if not qpdf.check(options.output_file, log): + if not check_pdf(options.output_file): log.warning('Output file: The generated PDF is INVALID') return ExitCode.invalid_output_pdf report_output_file_size(options, start_input_file, options.output_file) @@ -429,7 +353,7 @@ def run_pipeline(options, api=False): else: log.error(type(e).__name__) return e.exit_code - except (Exception if not api else NeverRaise) as e: + except (Exception if not api else NeverRaise) as e: # pylint: disable=broad-except log.exception("An exception occurred while executing the pipeline") return ExitCode.other_error finally: diff --git a/src/ocrmypdf/_validation.py b/src/ocrmypdf/_validation.py index 9006e115..2d688afa 100644 --- a/src/ocrmypdf/_validation.py +++ b/src/ocrmypdf/_validation.py @@ -21,29 +21,28 @@ import locale import logging import os import sys +import unicodedata from pathlib import Path from shutil import copyfileobj import pikepdf import PIL -from ._unicodefun import verify_python3_env -from .exceptions import ( +from ocrmypdf._exec import jbig2enc, pngquant, unpaper +from ocrmypdf._unicodefun import verify_python3_env +from ocrmypdf.exceptions import ( BadArgsError, InputFileError, MissingDependencyError, OutputFileAccessError, ) -from .exec import ( - check_external_program, - ghostscript, - jbig2enc, - pngquant, - qpdf, - tesseract, - unpaper, +from ocrmypdf.helpers import ( + is_file_writable, + is_iterable_notstr, + monotonic, + safe_symlink, ) -from .helpers import is_file_writable, is_iterable_notstr, monotonic, safe_symlink +from ocrmypdf.subprocess import check_external_program # ------------- # External dependencies @@ -68,35 +67,26 @@ def check_platform(): ) -def check_options_languages(options): - if not options.language: - options.language = [DEFAULT_LANGUAGE] +def check_options_languages(options, plugin_manager): + if not options.languages: + options.languages = {DEFAULT_LANGUAGE} system_lang = locale.getlocale()[0] if system_lang and not system_lang.startswith('en'): log.debug("No language specified; assuming --language %s", DEFAULT_LANGUAGE) - # Support v2.x "eng+deu" language syntax - if '+' in options.language[0]: - options.language = options.language[0].split('+') - - languages = set(options.language) - if not languages.issubset(tesseract.languages()): + ocr_engine = plugin_manager.hook.get_ocr_engine() + if not options.languages.issubset(ocr_engine.languages(options)): msg = ( - "The installed version of tesseract does not have language " - "data for the following requested languages: \n" + f"{ocr_engine} does not have language data for the following " + "requested languages: \n" ) - for lang in languages - tesseract.languages(): + for lang in options.languages - ocr_engine.languages(options): msg += lang + '\n' raise MissingDependencyError(msg) def check_options_output(options): - # We have these constraints to check for. - # 1. Ghostscript < 9.20 mangles multibyte Unicode - # 2. hocr doesn't work on non-Latin languages (so don't select it) - - languages = set(options.language) - is_latin = languages.issubset(HOCR_OK_LANGS) + is_latin = options.languages.issubset(HOCR_OK_LANGS) if options.pdf_renderer == 'hocr' and not is_latin: msg = ( @@ -106,37 +96,6 @@ def check_options_output(options): ) log.warning(msg) - if ghostscript.version() < '9.20' and options.output_type != 'pdf' and not is_latin: - # https://bugs.ghostscript.com/show_bug.cgi?id=696874 - # Ghostscript < 9.20 fails to encode multibyte characters properly - msg = ( - "The installed version of Ghostscript does not work correctly " - "with the OCR languages you specified. Use --output-type pdf or " - "upgrade to Ghostscript 9.20 or later to avoid this issue." - ) - msg += f"Found Ghostscript {ghostscript.version()}" - log.warning(msg) - - # Decide on what renderer to use - if options.pdf_renderer == 'auto': - options.pdf_renderer = 'sandwich' - - if options.pdf_renderer == 'sandwich' and not tesseract.has_textonly_pdf( - options.tesseract_env, languages - ): - raise MissingDependencyError( - "You are using an alpha version of Tesseract 4.0 that does not support " - "the textonly_pdf parameter. We don't support versions this old." - ) - - if options.output_type == 'pdfa': - options.output_type = 'pdfa-2' - - if options.output_type == 'pdfa-3' and ghostscript.version() < '9.19': - raise MissingDependencyError( - "--output-type pdfa-3 requires Ghostscript 9.19 or later" - ) - lossless_reconstruction = False if not any( ( @@ -271,18 +230,9 @@ def check_options_advanced(options): "--pdfa-image-compression argument has no effect when " "--output-type is not 'pdfa', 'pdfa-1', or 'pdfa-2'" ) - if not tesseract.has_user_words(options.tesseract_env) and ( - options.user_words or options.user_patterns - ): - log.warning( - "Tesseract 4.0 ignores --user-words and --user-patterns, so these " - "arguments have no effect." - ) def check_options_metadata(options): - import unicodedata - docinfo = [options.title, options.author, options.keywords, options.subject] for s in (m for m in docinfo if m): for c in s: @@ -301,9 +251,9 @@ def check_options_pillow(options): PIL.Image.MAX_IMAGE_PIXELS = None -def check_options(options): +def check_options(options, plugin_manager): check_platform() - check_options_languages(options) + check_options_languages(options, plugin_manager) check_options_metadata(options) check_options_output(options) check_options_sidecar(options) @@ -312,7 +262,7 @@ def check_options(options): check_options_optimizing(options) check_options_advanced(options) check_options_pillow(options) - check_dependency_versions(options) + plugin_manager.hook.check_options(options=options) def check_closed_streams(options): # pragma: no cover @@ -374,17 +324,17 @@ def log_page_orientations(pdfinfo): log.info('Page orientations detected: %s', ' '.join(orientations)) -def create_input_file(options, work_folder): +def create_input_file(options, work_folder: Path) -> (Path, str): if options.input_file == '-': # stdin log.info('reading file from standard input') - target = os.path.join(work_folder, 'stdin') + target = work_folder / 'stdin' with open(target, 'wb') as stream_buffer: copyfileobj(sys.stdin.buffer, stream_buffer) return target, "" else: try: - target = os.path.join(work_folder, 'origin') + target = work_folder / 'origin' safe_symlink(options.input_file, target) return target, os.fspath(options.input_file) except FileNotFoundError: @@ -459,31 +409,3 @@ def report_output_file_size(options, input_file, output_file): f"The output file size is {ratio:.2f}× larger than the input file.\n" f"{explanation}" ) - - -def check_dependency_versions(options): - check_external_program( - program='tesseract', - package={'linux': 'tesseract-ocr'}, - version_checker=tesseract.version, - need_version='4.0.0', # using backport for Travis CI - ) - check_external_program( - program='gs', - package='ghostscript', - version_checker=ghostscript.version, - need_version='9.15', # limited by Travis CI / Ubuntu 14.04 backports - ) - gs_version = ghostscript.version() - if gs_version in ('9.24', '9.51'): - raise MissingDependencyError( - f"Ghostscript {gs_version} contains serious regressions and is not " - "supported. Please upgrade to a newer version, or downgrade to the " - "previous version." - ) - check_external_program( - program='qpdf', - package='qpdf', - version_checker=qpdf.version, - need_version='8.0.2', - ) diff --git a/src/ocrmypdf/api.py b/src/ocrmypdf/api.py index 8fbcc2b9..7e389d88 100644 --- a/src/ocrmypdf/api.py +++ b/src/ocrmypdf/api.py @@ -15,47 +15,25 @@ # You should have received a copy of the GNU General Public License # along with OCRmyPDF. If not, see . +import inspect import logging import os import sys -from contextlib import suppress +from argparse import ArgumentParser from enum import IntEnum from pathlib import Path from typing import Dict, Iterable -from tqdm import tqdm +from ocrmypdf._logging import PageNumberFilter, TqdmConsole +from ocrmypdf._plugin_manager import get_plugin_manager +from ocrmypdf._sync import run_pipeline +from ocrmypdf._validation import check_options +from ocrmypdf.cli import get_parser -from ._sync import run_pipeline -from ._validation import check_options -from .cli import parser - - -class TqdmConsole: - """Wrapper to log messages in a way that is compatible with tqdm progress bar - - This routes log messages through tqdm so that it can print them above the - progress bar, and then refresh the progress bar, rather than overwriting - it which looks messy. - - For some reason Python 3.6 prints extra empty messages from time to time, - so we suppress those. - """ - - def __init__(self, file): - self.file = file - self.py36 = sys.version_info[0:2] == (3, 6) - - def write(self, msg): - # When no progress bar is active, tqdm.write() routes to print() - if self.py36: - if msg.strip() != '': - tqdm.write(msg.rstrip(), end='\n', file=self.file) - else: - tqdm.write(msg.rstrip(), end='\n', file=self.file) - - def flush(self): - with suppress(AttributeError): - self.file.flush() +try: + import coloredlogs +except ModuleNotFoundError: + coloredlogs = None class Verbosity(IntEnum): @@ -98,6 +76,7 @@ def configure_logging( """ prefix = '' if manage_root_logger else 'ocrmypdf' + log = logging.getLogger(prefix) log.setLevel(logging.DEBUG) @@ -113,9 +92,25 @@ def configure_logging( else: console.setLevel(logging.INFO) - formatter = logging.Formatter('%(levelname)7s - %(message)s') + console.addFilter(PageNumberFilter()) + if verbosity >= 2: - formatter = logging.Formatter('%(name)s - %(levelname)7s - %(message)s') + fmt = '%(levelname)7s %(name)s -%(pageno)s %(message)s' + else: + fmt = '%(pageno)s%(message)s' + + use_colors = progress_bar_friendly + if not coloredlogs: + use_colors = False + if use_colors: + if os.name == 'nt': + use_colors = coloredlogs.enable_ansi_support() + if use_colors: + use_colors = coloredlogs.terminal_supports_colors() + if use_colors: + formatter = coloredlogs.ColoredFormatter(fmt=fmt) + else: + formatter = logging.Formatter(fmt=fmt) console.setFormatter(formatter) log.addHandler(console) @@ -132,7 +127,13 @@ def configure_logging( return log -def create_options(*, input_file: os.PathLike, output_file: os.PathLike, **kwargs): +def create_options( + *, + input_file: os.PathLike, + output_file: os.PathLike, + parser: ArgumentParser, + **kwargs, +): cmdline = [] deferred = [] @@ -142,7 +143,7 @@ def create_options(*, input_file: os.PathLike, output_file: os.PathLike, **kwarg # These arguments with special handling for which we bypass # argparse - if arg in {'tesseract_env', 'progress_bar'}: + if arg in {'progress_bar', 'plugins'}: deferred.append((arg, val)) continue @@ -174,15 +175,10 @@ def create_options(*, input_file: os.PathLike, output_file: os.PathLike, **kwarg cmdline.append(str(input_file)) cmdline.append(str(output_file)) - parser.api_mode = True + parser._api_mode = True options = parser.parse_args(cmdline) for keyword, val in deferred: setattr(options, keyword, val) - - # If we are running a Tesseract spoof, ensure it knows what the input file is - if os.environ.get('PYTEST_CURRENT_TEST') and options.tesseract_env: - options.tesseract_env['_OCRMYPDF_TEST_INFILE'] = os.fspath(input_file) - return options @@ -230,9 +226,10 @@ def ocr( # pylint: disable=unused-argument user_words: os.PathLike = None, user_patterns: os.PathLike = None, fast_web_view: float = None, + plugins: Iterable[str] = None, keep_temporary_files: bool = None, progress_bar: bool = None, - tesseract_env: Dict[str, str] = None, + **kwargs, ): """Run OCRmyPDF on one PDF or image. @@ -243,7 +240,6 @@ def ocr( # pylint: disable=unused-argument use_threads (bool): Use worker threads instead of processes. This reduces performance but may make debugging easier since it is easier to set breakpoints. - tesseract_env (dict): Override environment variables for Tesseract Raises: ocrmypdf.PdfMergeFailedError: If the input PDF is malformed, preventing merging with the OCR layer. @@ -267,7 +263,18 @@ def ocr( # pylint: disable=unused-argument Returns: :class:`ocrmypdf.ExitCode` """ + if not plugins: + plugins = [] - options = create_options(**locals()) - check_options(options) - return run_pipeline(options, api=True) + parser = get_parser() + _plugin_manager = get_plugin_manager(plugins) + _plugin_manager.hook.add_options(parser=parser) # pylint: disable=no-member + + create_options_kwargs = { + k: v for k, v in locals().items() if not k.startswith('_') and k != 'kwargs' + } + create_options_kwargs.update(kwargs) + + options = create_options(**create_options_kwargs) + check_options(options, _plugin_manager) + return run_pipeline(options=options, plugin_manager=_plugin_manager, api=True) diff --git a/src/ocrmypdf/builtin_plugins/__init__.py b/src/ocrmypdf/builtin_plugins/__init__.py new file mode 100644 index 00000000..0ed32bc2 --- /dev/null +++ b/src/ocrmypdf/builtin_plugins/__init__.py @@ -0,0 +1,16 @@ +# © 2020 James R. Barlow: github.com/jbarlow83 +# +# This file is part of OCRmyPDF. +# +# OCRmyPDF is free software: you can redistribute it and/or modify +# it under the terms of the GNU General Public License as published by +# the Free Software Foundation, either version 3 of the License, or +# (at your option) any later version. +# +# OCRmyPDF is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with OCRmyPDF. If not, see . diff --git a/src/ocrmypdf/builtin_plugins/ghostscript.py b/src/ocrmypdf/builtin_plugins/ghostscript.py new file mode 100644 index 00000000..e451c771 --- /dev/null +++ b/src/ocrmypdf/builtin_plugins/ghostscript.py @@ -0,0 +1,104 @@ +# © 2020 James R. Barlow: github.com/jbarlow83 +# +# This file is part of OCRmyPDF. +# +# OCRmyPDF is free software: you can redistribute it and/or modify +# it under the terms of the GNU General Public License as published by +# the Free Software Foundation, either version 3 of the License, or +# (at your option) any later version. +# +# OCRmyPDF is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with OCRmyPDF. If not, see . + +import logging +from pathlib import Path + +from ocrmypdf import hookimpl +from ocrmypdf._exec import ghostscript +from ocrmypdf._validation import HOCR_OK_LANGS +from ocrmypdf.exceptions import MissingDependencyError +from ocrmypdf.helpers import Resolution +from ocrmypdf.subprocess import check_external_program + +log = logging.getLogger(__name__) + + +@hookimpl +def check_options(options): + gs_version = ghostscript.version() + check_external_program( + program='gs', + package='ghostscript', + version_checker=gs_version, + need_version='9.15', # limited by Travis CI / Ubuntu 14.04 backports + ) + if gs_version in ('9.24', '9.51'): + raise MissingDependencyError( + f"Ghostscript {gs_version} contains serious regressions and is not " + "supported. Please upgrade to a newer version, or downgrade to the " + "previous version." + ) + + # We have these constraints to check for. + # 1. Ghostscript < 9.20 mangles multibyte Unicode + # 2. hocr doesn't work on non-Latin languages (so don't select it) + is_latin = options.languages.issubset(HOCR_OK_LANGS) + if gs_version < '9.20' and options.output_type != 'pdf' and not is_latin: + # https://bugs.ghostscript.com/show_bug.cgi?id=696874 + # Ghostscript < 9.20 fails to encode multibyte characters properly + msg = ( + "The installed version of Ghostscript does not work correctly " + "with the OCR languages you specified. Use --output-type pdf or " + "upgrade to Ghostscript 9.20 or later to avoid this issue." + ) + msg += f"Found Ghostscript {gs_version}" + log.warning(msg) + + if options.output_type == 'pdfa': + options.output_type = 'pdfa-2' + + if options.output_type == 'pdfa-3' and ghostscript.version() < '9.19': + raise MissingDependencyError( + "--output-type pdfa-3 requires Ghostscript 9.19 or later" + ) + + +@hookimpl +def rasterize_pdf_page( + input_file, + output_file, + raster_device, + raster_dpi, + pageno, + page_dpi=None, + rotation=None, + filter_vector=False, +): + ghostscript.rasterize_pdf( + input_file, + output_file, + raster_device=raster_device, + raster_dpi=raster_dpi, + pageno=pageno, + page_dpi=page_dpi, + rotation=rotation, + filter_vector=filter_vector, + ) + return output_file + + +@hookimpl +def generate_pdfa(pdf_pages, pdfmark, output_file, compression, pdf_version, pdfa_part): + ghostscript.generate_pdfa( + pdf_pages=[*pdf_pages, pdfmark], + output_file=output_file, + compression=compression, + pdf_version=pdf_version, + pdfa_part=pdfa_part, + ) + return output_file diff --git a/src/ocrmypdf/builtin_plugins/tesseract_ocr.py b/src/ocrmypdf/builtin_plugins/tesseract_ocr.py new file mode 100644 index 00000000..bd15ddbe --- /dev/null +++ b/src/ocrmypdf/builtin_plugins/tesseract_ocr.py @@ -0,0 +1,197 @@ +# © 2020 James R. Barlow: github.com/jbarlow83 +# +# This file is part of OCRmyPDF. +# +# OCRmyPDF is free software: you can redistribute it and/or modify +# it under the terms of the GNU General Public License as published by +# the Free Software Foundation, either version 3 of the License, or +# (at your option) any later version. +# +# OCRmyPDF is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with OCRmyPDF. If not, see . + +import argparse +import logging +import os + +from ocrmypdf import hookimpl +from ocrmypdf._exec import tesseract +from ocrmypdf.cli import numeric +from ocrmypdf.exceptions import MissingDependencyError +from ocrmypdf.pluginspec import OcrEngine +from ocrmypdf.subprocess import check_external_program + +log = logging.getLogger(__name__) + + +@hookimpl +def add_options(parser): + tess = parser.add_argument_group("Tesseract", "Advanced control of Tesseract OCR") + tess.add_argument( + '--tesseract-config', + action='append', + metavar='CFG', + default=[], + help="Additional Tesseract configuration files -- see documentation", + ) + tess.add_argument( + '--tesseract-pagesegmode', + action='store', + type=int, + metavar='PSM', + choices=range(0, 14), + help="Set Tesseract page segmentation mode (see tesseract --help)", + ) + tess.add_argument( + '--tesseract-oem', + action='store', + type=int, + metavar='MODE', + choices=range(0, 4), + help=( + "Set Tesseract 4.0 OCR engine mode: " + "0 - original Tesseract only; " + "1 - neural nets LSTM only; " + "2 - Tesseract + LSTM; " + "3 - default." + ), + ) + tess.add_argument( + '--tesseract-timeout', + default=180.0, + type=numeric(float, 0), + metavar='SECONDS', + help='Give up on OCR after the timeout, but copy the preprocessed page ' + 'into the final output', + ) + tess.add_argument( + '--user-words', + metavar='FILE', + help="Specify the location of the Tesseract user words file. This is a " + "list of words Tesseract should consider while performing OCR in " + "addition to its standard language dictionaries. This can improve " + "OCR quality especially for specialized and technical documents.", + ) + tess.add_argument( + '--user-patterns', + metavar='FILE', + help="Specify the location of the Tesseract user patterns file.", + ) + + +@hookimpl +def check_options(options): + check_external_program( + program='tesseract', + package={'linux': 'tesseract-ocr'}, + version_checker=tesseract.version, + need_version='4.0.0', # using backport for Travis CI + ) + + # Decide on what renderer to use + if options.pdf_renderer == 'auto': + options.pdf_renderer = 'sandwich' + + if options.pdf_renderer == 'sandwich' and not tesseract.has_textonly_pdf( + set(options.languages) + ): + raise MissingDependencyError( + "You are using an alpha version of Tesseract 4.0 that does not support " + "the textonly_pdf parameter. We don't support versions this old." + ) + if not tesseract.has_user_words() and (options.user_words or options.user_patterns): + log.warning( + "Tesseract 4.0 ignores --user-words and --user-patterns, so these " + "arguments have no effect." + ) + if options.tesseract_pagesegmode in (0, 2): + log.warning( + "The --tesseract-pagesegmode argument you select will disable OCR. " + "This may cause processing to fail." + ) + + +@hookimpl +def validate(pdfinfo, options): + # Tesseract 4.x can be multithreaded, and we also run multiple workers. We want + # to manage how many threads it uses to avoid creating total threads than cores. + # Performance testing shows we're better off + # parallelizing ocrmypdf and forcing Tesseract to be single threaded, which we + # get by setting the envvar OMP_THREAD_LIMIT to 1. But if the page count of the + # input file is small, then we allow Tesseract to use threads, subject to the + # constraint: (ocrmypdf workers) * (tesseract threads) <= max_workers. + # As of Tesseract 4.1, 3 threads is the most effective on a 4 core/8 thread system. + if not os.environ.get('OMP_THREAD_LIMIT', '').isnumeric(): + tess_threads = min(3, options.jobs // len(pdfinfo), len(pdfinfo)) + os.environ['OMP_THREAD_LIMIT'] = str(tess_threads) + else: + tess_threads = int(os.environ['OMP_THREAD_LIMIT']) + + if tess_threads > 1: + log.info("Using Tesseract OpenMP thread limit %d", tess_threads) + + +class TesseractOcrEngine(OcrEngine): + @staticmethod + def version(): + return tesseract.version() + + @staticmethod + def creator_tag(options): + tag = '-PDF' if options.pdf_renderer == 'sandwich' else '' + return f"Tesseract OCR{tag} {TesseractOcrEngine.version()}" + + def __str__(self): + return f"Tesseract OCR {TesseractOcrEngine.version()}" + + @staticmethod + def languages(options): + return tesseract.get_languages() + + @staticmethod + def get_orientation(input_file, options): + return tesseract.get_orientation( + input_file, + engine_mode=options.tesseract_oem, + timeout=options.tesseract_timeout, + ) + + @staticmethod + def generate_hocr(input_file, output_hocr, output_text, options): + tesseract.generate_hocr( + input_file=input_file, + output_hocr=output_hocr, + output_text=output_text, + languages=options.languages, + engine_mode=options.tesseract_oem, + tessconfig=options.tesseract_config, + timeout=options.tesseract_timeout, + pagesegmode=options.tesseract_pagesegmode, + user_words=options.user_words, + user_patterns=options.user_patterns, + ) + + @staticmethod + def generate_pdf(input_file, output_pdf, output_text, options): + tesseract.generate_pdf( + input_file=input_file, + output_pdf=output_pdf, + output_text=output_text, + languages=options.languages, + engine_mode=options.tesseract_oem, + tessconfig=options.tesseract_config, + timeout=options.tesseract_timeout, + pagesegmode=options.tesseract_pagesegmode, + user_words=options.user_words, + user_patterns=options.user_patterns, + ) + + +@hookimpl +def get_ocr_engine(): + return TesseractOcrEngine() diff --git a/src/ocrmypdf/cli.py b/src/ocrmypdf/cli.py index e28162e7..a34a2108 100644 --- a/src/ocrmypdf/cli.py +++ b/src/ocrmypdf/cli.py @@ -17,10 +17,8 @@ import argparse -from ._version import PROGRAM_NAME as _PROGRAM_NAME -from ._version import __version__ as _VERSION - -__all__ = ['parser'] +from ocrmypdf._version import PROGRAM_NAME as _PROGRAM_NAME +from ocrmypdf._version import __version__ as _VERSION def numeric(basetype, min_=None, max_=None): @@ -47,27 +45,43 @@ class ArgumentParser(argparse.ArgumentParser): def __init__(self, *args, **kwargs): super().__init__(*args, **kwargs) - self.api_mode = False + self._api_mode = False def error(self, message): - if not self.api_mode: + if not self._api_mode: super().error(message) return raise ValueError(message) -parser = ArgumentParser( - prog=_PROGRAM_NAME, - fromfile_prefix_chars='@', - formatter_class=argparse.RawDescriptionHelpFormatter, - description="""\ +class LanguageSetAction(argparse.Action): + def __init__(self, option_strings, dest, default=None, **kwargs): + if default is None: + default = set() + super().__init__(option_strings, dest, default=default, **kwargs) + + def __call__(self, parser, namespace, values, option_string=None): + dest = getattr(namespace, self.dest) + if '+' in values: + dest.add(lang for lang in values.split('+')) + else: + dest.add(values) + + +def get_parser(): + parser = ArgumentParser( + prog=_PROGRAM_NAME, + allow_abbrev=True, + fromfile_prefix_chars='@', + formatter_class=argparse.RawDescriptionHelpFormatter, + description="""\ Generates a searchable PDF or PDF/A from a regular PDF. OCRmyPDF rasterizes each page of the input PDF, optionally corrects page rotation and performs image processing, runs the Tesseract OCR engine on the image, and then creates a PDF from the OCR information. """, - epilog="""\ + epilog="""\ OCRmyPDF attempts to keep the output file at about the same size. If a file contains losslessly compressed images, and output file will be losslessly compressed as well. @@ -108,386 +122,368 @@ Online documentation is located at: https://ocrmypdf.readthedocs.io/en/latest/introduction.html """, -) + ) -parser.add_argument( - 'input_file', - metavar="input_pdf_or_image", - help="PDF file containing the images to be OCRed (or '-' to read from " - "standard input)", + parser.add_argument( + 'input_file', + metavar="input_pdf_or_image", + help="PDF file containing the images to be OCRed (or '-' to read from " + "standard input)", + ) + parser.add_argument( + 'output_file', + metavar="output_pdf", + help="Output searchable PDF file (or '-' to write to standard output). " + "Existing files will be ovewritten. If same as input file, the " + "input file will be updated only if processing is successful.", + ) + parser.add_argument( + '-l', + '--language', + dest='languages', + action=LanguageSetAction, + help="Language(s) of the file to be OCRed (see tesseract --list-langs for " + "all language packs installed in your system). Use -l eng+deu for " + "multiple languages.", + ) + parser.add_argument( + '--image-dpi', + metavar='DPI', + type=int, + help="For input image instead of PDF, use this DPI instead of file's.", + ) + parser.add_argument( + '--output-type', + choices=['pdfa', 'pdf', 'pdfa-1', 'pdfa-2', 'pdfa-3'], + default='pdfa', + help="Choose output type. 'pdfa' creates a PDF/A-2b compliant file for " + "long term archiving (default, recommended) but may not suitable " + "for users who want their file altered as little as possible. 'pdfa' " + "also has problems with full Unicode text. 'pdf' attempts to " + "preserve file contents as much as possible. 'pdf-a1' creates a " + "PDF/A1-b file. 'pdf-a2' is equivalent to 'pdfa'. 'pdf-a3' creates a " + "PDF/A3-b file.", + ) + + # Use null string '\0' as sentinel to indicate the user supplied no argument, + # since that is the only invalid character for filepaths on all platforms + # bool('\0') is True in Python + parser.add_argument( + '--sidecar', + nargs='?', + const='\0', + default=None, + metavar='FILE', + help="Generate sidecar text files that contain the same text recognized " + "by Tesseract. This may be useful for building a OCR text database. " + "If FILE is omitted, the sidecar file be named {output_file}.txt " + "If FILE is set to '-', the sidecar is written to stdout (a " + "convenient way to preview OCR quality). The output file and sidecar " + "may not both use stdout at the same time.", + ) + + parser.add_argument( + '--version', + action='version', + version=_VERSION, + help="Print program version and exit", + ) + + jobcontrol = parser.add_argument_group("Job control options") + jobcontrol.add_argument( + '-j', + '--jobs', + metavar='N', + type=numeric(int, 0, 256), + help="Use up to N CPU cores simultaneously (default: use all).", + ) + jobcontrol.add_argument( + '-q', '--quiet', action='store_true', help="Suppress INFO messages" + ) + jobcontrol.add_argument( + '-v', + '--verbose', + type=numeric(int, 0, 2), + default=0, + const=1, + nargs='?', + help="Print more verbose messages for each additional verbose level. Use " + "`-v 1` typically for much more detailed logging. Higher numbers " + "are probably only useful in debugging.", + ) + jobcontrol.add_argument( + '--no-progress-bar', + action='store_false', + dest='progress_bar', + help=argparse.SUPPRESS, + ) + jobcontrol.add_argument( + '--use-threads', action='store_true', help=argparse.SUPPRESS + ) + + metadata = parser.add_argument_group( + "Metadata options", + "Set output PDF/A metadata (default: copy input document's metadata)", + ) + metadata.add_argument( + '--title', type=str, help="Set document title (place multiple words in quotes)" + ) + metadata.add_argument('--author', type=str, help="Set document author") + metadata.add_argument( + '--subject', type=str, help="Set document subject description" + ) + metadata.add_argument('--keywords', type=str, help="Set document keywords") + + preprocessing = parser.add_argument_group( + "Image preprocessing options", + "Options to improve the quality of the final PDF and OCR", + ) + preprocessing.add_argument( + '-r', + '--rotate-pages', + action='store_true', + help="Automatically rotate pages based on detected text orientation", + ) + preprocessing.add_argument( + '--remove-background', + action='store_true', + help="Attempt to remove background from gray or color pages, setting it " + "to white ", + ) + preprocessing.add_argument( + '-d', + '--deskew', + action='store_true', + help="Deskew each page before performing OCR", + ) + preprocessing.add_argument( + '-c', + '--clean', + action='store_true', + help="Clean pages from scanning artifacts before performing OCR, and send " + "the cleaned page to OCR, but do not include the cleaned page in " + "the output", + ) + preprocessing.add_argument( + '-i', + '--clean-final', + action='store_true', + help="Clean page as above, and incorporate the cleaned image in the final " + "PDF. Might remove desired content.", + ) + preprocessing.add_argument( + '--unpaper-args', + type=str, + default=None, + help="A quoted string of arguments to pass to unpaper. Requires --clean. " + "Example: --unpaper-args '--layout double'.", + ) + preprocessing.add_argument( + '--oversample', + metavar='DPI', + type=numeric(int, 0, 5000), + default=0, + help="Oversample images to at least the specified DPI, to improve OCR " + "results slightly", + ) + preprocessing.add_argument( + '--remove-vectors', + action='store_true', + help="EXPERIMENTAL. Mask out any vector objects in the PDF so that they " + "will not be included in OCR. This can eliminate false characters.", + ) + preprocessing.add_argument( + '--threshold', + action='store_true', + help=( + "EXPERIMENTAL. Threshold image to 1bpp before sending it to Tesseract " + "for OCR. Can improve OCR quality compared to Tesseract's thresholder." + ), + ) + + ocrsettings = parser.add_argument_group("OCR options", "Control how OCR is applied") + ocrsettings.add_argument( + '-f', + '--force-ocr', + action='store_true', + help="Rasterize any text or vector objects on each page, apply OCR, and " + "save the rastered output (this rewrites the PDF)", + ) + ocrsettings.add_argument( + '-s', + '--skip-text', + action='store_true', + help="Skip OCR on any pages that already contain text, but include the " + "page in final output; useful for PDFs that contain a mix of " + "images, text pages, and/or previously OCRed pages", + ) + ocrsettings.add_argument( + '--redo-ocr', + action='store_true', + help="Attempt to detect and remove the hidden OCR layer from files that " + "were previously OCRed with OCRmyPDF or another program. Apply OCR " + "to text found in raster images. Existing visible text objects will " + "not be changed. If there is no existing OCR, OCR will be added.", + ) + ocrsettings.add_argument( + '--skip-big', + type=numeric(float, 0, 5000), + metavar='MPixels', + help="Skip OCR on pages larger than the specified amount of megapixels, " + "but include skipped pages in final output", + ) + + optimizing = parser.add_argument_group( + "Optimization options", "Control how the PDF is optimized after OCR" + ) + optimizing.add_argument( + '-O', + '--optimize', + type=int, + choices=range(0, 4), + default=1, + help=( + "Control how PDF is optimized after processing:" + "0 - do not optimize; " + "1 - do safe, lossless optimizations (default); " + "2 - do some lossy optimizations; " + "3 - do aggressive lossy optimizations (including lossy JBIG2)" + ), + ) + optimizing.add_argument( + '--jpeg-quality', + type=numeric(int, 0, 100), + default=0, + metavar='Q', + help=( + "Adjust JPEG quality level for JPEG optimization. " + "100 is best quality and largest output size; " + "1 is lowest quality and smallest output; " + "0 uses the default." + ), + ) + optimizing.add_argument( + '--jpg-quality', + type=numeric(int, 0, 100), + default=0, + metavar='Q', + dest='jpeg_quality', + help=argparse.SUPPRESS, # Alias for --jpeg-quality + ) + optimizing.add_argument( + '--png-quality', + type=numeric(int, 0, 100), + default=0, + metavar='Q', + help=( + "Adjust PNG quality level to use when quantizing PNGs. " + "Values have same meaning as with --jpeg-quality" + ), + ) + optimizing.add_argument( + '--jbig2-lossy', + action='store_true', + help=( + "Enable JBIG2 lossy mode (better compression, not suitable for some " + "use cases - see documentation)." + ), + ) + optimizing.add_argument( + '--jbig2-page-group-size', + type=numeric(int, 1, 10000), + default=0, + metavar='N', + # Adjust number of pages to consider at once for JBIG2 compression + help=argparse.SUPPRESS, + ) + + advanced = parser.add_argument_group( + "Advanced", "Advanced options to control OCRmyPDF" + ) + advanced.add_argument( + '--pages', + type=str, + help=( + "Limit OCR to the specified pages (ranges or comma separated), " + "skipping others" + ), + ) + advanced.add_argument( + '--max-image-mpixels', + action='store', + type=numeric(float, 0), + metavar='MPixels', + help="Set maximum number of pixels to unpack before treating an image as a " + "decompression bomb", + default=128.0, + ) + advanced.add_argument( + '--pdf-renderer', + choices=['auto', 'hocr', 'sandwich'], + default='auto', + help="Choose OCR PDF renderer - the default option is to let OCRmyPDF " + "choose. See documentation for discussion.", + ) + advanced.add_argument( + '--rotate-pages-threshold', + default=14.0, + type=numeric(float, 0, 1000), + metavar='CONFIDENCE', + help="Only rotate pages when confidence is above this value (arbitrary " + "units reported by tesseract)", + ) + advanced.add_argument( + '--pdfa-image-compression', + choices=['auto', 'jpeg', 'lossless'], + default='auto', + help="Specify how to compress images in the output PDF/A. 'auto' lets " + "OCRmyPDF decide. 'jpeg' changes all grayscale and color images to " + "JPEG compression. 'lossless' uses PNG-style lossless compression " + "for all images. Monochrome images are always compressed using a " + "lossless codec. Compression settings " + "are applied to all pages, including those for which OCR was " + "skipped. Not supported for --output-type=pdf ; that setting " + "preserves the original compression of all images.", + ) + advanced.add_argument( + '--fast-web-view', + type=numeric(float, 0), + default=1.0, + metavar="MEGABYTES", + help="If the size of file is more than this threshold (in MB), then " + "linearize the PDF for fast web viewing. This allows the PDF to be " + "displayed before it is fully downloaded in web browsers, but increases " + "the space required slightly. By default we skip this for small files " + "which do not benefit. If the threshold is 0 it will be apply to all files. " + "Set the threshold very high to disable.", + ) + advanced.add_argument( + '--plugin', + dest='plugins', + action='append', + default=[], + help="Name of plugin to import.", + ) + + debugging = parser.add_argument_group( + "Debugging", "Arguments to help with troubleshooting and debugging" + ) + debugging.add_argument( + '-k', + '--keep-temporary-files', + action='store_true', + help="Keep temporary files (helpful for debugging)", + ) + return parser + + +plugins_only_parser = ArgumentParser( + prog=_PROGRAM_NAME, fromfile_prefix_chars='@', add_help=False, allow_abbrev=False ) -parser.add_argument( - 'output_file', - metavar="output_pdf", - help="Output searchable PDF file (or '-' to write to standard output). " - "Existing files will be ovewritten. If same as input file, the " - "input file will be updated only if processing is successful.", -) -parser.add_argument( - '-l', - '--language', +plugins_only_parser.add_argument( + '--plugin', + dest='plugins', action='append', - help="Language(s) of the file to be OCRed (see tesseract --list-langs for " - "all language packs installed in your system). Use -l eng+deu for " - "multiple languages.", -) -parser.add_argument( - '--image-dpi', - metavar='DPI', - type=int, - help="For input image instead of PDF, use this DPI instead of file's.", -) -parser.add_argument( - '--output-type', - choices=['pdfa', 'pdf', 'pdfa-1', 'pdfa-2', 'pdfa-3'], - default='pdfa', - help="Choose output type. 'pdfa' creates a PDF/A-2b compliant file for " - "long term archiving (default, recommended) but may not suitable " - "for users who want their file altered as little as possible. 'pdfa' " - "also has problems with full Unicode text. 'pdf' attempts to " - "preserve file contents as much as possible. 'pdf-a1' creates a " - "PDF/A1-b file. 'pdf-a2' is equivalent to 'pdfa'. 'pdf-a3' creates a " - "PDF/A3-b file.", -) - -# Use null string '\0' as sentinel to indicate the user supplied no argument, -# since that is the only invalid character for filepaths on all platforms -# bool('\0') is True in Python -parser.add_argument( - '--sidecar', - nargs='?', - const='\0', - default=None, - metavar='FILE', - help="Generate sidecar text files that contain the same text recognized " - "by Tesseract. This may be useful for building a OCR text database. " - "If FILE is omitted, the sidecar file be named {output_file}.txt " - "If FILE is set to '-', the sidecar is written to stdout (a " - "convenient way to preview OCR quality). The output file and sidecar " - "may not both use stdout at the same time.", -) - -parser.add_argument( - '--version', - action='version', - version=_VERSION, - help="Print program version and exit", -) - -jobcontrol = parser.add_argument_group("Job control options") -jobcontrol.add_argument( - '-j', - '--jobs', - metavar='N', - type=numeric(int, 0, 256), - help="Use up to N CPU cores simultaneously (default: use all).", -) -jobcontrol.add_argument( - '-q', '--quiet', action='store_true', help="Suppress INFO messages" -) -jobcontrol.add_argument( - '-v', - '--verbose', - type=numeric(int, 0, 2), - default=0, - const=1, - nargs='?', - help="Print more verbose messages for each additional verbose level. Use " - "`-v 1` typically for much more detailed logging. Higher numbers " - "are probably only useful in debugging.", -) -jobcontrol.add_argument( - '--no-progress-bar', - action='store_false', - dest='progress_bar', - help=argparse.SUPPRESS, -) -jobcontrol.add_argument('--use-threads', action='store_true', help=argparse.SUPPRESS) - -metadata = parser.add_argument_group( - "Metadata options", - "Set output PDF/A metadata (default: copy input document's metadata)", -) -metadata.add_argument( - '--title', type=str, help="Set document title (place multiple words in quotes)" -) -metadata.add_argument('--author', type=str, help="Set document author") -metadata.add_argument('--subject', type=str, help="Set document subject description") -metadata.add_argument('--keywords', type=str, help="Set document keywords") - -preprocessing = parser.add_argument_group( - "Image preprocessing options", - "Options to improve the quality of the final PDF and OCR", -) -preprocessing.add_argument( - '-r', - '--rotate-pages', - action='store_true', - help="Automatically rotate pages based on detected text orientation", -) -preprocessing.add_argument( - '--remove-background', - action='store_true', - help="Attempt to remove background from gray or color pages, setting it " - "to white ", -) -preprocessing.add_argument( - '-d', '--deskew', action='store_true', help="Deskew each page before performing OCR" -) -preprocessing.add_argument( - '-c', - '--clean', - action='store_true', - help="Clean pages from scanning artifacts before performing OCR, and send " - "the cleaned page to OCR, but do not include the cleaned page in " - "the output", -) -preprocessing.add_argument( - '-i', - '--clean-final', - action='store_true', - help="Clean page as above, and incorporate the cleaned image in the final " - "PDF. Might remove desired content.", -) -preprocessing.add_argument( - '--unpaper-args', - type=str, - default=None, - help="A quoted string of arguments to pass to unpaper. Requires --clean. " - "Example: --unpaper-args '--layout double'.", -) -preprocessing.add_argument( - '--oversample', - metavar='DPI', - type=numeric(int, 0, 5000), - default=0, - help="Oversample images to at least the specified DPI, to improve OCR " - "results slightly", -) -preprocessing.add_argument( - '--remove-vectors', - action='store_true', - help="EXPERIMENTAL. Mask out any vector objects in the PDF so that they " - "will not be included in OCR. This can eliminate false characters.", -) -preprocessing.add_argument( - '--threshold', - action='store_true', - help="EXPERIMENTAL. Threshold image to 1bpp before sending it to Tesseract for OCR. Can " - "improve OCR quality compared to Tesseract's thresholder.", -) - -ocrsettings = parser.add_argument_group("OCR options", "Control how OCR is applied") -ocrsettings.add_argument( - '-f', - '--force-ocr', - action='store_true', - help="Rasterize any text or vector objects on each page, apply OCR, and " - "save the rastered output (this rewrites the PDF)", -) -ocrsettings.add_argument( - '-s', - '--skip-text', - action='store_true', - help="Skip OCR on any pages that already contain text, but include the " - "page in final output; useful for PDFs that contain a mix of " - "images, text pages, and/or previously OCRed pages", -) -ocrsettings.add_argument( - '--redo-ocr', - action='store_true', - help="Attempt to detect and remove the hidden OCR layer from files that " - "were previously OCRed with OCRmyPDF or another program. Apply OCR " - "to text found in raster images. Existing visible text objects will " - "not be changed. If there is no existing OCR, OCR will be added.", -) -ocrsettings.add_argument( - '--skip-big', - type=numeric(float, 0, 5000), - metavar='MPixels', - help="Skip OCR on pages larger than the specified amount of megapixels, " - "but include skipped pages in final output", -) - -optimizing = parser.add_argument_group( - "Optimization options", "Control how the PDF is optimized after OCR" -) -optimizing.add_argument( - '-O', - '--optimize', - type=int, - choices=range(0, 4), - default=1, - help=( - "Control how PDF is optimized after processing:" - "0 - do not optimize; " - "1 - do safe, lossless optimizations (default); " - "2 - do some lossy optimizations; " - "3 - do aggressive lossy optimizations (including lossy JBIG2)" - ), -) -optimizing.add_argument( - '--jpeg-quality', - type=numeric(int, 0, 100), - default=0, - metavar='Q', - help=( - "Adjust JPEG quality level for JPEG optimization. " - "100 is best quality and largest output size; " - "1 is lowest quality and smallest output; " - "0 uses the default." - ), -) -optimizing.add_argument( - '--jpg-quality', - type=numeric(int, 0, 100), - default=0, - metavar='Q', - dest='jpeg_quality', - help=argparse.SUPPRESS, # Alias for --jpeg-quality -) -optimizing.add_argument( - '--png-quality', - type=numeric(int, 0, 100), - default=0, - metavar='Q', - help=( - "Adjust PNG quality level to use when quantizing PNGs. " - "Values have same meaning as with --jpeg-quality" - ), -) -optimizing.add_argument( - '--jbig2-lossy', - action='store_true', - help=( - "Enable JBIG2 lossy mode (better compression, not suitable for some " - "use cases - see documentation)." - ), -) -optimizing.add_argument( - '--jbig2-page-group-size', - type=numeric(int, 1, 10000), - default=0, - metavar='N', - # Adjust number of pages to consider at once for JBIG2 compression - help=argparse.SUPPRESS, -) - -advanced = parser.add_argument_group( - "Advanced", "Advanced options to control Tesseract's OCR behavior" -) -advanced.add_argument( - '--pages', - type=str, - help="Limit OCR to the specified pages (ranges or comma separated), skipping others", -) -advanced.add_argument( - '--max-image-mpixels', - action='store', - type=numeric(float, 0), - metavar='MPixels', - help="Set maximum number of pixels to unpack before treating an image as a " - "decompression bomb", - default=128.0, -) -advanced.add_argument( - '--tesseract-config', - action='append', - metavar='CFG', default=[], - help="Additional Tesseract configuration files -- see documentation", + help="Name of plugin to import.", ) -advanced.add_argument( - '--tesseract-pagesegmode', - action='store', - type=int, - metavar='PSM', - choices=range(0, 14), - help="Set Tesseract page segmentation mode (see tesseract --help)", -) -advanced.add_argument( - '--tesseract-oem', - action='store', - type=int, - metavar='MODE', - choices=range(0, 4), - help=( - "Set Tesseract 4.0 OCR engine mode: " - "0 - original Tesseract only; " - "1 - neural nets LSTM only; " - "2 - Tesseract + LSTM; " - "3 - default." - ), -) -advanced.add_argument( - '--pdf-renderer', - choices=['auto', 'hocr', 'sandwich'], - default='auto', - help="Choose OCR PDF renderer - the default option is to let OCRmyPDF " - "choose. See documentation for discussion.", -) -advanced.add_argument( - '--tesseract-timeout', - default=180.0, - type=numeric(float, 0), - metavar='SECONDS', - help='Give up on OCR after the timeout, but copy the preprocessed page ' - 'into the final output', -) -advanced.add_argument( - '--rotate-pages-threshold', - default=14.0, - type=numeric(float, 0, 1000), - metavar='CONFIDENCE', - help="Only rotate pages when confidence is above this value (arbitrary " - "units reported by tesseract)", -) -advanced.add_argument( - '--pdfa-image-compression', - choices=['auto', 'jpeg', 'lossless'], - default='auto', - help="Specify how to compress images in the output PDF/A. 'auto' lets " - "OCRmyPDF decide. 'jpeg' changes all grayscale and color images to " - "JPEG compression. 'lossless' uses PNG-style lossless compression " - "for all images. Monochrome images are always compressed using a " - "lossless codec. Compression settings " - "are applied to all pages, including those for which OCR was " - "skipped. Not supported for --output-type=pdf ; that setting " - "preserves the original compression of all images.", -) -advanced.add_argument( - '--user-words', - metavar='FILE', - help="Specify the location of the Tesseract user words file. This is a " - "list of words Tesseract should consider while performing OCR in " - "addition to its standard language dictionaries. This can improve " - "OCR quality especially for specialized and technical documents.", -) -advanced.add_argument( - '--user-patterns', - metavar='FILE', - help="Specify the location of the Tesseract user patterns file.", -) -advanced.add_argument( - '--fast-web-view', - type=numeric(float, 0), - default=1.0, - metavar="MEGABYTES", - help="If the size of file is more than this threshold (in MB), then " - "linearize the PDF for fast web viewing. This allows the PDF to be " - "displayed before it is fully downloaded in web browsers, but increases " - "the space required slightly. By default we skip this for small files " - "which do not benefit. If the threshold is 0 it will be apply to all files. " - "Set the threshold very high to disable.", -) - -debugging = parser.add_argument_group( - "Debugging", "Arguments to help with troubleshooting and debugging" -) -debugging.add_argument( - '-k', - '--keep-temporary-files', - action='store_true', - help="Keep temporary files (helpful for debugging)", -) -debugging.add_argument('--tesseract-env', type=str, help=argparse.SUPPRESS) diff --git a/src/ocrmypdf/exec/qpdf.py b/src/ocrmypdf/exec/qpdf.py deleted file mode 100644 index d46baf1c..00000000 --- a/src/ocrmypdf/exec/qpdf.py +++ /dev/null @@ -1,61 +0,0 @@ -# © 2017 James R. Barlow: github.com/jbarlow83 -# -# This file is part of OCRmyPDF. -# -# OCRmyPDF is free software: you can redistribute it and/or modify -# it under the terms of the GNU General Public License as published by -# the Free Software Foundation, either version 3 of the License, or -# (at your option) any later version. -# -# OCRmyPDF is distributed in the hope that it will be useful, -# but WITHOUT ANY WARRANTY; without even the implied warranty of -# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the -# GNU General Public License for more details. -# -# You should have received a copy of the GNU General Public License -# along with OCRmyPDF. If not, see . - -"""Interface to qpdf executable""" - -from io import StringIO - -import pikepdf - - -def version(): - return pikepdf.__libqpdf_version__ - - -def check(input_file, log=None): - pdf = None - try: - pdf = pikepdf.open(input_file) - except pikepdf.PdfError as e: - if log: - log.error(e) - return False - else: - messages = pdf.check() - for msg in messages: - if 'error' in msg.lower(): - log.error(msg) - else: - log.warning(msg) - - sio = StringIO() - linearize = None - try: - pdf.check_linearization(sio) - except RuntimeError: - pass - else: - linearize = sio.getvalue() - if linearize: - log.warning(linearize) - - if not messages and not linearize: - return True - return False - finally: - if pdf: - pdf.close() diff --git a/src/ocrmypdf/helpers.py b/src/ocrmypdf/helpers.py index ae326331..964f7670 100644 --- a/src/ocrmypdf/helpers.py +++ b/src/ocrmypdf/helpers.py @@ -20,23 +20,55 @@ import multiprocessing import os import shutil import warnings +from collections import namedtuple from collections.abc import Iterable from contextlib import suppress from functools import wraps +from io import StringIO +from math import isclose from pathlib import Path +import pikepdf + log = logging.getLogger(__name__) -def safe_symlink(input_file: os.PathLike, soft_link_name: os.PathLike, *args, **kwargs): +class Resolution(namedtuple('Resolution', ('x', 'y'))): + __slots__ = () + + def round(self, ndigits): + return Resolution(round(self.x, ndigits), round(self.y, ndigits)) + + def to_int(self): + return Resolution(int(round(self.x)), int(round(self.y))) + + @property + def is_square(self): + return isclose(self.x, self.y, rel_tol=1e-3) + + def take_max(self, vals, yvals=None): + if yvals is not None: + return Resolution(max(self.x, *vals), max(self.y, *yvals)) + max_x, max_y = self.x, self.y + for x, y in vals: + max_x = max(x, max_x) + max_y = max(y, max_y) + return Resolution(max_x, max_y) + + def flip_axis(self): + return Resolution(self.y, self.x) + + def __str__(self): + return f"{self.x:f}x{self.y:f}" + + def __repr__(self): + return f"Resolution({self.x}x{self.y} dpi)" + + +def safe_symlink(input_file: os.PathLike, soft_link_name: os.PathLike): """ Helper function: relinks soft symbolic link if necessary """ - if len(args) == 1 and isinstance(args[0], logging.Logger): - log.warning("Deprecated: safe_symlink(,log)") - if 'log' in kwargs: - log.warning('Deprecated: safe_symlink(...log=)') - input_file = os.fspath(input_file) soft_link_name = os.fspath(soft_link_name) @@ -72,6 +104,13 @@ def safe_symlink(input_file: os.PathLike, soft_link_name: os.PathLike, *args, ** os.symlink(os.path.abspath(input_file), soft_link_name) +def samefile(f1, f2): + if os.name == 'nt': + return f1 == f2 + else: + return os.path.samefile(f1, f2) + + def is_iterable_notstr(thing): return isinstance(thing, Iterable) and not isinstance(thing, str) @@ -105,11 +144,7 @@ def is_file_writable(test_file: os.PathLike): the location is writable. """ try: - if not isinstance(test_file, Path): - p = Path(test_file) - else: - p = test_file - + p = Path(test_file) if p.is_symlink(): p = p.resolve(strict=False) @@ -136,6 +171,40 @@ def is_file_writable(test_file: os.PathLike): return False +def check_pdf(input_file): + pdf = None + try: + pdf = pikepdf.open(input_file) + except pikepdf.PdfError as e: + log.error(e) + return False + else: + messages = pdf.check() + for msg in messages: + if 'error' in msg.lower(): + log.error(msg) + else: + log.warning(msg) + + sio = StringIO() + linearize = None + try: + pdf.check_linearization(sio) + except RuntimeError: + pass + else: + linearize = sio.getvalue() + if linearize: + log.warning(linearize) + + if not messages and not linearize: + return True + return False + finally: + if pdf: + pdf.close() + + def deprecated(func): """Warn that function is deprecated""" diff --git a/src/ocrmypdf/hocrtransform.py b/src/ocrmypdf/hocrtransform.py index 809f858e..6240d7ee 100755 --- a/src/ocrmypdf/hocrtransform.py +++ b/src/ocrmypdf/hocrtransform.py @@ -29,9 +29,11 @@ # SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. import argparse +import os import re from collections import namedtuple from math import atan, cos, sin +from pathlib import Path from xml.etree import ElementTree from reportlab.lib.units import inch @@ -64,9 +66,9 @@ class HocrTransform: {'ff': 'ff', 'ffi': 'f‌f‌i', 'ffl': 'f‌f‌l', 'fi': 'fi', 'fl': 'fl'} ) - def __init__(self, hocrFileName, dpi): + def __init__(self, hocr_filename: str, dpi: float): self.dpi = dpi - self.hocr = ElementTree.parse(hocrFileName) + self.hocr = ElementTree.parse(hocr_filename) # if the hOCR file has a namespace, ElementTree requires its use to # find elements @@ -114,12 +116,12 @@ class HocrTransform: return text @classmethod - def element_coordinates(cls, element): + def element_coordinates(cls, element) -> Rect: """ Returns a tuple containing the coordinates of the bounding box around an element """ - out = (0, 0, 0, 0) + out = Rect._make(0 for _ in range(4)) if 'title' in element.attrib: matches = cls.box_pattern.search(element.attrib['title']) if matches: @@ -136,7 +138,7 @@ class HocrTransform: matches = cls.baseline_pattern.search(element.attrib['title']) if matches: return float(matches.group(1)), int(matches.group(2)) - return (0, 0) + return (0.0, 0.0) def pt_from_pixel(self, pxl): """ @@ -145,7 +147,7 @@ class HocrTransform: return Rect._make((c / self.dpi * inch) for c in pxl) @classmethod - def replace_unsupported_chars(cls, s): + def replace_unsupported_chars(cls, s: str): """ Given an input string, returns the corresponding string that: - is available in the helvetica facetype @@ -155,12 +157,12 @@ class HocrTransform: def to_pdf( self, - outFileName, - imageFileName=None, - showBoundingboxes=False, - fontname="Helvetica", - invisibleText=False, - interwordSpaces=False, + out_filename: Path, + image_filename: Path = None, + show_bounding_boxes: bool = False, + fontname: str = "Helvetica", + invisible_text: bool = False, + interword_spaces: bool = False, ): """ Creates a PDF file with an image superimposed on top of the text. @@ -172,7 +174,11 @@ class HocrTransform: """ # create the PDF file # page size in points (1/72 in.) - pdf = Canvas(outFileName, pagesize=(self.width, self.height), pageCompression=1) + pdf = Canvas( + os.fspath(out_filename), + pagesize=(self.width, self.height), + pageCompression=1, + ) # draw bounding box for each paragraph # light blue for bounding box of paragraph @@ -190,7 +196,7 @@ class HocrTransform: pt = self.pt_from_pixel(pxl_coords) # draw the bbox border - if showBoundingboxes: # pragma: no cover + if show_bounding_boxes: # pragma: no cover pdf.rect( pt.x1, self.height - pt.y2, pt.x2 - pt.x1, pt.y2 - pt.y1, fill=1 ) @@ -205,9 +211,9 @@ class HocrTransform: line, "ocrx_word", fontname, - invisibleText, - interwordSpaces, - showBoundingboxes, + invisible_text, + interword_spaces, + show_bounding_boxes, ) if not found_lines: @@ -218,13 +224,15 @@ class HocrTransform: root, "ocrx_word", fontname, - invisibleText, - interwordSpaces, - showBoundingboxes, + invisible_text, + interword_spaces, + show_bounding_boxes, ) # put the image on the page, scaled to fill the page - if imageFileName is not None: - pdf.drawImage(imageFileName, 0, 0, width=self.width, height=self.height) + if image_filename is not None: + pdf.drawImage( + os.fspath(image_filename), 0, 0, width=self.width, height=self.height + ) # finish up the page and save it pdf.showPage() @@ -236,13 +244,13 @@ class HocrTransform: def _do_line( self, - pdf, + pdf: Canvas, line, - elemclass, - fontname, - invisibleText, - interwordSpaces, - showBoundingboxes, + elemclass: str, + fontname: str, + invisible_text: bool, + interword_spaces: bool, + show_bounding_boxes: bool, ): pxl_line_coords = self.element_coordinates(line) line_box = self.pt_from_pixel(pxl_line_coords) @@ -262,14 +270,14 @@ class HocrTransform: # on a sloped baseline and the edge of the bounding box. fontsize = (line_height - abs(intercept)) / cos_a text.setFont(fontname, fontsize) - if invisibleText: + if invisible_text: text.setTextRenderMode(3) # Invisible (indicates OCR text) # Intercept is normally negative, so this places it above the bottom # of the line box baseline_y2 = self.height - (line_box.y2 + intercept) - if showBoundingboxes: # pragma: no cover + if show_bounding_boxes: # pragma: no cover # draw the baseline in magenta, dashed pdf.setDash() pdf.setStrokeColorRGB(0.95, 0.65, 0.95) @@ -298,7 +306,7 @@ class HocrTransform: pxl_coords = self.element_coordinates(elem) box = self.pt_from_pixel(pxl_coords) - if interwordSpaces: + if interword_spaces: # if `--interword-spaces` is true, append a space # to the end of each text element to allow simpler PDF viewers # such as PDF.js to better recognize words in search and copy @@ -318,7 +326,7 @@ class HocrTransform: font_width = pdf.stringWidth(elemtxt, fontname, fontsize) # draw the bbox border - if showBoundingboxes: # pragma: no cover + if show_bounding_boxes: # pragma: no cover pdf.rect( box.x1, self.height - line_box.y2, box_width, line_height, fill=0 ) @@ -385,5 +393,5 @@ if __name__ == "__main__": args.outputfile, args.image, args.boundingboxes, - interwordSpaces=args.interword_spaces, + interword_spaces=args.interword_spaces, ) diff --git a/src/ocrmypdf/leptonica.py b/src/ocrmypdf/leptonica.py index 328b0630..6af38089 100644 --- a/src/ocrmypdf/leptonica.py +++ b/src/ocrmypdf/leptonica.py @@ -29,13 +29,13 @@ from collections.abc import Sequence from contextlib import suppress from ctypes.util import find_library from functools import lru_cache -from io import BytesIO +from io import BytesIO, UnsupportedOperation from os import fspath from tempfile import TemporaryFile -from .exceptions import MissingDependencyError -from .exec import shim_paths_with_program_files -from .lib._leptonica import ffi +from ocrmypdf.exceptions import MissingDependencyError +from ocrmypdf.lib._leptonica import ffi +from ocrmypdf.subprocess import shim_paths_with_program_files # pylint: disable=protected-access @@ -96,7 +96,6 @@ class _LeptonicaErrorTrap: self.no_stderr = False def __enter__(self): - from io import UnsupportedOperation self.tmpfile = TemporaryFile() @@ -351,7 +350,7 @@ class Pix(LeptonicaObject): py_file.write(buffer) @classmethod - def frompil(self, pillow_image): + def frompil(cls, pillow_image): """Create a copy of a PIL.Image from this Pix""" bio = BytesIO() pillow_image.save(bio, format='png', compress_level=1) @@ -363,7 +362,7 @@ class Pix(LeptonicaObject): def topil(self): """Returns a PIL.Image version of this Pix""" - from PIL import Image + from PIL import Image # pylint: disable=import-outside-toplevel # Leptonica manages data in words, so it implicitly does an endian # swap. Tell Pillow about this when it reads the data. @@ -534,16 +533,7 @@ class Pix(LeptonicaObject): ) return Pix(thresh_pix) - def crop_to_foreground( - self, - threshold=128, - mindist=70, - erasedist=30, - pagenum=0, - showmorph=0, - display=0, - pdfdir=ffi.NULL, - ): + def crop_to_foreground(self, threshold=128, mindist=70, erasedist=30, showmorph=0): if get_leptonica_version() < 'leptonica-1.76': # Leptonica 1.76 changed the API for pixFindPageForeground; we don't # support the old version diff --git a/src/ocrmypdf/optimize.py b/src/ocrmypdf/optimize.py index 99211698..d1a346ab 100644 --- a/src/ocrmypdf/optimize.py +++ b/src/ocrmypdf/optimize.py @@ -15,10 +15,11 @@ # You should have received a copy of the GNU General Public License # along with OCRmyPDF. If not, see . -import concurrent.futures +import logging import sys import tempfile from collections import defaultdict +from functools import partial from os import fspath from pathlib import Path @@ -27,11 +28,14 @@ from pikepdf import Dictionary, Name from PIL import Image from tqdm import tqdm -from . import leptonica -from ._jobcontext import PDFContext -from .exceptions import OutputFileAccessError -from .exec import jbig2enc, pngquant -from .helpers import safe_symlink +from ocrmypdf import leptonica +from ocrmypdf._concurrent import exec_progress_pool +from ocrmypdf._exec import jbig2enc, pngquant +from ocrmypdf._jobcontext import PdfContext +from ocrmypdf.exceptions import OutputFileAccessError +from ocrmypdf.helpers import safe_symlink + +log = logging.getLogger(__name__) DEFAULT_JPEG_QUALITY = 75 DEFAULT_PNG_QUALITY = 70 @@ -53,7 +57,7 @@ def tif_name(root, xref): return img_name(root, xref, '.tif') -def extract_image_filter(pike, root, log, image, xref): +def extract_image_filter(pike, root, image, xref): if image.Subtype != Name.Image: return None if image.Length < 100: @@ -79,8 +83,8 @@ def extract_image_filter(pike, root, log, image, xref): return pim, filtdp -def extract_image_jbig2(*, pike, root, log, image, xref, options): - result = extract_image_filter(pike, root, log, image, xref) +def extract_image_jbig2(*, pike, root, image, xref, options): + result = extract_image_filter(pike, root, image, xref) if result is None: return None pim, filtdp = result @@ -101,8 +105,8 @@ def extract_image_jbig2(*, pike, root, log, image, xref, options): return None -def extract_image_generic(*, pike, root, log, image, xref, options): - result = extract_image_filter(pike, root, log, image, xref) +def extract_image_generic(*, pike, root, image, xref, options): + result = extract_image_filter(pike, root, image, xref) if result is None: return None pim, filtdp = result @@ -170,7 +174,7 @@ def extract_image_generic(*, pike, root, log, image, xref, options): return None -def extract_images(pike, root, log, options, extract_fn): +def extract_images(pike, root, options, extract_fn): """Extract image using extract_fn Enumerate images on each page, lookup their xref/ID number in the PDF. @@ -212,9 +216,9 @@ def extract_images(pike, root, log, options, extract_fn): image = pike.get_object((xref, 0)) try: result = extract_fn( - pike=pike, root=root, log=log, image=image, xref=xref, options=options + pike=pike, root=root, image=image, xref=xref, options=options ) - except Exception as e: + except Exception as e: # pylint: disable=broad-except log.debug("Image xref %s, error %s", xref, repr(e)) errors += 1 else: @@ -223,12 +227,12 @@ def extract_images(pike, root, log, options, extract_fn): yield pageno_for_xref[xref], xref, ext -def extract_images_generic(pike, root, log, options): +def extract_images_generic(pike, root, options): """Extract any >=2bpp image we think we can improve""" jpegs = [] pngs = [] - for _, xref, ext in extract_images(pike, root, log, options, extract_image_generic): + for _, xref, ext in extract_images(pike, root, options, extract_image_generic): log.debug('xref = %s ext = %s', xref, ext) if ext == '.png': pngs.append(xref) @@ -238,13 +242,11 @@ def extract_images_generic(pike, root, log, options): return jpegs, pngs -def extract_images_jbig2(pike, root, log, options): +def extract_images_jbig2(pike, root, options): """Extract any bitonal image that we think we can improve as JBIG2""" jbig2_groups = defaultdict(list) - for pageno, xref, ext in extract_images( - pike, root, log, options, extract_image_jbig2 - ): + for pageno, xref, ext in extract_images(pike, root, options, extract_image_jbig2): group = pageno // options.jbig2_page_group_size jbig2_groups[group].append((xref, ext)) @@ -256,55 +258,55 @@ def extract_images_jbig2(pike, root, log, options): return jbig2_groups -def _produce_jbig2_images(jbig2_groups, root, log, options): +def _produce_jbig2_images(jbig2_groups, root, options): """Produce JBIG2 images from their groups""" - def jbig2_group_futures(executor, root, groups): + def jbig2_group_args(root, groups): for group, xref_exts in groups.items(): prefix = f'group{group:08d}' - future = executor.submit( - jbig2enc.convert_group, + yield dict( cwd=fspath(root), infiles=(img_name(root, xref, ext) for xref, ext in xref_exts), out_prefix=prefix, ) - yield future - def jbig2_single_futures(executor, root, groups): + def jbig2_single_args(root, groups): for group, xref_exts in groups.items(): prefix = f'group{group:08d}' # Second loop is to ensure multiple images per page are unpacked for n, xref_ext in enumerate(xref_exts): xref, ext = xref_ext - future = executor.submit( - jbig2enc.convert_single, + yield dict( cwd=fspath(root), infile=img_name(root, xref, ext), outfile=root / f'{prefix}.{n:04d}', ) - yield future + + def convert_generic(fn, kwargs_dict): + return fn(**kwargs_dict) if options.jbig2_page_group_size > 1: - jbig2_futures = jbig2_group_futures + jbig2_args = jbig2_group_args + jbig2_convert = partial(convert_generic, jbig2enc.convert_group) else: - jbig2_futures = jbig2_single_futures + jbig2_args = jbig2_single_args + jbig2_convert = partial(convert_generic, jbig2enc.convert_single) - with concurrent.futures.ThreadPoolExecutor(max_workers=options.jobs) as executor: - futures = jbig2_futures(executor, root, jbig2_groups) - with tqdm( + exec_progress_pool( + use_threads=True, + max_workers=options.jobs, + tqdm_kwargs=dict( total=len(jbig2_groups), desc="JBIG2", unit='item', disable=not options.progress_bar, - ) as pbar: - for future in concurrent.futures.as_completed(futures): - proc = future.result() - if proc.stderr: - log.debug(proc.stderr.decode()) - pbar.update() + ), + task=jbig2_convert, + task_arguments=jbig2_args(root, jbig2_groups), + ) -def convert_to_jbig2(pike, jbig2_groups, root, log, options): +def convert_to_jbig2(pike, jbig2_groups, root, options): """Convert images to JBIG2 and insert into PDF. When the JBIG2 page group size is > 1 we do several JBIG2 images at once @@ -318,7 +320,7 @@ def convert_to_jbig2(pike, jbig2_groups, root, log, options): and needs no dictionary. Currently this must be lossless JBIG2. """ - _produce_jbig2_images(jbig2_groups, root, log, options) + _produce_jbig2_images(jbig2_groups, root, options) for group, xref_exts in jbig2_groups.items(): prefix = f'group{group:08d}' @@ -342,7 +344,7 @@ def convert_to_jbig2(pike, jbig2_groups, root, log, options): ) -def transcode_jpegs(pike, jpegs, root, log, options): +def transcode_jpegs(pike, jpegs, root, options): for xref in tqdm( jpegs, desc="JPEGs", unit='image', disable=not options.progress_bar ): @@ -365,37 +367,40 @@ def transcode_jpegs(pike, jpegs, root, log, options): im_obj.write(compdata.read(), filter=Name.DCTDecode) -def transcode_pngs(pike, images, image_name_fn, root, log, options): +def transcode_pngs(pike, images, image_name_fn, root, options): modified = set() if options.optimize >= 2: png_quality = ( max(10, options.png_quality - 10), min(100, options.png_quality + 10), ) - with concurrent.futures.ThreadPoolExecutor( - max_workers=options.jobs - ) as executor: - futures = [] + + def pngquant_args(): for xref in images: log.debug(image_name_fn(root, xref)) - futures.append( - executor.submit( - pngquant.quantize, - image_name_fn(root, xref), - png_name(root, xref), - png_quality[0], - png_quality[1], - ) + yield ( + image_name_fn(root, xref), + png_name(root, xref), + png_quality[0], + png_quality[1], ) modified.add(xref) - with tqdm( + + def pngquant_fn(args): + pngquant.quantize(*args) + + exec_progress_pool( + use_threads=True, + max_workers=options.jobs, + tqdm_kwargs=dict( desc="PNGs", - total=len(futures), + total=len(images), unit='image', disable=not options.progress_bar, - ) as pbar: - for _future in concurrent.futures.as_completed(futures): - pbar.update() + ), + task=pngquant_fn, + task_arguments=pngquant_args(), + ) for xref in modified: im_obj = pike.get_object(xref, 0) @@ -421,12 +426,12 @@ def transcode_pngs(pike, images, image_name_fn, root, log, options): ) continue if compdata.type == leptonica.lept.L_FLATE_ENCODE: - rewrite_png(pike, im_obj, compdata, log) + rewrite_png(pike, im_obj, compdata) elif compdata.type == leptonica.lept.L_G4_ENCODE: - rewrite_png_as_g4(pike, im_obj, compdata, log) + rewrite_png_as_g4(pike, im_obj, compdata) -def rewrite_png_as_g4(pike, im_obj, compdata, log): +def rewrite_png_as_g4(pike, im_obj, compdata): im_obj.BitsPerComponent = 1 im_obj.Width = compdata.w im_obj.Height = compdata.h @@ -446,7 +451,7 @@ def rewrite_png_as_g4(pike, im_obj, compdata, log): return -def rewrite_png(pike, im_obj, compdata, log): +def rewrite_png(pike, im_obj, compdata): # When a PNG is inserted into a PDF, we more or less copy the IDAT section from # the PDF and transfer the rest of the PNG headers to PDF image metadata. # One thing we have to do is tell the PDF reader whether a predictor was used @@ -500,7 +505,6 @@ def rewrite_png(pike, im_obj, compdata, log): def optimize(input_file, output_file, context, save_settings): - log = context.log options = context.options if options.optimize == 0: safe_symlink(input_file, output_file) @@ -517,15 +521,15 @@ def optimize(input_file, output_file, context, save_settings): root = Path(output_file).parent / 'images' root.mkdir(exist_ok=True) - jpegs, pngs = extract_images_generic(pike, root, log, options) - transcode_jpegs(pike, jpegs, root, log, options) + jpegs, pngs = extract_images_generic(pike, root, options) + transcode_jpegs(pike, jpegs, root, options) # if options.optimize >= 2: # Try pngifying the jpegs - # transcode_pngs(pike, jpegs, jpg_name, root, log, options) - transcode_pngs(pike, pngs, png_name, root, log, options) + # transcode_pngs(pike, jpegs, jpg_name, root, options) + transcode_pngs(pike, pngs, png_name, root, options) - jbig2_groups = extract_images_jbig2(pike, root, log, options) - convert_to_jbig2(pike, jbig2_groups, root, log, options) + jbig2_groups = extract_images_jbig2(pike, root, options) + convert_to_jbig2(pike, jbig2_groups, root, options) target_file = Path(output_file).with_suffix('.opt.pdf') pike.remove_unreferenced_resources() @@ -553,8 +557,8 @@ def optimize(input_file, output_file, context, save_settings): def main(infile, outfile, level, jobs=1): - from tempfile import TemporaryDirectory - from shutil import copy + from tempfile import TemporaryDirectory # pylint: disable=import-outside-toplevel + from shutil import copy # pylint: disable=import-outside-toplevel class OptimizeOptions: """Emulate ocrmypdf's options""" @@ -582,7 +586,7 @@ def main(infile, outfile, level, jobs=1): ) with TemporaryDirectory() as td: - context = PDFContext(options, td, infile, None) + context = PdfContext(options, td, infile, None, None) tmpout = Path(td) / 'out.pdf' optimize( infile, diff --git a/src/ocrmypdf/pdfinfo/__init__.py b/src/ocrmypdf/pdfinfo/__init__.py index 093fea5e..0e8b8750 100644 --- a/src/ocrmypdf/pdfinfo/__init__.py +++ b/src/ocrmypdf/pdfinfo/__init__.py @@ -16,4 +16,4 @@ # You should have received a copy of the GNU General Public License # along with OCRmyPDF. If not, see . -from .info import Colorspace, Encoding, PdfInfo +from ocrmypdf.pdfinfo.info import Colorspace, Encoding, PdfInfo diff --git a/src/ocrmypdf/pdfinfo/ghosttext.py b/src/ocrmypdf/pdfinfo/ghosttext.py deleted file mode 100644 index 9626fad7..00000000 --- a/src/ocrmypdf/pdfinfo/ghosttext.py +++ /dev/null @@ -1,102 +0,0 @@ -# © 2018 James R. Barlow: github.com/jbarlow83 -# -# This file is part of OCRmyPDF. -# -# OCRmyPDF is free software: you can redistribute it and/or modify -# it under the terms of the GNU General Public License as published by -# the Free Software Foundation, either version 3 of the License, or -# (at your option) any later version. -# -# OCRmyPDF is distributed in the hope that it will be useful, -# but WITHOUT ANY WARRANTY; without even the implied warranty of -# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the -# GNU General Public License for more details. -# -# You should have received a copy of the GNU General Public License -# along with OCRmyPDF. If not, see . - -import logging -import re -import xml.etree.ElementTree as ET - -from ..exec import ghostscript - -gslog = logging.getLogger() - -# Forgive me for I have sinned -# I am using regular expressions to parse XML. However the XML in this case, -# generated by Ghostscript, is self-consistent enough to be parseable. -regex_remove_char_tags = re.compile( - br""" - ] # anything single character but > - | \">\" # special case: trap ">" - )* - /> # terminate with '/>' -""", - re.VERBOSE, -) - - -def page_get_textblocks(infile, pageno, xmltext, height): - """Get text boxes out of Ghostscript txtwrite xml""" - - root = xmltext - if not hasattr(xmltext, 'findall'): - return [] - - def blocks(): - for span in root.findall('.//span'): - bbox_str = span.attrib['bbox'] - font_size = span.attrib['size'] - pts = [int(pt) for pt in bbox_str.split()] - pts[1] = pts[1] - int(float(font_size) + 0.5) - bbox_topdown = tuple(pts) - bb = bbox_topdown - bbox_bottomup = (bb[0], height - bb[3], bb[2], height - bb[1]) - yield bbox_bottomup - - def joined_blocks(): - prev = None - for bbox in blocks(): - if prev is None: - prev = bbox - if bbox[1] == prev[1] and bbox[3] == prev[3]: - gap = prev[2] - bbox[0] - height = abs(bbox[3] - bbox[1]) - if gap < height: - # Join boxes - prev = (prev[0], prev[1], bbox[2], bbox[3]) - continue - # yield previously joined bboxes and start anew - yield prev - prev = bbox - if prev is not None: - yield prev - - return [block for block in joined_blocks()] - - -def extract_text_xml(infile, pdf, pageno=None, log=gslog): - existing_text = ghostscript.extract_text(infile, pageno=None) - existing_text = regex_remove_char_tags.sub(b' ', existing_text) - - try: - root = ET.fromstringlist([b'\n', existing_text, b'\n']) - page_xml = root.findall('page') - except ET.ParseError as e: - log.error( - "An error occurred while attempting to retrieve existing text in " - "the input file. Will attempt to continue assuming that there is " - "no existing text in the file. The error was:" - ) - log.error(e) - page_xml = [None] * len(pdf.pages) - - page_count_difference = len(pdf.pages) - len(page_xml) - if page_count_difference != 0: - log.error("The number of pages in the input file is inconsistent.") - log.error(f"Expected {len(pdf.pages)}, txtwrite says {len(page_xml)}") - if page_count_difference > 0: - page_xml.extend([None] * page_count_difference) - return page_xml diff --git a/src/ocrmypdf/pdfinfo/info.py b/src/ocrmypdf/pdfinfo/info.py index d304688b..fa2dcc05 100644 --- a/src/ocrmypdf/pdfinfo/info.py +++ b/src/ocrmypdf/pdfinfo/info.py @@ -21,18 +21,18 @@ import re from collections import defaultdict, namedtuple from decimal import Decimal from enum import Enum +from functools import partial from math import hypot, isclose -from os import PathLike, fspath +from os import PathLike from pathlib import Path from warnings import warn import pikepdf from pikepdf import PdfMatrix -from tqdm import tqdm +from ocrmypdf._concurrent import exec_progress_pool from ocrmypdf.exceptions import EncryptedPdfError -from ocrmypdf.exec import ghostscript -from ocrmypdf.pdfinfo import ghosttext +from ocrmypdf.helpers import Resolution, available_cpu_count from ocrmypdf.pdfinfo.layout import get_page_analysis, get_text_boxes logger = logging.getLogger() @@ -265,7 +265,7 @@ def _get_dpi(ctm_shorthand, image_size): dpi_w = scale_w * 72.0 dpi_h = scale_h * 72.0 - return dpi_w, dpi_h + return Resolution(dpi_w, dpi_h) class ImageInfo: @@ -356,12 +356,8 @@ class ImageInfo: return self._enc @property - def xres(self): - return _get_dpi(self._shorthand, (self._width, self._height))[0] - - @property - def yres(self): - return _get_dpi(self._shorthand, (self._width, self._height))[1] + def dpi(self): + return _get_dpi(self._shorthand, (self._width, self._height)) def __repr__(self): class_locals = { @@ -371,7 +367,7 @@ class ImageInfo: } return ( "" + "{comp} {bpc} {enc} {dpi}>" ).format(**class_locals) @@ -558,7 +554,7 @@ def simplify_textboxes(miner, textbox_getter): yield TextboxInfo(box.bbox, visible, corrupt) -def _pdf_get_pageinfo(pdf, pageno: int, infile: PathLike, xmltext: str): +def _pdf_get_pageinfo(pdf, pageno: int, infile: PathLike): pageinfo = {} pageinfo['pageno'] = pageno pageinfo['images'] = [] @@ -568,16 +564,10 @@ def _pdf_get_pageinfo(pdf, pageno: int, infile: PathLike, xmltext: str): width_pt = mediabox[2] - mediabox[0] height_pt = mediabox[3] - mediabox[1] - if xmltext is not None: - bboxes = ghosttext.page_get_textblocks( - fspath(infile), pageno, xmltext=xmltext, height=height_pt - ) - pageinfo['bboxes'] = bboxes - else: - pscript5_mode = str(pdf.docinfo.get('/Creator')).startswith('PScript5') - miner = get_page_analysis(infile, pageno, pscript5_mode) - pageinfo['textboxes'] = list(simplify_textboxes(miner, get_text_boxes)) - bboxes = (box.bbox for box in pageinfo['textboxes']) + pscript5_mode = str(pdf.docinfo.get('/Creator')).startswith('PScript5') + miner = get_page_analysis(infile, pageno, pscript5_mode) + pageinfo['textboxes'] = list(simplify_textboxes(miner, get_text_boxes)) + bboxes = (box.bbox for box in pageinfo['textboxes']) pageinfo['has_text'] = _page_has_text(bboxes, width_pt, height_pt) @@ -607,36 +597,69 @@ def _pdf_get_pageinfo(pdf, pageno: int, infile: PathLike, xmltext: str): pageinfo['images'] = [im for im in contentsinfo if isinstance(im, ImageInfo)] if pageinfo['images']: - xres = Decimal(max(image.xres for image in pageinfo['images'])) - yres = Decimal(max(image.yres for image in pageinfo['images'])) - pageinfo['xres'], pageinfo['yres'] = xres, yres - pageinfo['width_pixels'] = int(round(xres * pageinfo['width_inches'])) - pageinfo['height_pixels'] = int(round(yres * pageinfo['height_inches'])) + dpi = Resolution(0.0, 0.0).take_max(image.dpi for image in pageinfo['images']) + pageinfo['dpi'] = dpi + pageinfo['width_pixels'] = int(round(dpi.x * float(pageinfo['width_inches']))) + pageinfo['height_pixels'] = int(round(dpi.y * float(pageinfo['height_inches']))) return pageinfo -def _pdf_get_all_pageinfo(infile, detailed_analysis=False, log=None, progbar=False): +worker_pdf = None + + +def _pdf_pageinfo_sync_init(infile): + global worker_pdf # pylint: disable=global-statement + worker_pdf = pikepdf.open(infile) + + +def _pdf_pageinfo_sync(args): + global worker_pdf # pylint: disable=global-statement + pageno, infile = args + page = PageInfo(worker_pdf, pageno, infile) + return page + + +def _pdf_pageinfo_concurrent(pdf, infile, progbar, max_workers): + pages = [None] * len(pdf.pages) + + def update_pageinfo(result, pbar): + page = result + pages[page.pageno] = page + pbar.update() + + if max_workers is None: + max_workers = available_cpu_count() + + contexts = ((n, infile) for n in range(len(pdf.pages))) + + use_threads = False # No performance gain if threaded due to GIL + n_workers = min(1 + len(pages) // 4, max_workers) + if n_workers == 1: + # But if we decided on only one worker, there is no point in using + # a separate process. + use_threads = True + + exec_progress_pool( + use_threads=use_threads, + max_workers=n_workers, + tqdm_kwargs=dict( + total=len(pdf.pages), desc="Scan", unit='page', disable=not progbar + ), + task_initializer=partial(_pdf_pageinfo_sync_init, infile), + task=_pdf_pageinfo_sync, + task_arguments=contexts, + task_finished=update_pageinfo, + ) + return pages + + +def _pdf_get_all_pageinfo(infile, progbar=False, max_workers=None): pdf = pikepdf.open(infile) # Do not close in this function try: if pdf.is_encrypted: raise EncryptedPdfError() # Triggered by encryption with empty passwd - if detailed_analysis: - pages_xml = None - else: - pages_xml = ghosttext.extract_text_xml(infile, pdf, pageno=None, log=log) - - pages = [] - for n, _ in tqdm( - enumerate(pdf.pages), - total=len(pdf.pages), - desc="Scan", - unit='page', - disable=not progbar, - ): - page_xml = pages_xml[n] if pages_xml else None - page = PageInfo(pdf, n, infile, page_xml, detailed_analysis) - pages.append(page) + pages = _pdf_pageinfo_concurrent(pdf, infile, progbar, max_workers) except Exception: pdf.close() raise @@ -645,11 +668,10 @@ def _pdf_get_all_pageinfo(infile, detailed_analysis=False, log=None, progbar=Fal class PageInfo: - def __init__(self, pdf, pageno, infile, xmltext, detailed_analysis=False): + def __init__(self, pdf, pageno, infile): self._pageno = pageno self._infile = infile - self._pageinfo = _pdf_get_pageinfo(pdf, pageno, infile, xmltext) - self._detailed_analysis = detailed_analysis + self._pageinfo = _pdf_get_pageinfo(pdf, pageno, infile) @property def pageno(self): @@ -661,8 +683,6 @@ class PageInfo: @property def has_corrupt_text(self): - if not self._detailed_analysis: - raise NotImplementedError('Did not do detailed analysis') return any(tbox.is_corrupt for tbox in self._pageinfo['textboxes']) @property @@ -679,11 +699,11 @@ class PageInfo: @property def width_pixels(self): - return int(round(self.width_inches * self.xres)) + return int(round(float(self.width_inches) * self.dpi.x)) @property def height_pixels(self): - return int(round(self.height_inches * self.yres)) + return int(round(float(self.height_inches) * self.dpi.y)) @property def rotation(self): @@ -713,7 +733,7 @@ class PageInfo: if 'textboxes' not in self._pageinfo: if visible is not None and corrupt is not None: - raise NotImplementedError('Ghostscript textboxes cannot be classified') + raise NotImplementedError('Incomplete information on textboxes') return self._pageinfo['bboxes'] return ( @@ -723,12 +743,8 @@ class PageInfo: ) @property - def xres(self): - return self._pageinfo.get('xres', None) - - @property - def yres(self): - return self._pageinfo.get('yres', None) + def dpi(self): + return self._pageinfo.get('dpi', Resolution(0.0, 0.0)) @property def userunit(self): @@ -743,27 +759,19 @@ class PageInfo: def __repr__(self): return ( - '' - ).format( - self.pageno, - self.width_inches, - self.height_inches, - self.rotation, - self.xres, - self.yres, - self.has_text, + f'' ) class PdfInfo: """Get summary information about a PDF""" - def __init__(self, infile, detailed_page_analysis=False, log=logger, progbar=False): + def __init__(self, infile, progbar=False, max_workers=None): self._infile = infile - if ghostscript.version() in ('9.52',): - detailed_page_analysis = True # txtwrite doesn't work in these versions self._pages, pdf = _pdf_get_all_pageinfo( - infile, detailed_page_analysis, log=log, progbar=progbar + infile, progbar=progbar, max_workers=max_workers ) self._needs_rendering = pdf.root.get('/NeedsRendering', False) self._has_acroform = False @@ -812,13 +820,13 @@ class PdfInfo: def main(): - import argparse + import argparse # pylint: disable=import-outside-toplevel + from pprint import pprint # pylint: disable=import-outside-toplevel parser = argparse.ArgumentParser() parser.add_argument('infile') args = parser.parse_args() pagesinfo, pdfinfo = _pdf_get_all_pageinfo(args.infile) - from pprint import pprint pprint(pdfinfo) for page in pagesinfo: diff --git a/src/ocrmypdf/pdfinfo/layout.py b/src/ocrmypdf/pdfinfo/layout.py index 1368a75b..9eb8306a 100644 --- a/src/ocrmypdf/pdfinfo/layout.py +++ b/src/ocrmypdf/pdfinfo/layout.py @@ -25,65 +25,16 @@ import pdfminer.encodingdb import pdfminer.pdfdevice import pdfminer.pdfinterp from pdfminer.converter import PDFLayoutAnalyzer -from pdfminer.glyphlist import glyphname2unicode from pdfminer.layout import LAParams, LTChar, LTPage, LTTextBox from pdfminer.pdfdocument import PDFTextExtractionNotAllowed -from pdfminer.pdffont import PDFFont, PDFSimpleFont, PDFUnicodeNotDefined +from pdfminer.pdffont import PDFSimpleFont, PDFUnicodeNotDefined from pdfminer.pdfpage import PDFPage from pdfminer.utils import bbox2str, matrix2str -from ..exceptions import EncryptedPdfError +from ocrmypdf.exceptions import EncryptedPdfError STRIP_NAME = re.compile(r'[0-9]+') -# -# pdfminer 20181108 patches -# - -if pdfminer.__version__ == '20181108': - - def name2unicode(name): - """Fix pdfminer's name2unicode function - - Font cids that are mapped to names of the form /g123 seem to be, by convention - characters with no corresponding Unicode entry. These can be subsetted fonts - or symbolic fonts. There seems to be no way to map /g123 fonts to Unicode, - barring a ToUnicode data structure. - """ - if name in glyphname2unicode: - return glyphname2unicode[name] - if name.startswith('g') or name.startswith('a'): - raise KeyError(name) - if name.startswith('uni'): - try: - return chr(int(name[3:], 16)) - except ValueError: # Not hexadecimal - raise KeyError(name) - m = STRIP_NAME.search(name) - if not m: - raise KeyError(name) - return chr(int(m.group(0))) - - pdfminer.encodingdb.name2unicode = name2unicode - - original_PDFFont_init = PDFFont.__init__ - - def PDFFont__init__(self, descriptor, widths, default_width=None): - original_PDFFont_init(self, descriptor, widths, default_width) - # PDF spec says descent should be negative - # A font with a positive descent implies it floats entirely above the - # baseline, i.e. it's not really a baseline anymore. I have fonts that - # claim a positive descent, but treating descent as positive always seems - # to misposition text. - if self.descent > 0: - self.descent = -self.descent - - PDFFont.__init__ = PDFFont__init__ - -# -# end of pdfminer 20181108 patches -# - original_PDFSimpleFont_init = PDFSimpleFont.__init__ diff --git a/src/ocrmypdf/pluginspec.py b/src/ocrmypdf/pluginspec.py new file mode 100644 index 00000000..e3961fbf --- /dev/null +++ b/src/ocrmypdf/pluginspec.py @@ -0,0 +1,195 @@ +# © 2020 James R. Barlow: github.com/jbarlow83 +# +# This file is part of OCRmyPDF. +# +# OCRmyPDF is free software: you can redistribute it and/or modify +# it under the terms of the GNU General Public License as published by +# the Free Software Foundation, either version 3 of the License, or +# (at your option) any later version. +# +# OCRmyPDF is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with OCRmyPDF. If not, see . + +from abc import ABC, abstractstaticmethod +from argparse import ArgumentParser, Namespace +from collections import namedtuple +from pathlib import Path +from typing import AbstractSet, List, Optional + +import pluggy +from PIL import Image + +from ocrmypdf.helpers import Resolution + +hookspec = pluggy.HookspecMarker('ocrmypdf') + +# pylint: disable=unused-argument + + +@hookspec +def add_options(parser: ArgumentParser) -> None: + """Allows the plugin to add its own command line arguments. + + Even if you do not intend to use plugins in a command line context, you + should use this function to create your options. + """ + + +@hookspec +def check_options(options: Namespace) -> None: + """Called to ask the plugin to check all of its options. + + The plugin may modify the *options*. All objects that are in options must + be picklable so they can be marshalled to child worker processes. + """ + + +@hookspec +def validate(pdfinfo: 'PdfInfo', options: Namespace) -> None: + """Called to give a plugin an opportunity to review *options* and *pdfinfo*. + + *options* contains the "work order" to process a particular file. *pdfinfo* + contains information about the input file obtained after loading and + parsing. The plugin may modify the *options*. For example, you could decide + that a certain type of file should be treated with ``options.force_ocr = True`` + based on information in its *pdfinfo*. + + The plugin may raise :class:`ocrmypdf.exceptions.InputFileError` or any + :class:`ocrmypdf.exceptions.ExitCodeException` to request + normal termination. ocrmypdf will hold the plugin responsible for raising + exceptions of any other type. + + The return value is ignored. To abort processing, raise an ``ExitCodeException``. + """ + + +@hookspec(firstresult=True) +def rasterize_pdf_page( + input_file: Path, + output_file: Path, + raster_device: str, + raster_dpi: Resolution, + pageno: int, + page_dpi: Optional[Resolution] = None, + rotation: Optional[int] = None, + filter_vector: bool = False, +) -> Path: + """Rasterize one page of a PDF at resolution raster_dpi in canvas units. + + The image is sized to match the integer pixels dimensions implied by + raster_dpi even if those numbers are noninteger. The image's DPI will + be overridden with the values in page_dpi. + + Args: + raster_device: type of image to produce at output_file + raster_dpi: resolution at which to rasterize page + pageno: page number to rasterize (beginning at page 1) + page_dpi: resolution, overriding output image DPI + rotation: cardinal angle, clockwise, to rotate page + filter_vector: if True, remove vector graphics objects + Returns: + output_file + """ + + +@hookspec(firstresult=True) +def filter_ocr_image(page: 'PageContext', image: Image) -> Image: + """Called to filter the image before it is sent to OCR. + + This is the image that OCR sees, not what the user sees when they view the + PDF. + """ + + +@hookspec(firstresult=True) +def filter_page_image(page: 'PageContext', image_filename: Path) -> Path: + """Called to filter the whole page before it is inserted into the PDF. + + A whole page image is only produced when preprocessing command line arguments + are issued or when ``--force-ocr`` is issued. If no whole page is image is + produced for a given page, this function will not be called. This is not + the image that will be shown to OCR. + + ocrmypdf will create the PDF page based on the image format used. If you + convert the image to a JPEG, the output page will be created as a JPEG, etc. + Note that the ocrmypdf image optimization stage may ultimately chose a + different format. + """ + + +OrientationConfidence = namedtuple('OrientationConfidence', ('angle', 'confidence')) + + +class OcrEngine(ABC): + @abstractstaticmethod + def version() -> str: + """Returns the version of the OCR engine.""" + + @abstractstaticmethod + def creator_tag(options: Namespace) -> str: + """Returns the creator tag to identify this software's role in creating the PDF.""" + + @abstractstaticmethod + def __str__(self): + """Returns name of OCR engine and version.""" + + @abstractstaticmethod + def languages(options: Namespace) -> AbstractSet[str]: + """Returns set of languages that are supported.""" + + @abstractstaticmethod + def get_orientation(input_file: Path, options: Namespace) -> OrientationConfidence: + """Returns the orientation of the image.""" + + @abstractstaticmethod + def generate_hocr( + input_file: Path, output_hocr: Path, output_text: Path, options: Namespace + ) -> None: + """Called to produce a hOCR file.""" + + @abstractstaticmethod + def generate_pdf( + input_file: Path, output_pdf: Path, output_text: Path, options: Namespace + ) -> None: + """Called to produce a text only PDF (no image, invisible text).""" + + +@hookspec(firstresult=True) +def get_ocr_engine() -> OcrEngine: + pass + + +@hookspec(firstresult=True) +def generate_pdfa( + pdf_pages: List[Path], + pdfmark: Path, + output_file: Path, + compression: str, + pdf_version: str, + pdfa_part: str, +) -> Path: + """Generate a PDF/A. + + The pdf_pages, a list of files, will be merged into output_file. One or more + PDF files may be merged. The pdfmark file is a PostScript.ps file that + provides Ghostscript with details on how to perform the PDF/A + conversion. By default with we pick PDF/A-2b, but this works for 1 or 3. + + compression can be 'jpeg', 'lossless', or an empty string. In 'jpeg', + Ghostscript is instructed to convert color and grayscale images to DCT + (JPEG encoding). In 'lossless' Ghostscript is told to convert images to + Flate (lossless/PNG). If the parameter is omitted Ghostscript is left to + make its own decisions about how to encode images; it appears to use a + heuristic to decide how to encode images. As of Ghostscript 9.25, we + support passthrough JPEG which allows Ghostscript to avoid transcoding + images entirely. (The feature was added in 9.23 but broken, and the 9.24 + release of Ghostscript had regressions, so we don't support it until 9.25.) + + Returns: + output_file + """ diff --git a/src/ocrmypdf/exec/__init__.py b/src/ocrmypdf/subprocess.py similarity index 82% rename from src/ocrmypdf/exec/__init__.py rename to src/ocrmypdf/subprocess.py index 145cc43b..96a0afb6 100644 --- a/src/ocrmypdf/exec/__init__.py +++ b/src/ocrmypdf/subprocess.py @@ -1,4 +1,4 @@ -# © 2016 James R. Barlow: github.com/jbarlow83 +# © 2020 James R. Barlow: github.com/jbarlow83 # # This file is part of OCRmyPDF. # @@ -23,31 +23,22 @@ import re import shutil import sys from collections.abc import Mapping +from contextlib import suppress from distutils.version import LooseVersion from functools import lru_cache from pathlib import Path from subprocess import PIPE, STDOUT, CalledProcessError from subprocess import run as subprocess_run -from ..exceptions import ExitCode, MissingDependencyError +from ocrmypdf.exceptions import MissingDependencyError log = logging.getLogger(__name__) -def _get_program(args, env=None): - program = args[0] - test_path = env.get('_OCRMYPDF_TEST_PATH', '') - if test_path: - program = shutil.which(program, path=test_path) - return program - - def run(args, *, env=None, **kwargs): """Wrapper around subprocess.run() - The main purpose of this wrapper is to allow us to substitute the main program - for a spoof in the test suite. The hidden variable _OCRMYPDF_TEST_PATH replaces - the main PATH as a location to check for programs to run. + The main purpose of this wrapper is to log subprocess output. Secondly we have to account for behavioral differences in Windows in particular. Creating symbolic links in Windows requires administrator privileges and @@ -62,40 +53,58 @@ def run(args, *, env=None, **kwargs): env = os.environ # Search in spoof path if necessary - program = _get_program(args, env) - - # If we are running a .py on Windows, ensure we call it with this Python - # (to support test suite shims) - if os.name == 'nt' and program.lower().endswith('.py'): - args = [sys.executable, program] + args[1:] - else: - args = [program] + args[1:] + program = args[0] if os.name == 'nt': - paths = os.pathsep.join(os.get_exec_path(env)) - if not shutil.which(args[0], path=paths): - shimmed_path = shim_paths_with_program_files(env) - new_args0 = shutil.which(args[0], path=shimmed_path) - if new_args0: - args[0] = new_args0 + args = _fix_windows_args(program, args, env) - process_log = log.getChild(os.path.basename(program)) - process_log.debug("Running: %s", args) + log.debug("Running: %s", args) + process_log = log.getChild('subprocess.' + os.path.basename(program)) if sys.version_info < (3, 7) and os.name == 'nt': # Can't use close_fds=True on Windows with Python 3.6 or older # https://bugs.python.org/issue19575, etc. kwargs['close_fds'] = False - proc = subprocess_run(args, env=env, **kwargs) - if process_log.isEnabledFor(logging.DEBUG): - try: - stderr = proc.stderr.decode('utf-8', 'replace') - except AttributeError: - stderr = proc.stderr - if stderr: + + stderr = None + try: + proc = subprocess_run(args, env=env, **kwargs) + except CalledProcessError as e: + stderr = getattr(e, 'stderr', None) + raise + else: + stderr = getattr(proc, 'stderr', None) + finally: + if process_log.isEnabledFor(logging.DEBUG) and stderr: + with suppress(AttributeError, UnicodeDecodeError): + stderr = stderr.decode('utf-8', 'replace') process_log.debug("stderr = %s", stderr) return proc +def _fix_windows_args(program, args, env): + """Adjust our desired program and command line arguments for use on Windows""" + + if sys.version_info < (3, 8): + # bpo-33617 - Windows needs manual Path -> str conversion + args = [os.fspath(arg) for arg in args] + program = os.fspath(program) + + # If we are running a .py on Windows, ensure we call it with this Python + # (to support test suite shims) + if program.lower().endswith('.py'): + args = [sys.executable] + args + + paths = os.pathsep.join(os.get_exec_path(env)) + if not shutil.which(args[0], path=paths): + # If the program we want is not on the PATH, add some interesting + # locations in %PROGRAMFILES% to the PATH and try again + shimmed_path = shim_paths_with_program_files(env) + new_args0 = shutil.which(args[0], path=shimmed_path) + if new_args0: + args[0] = new_args0 + return args + + def get_version(program, *, version_arg='--version', regex=r'(\d+(\.\d+)*)', env=None): """Get the version of the specified program""" args_prog = [program, version_arg] @@ -260,13 +269,12 @@ def check_external_program( need_version, required_for=None, recommended=False, - **kwargs, # To consume log parameter ): - if kwargs: - if not 'log' in kwargs: - log.warning('check_external_program(log=...) is deprecated') try: - found_version = version_checker() + if callable(version_checker): + found_version = version_checker() + else: + found_version = version_checker except (CalledProcessError, FileNotFoundError, MissingDependencyError): _error_missing_program(program, package, required_for, recommended) if not recommended: diff --git a/tests/cache/manifest.jsonl b/tests/cache/manifest.jsonl index 05a0e86c..23e50166 100644 --- a/tests/cache/manifest.jsonl +++ b/tests/cache/manifest.jsonl @@ -69,3 +69,4 @@ {"tesseract_version": "tesseract 4.1.0 leptonica-1.78.0 libgif 5.1.4 : libjpeg 9c : libpng 1.6.37 : libtiff 4.1.0 : zlib 1.2.11 : libwebp 1.0.3 : libopenjp2 2.3.1 Found AVX2 Found AVX Found SSE ", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.5", "argv_slug": "__-l__eng__000002_ocr.png__000002_ocr_tess__pdf__txt", "sourcefile": "resources/3small.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000002_ocr.png", "$TMPDIR/000002_ocr_tess", "pdf", "txt"]} {"tesseract_version": "tesseract 4.1.0 leptonica-1.78.0 libgif 5.1.4 : libjpeg 9c : libpng 1.6.37 : libtiff 4.1.0 : zlib 1.2.11 : libwebp 1.0.3 : libopenjp2 2.3.1 Found AVX2 Found AVX Found SSE ", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.5", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/3small.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]} {"tesseract_version": "tesseract 4.1.0 leptonica-1.78.0 libgif 5.1.4 : libjpeg 9c : libpng 1.6.37 : libtiff 4.1.0 : zlib 1.2.11 : libwebp 1.0.3 : libopenjp2 2.3.1 Found AVX2 Found AVX Found SSE ", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.5", "argv_slug": "__-l__eng__000003_ocr.png__000003_ocr_tess__pdf__txt", "sourcefile": "resources/3small.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000003_ocr.png", "$TMPDIR/000003_ocr_tess", "pdf", "txt"]} +{"tesseract_version": "tesseract 4.1.1 leptonica-1.79.0 libgif 5.2.1 : libjpeg 9d : libpng 1.6.37 : libtiff 4.1.0 : zlib 1.2.11 : libwebp 1.1.0 : libopenjp2 2.3.1 Found AVX2 Found AVX Found FMA Found SSE ", "platform": "Darwin-18.7.0-x86_64-i386-64bit", "python": "3.7.7", "argv_slug": "__-l__eng__000002_ocr.png__000002_ocr_hocr__hocr__txt", "sourcefile": "resources/multipage.pdf", "args": ["-l", "eng", "$TMPDIR/000002_ocr.png", "$TMPDIR/000002_ocr_hocr", "hocr", "txt"]} diff --git a/tests/cache/multipage/__-l__eng__000002_ocr.png__000002_ocr_hocr__hocr__txt/hocr.bin b/tests/cache/multipage/__-l__eng__000002_ocr.png__000002_ocr_hocr__hocr__txt/hocr.bin new file mode 100644 index 00000000..27e769f6 --- /dev/null +++ b/tests/cache/multipage/__-l__eng__000002_ocr.png__000002_ocr_hocr__hocr__txt/hocr.bin @@ -0,0 +1,30 @@ + + + + + + + + + + +
+
+

+ + YOOOxXYOO0O + pixels + at + GOO + DPI + + + oO] + megapixels + +

+
+
+ + diff --git a/tests/cache/multipage/__-l__eng__000002_ocr.png__000002_ocr_hocr__hocr__txt/stderr.bin b/tests/cache/multipage/__-l__eng__000002_ocr.png__000002_ocr_hocr__hocr__txt/stderr.bin new file mode 100644 index 00000000..16b617e5 --- /dev/null +++ b/tests/cache/multipage/__-l__eng__000002_ocr.png__000002_ocr_hocr__hocr__txt/stderr.bin @@ -0,0 +1 @@ +Tesseract Open Source OCR Engine v4.1.1 with Leptonica diff --git a/tests/cache/multipage/__-l__eng__000002_ocr.png__000002_ocr_hocr__hocr__txt/stdout.bin b/tests/cache/multipage/__-l__eng__000002_ocr.png__000002_ocr_hocr__hocr__txt/stdout.bin new file mode 100644 index 00000000..e69de29b diff --git a/tests/cache/multipage/__-l__eng__000002_ocr.png__000002_ocr_hocr__hocr__txt/txt.bin b/tests/cache/multipage/__-l__eng__000002_ocr.png__000002_ocr_hocr__hocr__txt/txt.bin new file mode 100644 index 00000000..21e1e995 --- /dev/null +++ b/tests/cache/multipage/__-l__eng__000002_ocr.png__000002_ocr_hocr__hocr__txt/txt.bin @@ -0,0 +1,3 @@ +YOOOxXYOO0O pixels at GOO DPI +oO] megapixels + \ No newline at end of file diff --git a/tests/conftest.py b/tests/conftest.py index 8724ead5..adcd1354 100644 --- a/tests/conftest.py +++ b/tests/conftest.py @@ -24,7 +24,9 @@ from subprocess import PIPE, run import pytest -from ocrmypdf import api, cli +from ocrmypdf import api, cli, pdfinfo +from ocrmypdf._exec import unpaper +from ocrmypdf._plugin_manager import get_parser_options_plugins pytest_plugins = ['helpers_namespace'] @@ -51,7 +53,7 @@ def is_macos(): def running_in_docker(): # Docker creates a file named /.dockerenv (newer versions) or # /.dockerinit (older) -- this is undocumented, not an offical test - return os.path.exists('/.dockerenv') or os.path.exists('/.dockerinit') + return Path('/.dockerenv').exists() or Path('/.dockerinit').exists() @pytest.helpers.register @@ -62,125 +64,17 @@ def running_in_travis(): @pytest.helpers.register def have_unpaper(): try: - from ocrmypdf.exec import unpaper - unpaper.version() - except Exception: + except Exception: # pylint: disable=broad-except return False return True -TESTS_ROOT = os.path.abspath(os.path.dirname(__file__)) -SPOOF_PATH = os.path.join(TESTS_ROOT, 'spoof') -PROJECT_ROOT = os.path.dirname(TESTS_ROOT) +TESTS_ROOT = Path(__file__).parent.resolve() +PROJECT_ROOT = TESTS_ROOT OCRMYPDF = [sys.executable, '-m', 'ocrmypdf'] -WINDOWS_SHIM_TEMPLATE = """ -# This is a shim for Windows that has the same effect as a symlink to the target .py -# file -import os -import subprocess -import sys - -args = [sys.executable, {spoofer}, *sys.argv[1:]] -p = subprocess.run(args, check=False, stdout=subprocess.PIPE, stderr=subprocess.PIPE) -sys.stdout.buffer.write(p.stdout) -sys.stderr.buffer.write(p.stderr) -sys.exit(p.returncode) -""" - -assert ast.parse(WINDOWS_SHIM_TEMPLATE.format(spoofer=repr(r"C:\\Temp\\file.py"))) - - -@pytest.helpers.register -def spoof(tmp_path_factory, **kwargs): - """Modify PATH to override subprocess executables - - spoof(tmp_path_factory, program1='replacement', ...) - - For the test suite we need a way override executables, so that we can - substitute desired results such as errors or just speed up OCR. - - On POSIXish platforms we create a temporary folder with overrides that - are symlinks to the executables we want to run. We do not actually override - PATH. We also set an environment variable _OCRMYPDF_TEST_PATH, which - OCRmyPDF's subprocess wrapper will check before they use regular PATH. The - output is a folder full of executables we are overriding. We can override - multiple executables. The end result is a folder we can use in a PATH-style - lookup to override some executables: - - /tmp/abcxyz/tesseract -> ocrmypdf/tests/resources/spoof/tesseract_crash.py - /tmp/abcxyz/gs -> ocrmypdf/tests/resources/spoof/gs_backflip.py - - Windows needs extra help from us because usually, only the Administrator - can create symlinks. Instead we create small Python scripts that call - the programs we want, implementing the effect of a symlink. This is cleaner - than creating Windows executables or trying to use non-Python scripts. - The temporary folder generated for Windows could like: - - %TEMP%\abcxyz\tesseract.py: - (script that runs ocrmypdf/tests/resources/spoof/tesseract_crash.py) - %TEMP%\abcxyz\gswin32c.py: - (script that runs ocrmypdf/tests/resources/spoof/gs_backflip.py) - %TEMP%\abcxyz\gswin64c.py: - (script that runs ocrmypdf/tests/resources/spoof/gs_backflip.py) - - We also address one quirk here, that Ghostscript may be known as gswin32c - or gswin64c, depending on what the user installed (regardless of Windows - itself). On POSIX, Ghostscript is just 'gs'. We handle the special case here - too. - - All of this is intimately dependent on the machinery in ocrmypdf.exec.run(). - In particular, for Windows, that code has to know that if there is a .py - file, it needs to run it with Python, since Windows does not like being - asked to execute files. - - We don't overload PATH directly because we have some tests where we call - ocrmypdf as a subprocess (to exercise the command line interface) and some - tests where we call it as an API. - """ - env = os.environ.copy() - slug = '-'.join(v.replace('.py', '') for v in sorted(kwargs.values())) - spoofer_base = tmp_path_factory.mktemp('spoofers') - tmpdir = Path(spoofer_base / slug) - tmpdir.mkdir(parents=True) - - for replace_program, with_spoof in kwargs.items(): - spoofer = Path(SPOOF_PATH) / with_spoof - if os.name != 'nt': - spoofer.chmod(0o755) - (tmpdir / replace_program).symlink_to(spoofer) - else: - py_file = WINDOWS_SHIM_TEMPLATE.format( - spoofer=repr(os.fspath(spoofer.absolute())) - ) - if replace_program == 'gs': - programs = ['gswin64c', 'gswin32c'] - else: - programs = [replace_program] - for prog in programs: - (tmpdir / f'{prog}.py').write_text(py_file, encoding='utf-8') - - env['_OCRMYPDF_TEST_PATH'] = str(tmpdir) + os.pathsep + env['PATH'] - if os.name == 'nt': - if '.py' not in env['PATHEXT'].lower(): - raise EnvironmentError("PATHEXT is not configured to support .py") - return env - - -@pytest.fixture -def spoof_tesseract_noop(tmp_path_factory): - return spoof(tmp_path_factory, tesseract='tesseract_noop.py') - - -@pytest.fixture -def spoof_tesseract_cache(tmp_path_factory): - if running_in_docker(): - return os.environ.copy() - return spoof(tmp_path_factory, tesseract="tesseract_cache.py") - - @pytest.fixture def resources(): return Path(TESTS_ROOT) / 'resources' @@ -211,58 +105,43 @@ def no_outpdf(tmp_path): @pytest.helpers.register -def check_ocrmypdf(input_file, output_file, *args, env=None): +def check_ocrmypdf(input_file, output_file, *args): """Run ocrmypdf and confirmed that a valid file was created""" + args = [str(input_file), str(output_file)] + [ + str(arg) for arg in args if arg is not None + ] - options = cli.parser.parse_args( - [str(input_file), str(output_file)] - + [str(arg) for arg in args if arg is not None] - ) - api.check_options(options) - if env: - options.tesseract_env = env - options.tesseract_env['_OCRMYPDF_TEST_INFILE'] = os.fspath(input_file) - result = api.run_pipeline(options, api=True) + _parser, options, plugin_manager = get_parser_options_plugins(args=args) + api.check_options(options, plugin_manager) + result = api.run_pipeline(options, plugin_manager=plugin_manager, api=True) assert result == 0 - assert os.path.exists(str(output_file)), "Output file not created" - assert os.stat(str(output_file)).st_size > 100, "PDF too small or empty" + assert output_file.exists(), "Output file not created" + assert output_file.stat().st_size > 100, "PDF too small or empty" return output_file @pytest.helpers.register -def run_ocrmypdf_api(input_file, output_file, *args, env=None): +def run_ocrmypdf_api(input_file, output_file, *args): """Run ocrmypdf via API and let caller deal with results Does not currently have a way to manipulate the PATH except for Tesseract. """ - options = cli.parser.parse_args( - [str(input_file), str(output_file)] - + [str(arg) for arg in args if arg is not None] - ) - api.check_options(options) - if env: - options.tesseract_env = env.copy() - options.tesseract_env['_OCRMYPDF_TEST_INFILE'] = os.fspath(input_file) - first_path = env.get('_OCRMYPDF_TEST_PATH', '').split(os.pathsep)[0] - if 'spoof' in first_path: - assert 'gs' not in first_path, "use run_ocrmypdf() for gs" - assert 'tesseract' in first_path - if options.tesseract_env: - assert all(isinstance(v, (str, bytes)) for v in options.tesseract_env.values()) + args = [str(input_file), str(output_file)] + [ + str(arg) for arg in args if arg is not None + ] + _parser, options, plugin_manager = get_parser_options_plugins(args=args) - return api.run_pipeline(options, api=False) + api.check_options(options, plugin_manager) + return api.run_pipeline(options, plugin_manager=None, api=False) @pytest.helpers.register -def run_ocrmypdf(input_file, output_file, *args, env=None, universal_newlines=True): +def run_ocrmypdf(input_file, output_file, *args, universal_newlines=True): "Run ocrmypdf and let caller deal with results" - if env is None: - env = os.environ.copy() - p_args = ( OCRMYPDF + [str(arg) for arg in args if arg is not None] @@ -274,10 +153,16 @@ def run_ocrmypdf(input_file, output_file, *args, env=None, universal_newlines=Tr # Details: https://coverage.readthedocs.io/en/coverage-5.0/subprocess.html coverage_rc = Path(__file__).parent.parent / '.coveragerc' assert coverage_rc.exists() + env = os.environ.copy() env['COVERAGE_PROCESS_START'] = os.fspath(coverage_rc) p = run( - p_args, stdout=PIPE, stderr=PIPE, universal_newlines=universal_newlines, env=env + p_args, + stdout=PIPE, + stderr=PIPE, + universal_newlines=universal_newlines, + env=env, + check=False, ) # print(p.stderr) return p, p.stdout, p.stderr @@ -285,8 +170,6 @@ def run_ocrmypdf(input_file, output_file, *args, env=None, universal_newlines=Tr @pytest.helpers.register def first_page_dimensions(pdf): - from ocrmypdf import pdfinfo - info = pdfinfo.PdfInfo(pdf) page0 = info[0] return (page0.width_inches, page0.height_inches) diff --git a/tests/spoof/gs_feature_elision.py b/tests/plugins/gs_feature_elision.py old mode 100755 new mode 100644 similarity index 59% rename from tests/spoof/gs_feature_elision.py rename to tests/plugins/gs_feature_elision.py index a06deaf3..419855cb --- a/tests/spoof/gs_feature_elision.py +++ b/tests/plugins/gs_feature_elision.py @@ -1,5 +1,4 @@ -#!/usr/bin/env python3 -# © 2016 James R. Barlow: github.com/jbarlow83 +# © 2020 James R. Barlow: github.com/jbarlow83 # # Permission is hereby granted, free of charge, to any person obtaining a # copy of this software and associated documentation files (the @@ -20,34 +19,31 @@ # TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE # SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. +from unittest.mock import patch -import os -import sys -from subprocess import check_call - -from gs import real_ghostscript - -"""Replicate one type of Ghostscript feature elision warning during -PDF/A creation.""" - +from ocrmypdf import hookimpl +from ocrmypdf.builtin_plugins import ghostscript +from ocrmypdf.subprocess import run elision_warning = """GPL Ghostscript 9.20: Setting Overprint Mode to 1 not permitted in PDF/A-2, overprint mode not set""" -def main(): - if '--version' in sys.argv: - print('9.20') - print('SPOOFED: ' + os.path.basename(__file__)) - sys.exit(0) - gs_args = ['gs'] + sys.argv[1:] - check_call(gs_args) - - if '-sDEVICE=pdfwrite' in sys.argv[1:]: - print(elision_warning) - - sys.exit(0) +def run_append_stderr(*args, **kwargs): + proc = run(*args, **kwargs) + proc.stderr = b'\n'.join([proc.stderr, elision_warning.encode('utf-8')]) + return proc -if __name__ == '__main__': - main() +@hookimpl +def generate_pdfa(pdf_pages, pdfmark, output_file, compression, pdf_version, pdfa_part): + with patch('ocrmypdf._exec.ghostscript.run', new=run_append_stderr): + ghostscript.generate_pdfa( + pdf_pages=pdf_pages, + pdfmark=pdfmark, + output_file=output_file, + compression=compression, + pdf_version=pdf_version, + pdfa_part=pdfa_part, + ) + return output_file diff --git a/tests/spoof/gs_pdfa_failure.py b/tests/plugins/gs_pdfa_failure.py old mode 100755 new mode 100644 similarity index 59% rename from tests/spoof/gs_pdfa_failure.py rename to tests/plugins/gs_pdfa_failure.py index 1d9fdf7d..dcad94f6 --- a/tests/spoof/gs_pdfa_failure.py +++ b/tests/plugins/gs_pdfa_failure.py @@ -1,5 +1,4 @@ -#!/usr/bin/env python3 -# © 2016 James R. Barlow: github.com/jbarlow83 +# © 2020 James R. Barlow: github.com/jbarlow83 # # Permission is hereby granted, free of charge, to any person obtaining a # copy of this software and associated documentation files (the @@ -20,41 +19,33 @@ # TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE # SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. -import os -import sys +from unittest.mock import patch -from gs import real_ghostscript +from ocrmypdf import hookimpl +from ocrmypdf.builtin_plugins import ghostscript +from ocrmypdf.subprocess import run -"""Replicate Ghostscript PDF/A conversion failure by suppressing some -arguments""" - - -def main(): - if '--version' in sys.argv: - print('9.20') - print('SPOOFED: ' + os.path.basename(__file__)) - sys.exit(0) - - # Unless some argument is calling for PDFA generation, forward to - # real ghostscript - if not any(arg.startswith('-dPDFA') for arg in sys.argv): - real_ghostscript(sys.argv) - return - +def run_rig_args(args, **kwargs): # Remove the two arguments that tell ghostscript to create a PDF/A # Does not remove the Postscript definition file - not necessary # to cause PDF/A creation failure - argv = [] - for arg in sys.argv: - if arg.startswith('-dPDFA'): - continue - elif arg.startswith('-dPDFACompatibilityPolicy'): - continue - argv.append(arg) - - real_ghostscript(argv) + new_args = [ + arg for arg in args if not arg.startswith('-dPDFA') and not arg.endswith('.ps') + ] + proc = run(new_args, **kwargs) + return proc -if __name__ == '__main__': - main() +@hookimpl +def generate_pdfa(pdf_pages, pdfmark, output_file, compression, pdf_version, pdfa_part): + with patch('ocrmypdf._exec.ghostscript.run', new=run_rig_args): + ghostscript.generate_pdfa( + pdf_pages=pdf_pages, + pdfmark=pdfmark, + output_file=output_file, + compression=compression, + pdf_version=pdf_version, + pdfa_part=pdfa_part, + ) + return output_file diff --git a/tests/spoof/gs_raster_failure.py b/tests/plugins/gs_raster_failure.py old mode 100755 new mode 100644 similarity index 51% rename from tests/spoof/gs_raster_failure.py rename to tests/plugins/gs_raster_failure.py index 7619aae2..98b1984c --- a/tests/spoof/gs_raster_failure.py +++ b/tests/plugins/gs_raster_failure.py @@ -1,5 +1,4 @@ -#!/usr/bin/env python3 -# © 2016 James R. Barlow: github.com/jbarlow83 +# © 2020 James R. Barlow: github.com/jbarlow83 # # Permission is hereby granted, free of charge, to any person obtaining a # copy of this software and associated documentation files (the @@ -20,30 +19,41 @@ # TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE # SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. +from pathlib import Path +from subprocess import CalledProcessError +from unittest.mock import patch -import os -import sys - -from gs import real_ghostscript - -"""Replicate Ghostscript raster failure while allowing rendering""" +from ocrmypdf import hookimpl +from ocrmypdf.builtin_plugins import ghostscript +from ocrmypdf.subprocess import run -def main(): - if '--version' in sys.argv: - print('9.20') - print('SPOOFED: ' + os.path.basename(__file__)) - sys.exit(0) - - # For non-image rastering calls, use real ghostscript - if '-sDEVICE=pdfwrite' in sys.argv or '-sDEVICE=txtwrite' in sys.argv: - real_ghostscript(sys.argv) - return - - # Fail - print("ERROR: Ghost story archive not found", file=sys.stderr) - sys.exit(1) +def raise_gs_fail(*args, **kwargs): + raise CalledProcessError( + 1, 'gs', output=b"", stderr=b"ERROR: Ghost story archive not found" + ) -if __name__ == '__main__': - main() +@hookimpl +def rasterize_pdf_page( + input_file, + output_file, + raster_device, + raster_dpi, + pageno, + page_dpi=None, + rotation=None, + filter_vector=False, +) -> Path: + with patch('ocrmypdf._exec.ghostscript.run', new=raise_gs_fail): + ghostscript.rasterize_pdf_page( + input_file=input_file, + output_file=output_file, + raster_device=raster_device, + raster_dpi=raster_dpi, + pageno=pageno, + page_dpi=page_dpi, + rotation=rotation, + filter_vector=filter_vector, + ) + return output_file diff --git a/tests/spoof/gs_render_failure.py b/tests/plugins/gs_render_failure.py old mode 100755 new mode 100644 similarity index 55% rename from tests/spoof/gs_render_failure.py rename to tests/plugins/gs_render_failure.py index d0c1d60d..c27a5801 --- a/tests/spoof/gs_render_failure.py +++ b/tests/plugins/gs_render_failure.py @@ -1,5 +1,4 @@ -#!/usr/bin/env python3 -# © 2016-18 James R. Barlow: github.com/jbarlow83 +# © 2020 James R. Barlow: github.com/jbarlow83 # # Permission is hereby granted, free of charge, to any person obtaining a # copy of this software and associated documentation files (the @@ -20,29 +19,30 @@ # TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE # SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. -"""Replicate Ghostscript render failure while allowing rasterizing""" +from pathlib import Path +from subprocess import CalledProcessError +from unittest.mock import patch -import os -import sys - -from gs import real_ghostscript +from ocrmypdf import hookimpl +from ocrmypdf.builtin_plugins import ghostscript +from ocrmypdf.subprocess import run -def main(): - if '--version' in sys.argv: - print('9.20') - print('SPOOFED: ' + os.path.basename(__file__)) - sys.exit(0) - - # For any rasterize calls (device != pdfwrite) call real ghostscript - if '-sDEVICE=pdfwrite' not in sys.argv: - real_ghostscript(sys.argv) - return - - # Fail - print("ERROR: Casper is not a friendly ghost", file=sys.stderr) - sys.exit(1) +def raise_gs_fail(*args, **kwargs): + raise CalledProcessError( + 1, 'gs', output=b"", stderr=b"ERROR: Casper is not a friendly ghost" + ) -if __name__ == '__main__': - main() +@hookimpl +def generate_pdfa(pdf_pages, pdfmark, output_file, compression, pdf_version, pdfa_part): + with patch('ocrmypdf._exec.ghostscript.run', new=raise_gs_fail): + ghostscript.generate_pdfa( + pdf_pages=pdf_pages, + pdfmark=pdfmark, + output_file=output_file, + compression=compression, + pdf_version=pdf_version, + pdfa_part=pdfa_part, + ) + return output_file diff --git a/tests/plugins/tesseract_badutf8.py b/tests/plugins/tesseract_badutf8.py new file mode 100644 index 00000000..3511938d --- /dev/null +++ b/tests/plugins/tesseract_badutf8.py @@ -0,0 +1,63 @@ +# © 2020 James R. Barlow: github.com/jbarlow83 +# +# Permission is hereby granted, free of charge, to any person obtaining a +# copy of this software and associated documentation files (the +# "Software"), to deal in the Software without restriction, including +# without limitation the rights to use, copy, modify, merge, publish, +# distribute, sublicense, and/or sell copies of the Software, and to +# permit persons to whom the Software is furnished to do so, subject to +# the following conditions: +# +# The above copyright notice and this permission notice shall be included +# in all copies or substantial portions of the Software. +# +# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS +# OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF +# MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. +# IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY +# CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, +# TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE +# SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. + +"""Tesseract bad utf8 + +In some cases, some versions of Tesseract can output binary gibberish or data +that is not UTF-8 compatible, so we are forced to check that we can convert it +and present it to the user. +""" + +from subprocess import CalledProcessError +from unittest.mock import patch + +from ocrmypdf import hookimpl +from ocrmypdf.builtin_plugins.tesseract_ocr import TesseractOcrEngine + + +def bad_utf8(*args, **kwargs): + raise CalledProcessError( + 1, + 'tesseract', + output=b'\x96\xb3\x8c\xf8\x82\xc8UTF-8\x0a', # "Invalid UTF-8" in Shift JIS + stderr=b"", + ) + + +class BadUtf8OcrEngine(TesseractOcrEngine): + @staticmethod + def generate_hocr(input_file, output_hocr, output_text, options): + with patch('ocrmypdf._exec.tesseract.run', new=bad_utf8): + TesseractOcrEngine.generate_hocr( + input_file, output_hocr, output_text, options + ) + + @staticmethod + def generate_pdf(input_file, output_pdf, output_text, options): + with patch('ocrmypdf._exec.tesseract.run', new=bad_utf8): + TesseractOcrEngine.generate_pdf( + input_file, output_pdf, output_text, options + ) + + +@hookimpl +def get_ocr_engine(): + return BadUtf8OcrEngine() diff --git a/tests/plugins/tesseract_big_image_error.py b/tests/plugins/tesseract_big_image_error.py new file mode 100644 index 00000000..04d0e0cd --- /dev/null +++ b/tests/plugins/tesseract_big_image_error.py @@ -0,0 +1,61 @@ +# © 2020 James R. Barlow: github.com/jbarlow83 +# +# Permission is hereby granted, free of charge, to any person obtaining a +# copy of this software and associated documentation files (the +# "Software"), to deal in the Software without restriction, including +# without limitation the rights to use, copy, modify, merge, publish, +# distribute, sublicense, and/or sell copies of the Software, and to +# permit persons to whom the Software is furnished to do so, subject to +# the following conditions: +# +# The above copyright notice and this permission notice shall be included +# in all copies or substantial portions of the Software. +# +# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS +# OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF +# MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. +# IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY +# CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, +# TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE +# SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. + +from subprocess import CalledProcessError +from unittest.mock import patch + +from ocrmypdf import hookimpl +from ocrmypdf.builtin_plugins.tesseract_ocr import TesseractOcrEngine + + +def raise_size_exception(*args, **kwargs): + raise CalledProcessError( + 1, + 'tesseract', + output=b"Image too large: (33830, 14959)\nError during processing.", + stderr=b"", + ) + + +class BigImageErrorOcrEngine(TesseractOcrEngine): + @staticmethod + def get_orientation(input_file, options): + with patch('ocrmypdf._exec.tesseract.run', new=raise_size_exception): + return TesseractOcrEngine.get_orientation(input_file, options) + + @staticmethod + def generate_hocr(input_file, output_hocr, output_text, options): + with patch('ocrmypdf._exec.tesseract.run', new=raise_size_exception): + TesseractOcrEngine.generate_hocr( + input_file, output_hocr, output_text, options + ) + + @staticmethod + def generate_pdf(input_file, output_pdf, output_text, options): + with patch('ocrmypdf._exec.tesseract.run', new=raise_size_exception): + TesseractOcrEngine.generate_pdf( + input_file, output_pdf, output_text, options + ) + + +@hookimpl +def get_ocr_engine(): + return BigImageErrorOcrEngine() diff --git a/tests/spoof/tesseract_cache.py b/tests/plugins/tesseract_cache.py old mode 100755 new mode 100644 similarity index 55% rename from tests/spoof/tesseract_cache.py rename to tests/plugins/tesseract_cache.py index adf3e257..1df3fd98 --- a/tests/spoof/tesseract_cache.py +++ b/tests/plugins/tesseract_cache.py @@ -1,5 +1,4 @@ -#!/usr/bin/env python3 -# © 2016 James R. Barlow: github.com/jbarlow83 +# © 2020 James R. Barlow: github.com/jbarlow83 # # Permission is hereby granted, free of charge, to any person obtaining a # copy of this software and associated documentation files (the @@ -22,11 +21,10 @@ """Cache output of tesseract to speed up test suite -The cache is keyed by an environment variable that slips the input test file -from tests/resources/ to us. The input arguments are slugged into a hideous -filename that more or less represents them literally. Joined together, this -becomes the name of the cache folder. A few name files like stdout, stderr, -hocr, pdf, describe the output to reproduce. +The cache is keyed by by the input test file The input arguments are slugged +into a hideous filename that more or less represents them literally. Joined +together, this becomes the name of the cache folder. A few name files like +stdout, stderr, hocr, pdf, describe the output to reproduce. Changes to tests/resources/ or image processing algorithms don't trigger a cache miss. By design, an input image that varies according to platform @@ -40,10 +38,7 @@ information about the system that produced the results used when cache was generated. This mainly a log to answer questions about how the files were produced. -For performance reasons, especially the slow performance of Tesseract on -machines with AVX2, the cache is now bundled. - -Certain operations are not cached and routed to tesseract directly. +Certain operations are not cached and routed to Tesseract OCR directly. Assumes Tesseract 4.0.0-alpha or higher. @@ -51,17 +46,23 @@ Assumes Tesseract 4.0.0-alpha or higher. import argparse import json -import os +import logging import platform import re import shutil -import subprocess -import sys +from functools import partial from pathlib import Path +from subprocess import PIPE, CalledProcessError, CompletedProcess +from unittest.mock import patch -__version__ = subprocess.check_output( - ['tesseract', '--version'], stderr=subprocess.STDOUT -).decode() +from ocrmypdf import hookimpl +from ocrmypdf.builtin_plugins.tesseract_ocr import TesseractOcrEngine +from ocrmypdf.subprocess import run + +log = logging.getLogger(__name__) + +TESTS_ROOT = Path(__file__).resolve().parent.parent +CACHE_ROOT = TESTS_ROOT / 'cache' parser = argparse.ArgumentParser( @@ -77,43 +78,15 @@ parser.add_argument('-c', action='append') parser.add_argument('--psm', type=int) parser.add_argument('--oem', type=int) -TESTS_ROOT = Path(__file__).resolve().parent.parent -CACHE_ROOT = TESTS_ROOT / 'cache' - - -def real_tesseract(): - tess_args = ['tesseract'] + sys.argv[1:] - os.execvp("tesseract", tess_args) - return # Not reachable - - -def main(): - if any( - opt in sys.argv[1:] - for opt in ('--print-parameters', '--list-langs', '--version') - ): - real_tesseract() # jump into real tesseract, replacing this process - - # Convert non-standard but supported -psm to --psm - sys.argv = ['--psm' if arg == '-psm' else arg for arg in sys.argv] - - if '_OCRMYPDF_TEST_INFILE' not in os.environ: - real_tesseract() # test not properly set up - source = os.environ['_OCRMYPDF_TEST_INFILE'] # required - args = parser.parse_args() - - cache_disabled = os.environ.get('_OCRMYPDF_CACHE_DISABLED', False) - - if args.imagename == 'stdin': - real_tesseract() +def get_cache_folder(source_pdf, run_args, parsed_args): def slugs(): yield '' # so we don't start with a '-' which makes rm difficult - for arg in sys.argv[1:]: - if arg == args.imagename: - yield Path(args.imagename).name - elif arg == args.outputbase: - yield Path(args.outputbase).name + for arg in run_args[1:]: + if arg == parsed_args.imagename: + yield Path(parsed_args.imagename).name + elif arg == parsed_args.outputbase: + yield Path(parsed_args.outputbase).name elif arg == '-c' or arg.startswith('textonly'): pass else: @@ -122,18 +95,26 @@ def main(): argv_slug = '__'.join(slugs()) argv_slug = argv_slug.replace('/', '___') - cache_folder = Path(CACHE_ROOT) / Path(source).stem / argv_slug + return Path(CACHE_ROOT) / Path(source_pdf).stem / argv_slug + + +def cached_run(options, run_args, **run_kwargs): + run_args = [str(arg) for arg in run_args] # flatten PosixPaths + args = parser.parse_args(run_args[1:]) + + if args.imagename in ('stdin', '-'): + return run(run_args, **run_kwargs) + + source_file = options.input_file + cache_folder = get_cache_folder(source_file, run_args, args) cache_folder.mkdir(parents=True, exist_ok=True) - print(f"Tesseract cache folder {cache_folder} - ", end='', file=sys.stderr) + log.debug("Using Tesseract cache {cache_folder}") - if (cache_folder / 'stderr.bin').exists() and not cache_disabled: - # Cache hit - print("HIT", file=sys.stderr) + if (cache_folder / 'stderr.bin').exists(): + log.debug("Cache HIT") # Replicate stdout/err - sys.stdout.buffer.write((cache_folder / 'stdout.bin').read_bytes()) - sys.stderr.buffer.write((cache_folder / 'stderr.bin').read_bytes()) if args.outputbase != 'stdout': if not args.configfiles: args.configfiles.append('txt') @@ -141,25 +122,28 @@ def main(): # cp cache -> output tessfile = args.outputbase + '.' + configfile shutil.copy(str(cache_folder / configfile) + '.bin', tessfile) - sys.exit(0) + return CompletedProcess( + args=run_args, + returncode=0, + stdout=(cache_folder / 'stdout.bin').read_bytes(), + stderr=(cache_folder / 'stderr.bin').read_bytes(), + ) - # Cache miss - print("MISS", file=sys.stderr) + log.debug("Cache MISS") - # Call tesseract - print(sys.argv[1:]) - p = subprocess.run( - ['tesseract'] + sys.argv[1:], stdout=subprocess.PIPE, stderr=subprocess.PIPE - ) - sys.stdout.buffer.write(p.stdout) - sys.stderr.buffer.write(p.stderr) - - if p.returncode != 0: - # Do not cache errors or crashes - print("Tesseract error", file=sys.stderr) - return p.returncode + cache_kwargs = { + k: v for k, v in run_kwargs.items() if k not in ('stdout', 'stderr') + } + assert cache_kwargs['check'] + try: + p = run(run_args, stdout=PIPE, stderr=PIPE, **cache_kwargs) + except CalledProcessError as e: + log.exception(e) + raise # Pass exception onward + # Update cache (cache_folder / 'stdout.bin').write_bytes(p.stdout) + (cache_folder / 'stderr.bin').write_bytes(p.stderr) if args.outputbase != 'stdout': if not args.configfiles: @@ -172,27 +156,46 @@ def main(): tessfile = args.outputbase + '.' + configfile shutil.copy(tessfile, str(cache_folder / configfile) + '.bin') - (cache_folder / 'stderr.bin').write_bytes(p.stderr) - manifest = {} - manifest['tesseract_version'] = __version__.replace('\n', ' ') + manifest['tesseract_version'] = TesseractOcrEngine.version().replace('\n', ' ') manifest['platform'] = platform.platform() manifest['python'] = platform.python_version() - manifest['argv_slug'] = argv_slug - manifest['sourcefile'] = str(Path(source).relative_to(TESTS_ROOT)) + manifest['argv_slug'] = cache_folder.name + manifest['sourcefile'] = str(Path(source_file).relative_to(TESTS_ROOT)) def clean_sys_argv(): - for arg in sys.argv[1:]: + for arg in run_args[1:]: yield re.sub(r'.*/com.github.ocrmypdf[^/]+[/](.*)', r'$TMPDIR/\1', arg) manifest['args'] = list(clean_sys_argv()) - - # pylint: disable=E1101 with (Path(CACHE_ROOT) / 'manifest.jsonl').open('a') as f: json.dump(manifest, f) f.write('\n') f.flush() + return p -if __name__ == '__main__': - main() +class CacheOcrEngine(TesseractOcrEngine): + @staticmethod + def get_orientation(input_file, options): + with patch('ocrmypdf._exec.tesseract.run', new=partial(cached_run, options)): + return TesseractOcrEngine.get_orientation(input_file, options) + + @staticmethod + def generate_hocr(input_file, output_hocr, output_text, options): + with patch('ocrmypdf._exec.tesseract.run', new=partial(cached_run, options)): + TesseractOcrEngine.generate_hocr( + input_file, output_hocr, output_text, options + ) + + @staticmethod + def generate_pdf(input_file, output_pdf, output_text, options): + with patch('ocrmypdf._exec.tesseract.run', new=partial(cached_run, options)): + TesseractOcrEngine.generate_pdf( + input_file, output_pdf, output_text, options + ) + + +@hookimpl +def get_ocr_engine(): + return CacheOcrEngine() diff --git a/tests/plugins/tesseract_crash.py b/tests/plugins/tesseract_crash.py new file mode 100755 index 00000000..74c3970a --- /dev/null +++ b/tests/plugins/tesseract_crash.py @@ -0,0 +1,64 @@ +# © 2020 James R. Barlow: github.com/jbarlow83 +# +# Permission is hereby granted, free of charge, to any person obtaining a +# copy of this software and associated documentation files (the +# "Software"), to deal in the Software without restriction, including +# without limitation the rights to use, copy, modify, merge, publish, +# distribute, sublicense, and/or sell copies of the Software, and to +# permit persons to whom the Software is furnished to do so, subject to +# the following conditions: +# +# The above copyright notice and this permission notice shall be included +# in all copies or substantial portions of the Software. +# +# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS +# OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF +# MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. +# IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY +# CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, +# TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE +# SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. + +import signal +import sys +from subprocess import CalledProcessError +from unittest.mock import patch + +from ocrmypdf import hookimpl +from ocrmypdf.builtin_plugins.tesseract_ocr import TesseractOcrEngine + + +def raise_crash(*args, **kwargs): + raise CalledProcessError( + 128 + signal.SIGABRT, + 'tesseract', + output=b"", + stderr=b"libc++abi.dylib: terminating with uncaught exception of type " + + b"std::bad_alloc: std::bad_alloc", + ) + + +class CrashOcrEngine(TesseractOcrEngine): + @staticmethod + def get_orientation(input_file, options): + with patch('ocrmypdf._exec.tesseract.run', new=raise_crash): + return TesseractOcrEngine.get_orientation(input_file, options) + + @staticmethod + def generate_hocr(input_file, output_hocr, output_text, options): + with patch('ocrmypdf._exec.tesseract.run', new=raise_crash): + TesseractOcrEngine.generate_hocr( + input_file, output_hocr, output_text, options + ) + + @staticmethod + def generate_pdf(input_file, output_pdf, output_text, options): + with patch('ocrmypdf._exec.tesseract.run', new=raise_crash): + TesseractOcrEngine.generate_pdf( + input_file, output_pdf, output_text, options + ) + + +@hookimpl +def get_ocr_engine(): + return CrashOcrEngine() diff --git a/tests/spoof/tesseract_noop.py b/tests/plugins/tesseract_noop.py old mode 100755 new mode 100644 similarity index 51% rename from tests/spoof/tesseract_noop.py rename to tests/plugins/tesseract_noop.py index 30f97209..26bfe1df --- a/tests/spoof/tesseract_noop.py +++ b/tests/plugins/tesseract_noop.py @@ -1,5 +1,4 @@ -#!/usr/bin/env python3 -# © 2016 James R. Barlow: github.com/jbarlow83 +# © 2020 James R. Barlow: github.com/jbarlow83 # # Permission is hereby granted, free of charge, to any person obtaining a # copy of this software and associated documentation files (the @@ -20,7 +19,7 @@ # TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE # SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. -"""Tesseract no-op spoof +"""Tesseract no-op plugin To quickly run tests where getting OCR output is not necessary. @@ -31,21 +30,10 @@ In 'pdf' mode, convert the image to PDF using another program. In orientation check mode, report the orientation is upright. """ -import sys -from pathlib import Path - -import img2pdf import pikepdf from PIL import Image -VERSION_STRING = '''tesseract 4.0.0 - leptonica-1.77.0 - libjpeg 9c : libpng 1.6.35 : libtiff 4.0.10 : zlib 1.2.11 : libopenjp2 2.3.0 - Found AVX2 - Found AVX - Found SSE -SPOOFED -''' +from ocrmypdf import OcrEngine, OrientationConfidence, hookimpl HOCR_TEMPLATE = ''' ''' -def main(): - if sys.argv[1] == '--version': - print(VERSION_STRING, file=sys.stderr) - sys.exit(0) - elif sys.argv[1] == '--list-langs': - print('List of available languages (1):\neng', file=sys.stderr) - sys.exit(0) - elif sys.argv[-2] == '--print-parameters': - print("Some parameters", file=sys.stderr) - print("textonly_pdf\t1\tSome help text") - sys.exit(0) - elif sys.argv[-2] == 'hocr': - inputf = sys.argv[-4] - output = sys.argv[-3] - with Image.open(inputf) as im, open( - output + '.hocr', 'w', encoding='utf-8' +class NoopOcrEngine(OcrEngine): + @staticmethod + def version(): + return '4.0.0' + + @staticmethod + def creator_tag(options): + tag = '-PDF' if options.pdf_renderer == 'sandwich' else '' + return f"NO-OP {tag} {NoopOcrEngine.version()}" + + def __str__(self): + return f"NO-OP {NoopOcrEngine.version()}" + + @staticmethod + def languages(options): + return {'eng'} + + @staticmethod + def get_orientation(input_file, options): + return OrientationConfidence(angle=0, confidence=0.0) + + @staticmethod + def generate_hocr(input_file, output_hocr, output_text, options): + with Image.open(input_file) as im, open( + output_hocr, 'w', encoding='utf-8' ) as f: w, h = im.size f.write(HOCR_TEMPLATE.format(str(w), str(h))) - with open(output + '.txt', 'w') as f: + with open(output_text, 'w') as f: f.write('') - elif sys.argv[-2] == 'pdf': - if 'textonly_pdf=1' in sys.argv: - inputf = sys.argv[-4] - output = sys.argv[-3] - with Image.open(inputf) as im: - dpi = im.info['dpi'] - pagesize = im.size[0] / dpi[0], im.size[1] / dpi[1] - ptsize = pagesize[0] * 72, pagesize[1] * 72 - pdf_out = pikepdf.new() - pdf_out.add_blank_page(page_size=ptsize) - pdf_out.save(Path(output).with_suffix('.pdf'), static_id=True) - Path(output).with_suffix('.txt').write_text('') - else: - inputf = sys.argv[-4] - output = sys.argv[-3] - pdf_bytes = img2pdf.convert([inputf], dpi=300) - with open(output + '.pdf', 'wb') as f: - f.write(pdf_bytes) - with open(output + '.txt', 'w') as f: - f.write('') - elif sys.argv[-1] == 'stdout': - inputf = sys.argv[-2] - print( - """Orientation: 0 -Orientation in degrees: 0 -Orientation confidence: 100.00 -Script: 1 -Script confidence: 100.00""", - file=sys.stderr, - ) - else: - print("Spoof doesn't understand arguments", file=sys.stderr) - print(sys.argv, file=sys.stderr) - sys.exit(1) - - sys.exit(0) + @staticmethod + def generate_pdf(input_file, output_pdf, output_text, options): + with Image.open(input_file) as im: + dpi = im.info['dpi'] + pagesize = im.size[0] / dpi[0], im.size[1] / dpi[1] + ptsize = pagesize[0] * 72, pagesize[1] * 72 + pdf = pikepdf.new() + pdf.add_blank_page(page_size=ptsize) + pdf.save(output_pdf, static_id=True) + output_text.write_text('') -if __name__ == '__main__': - main() +@hookimpl +def get_ocr_engine(): + return NoopOcrEngine() diff --git a/tests/spoof/tesseract_big_image_error.py b/tests/spoof/tesseract_big_image_error.py deleted file mode 100755 index 8b710bee..00000000 --- a/tests/spoof/tesseract_big_image_error.py +++ /dev/null @@ -1,77 +0,0 @@ -#!/usr/bin/env python3 -# © 2016 James R. Barlow: github.com/jbarlow83 -# -# Permission is hereby granted, free of charge, to any person obtaining a -# copy of this software and associated documentation files (the -# "Software"), to deal in the Software without restriction, including -# without limitation the rights to use, copy, modify, merge, publish, -# distribute, sublicense, and/or sell copies of the Software, and to -# permit persons to whom the Software is furnished to do so, subject to -# the following conditions: -# -# The above copyright notice and this permission notice shall be included -# in all copies or substantial portions of the Software. -# -# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS -# OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF -# MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. -# IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY -# CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, -# TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE -# SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. - - -import sys - -VERSION_STRING = '''tesseract 4.0.0 - leptonica-1.77.0 - libjpeg 9c : libpng 1.6.35 : libtiff 4.0.10 : zlib 1.2.11 : libopenjp2 2.3.0 - Found AVX2 - Found AVX - Found SSE -SPOOFED: return error claiming image too big -''' - -"""Simulates an error of Tesseract failing on attempts to process large images - -""" - - -def main(): - if sys.argv[1] == '--version': - print(VERSION_STRING, file=sys.stderr) - sys.exit(0) - elif sys.argv[1] == '--list-langs': - print('List of available languages (1):\neng\n', file=sys.stderr) - sys.exit(0) - elif sys.argv[-2] == '--print-parameters': - print('A parameter list would go here\ntextonly_pdf 0\n', file=sys.stderr) - sys.exit(0) - elif sys.argv[-2] == 'hocr': - print( - "Image too large: (33830, 14959)\n" "Error during processing.", - file=sys.stderr, - ) - sys.exit(1) - elif sys.argv[-2] == 'pdf': - print( - "Image too large: (33830, 14959)\n" "Error during processing.", - file=sys.stderr, - ) - sys.exit(1) - elif sys.argv[-1] == 'stdout': - print( - "Image too large: (33830, 14959)\n" "Error during processing.", - file=sys.stderr, - ) - sys.exit(1) - else: - print("Spoof doesn't understand arguments", file=sys.stderr) - print(sys.argv, file=sys.stderr) - sys.exit(1) - - sys.exit(0) - - -if __name__ == '__main__': - main() diff --git a/tests/spoof/tesseract_crash.py b/tests/spoof/tesseract_crash.py deleted file mode 100755 index 03c7dbde..00000000 --- a/tests/spoof/tesseract_crash.py +++ /dev/null @@ -1,78 +0,0 @@ -#!/usr/bin/env python3 -# © 2016 James R. Barlow: github.com/jbarlow83 -# -# Permission is hereby granted, free of charge, to any person obtaining a -# copy of this software and associated documentation files (the -# "Software"), to deal in the Software without restriction, including -# without limitation the rights to use, copy, modify, merge, publish, -# distribute, sublicense, and/or sell copies of the Software, and to -# permit persons to whom the Software is furnished to do so, subject to -# the following conditions: -# -# The above copyright notice and this permission notice shall be included -# in all copies or substantial portions of the Software. -# -# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS -# OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF -# MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. -# IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY -# CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, -# TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE -# SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. - -import signal -import sys - -VERSION_STRING = '''tesseract 4.0.0 - leptonica-1.77.0 - libjpeg 9c : libpng 1.6.35 : libtiff 4.0.10 : zlib 1.2.11 : libopenjp2 2.3.0 - Found AVX2 - Found AVX - Found SSE -SPOOFED: CRASH ON OCR or --psm 0 -''' - -"""Simulates a Tesseract crash when asked to run OCR - -It isn't strictly necessary to crash the process and that has unwanted -side effects like triggering core dumps or error reporting, logging and such. -It's enough to dump some text to stderr and return an error code. - -Follows the POSIX(?) convention of returning 128 + signal number. - -""" - - -def main(): - if sys.argv[1] == '--version': - print(VERSION_STRING, file=sys.stderr) - sys.exit(0) - elif sys.argv[1] == '--list-langs': - print('List of available languages (1):\neng', file=sys.stderr) - sys.exit(0) - elif sys.argv[-2] == '--print-parameters': - print('A parameter list would go here\ntextonly_pdf 0\n', file=sys.stderr) - sys.exit(0) - elif sys.argv[-2] == 'hocr': - print("KABOOM! Tesseract failed for some reason", file=sys.stderr) - sys.exit(128 + signal.SIGSEGV) - elif sys.argv[-2] == 'pdf': - print("KABOOM! Tesseract failed for some reason", file=sys.stderr) - sys.exit(128 + signal.SIGSEGV) - elif sys.argv[-1] == 'stdout': - print( - "libc++abi.dylib: terminating with uncaught exception of type " - "std::bad_alloc: std::bad_alloc", - file=sys.stderr, - ) - sys.exit(128 + signal.SIGABRT) - else: - print("Spoof doesn't understand arguments", file=sys.stderr) - print(sys.argv, file=sys.stderr) - sys.exit(1) - - sys.exit(0) - - -if __name__ == '__main__': - main() diff --git a/tests/spoof/unpaper_oldversion.py b/tests/spoof/unpaper_oldversion.py deleted file mode 100755 index ff2e27ea..00000000 --- a/tests/spoof/unpaper_oldversion.py +++ /dev/null @@ -1,37 +0,0 @@ -#!/usr/bin/env python3 -# © 2016 James R. Barlow: github.com/jbarlow83 -# -# Permission is hereby granted, free of charge, to any person obtaining a -# copy of this software and associated documentation files (the -# "Software"), to deal in the Software without restriction, including -# without limitation the rights to use, copy, modify, merge, publish, -# distribute, sublicense, and/or sell copies of the Software, and to -# permit persons to whom the Software is furnished to do so, subject to -# the following conditions: -# -# The above copyright notice and this permission notice shall be included -# in all copies or substantial portions of the Software. -# -# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS -# OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF -# MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. -# IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY -# CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, -# TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE -# SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. - - -import sys - - -def main(): - if sys.argv[1] == '--version': - print('0.5') - sys.exit(0) - - print("Only supports --version") - sys.exit(1) - - -if __name__ == '__main__': - main() diff --git a/tests/test_acroform.py b/tests/test_acroform.py index 44de63da..4ab52406 100644 --- a/tests/test_acroform.py +++ b/tests/test_acroform.py @@ -35,8 +35,8 @@ def test_acroform_and_redo(acroform, caplog, no_outpdf): assert '--redo-ocr is not currently possible' in caplog.text -def test_acroform_message(acroform, caplog, spoof_tesseract_noop, outpdf): +def test_acroform_message(acroform, caplog, outpdf): caplog.set_level(logging.INFO) - check_ocrmypdf(acroform, outpdf, env=spoof_tesseract_noop) + check_ocrmypdf(acroform, outpdf, '--plugin', 'tests/plugins/tesseract_noop.py') assert 'fillable form' in caplog.text assert '--force-ocr' in caplog.text diff --git a/tests/test_qpdf.py b/tests/test_check_pdf.py similarity index 82% rename from tests/test_qpdf.py rename to tests/test_check_pdf.py index 0e925249..b3516e90 100644 --- a/tests/test_qpdf.py +++ b/tests/test_check_pdf.py @@ -17,9 +17,9 @@ import pytest -import ocrmypdf.exec.qpdf as qpdf +from ocrmypdf.helpers import check_pdf -def test_qpdf_error(resources): - assert qpdf.check(resources / 'blank.pdf') - assert not qpdf.check(__file__) +def test_pdf_error(resources): + assert check_pdf(resources / 'blank.pdf') + assert not check_pdf(__file__) diff --git a/tests/test_ghostscript.py b/tests/test_ghostscript.py index 3aed104a..a2dd90d7 100644 --- a/tests/test_ghostscript.py +++ b/tests/test_ghostscript.py @@ -22,41 +22,13 @@ import pikepdf import pytest from PIL import Image +from ocrmypdf._exec.ghostscript import rasterize_pdf from ocrmypdf.exceptions import ExitCode -from ocrmypdf.exec.ghostscript import rasterize_pdf +from ocrmypdf.helpers import Resolution check_ocrmypdf = pytest.helpers.check_ocrmypdf run_ocrmypdf = pytest.helpers.run_ocrmypdf run_ocrmypdf_api = pytest.helpers.run_ocrmypdf_api -spoof = pytest.helpers.spoof - - -@pytest.fixture -def spoof_no_tess_gs_render_fail(tmp_path_factory): - return spoof( - tmp_path_factory, tesseract='tesseract_noop.py', gs='gs_render_failure.py' - ) - - -@pytest.fixture -def spoof_no_tess_gs_raster_fail(tmp_path_factory): - return spoof( - tmp_path_factory, tesseract='tesseract_noop.py', gs='gs_raster_failure.py' - ) - - -@pytest.fixture -def spoof_no_tess_no_pdfa(tmp_path_factory): - return spoof( - tmp_path_factory, tesseract='tesseract_noop.py', gs='gs_pdfa_failure.py' - ) - - -@pytest.fixture -def spoof_no_tess_pdfa_warning(tmp_path_factory): - return spoof( - tmp_path_factory, tesseract='tesseract_noop.py', gs='gs_feature_elision.py' - ) @pytest.fixture @@ -71,16 +43,15 @@ def test_rasterize_size(francais, outdir, caplog): assert pdf.pages[0].MediaBox[0] == pdf.pages[0].MediaBox[1] == 0 page_size = (page_size_pts[0] / Decimal(72), page_size_pts[1] / Decimal(72)) target_size = Decimal('50.0'), Decimal('30.0') - forced_dpi = 42.0, 4242.0 + forced_dpi = Resolution(42.0, 4242.0) - log = logging.getLogger() rasterize_pdf( path, outdir / 'out.png', - target_size[0] / page_size[0], - target_size[1] / page_size[1], raster_device='pngmono', - log=log, + raster_dpi=Resolution( + target_size[0] / page_size[0], target_size[1] / page_size[1] + ), page_dpi=forced_dpi, ) @@ -95,17 +66,16 @@ def test_rasterize_rotated(francais, outdir, caplog): assert pdf.pages[0].MediaBox[0] == pdf.pages[0].MediaBox[1] == 0 page_size = (page_size_pts[0] / Decimal(72), page_size_pts[1] / Decimal(72)) target_size = Decimal('50.0'), Decimal('30.0') - forced_dpi = 42.0, 4242.0 + forced_dpi = Resolution(42.0, 4242.0) - log = logging.getLogger() caplog.set_level(logging.DEBUG) rasterize_pdf( path, outdir / 'out.png', - target_size[0] / page_size[0], - target_size[1] / page_size[1], raster_device='pngmono', - log=log, + raster_dpi=Resolution( + target_size[0] / page_size[0], target_size[1] / page_size[1] + ), page_dpi=forced_dpi, rotation=90, ) @@ -115,30 +85,52 @@ def test_rasterize_rotated(francais, outdir, caplog): assert im.info['dpi'] == (forced_dpi[1], forced_dpi[0]) -def test_gs_render_failure(spoof_no_tess_gs_render_fail, resources, outpdf): +def test_gs_render_failure(resources, outpdf): p, out, err = run_ocrmypdf( - resources / 'blank.pdf', outpdf, env=spoof_no_tess_gs_render_fail + resources / 'blank.pdf', + outpdf, + '--plugin', + 'tests/plugins/tesseract_noop.py', + '--plugin', + 'tests/plugins/gs_render_failure.py', ) assert 'Casper is not a friendly ghost' in err assert p.returncode == ExitCode.child_process_error -def test_gs_raster_failure(spoof_no_tess_gs_raster_fail, resources, outpdf): +def test_gs_raster_failure(resources, outpdf): p, out, err = run_ocrmypdf( - resources / 'francais.pdf', outpdf, env=spoof_no_tess_gs_raster_fail + resources / 'francais.pdf', + outpdf, + '--plugin', + 'tests/plugins/tesseract_noop.py', + '--plugin', + 'tests/plugins/gs_raster_failure.py', ) assert 'Ghost story archive not found' in err assert p.returncode == ExitCode.child_process_error -def test_ghostscript_pdfa_failure(spoof_no_tess_no_pdfa, resources, outpdf): +def test_ghostscript_pdfa_failure(resources, outpdf): p, out, err = run_ocrmypdf( - resources / 'francais.pdf', outpdf, env=spoof_no_tess_no_pdfa + resources / 'francais.pdf', + outpdf, + '--plugin', + 'tests/plugins/tesseract_noop.py', + '--plugin', + 'tests/plugins/gs_pdfa_failure.py', ) assert ( p.returncode == ExitCode.pdfa_conversion_failed ), "Unexpected return when PDF/A fails" -def test_ghostscript_feature_elision(spoof_no_tess_pdfa_warning, resources, outpdf): - check_ocrmypdf(resources / 'francais.pdf', outpdf, env=spoof_no_tess_pdfa_warning) +def test_ghostscript_feature_elision(resources, outpdf): + check_ocrmypdf( + resources / 'francais.pdf', + outpdf, + '--plugin', + 'tests/plugins/tesseract_noop.py', + '--plugin', + 'tests/plugins/gs_feature_elision.py', + ) diff --git a/tests/test_graft.py b/tests/test_graft.py index 65cba7c8..1329e869 100644 --- a/tests/test_graft.py +++ b/tests/test_graft.py @@ -15,7 +15,6 @@ # You should have received a copy of the GNU General Public License # along with OCRmyPDF. If not, see . -import os from unittest.mock import patch import pikepdf diff --git a/tests/test_helpers.py b/tests/test_helpers.py index c211477f..47139bec 100644 --- a/tests/test_helpers.py +++ b/tests/test_helpers.py @@ -24,6 +24,7 @@ from unittest.mock import MagicMock import pytest import ocrmypdf.helpers as helpers +from ocrmypdf.subprocess import shim_paths_with_program_files class TestSafeSymlink: @@ -106,7 +107,6 @@ def test_shim_paths(tmp_path): (progfiles / 'gs' / '9.52' / 'bin').mkdir(parents=True) syspath = tmp_path / 'bin' env = {'PROGRAMFILES': str(progfiles), 'PATH': str(syspath)} - from ocrmypdf.exec import shim_paths_with_program_files result_str = shim_paths_with_program_files(env=env) results = result_str.split(os.pathsep) diff --git a/tests/test_hocrtransform.py b/tests/test_hocrtransform.py index 19e4684d..1a1f817a 100644 --- a/tests/test_hocrtransform.py +++ b/tests/test_hocrtransform.py @@ -21,8 +21,8 @@ import pytest from PIL import Image from ocrmypdf import hocrtransform -from ocrmypdf.exec import qpdf -from ocrmypdf.exec.tesseract import HOCR_TEMPLATE +from ocrmypdf._exec.tesseract import HOCR_TEMPLATE +from ocrmypdf.helpers import check_pdf # pylint: disable=redefined-outer-name @@ -41,6 +41,6 @@ def test_mono_image(blank_hocr, outdir): im.save(outdir / 'mono.tif', format='TIFF') hocr = hocrtransform.HocrTransform(str(blank_hocr), 300) - hocr.to_pdf(str(outdir / 'mono.pdf'), imageFileName=str(outdir / 'mono.tif')) + hocr.to_pdf(str(outdir / 'mono.pdf'), image_filename=str(outdir / 'mono.tif')) - qpdf.check(str(outdir / 'mono.pdf')) + check_pdf(str(outdir / 'mono.pdf')) diff --git a/tests/test_image_input.py b/tests/test_image_input.py index ceb94cbe..c3636952 100644 --- a/tests/test_image_input.py +++ b/tests/test_image_input.py @@ -33,9 +33,14 @@ def baiona(resources): return Image.open(resources / 'baiona_gray.png') -def test_image_to_pdf(spoof_tesseract_noop, resources, outpdf): +def test_image_to_pdf(resources, outpdf): check_ocrmypdf( - resources / 'crom.png', outpdf, '--image-dpi', '200', env=spoof_tesseract_noop + resources / 'crom.png', + outpdf, + '--image-dpi', + '200', + '--plugin', + 'tests/plugins/tesseract_noop.py', ) @@ -77,7 +82,7 @@ def test_img2pdf_fails(resources, no_outpdf): assert rc == ocrmypdf.ExitCode.input_file -def test_jpeg_in_jpeg_out(resources, outpdf, spoof_tesseract_noop): +def test_jpeg_in_jpeg_out(resources, outpdf): check_ocrmypdf( resources / 'congress.jpg', outpdf, @@ -86,7 +91,8 @@ def test_jpeg_in_jpeg_out(resources, outpdf, spoof_tesseract_noop): '--output-type', 'pdf', # specifically check pdf because Ghostscript may convert to JPEG '--remove-background', - env=spoof_tesseract_noop, + '--plugin', + 'tests/plugins/tesseract_noop.py', ) with pikepdf.open(outpdf) as pdf: assert next(pdf.pages[0].images.values()).Filter == pikepdf.Name.DCTDecode diff --git a/tests/test_main.py b/tests/test_main.py index 39abbe98..289cbbcc 100644 --- a/tests/test_main.py +++ b/tests/test_main.py @@ -28,10 +28,11 @@ import pytest from PIL import Image import ocrmypdf +from ocrmypdf._exec import ghostscript, tesseract from ocrmypdf.exceptions import ExitCode, MissingDependencyError -from ocrmypdf.exec import ghostscript, qpdf, tesseract from ocrmypdf.pdfa import file_claims_pdfa from ocrmypdf.pdfinfo import Colorspace, Encoding, PdfInfo +from ocrmypdf.subprocess import get_version # pytest.helpers is dynamic # pylint: disable=no-member,redefined-outer-name @@ -39,28 +40,19 @@ from ocrmypdf.pdfinfo import Colorspace, Encoding, PdfInfo check_ocrmypdf = pytest.helpers.check_ocrmypdf run_ocrmypdf = pytest.helpers.run_ocrmypdf run_ocrmypdf_api = pytest.helpers.run_ocrmypdf_api -spoof = pytest.helpers.spoof RENDERERS = ['hocr', 'sandwich'] -@pytest.fixture -def spoof_tesseract_crash(tmp_path_factory): - return spoof(tmp_path_factory, tesseract='tesseract_crash.py') - - -@pytest.fixture -def spoof_tesseract_big_image_error(tmp_path_factory): - return spoof(tmp_path_factory, tesseract='tesseract_big_image_error.py') - - -def test_quick(spoof_tesseract_cache, resources, outpdf): - check_ocrmypdf(resources / 'ccitt.pdf', outpdf, env=spoof_tesseract_cache) +def test_quick(resources, outpdf): + check_ocrmypdf( + resources / 'ccitt.pdf', outpdf, '--plugin', 'tests/plugins/tesseract_cache.py' + ) @pytest.mark.parametrize('renderer', RENDERERS) -def test_oversample(spoof_tesseract_cache, renderer, resources, outpdf): +def test_oversample(renderer, resources, outpdf): oversampled_pdf = check_ocrmypdf( resources / 'skew.pdf', outpdf, @@ -69,13 +61,14 @@ def test_oversample(spoof_tesseract_cache, renderer, resources, outpdf): '-f', '--pdf-renderer', renderer, - env=spoof_tesseract_cache, + '--plugin', + 'tests/plugins/tesseract_cache.py', ) pdfinfo = PdfInfo(oversampled_pdf) - print(pdfinfo[0].xres) - assert abs(pdfinfo[0].xres - 350) < 1 + print(pdfinfo[0].dpi.x) + assert abs(pdfinfo[0].dpi.x - 350) < 1 def test_repeat_ocr(resources, no_outpdf): @@ -83,17 +76,25 @@ def test_repeat_ocr(resources, no_outpdf): assert result == ExitCode.already_done_ocr -def test_force_ocr(spoof_tesseract_cache, resources, outpdf): +def test_force_ocr(resources, outpdf): out = check_ocrmypdf( - resources / 'graph_ocred.pdf', outpdf, '-f', env=spoof_tesseract_cache + resources / 'graph_ocred.pdf', + outpdf, + '-f', + '--plugin', + 'tests/plugins/tesseract_cache.py', ) pdfinfo = PdfInfo(out) assert pdfinfo[0].has_text -def test_skip_ocr(spoof_tesseract_cache, resources, outpdf): +def test_skip_ocr(resources, outpdf): out = check_ocrmypdf( - resources / 'graph_ocred.pdf', outpdf, '-s', env=spoof_tesseract_cache + resources / 'graph_ocred.pdf', + outpdf, + '-s', + '--plugin', + 'tests/plugins/tesseract_cache.py', ) pdfinfo = PdfInfo(out) assert pdfinfo[0].has_text @@ -101,17 +102,17 @@ def test_skip_ocr(spoof_tesseract_cache, resources, outpdf): def test_redo_ocr(resources, outpdf): in_ = resources / 'graph_ocred.pdf' - before = PdfInfo(in_, detailed_page_analysis=True) + before = PdfInfo(in_) out = outpdf out = check_ocrmypdf(in_, out, '--redo-ocr') - after = PdfInfo(out, detailed_page_analysis=True) + after = PdfInfo(out) assert before[0].has_text and after[0].has_text assert ( before[0].get_textareas() != after[0].get_textareas() ), "Expected text to be different after re-OCR" -def test_argsfile(spoof_tesseract_noop, resources, outdir): +def test_argsfile(resources, outdir): path_argsfile = outdir / 'test_argsfile.txt' with open(str(path_argsfile), 'w') as argsfile: print( @@ -119,15 +120,14 @@ def test_argsfile(spoof_tesseract_noop, resources, outdir): 'ArgsFile Test', '--author', 'Test Cases', + '--plugin', + 'tests/plugins/tesseract_noop.py', sep='\n', end='\n', file=argsfile, ) check_ocrmypdf( - resources / 'graph.pdf', - path_argsfile, - '@' + str(outdir / 'test_argsfile.txt'), - env=spoof_tesseract_noop, + resources / 'graph.pdf', path_argsfile, '@' + str(outdir / 'test_argsfile.txt') ) @@ -145,9 +145,14 @@ def test_ocr_timeout(renderer, resources, outpdf): assert not pdfinfo[0].has_text -def test_skip_big(spoof_tesseract_cache, resources, outpdf): +def test_skip_big(resources, outpdf): out = check_ocrmypdf( - resources / 'jbig2.pdf', outpdf, '--skip-big', '1', env=spoof_tesseract_cache + resources / 'jbig2.pdf', + outpdf, + '--skip-big', + '1', + '--plugin', + 'tests/plugins/tesseract_cache.py', ) pdfinfo = PdfInfo(out) assert not pdfinfo[0].has_text @@ -155,9 +160,7 @@ def test_skip_big(spoof_tesseract_cache, resources, outpdf): @pytest.mark.parametrize('renderer', RENDERERS) @pytest.mark.parametrize('output_type', ['pdf', 'pdfa']) -def test_maximum_options( - spoof_tesseract_cache, renderer, output_type, resources, outpdf -): +def test_maximum_options(renderer, output_type, resources, outpdf): check_ocrmypdf( resources / 'multipage.pdf', outpdf, @@ -178,18 +181,15 @@ def test_maximum_options( renderer, '--output-type', output_type, - env=spoof_tesseract_cache, + '--plugin', + 'tests/plugins/tesseract_cache.py', ) -def test_tesseract_missing_tessdata(resources, no_outpdf, tmpdir): - env = os.environ.copy() - env['TESSDATA_PREFIX'] = os.fspath(tmpdir) - - returncode = run_ocrmypdf_api( - resources / 'graph.pdf', no_outpdf, '-v', '1', '--skip-text', env=env - ) - assert returncode == ExitCode.missing_dependency +def test_tesseract_missing_tessdata(monkeypatch, resources, no_outpdf, tmpdir): + monkeypatch.setenv("TESSDATA_PREFIX", os.fspath(tmpdir)) + with pytest.raises(MissingDependencyError): + run_ocrmypdf_api(resources / 'graph.pdf', no_outpdf, '-v', '1', '--skip-text') def test_invalid_input_pdf(resources, no_outpdf): @@ -202,22 +202,26 @@ def test_blank_input_pdf(resources, outpdf): assert result == ExitCode.ok -def test_force_ocr_on_pdf_with_no_images(spoof_tesseract_crash, resources, no_outpdf): +def test_force_ocr_on_pdf_with_no_images(resources, no_outpdf): # As a correctness test, make sure that --force-ocr on a PDF with no # content still triggers tesseract. If tesseract crashes, then it was # called. p, _, _ = run_ocrmypdf( - resources / 'blank.pdf', no_outpdf, '--force-ocr', env=spoof_tesseract_crash + resources / 'blank.pdf', + no_outpdf, + '--force-ocr', + '--plugin', + 'tests/plugins/tesseract_crash.py', ) assert p.returncode == ExitCode.child_process_error - assert not os.path.exists(no_outpdf) + assert not no_outpdf.exists() @pytest.mark.skipif( pytest.helpers.is_macos() and pytest.helpers.running_in_travis(), reason="takes too long to install language packs in Travis macOS homebrew", ) -def test_german(spoof_tesseract_cache, resources, outdir): +def test_german(resources, outdir): # Produce a sidecar too - implicit test that system locale is set up # properly. It is fine that we are testing -l deu on a French file because # we are exercising the functionality not going for accuracy. @@ -230,10 +234,11 @@ def test_german(spoof_tesseract_cache, resources, outdir): 'deu', # more commonly installed '--sidecar', sidecar, - env=spoof_tesseract_cache, + '--plugin', + 'tests/plugins/tesseract_cache.py', ) except MissingDependencyError: - if 'deu' not in tesseract.languages(): + if 'deu' not in tesseract.get_languages(): pytest.xfail(reason="tesseract-deu language pack not installed") raise @@ -243,23 +248,27 @@ def test_klingon(resources, outpdf): assert p.returncode == ExitCode.missing_dependency -def test_missing_docinfo(spoof_tesseract_noop, resources, outpdf): +def test_missing_docinfo(resources, outpdf): result = run_ocrmypdf_api( resources / 'missing_docinfo.pdf', outpdf, '-l', 'eng', '--skip-text', - env=spoof_tesseract_noop, + '--plugin', + 'tests/plugins/tesseract_noop.py', ) assert result == ExitCode.ok -def test_uppercase_extension(spoof_tesseract_noop, resources, outdir): +def test_uppercase_extension(resources, outdir): shutil.copy(str(resources / "skew.pdf"), str(outdir / "UPPERCASE.PDF")) check_ocrmypdf( - outdir / "UPPERCASE.PDF", outdir / "UPPERCASE_OUT.PDF", env=spoof_tesseract_noop + outdir / "UPPERCASE.PDF", + outdir / "UPPERCASE_OUT.PDF", + '--plugin', + 'tests/plugins/tesseract_noop.py', ) @@ -295,7 +304,7 @@ def test_encrypted(resources, caplog, no_outpdf): @pytest.mark.parametrize('renderer', RENDERERS) -def test_pagesegmode(renderer, spoof_tesseract_cache, resources, outpdf): +def test_pagesegmode(renderer, resources, outpdf): check_ocrmypdf( resources / 'skew.pdf', outpdf, @@ -305,12 +314,13 @@ def test_pagesegmode(renderer, spoof_tesseract_cache, resources, outpdf): '1', '--pdf-renderer', renderer, - env=spoof_tesseract_cache, + '--plugin', + 'tests/plugins/tesseract_cache.py', ) @pytest.mark.parametrize('renderer', RENDERERS) -def test_tesseract_crash(renderer, spoof_tesseract_crash, resources, no_outpdf, caplog): +def test_tesseract_crash(renderer, resources, no_outpdf): p, _, err = run_ocrmypdf( resources / 'ccitt.pdf', no_outpdf, @@ -318,29 +328,32 @@ def test_tesseract_crash(renderer, spoof_tesseract_crash, resources, no_outpdf, '1', '--pdf-renderer', renderer, - env=spoof_tesseract_crash, + '--plugin', + 'tests/plugins/tesseract_crash.py', ) assert p.returncode == ExitCode.child_process_error - assert not os.path.exists(no_outpdf) + assert not no_outpdf.exists() assert "SubprocessOutputError" in err -def test_tesseract_crash_autorotate(spoof_tesseract_crash, resources, no_outpdf): +def test_tesseract_crash_autorotate(resources, no_outpdf): p, out, err = run_ocrmypdf( - resources / 'ccitt.pdf', no_outpdf, '-r', env=spoof_tesseract_crash + resources / 'ccitt.pdf', + no_outpdf, + '-r', + '--plugin', + 'tests/plugins/tesseract_crash.py', ) assert p.returncode == ExitCode.child_process_error - assert not os.path.exists(no_outpdf) - assert "ERROR" in err + assert not no_outpdf.exists() + assert "uncaught exception" in err print(out) print(err) @pytest.mark.parametrize('renderer', RENDERERS) @pytest.mark.slow -def test_tesseract_image_too_big( - renderer, spoof_tesseract_big_image_error, resources, outpdf -): +def test_tesseract_image_too_big(renderer, resources, outpdf): check_ocrmypdf( resources / 'hugemono.pdf', outpdf, @@ -349,18 +362,22 @@ def test_tesseract_image_too_big( renderer, '--max-image-mpixels', '0', - env=spoof_tesseract_big_image_error, + '--plugin', + 'tests/plugins/tesseract_big_image_error.py', ) -def test_algo4(resources, spoof_tesseract_noop, outpdf): +def test_algo4(resources, outpdf): p, _, _ = run_ocrmypdf( - resources / 'encrypted_algo4.pdf', outpdf, env=spoof_tesseract_noop + resources / 'encrypted_algo4.pdf', + outpdf, + '--plugin', + 'tests/plugins/tesseract_noop.py', ) assert p.returncode == ExitCode.encrypted_pdf -def test_jbig2_passthrough(spoof_tesseract_cache, resources, outpdf): +def test_jbig2_passthrough(resources, outpdf): out = check_ocrmypdf( resources / 'jbig2.pdf', outpdf, @@ -368,49 +385,64 @@ def test_jbig2_passthrough(spoof_tesseract_cache, resources, outpdf): 'pdf', '--pdf-renderer', 'hocr', - env=spoof_tesseract_cache, + '--plugin', + 'tests/plugins/tesseract_cache.py', ) out_pageinfo = PdfInfo(out) assert out_pageinfo[0].images[0].enc == Encoding.jbig2 -def test_masks(spoof_tesseract_noop, resources, outpdf): +def test_masks(resources, outpdf): assert ( ocrmypdf.ocr( - resources / 'masks.pdf', outpdf, tesseract_env=spoof_tesseract_noop + resources / 'masks.pdf', outpdf, plugins=['tests/plugins/tesseract_noop.py'] ) == ExitCode.ok ) -def test_linearized_pdf_and_indirect_object(spoof_tesseract_noop, resources, outpdf): - check_ocrmypdf(resources / 'epson.pdf', outpdf, env=spoof_tesseract_noop) - - -def test_very_high_dpi(spoof_tesseract_cache, resources, outpdf): - "Checks for a Decimal quantize error with high DPI, etc" - check_ocrmypdf(resources / '2400dpi.pdf', outpdf, env=spoof_tesseract_cache) - pdfinfo = PdfInfo(outpdf) - - image = pdfinfo[0].images[0] - assert isclose(image.xres, image.yres) - assert isclose(image.xres, 2400) - - -def test_overlay(spoof_tesseract_noop, resources, outpdf): +def test_linearized_pdf_and_indirect_object(resources, outpdf): check_ocrmypdf( - resources / 'overlay.pdf', outpdf, '--skip-text', env=spoof_tesseract_noop + resources / 'epson.pdf', outpdf, '--plugin', 'tests/plugins/tesseract_noop.py' ) -def test_destination_not_writable(spoof_tesseract_noop, resources, outdir): +def test_very_high_dpi(resources, outpdf): + "Checks for a Decimal quantize error with high DPI, etc" + check_ocrmypdf( + resources / '2400dpi.pdf', + outpdf, + '--plugin', + 'tests/plugins/tesseract_cache.py', + ) + pdfinfo = PdfInfo(outpdf) + + image = pdfinfo[0].images[0] + assert isclose(image.dpi.x, image.dpi.y) + assert isclose(image.dpi.x, 2400) + + +def test_overlay(resources, outpdf): + check_ocrmypdf( + resources / 'overlay.pdf', + outpdf, + '--skip-text', + '--plugin', + 'tests/plugins/tesseract_noop.py', + ) + + +def test_destination_not_writable(resources, outdir): if os.name != 'nt' and (os.getuid() == 0 or os.geteuid() == 0): pytest.xfail(reason="root can write to anything") protected_file = outdir / 'protected.pdf' protected_file.touch() protected_file.chmod(0o400) # Read-only - p, out, err = run_ocrmypdf( - resources / 'jbig2.pdf', protected_file, env=spoof_tesseract_noop + p, _out, _err = run_ocrmypdf( + resources / 'jbig2.pdf', + protected_file, + '--plugin', + 'tests/plugins/tesseract_noop.py', ) assert p.returncode == ExitCode.file_access_error, "Expected error" @@ -447,7 +479,7 @@ THIS FILE IS INVALID ''' ) - p, out, err = run_ocrmypdf( + p, _out, err = run_ocrmypdf( resources / 'ccitt.pdf', outdir / 'out.pdf', '--pdf-renderer', @@ -483,9 +515,13 @@ def test_user_words_ocr(resources, outdir): ) -def test_form_xobject(spoof_tesseract_noop, resources, outpdf): +def test_form_xobject(resources, outpdf): check_ocrmypdf( - resources / 'formxobject.pdf', outpdf, '--force-ocr', env=spoof_tesseract_noop + resources / 'formxobject.pdf', + outpdf, + '--force-ocr', + '--plugin', + 'tests/plugins/tesseract_noop.py', ) @@ -517,33 +553,36 @@ def test_pagesize_consistency(renderer, resources, outpdf): assert isclose(before_dims[1], after_dims[1], rel_tol=1e-4) -def test_skip_big_with_no_images(spoof_tesseract_noop, resources, outpdf): +def test_skip_big_with_no_images(resources, outpdf): check_ocrmypdf( resources / 'blank.pdf', outpdf, '--skip-big', '5', '--force-ocr', - env=spoof_tesseract_noop, + '--plugin', + 'tests/plugins/tesseract_noop.py', ) @pytest.mark.skipif( - '8.0.0' <= qpdf.version() <= '8.0.1', - reason="qpdf regression on pages with no contents", + '8.0.0' <= pikepdf.__libqpdf_version__ <= '8.0.1', + reason="libqpdf regression on pages with no contents", ) -def test_no_contents(spoof_tesseract_noop, resources, outpdf): +def test_no_contents(resources, outpdf): check_ocrmypdf( - resources / 'no_contents.pdf', outpdf, '--force-ocr', env=spoof_tesseract_noop + resources / 'no_contents.pdf', + outpdf, + '--force-ocr', + '--plugin', + 'tests/plugins/tesseract_noop.py', ) @pytest.mark.parametrize( 'image', ['baiona.png', 'baiona_gray.png', 'baiona_alpha.png', 'congress.jpg'] ) -def test_compression_preserved( - spoof_tesseract_noop, ocrmypdf_exec, resources, image, outpdf -): +def test_compression_preserved(ocrmypdf_exec, resources, image, outpdf): input_file = str(resources / image) output_file = str(outpdf) @@ -557,6 +596,8 @@ def test_compression_preserved( '150', '--output-type', 'pdf', + '--plugin', + 'tests/plugins/tesseract_noop.py', '-', output_file, ] @@ -566,7 +607,7 @@ def test_compression_preserved( stderr=PIPE, stdin=input_stream, universal_newlines=True, - env=spoof_tesseract_noop, + check=False, ) if im.mode in ('RGBA', 'LA'): @@ -599,9 +640,7 @@ def test_compression_preserved( ('congress.jpg', 'lossless'), ], ) -def test_compression_changed( - spoof_tesseract_noop, ocrmypdf_exec, resources, image, compression, outpdf -): +def test_compression_changed(ocrmypdf_exec, resources, image, compression, outpdf): input_file = str(resources / image) output_file = str(outpdf) @@ -618,6 +657,8 @@ def test_compression_changed( '0', '--pdfa-image-compression', compression, + '--plugin', + 'tests/plugins/tesseract_noop.py', '-', output_file, ] @@ -627,7 +668,7 @@ def test_compression_changed( stderr=PIPE, stdin=input_stream, universal_newlines=True, - env=spoof_tesseract_noop, + check=False, ) assert p.returncode == ExitCode.ok, p.stderr @@ -653,7 +694,7 @@ def test_compression_changed( im.close() -def test_sidecar_pagecount(spoof_tesseract_cache, resources, outpdf): +def test_sidecar_pagecount(resources, outpdf): sidecar = outpdf.with_suffix('.txt') check_ocrmypdf( resources / '3small.pdf', @@ -661,7 +702,8 @@ def test_sidecar_pagecount(spoof_tesseract_cache, resources, outpdf): '--skip-text', '--sidecar', sidecar, - env=spoof_tesseract_cache, + '--plugin', + 'tests/plugins/tesseract_cache.py', ) pdfinfo = PdfInfo(resources / '3small.pdf') @@ -677,10 +719,15 @@ def test_sidecar_pagecount(spoof_tesseract_cache, resources, outpdf): ), "Sidecar page count does not match PDF page count" -def test_sidecar_nonempty(spoof_tesseract_cache, resources, outpdf): +def test_sidecar_nonempty(resources, outpdf): sidecar = outpdf.with_suffix('.txt') check_ocrmypdf( - resources / 'ccitt.pdf', outpdf, '--sidecar', sidecar, env=spoof_tesseract_cache + resources / 'ccitt.pdf', + outpdf, + '--sidecar', + sidecar, + '--plugin', + 'tests/plugins/tesseract_cache.py', ) with open(sidecar, 'r', encoding='utf-8') as f: @@ -689,7 +736,7 @@ def test_sidecar_nonempty(spoof_tesseract_cache, resources, outpdf): @pytest.mark.parametrize('pdfa_level', ['1', '2', '3']) -def test_pdfa_n(spoof_tesseract_cache, pdfa_level, resources, outpdf): +def test_pdfa_n(pdfa_level, resources, outpdf): if pdfa_level == '3' and ghostscript.version() < '9.19': pytest.xfail(reason='Ghostscript >= 9.19 required') @@ -698,7 +745,8 @@ def test_pdfa_n(spoof_tesseract_cache, pdfa_level, resources, outpdf): outpdf, '--output-type', 'pdfa-' + pdfa_level, - env=spoof_tesseract_cache, + '--plugin', + 'tests/plugins/tesseract_cache.py', ) pdfa_info = file_claims_pdfa(outpdf) @@ -710,44 +758,61 @@ def test_pdfa_n(spoof_tesseract_cache, pdfa_level, resources, outpdf): ) @pytest.mark.slow def test_decompression_bomb(resources, outpdf): - p, out, err = run_ocrmypdf(resources / 'hugemono.pdf', outpdf) + p, _out, err = run_ocrmypdf(resources / 'hugemono.pdf', outpdf) assert 'decompression bomb' in err - p, out, err = run_ocrmypdf( + p, _out, err = run_ocrmypdf( resources / 'hugemono.pdf', outpdf, '--max-image-mpixels', '2000' ) assert p.returncode == 0 -def test_text_curves(spoof_tesseract_noop, resources, outpdf): +def test_text_curves(resources, outpdf): with patch('ocrmypdf._pipeline.VECTOR_PAGE_DPI', 100): - check_ocrmypdf(resources / 'vector.pdf', outpdf, env=spoof_tesseract_noop) + check_ocrmypdf( + resources / 'vector.pdf', + outpdf, + '--plugin', + 'tests/plugins/tesseract_noop.py', + ) info = PdfInfo(outpdf) assert len(info.pages[0].images) == 0, "added images to the vector PDF" check_ocrmypdf( - resources / 'vector.pdf', outpdf, '--force-ocr', env=spoof_tesseract_noop + resources / 'vector.pdf', + outpdf, + '--force-ocr', + '--plugin', + 'tests/plugins/tesseract_noop.py', ) info = PdfInfo(outpdf) assert len(info.pages[0].images) != 0, "force did not rasterize" -def test_output_is_dir(spoof_tesseract_noop, resources, outdir): - p, out, err = run_ocrmypdf( - resources / 'trivial.pdf', outdir, '--force-ocr', env=spoof_tesseract_noop +def test_output_is_dir(resources, outdir): + p, _out, err = run_ocrmypdf( + resources / 'trivial.pdf', + outdir, + '--force-ocr', + '--plugin', + 'tests/plugins/tesseract_noop.py', ) assert p.returncode == ExitCode.file_access_error assert 'is not a writable file' in err @pytest.mark.skipif(os.name == 'nt', reason="symlink needs admin permissions") -def test_output_is_symlink(spoof_tesseract_noop, resources, outdir): +def test_output_is_symlink(resources, outdir): sym = Path(outdir / 'this_is_a_symlink') sym.symlink_to(outdir / 'out.pdf') - p, out, err = run_ocrmypdf( - resources / 'trivial.pdf', sym, '--force-ocr', env=spoof_tesseract_noop + p, _out, err = run_ocrmypdf( + resources / 'trivial.pdf', + sym, + '--force-ocr', + '--plugin', + 'tests/plugins/tesseract_noop.py', ) assert p.returncode == ExitCode.ok, err assert (outdir / 'out.pdf').stat().st_size > 0, 'target file not created' @@ -760,8 +825,6 @@ def test_livecycle(resources, no_outpdf): def test_version_check(): - from ocrmypdf.exec import get_version - with pytest.raises(MissingDependencyError): get_version('NOT_FOUND_UNLIKELY_ON_PATH') @@ -785,9 +848,7 @@ def test_version_check(): [0.0, 1, 'pdf', True], ], ) -def test_fast_web_view( - spoof_tesseract_noop, resources, outpdf, threshold, optimize, output_type, expected -): +def test_fast_web_view(resources, outpdf, threshold, optimize, output_type, expected): check_ocrmypdf( resources / 'trivial.pdf', outpdf, @@ -797,18 +858,20 @@ def test_fast_web_view( optimize, '--output-type', output_type, - env=spoof_tesseract_noop, + '--plugin', + 'tests/plugins/tesseract_noop.py', ) with pikepdf.open(outpdf) as pdf: assert pdf.is_linearized == expected -def test_image_dpi_not_image(caplog, spoof_tesseract_noop, resources, outpdf): +def test_image_dpi_not_image(caplog, resources, outpdf): check_ocrmypdf( resources / 'trivial.pdf', outpdf, '--image-dpi', '100', - env=spoof_tesseract_noop, + '--plugin', + 'tests/plugins/tesseract_noop.py', ) assert '--image-dpi is being ignored' in caplog.text diff --git a/tests/test_metadata.py b/tests/test_metadata.py index 270a8b62..1d310107 100644 --- a/tests/test_metadata.py +++ b/tests/test_metadata.py @@ -17,22 +17,20 @@ import datetime -import logging import mmap -import os from datetime import timezone from os import fspath -from pathlib import Path -from shutil import copyfile, move -from unittest.mock import MagicMock, patch +from shutil import copyfile +from unittest.mock import patch import pikepdf import pytest from pikepdf.models.metadata import decode_pdf_date -from ocrmypdf._jobcontext import PDFContext -from ocrmypdf._pipeline import convert_to_pdfa -from ocrmypdf.cli import parser +from ocrmypdf._jobcontext import PdfContext +from ocrmypdf._pipeline import convert_to_pdfa, metadata_fixup +from ocrmypdf._plugin_manager import get_plugin_manager +from ocrmypdf.cli import get_parser from ocrmypdf.exceptions import ExitCode from ocrmypdf.pdfa import SRGB_ICC_PROFILE, file_claims_pdfa, generate_pdfa_ps from ocrmypdf.pdfinfo import PdfInfo @@ -44,17 +42,15 @@ except ImportError: # pytest.helpers is dynamic # pylint: disable=no-member -# pylint: disable=w0612 pytestmark = pytest.mark.filterwarnings('ignore:.*XMLParser.*:DeprecationWarning') check_ocrmypdf = pytest.helpers.check_ocrmypdf run_ocrmypdf = pytest.helpers.run_ocrmypdf -spoof = pytest.helpers.spoof @pytest.mark.parametrize("output_type", ['pdfa', 'pdf']) -def test_preserve_metadata(spoof_tesseract_noop, output_type, resources, outpdf): +def test_preserve_metadata(output_type, resources, outpdf): pdf_before = pikepdf.open(resources / 'graph.pdf') output = check_ocrmypdf( @@ -62,7 +58,8 @@ def test_preserve_metadata(spoof_tesseract_noop, output_type, resources, outpdf) outpdf, '--output-type', output_type, - env=spoof_tesseract_noop, + '--plugin', + 'tests/plugins/tesseract_noop.py', ) pdf_after = pikepdf.open(output) @@ -75,12 +72,12 @@ def test_preserve_metadata(spoof_tesseract_noop, output_type, resources, outpdf) @pytest.mark.parametrize("output_type", ['pdfa', 'pdf']) -def test_override_metadata(spoof_tesseract_noop, output_type, resources, outpdf): +def test_override_metadata(output_type, resources, outpdf): input_file = resources / 'c02-22.pdf' german = 'Du siehst den Wald vor lauter Bäumen nicht.' chinese = '孔子' - p, out, err = run_ocrmypdf( + p, _out, err = run_ocrmypdf( input_file, outpdf, '--title', @@ -89,7 +86,8 @@ def test_override_metadata(spoof_tesseract_noop, output_type, resources, outpdf) chinese, '--output-type', output_type, - env=spoof_tesseract_noop, + '--plugin', + 'tests/plugins/tesseract_noop.py', ) assert p.returncode == ExitCode.ok, err @@ -109,21 +107,22 @@ def test_override_metadata(spoof_tesseract_noop, output_type, resources, outpdf) assert pdfa_info['output'] == output_type -def test_high_unicode(spoof_tesseract_noop, resources, no_outpdf): +def test_high_unicode(resources, no_outpdf): # Ghostscript doesn't support high Unicode, so neither do we, to be # safe input_file = resources / 'c02-22.pdf' high_unicode = 'U+1030C is: 𐌌' - p, out, err = run_ocrmypdf( + p, _out, err = run_ocrmypdf( input_file, no_outpdf, '--subject', high_unicode, '--output-type', 'pdfa', - env=spoof_tesseract_noop, + '--plugin', + 'tests/plugins/tesseract_noop.py', ) assert p.returncode == ExitCode.bad_args, err @@ -132,9 +131,7 @@ def test_high_unicode(spoof_tesseract_noop, resources, no_outpdf): @pytest.mark.skipif(not fitz, reason="test uses fitz") @pytest.mark.parametrize('ocr_option', ['--skip-text', '--force-ocr']) @pytest.mark.parametrize('output_type', ['pdf', 'pdfa']) -def test_bookmarks_preserved( - spoof_tesseract_noop, output_type, ocr_option, resources, outpdf -): +def test_bookmarks_preserved(output_type, ocr_option, resources, outpdf): input_file = resources / 'toc.pdf' before_toc = fitz.Document(str(input_file)).getToC() @@ -144,7 +141,8 @@ def test_bookmarks_preserved( ocr_option, '--output-type', output_type, - env=spoof_tesseract_noop, + '--plugin', + 'tests/plugins/tesseract_noop.py', ) after_toc = fitz.Document(str(outpdf)).getToC() @@ -159,13 +157,16 @@ def seconds_between_dates(date1, date2): @pytest.mark.parametrize('infile', ['trivial.pdf', 'jbig2.pdf']) @pytest.mark.parametrize('output_type', ['pdf', 'pdfa']) -def test_creation_date_preserved( - spoof_tesseract_noop, output_type, resources, infile, outpdf -): +def test_creation_date_preserved(output_type, resources, infile, outpdf): input_file = resources / infile check_ocrmypdf( - input_file, outpdf, '--output-type', output_type, env=spoof_tesseract_noop + input_file, + outpdf, + '--output-type', + output_type, + '--plugin', + 'tests/plugins/tesseract_noop.py', ) pdf_before = pikepdf.open(input_file) @@ -188,19 +189,23 @@ def test_creation_date_preserved( @pytest.mark.parametrize('output_type', ['pdf', 'pdfa']) -def test_xml_metadata_preserved(spoof_tesseract_noop, output_type, resources, outpdf): +def test_xml_metadata_preserved(output_type, resources, outpdf): input_file = resources / 'graph.pdf' try: - from libxmp import consts - from libxmp.utils import file_to_dict - except Exception: + from libxmp.utils import file_to_dict # pylint: disable=import-outside-toplevel + except Exception: # pylint: disable=broad-except pytest.skip("libxmp not available or libexempi3 not installed") before = file_to_dict(str(input_file)) check_ocrmypdf( - input_file, outpdf, '--output-type', output_type, env=spoof_tesseract_noop + input_file, + outpdf, + '--output-type', + output_type, + '--plugin', + 'tests/plugins/tesseract_noop.py', ) after = file_to_dict(str(outpdf)) @@ -278,9 +283,14 @@ def test_srgb_in_unicode_path(tmp_path): generate_pdfa_ps(dstdir / 'out.ps') -def test_kodak_toc(resources, outpdf, spoof_tesseract_noop): - output = check_ocrmypdf( - resources / 'kcs.pdf', outpdf, '--output-type', 'pdf', env=spoof_tesseract_noop +def test_kodak_toc(resources, outpdf): + _output = check_ocrmypdf( + resources / 'kcs.pdf', + outpdf, + '--output-type', + 'pdf', + '--plugin', + 'tests/plugins/tesseract_noop.py', ) p = pikepdf.open(outpdf) @@ -290,16 +300,15 @@ def test_kodak_toc(resources, outpdf, spoof_tesseract_noop): def test_metadata_fixup_warning(resources, outdir, caplog): - from ocrmypdf.__main__ import parser - from ocrmypdf._pipeline import metadata_fixup - - options = parser.parse_args( + options = get_parser().parse_args( args=['--output-type', 'pdfa-2', 'graph.pdf', 'out.pdf'] ) copyfile(resources / 'graph.pdf', outdir / 'graph.pdf') - context = PDFContext(options, outdir, outdir / 'graph.pdf', None) + context = PdfContext( + options, outdir, outdir / 'graph.pdf', None, get_plugin_manager([]) + ) metadata_fixup(working_file=outdir / 'graph.pdf', context=context) for record in caplog.records: assert record.levelname != 'WARNING' @@ -310,7 +319,9 @@ def test_metadata_fixup_warning(resources, outdir, caplog): meta['prism2:publicationName'] = 'OCRmyPDF Test' graph.save(outdir / 'graph_mod.pdf') - context = PDFContext(options, outdir, outdir / 'graph_mod.pdf', None) + context = PdfContext( + options, outdir, outdir / 'graph_mod.pdf', None, get_plugin_manager([]) + ) metadata_fixup(working_file=outdir / 'graph.pdf', context=context) assert any(record.levelname == 'WARNING' for record in caplog.records) @@ -326,11 +337,13 @@ def test_prevent_gs_invalid_xml(resources, outdir): Title=b'String with trailing nul\x00' ) - options = parser.parse_args( + options = get_parser().parse_args( args=['-j', '1', '--output-type', 'pdfa-2', 'a.pdf', 'b.pdf'] ) pdfinfo = PdfInfo(outdir / 'layers.rendered.pdf') - context = PDFContext(options, outdir, outdir / 'layers.rendered.pdf', pdfinfo) + context = PdfContext( + options, outdir, outdir / 'layers.rendered.pdf', pdfinfo, get_plugin_manager([]) + ) convert_to_pdfa( str(outdir / 'layers.rendered.pdf'), str(outdir / 'pdfa.ps'), context @@ -357,11 +370,13 @@ def test_malformed_docinfo(caplog, resources, outdir): pike.trailer.Info = pikepdf.Stream(pike, b"") pike.save(outdir / 'layers.rendered.pdf', fix_metadata_version=False) - options = parser.parse_args( + options = get_parser().parse_args( args=['-j', '1', '--output-type', 'pdfa-2', 'a.pdf', 'b.pdf'] ) pdfinfo = PdfInfo(outdir / 'layers.rendered.pdf') - context = PDFContext(options, outdir, outdir / 'layers.rendered.pdf', pdfinfo) + context = PdfContext( + options, outdir, outdir / 'layers.rendered.pdf', pdfinfo, get_plugin_manager([]) + ) convert_to_pdfa( str(outdir / 'layers.rendered.pdf'), str(outdir / 'pdfa.ps'), context diff --git a/tests/test_optimize.py b/tests/test_optimize.py index b9368849..b4fee041 100644 --- a/tests/test_optimize.py +++ b/tests/test_optimize.py @@ -15,7 +15,6 @@ # You should have received a copy of the GNU General Public License # along with OCRmyPDF. If not, see . -import logging from os import fspath from pathlib import Path from unittest.mock import patch @@ -26,8 +25,9 @@ import pytest from PIL import Image, ImageDraw from ocrmypdf import optimize as opt -from ocrmypdf.exec import jbig2enc, pngquant -from ocrmypdf.exec.ghostscript import rasterize_pdf +from ocrmypdf._exec import jbig2enc, pngquant +from ocrmypdf._exec.ghostscript import rasterize_pdf +from ocrmypdf.helpers import Resolution check_ocrmypdf = pytest.helpers.check_ocrmypdf # pylint: disable=e1101 @@ -47,10 +47,8 @@ def test_mono_not_inverted(resources, outdir): rasterize_pdf( outdir / 'out.pdf', outdir / 'im.png', - xres=10, - yres=10, raster_device='pnggray', - log=logging.getLogger(name='test_mono_not_inverted'), + raster_dpi=Resolution(10, 10), ) with Image.open(fspath(outdir / 'im.png')) as im: @@ -58,7 +56,7 @@ def test_mono_not_inverted(resources, outdir): @pytest.mark.skipif(not pngquant.available(), reason='need pngquant') -def test_jpg_png_params(resources, outpdf, spoof_tesseract_noop): +def test_jpg_png_params(resources, outpdf): check_ocrmypdf( resources / 'crom.png', outpdf, @@ -70,13 +68,14 @@ def test_jpg_png_params(resources, outpdf, spoof_tesseract_noop): '50', '--png-quality', '20', - env=spoof_tesseract_noop, + '--plugin', + 'tests/plugins/tesseract_noop.py', ) @pytest.mark.skipif(not jbig2enc.available(), reason='need jbig2enc') @pytest.mark.parametrize('lossy', [False, True]) -def test_jbig2_lossy(lossy, resources, outpdf, spoof_tesseract_noop): +def test_jbig2_lossy(lossy, resources, outpdf): args = [ resources / 'ccitt.pdf', outpdf, @@ -88,11 +87,13 @@ def test_jbig2_lossy(lossy, resources, outpdf, spoof_tesseract_noop): '50', '--png-quality', '20', + '--plugin', + 'tests/plugins/tesseract_noop.py', ] if lossy: args.append('--jbig2-lossy') - check_ocrmypdf(*args, env=spoof_tesseract_noop) + check_ocrmypdf(*args) pdf = pikepdf.open(outpdf) pim = pikepdf.PdfImage(next(iter(pdf.pages[0].images.values()))) @@ -108,7 +109,7 @@ def test_jbig2_lossy(lossy, resources, outpdf, spoof_tesseract_noop): not jbig2enc.available() or not pngquant.available(), reason='need jbig2enc and pngquant', ) -def test_flate_to_jbig2(resources, outdir, spoof_tesseract_noop): +def test_flate_to_jbig2(resources, outdir): # This test requires an image that pngquant is capable of converting to # to 1bpp - so use an existing 1bpp image, convert up, confirm it can # convert down @@ -126,7 +127,8 @@ def test_flate_to_jbig2(resources, outdir, spoof_tesseract_noop): '50', '--optimize', '3', - env=spoof_tesseract_noop, + '--plugin', + 'tests/plugins/tesseract_noop.py', ) pdf = pikepdf.open(outdir / 'out.pdf') @@ -134,7 +136,7 @@ def test_flate_to_jbig2(resources, outdir, spoof_tesseract_noop): assert pim.filters[0] == '/JBIG2Decode' -def test_multiple_pngs(resources, outdir, spoof_tesseract_noop): +def test_multiple_pngs(resources, outdir): with Path.open(outdir / 'in.pdf', 'wb') as inpdf: img2pdf.convert( fspath(resources / 'baiona_colormapped.png'), @@ -160,7 +162,8 @@ def test_multiple_pngs(resources, outdir, spoof_tesseract_noop): '--use-threads', '--output-type', 'pdf', - env=spoof_tesseract_noop, + '--plugin', + 'tests/plugins/tesseract_noop.py', ) with pikepdf.open(outdir / 'in.pdf') as inpdf, pikepdf.open( diff --git a/tests/test_page_numbers.py b/tests/test_page_numbers.py index 1fb494c4..733153fc 100644 --- a/tests/test_page_numbers.py +++ b/tests/test_page_numbers.py @@ -58,7 +58,7 @@ def test_list_range(): assert _pages_from_ranges([0, 1, 2]) == {0, 1, 2} -def test_limited_pages(resources, outpdf, spoof_tesseract_cache): +def test_limited_pages(resources, outpdf): multi = resources / 'multipage.pdf' ocrmypdf.ocr( multi, @@ -66,7 +66,7 @@ def test_limited_pages(resources, outpdf, spoof_tesseract_cache): pages='5-6', optimize=0, output_type='pdf', - tesseract_env=spoof_tesseract_cache, + plugins=['tests/plugins/tesseract_cache.py'], ) pi = PdfInfo(outpdf) assert not pi.pages[0].has_text diff --git a/tests/test_pdfinfo.py b/tests/test_pdfinfo.py index facf3b6f..558995fc 100644 --- a/tests/test_pdfinfo.py +++ b/tests/test_pdfinfo.py @@ -26,7 +26,7 @@ from PIL import Image from reportlab.pdfgen.canvas import Canvas from ocrmypdf import pdfinfo -from ocrmypdf.exec import ghostscript +from ocrmypdf._exec import ghostscript from ocrmypdf.pdfinfo import Colorspace, Encoding # pylint: disable=protected-access @@ -85,8 +85,8 @@ def test_single_page_image(outdir): assert pdfimage.color == Colorspace.gray # DPI in a 1"x1" is the image width - assert isclose(pdfimage.xres, 8) - assert isclose(pdfimage.yres, 8) + assert isclose(pdfimage.dpi.x, 8) + assert isclose(pdfimage.dpi.y, 8) def test_single_page_inline_image(outdir): @@ -105,7 +105,7 @@ def test_single_page_inline_image(outdir): info = pdfinfo.PdfInfo(filename) print(info) pdfimage = info[0].images[0] - assert isclose(pdfimage.xres, 8) + assert isclose(pdfimage.dpi.x, 8) assert pdfimage.color == Colorspace.gray assert pdfimage.width == 8 @@ -117,7 +117,7 @@ def test_jpeg(resources, outdir): pdfimage = pdf[0].images[0] assert pdfimage.enc == Encoding.jpeg - assert isclose(pdfimage.xres, 150) + assert isclose(pdfimage.dpi.x, 150) def test_form_xobject(resources): @@ -139,7 +139,7 @@ def test_no_contents(resources): def test_oversized_page(resources): pdf = pdfinfo.PdfInfo(resources / 'poster.pdf') image = pdf[0].images[0] - assert image.width * image.xres > 200, "this is supposed to be oversized" + assert image.width * image.dpi.x > 200, "this is supposed to be oversized" def test_pickle(resources): @@ -151,22 +151,6 @@ def test_pickle(resources): pickle.dumps(pdf) -def test_regex(): - rx = pdfinfo.ghosttext.regex_remove_char_tags - - must_match = [ - b'', - b'', - b'', - ] - must_not_match = [b'', b'', b'', b''] - - for s in must_match: - assert rx.match(s) - for s in must_not_match: - assert not rx.match(s) - - def test_vector(resources): filename = resources / 'vector.pdf' pdf = pdfinfo.PdfInfo(filename) @@ -184,16 +168,9 @@ def test_ocr_detection(resources): @pytest.mark.parametrize( 'testfile', ('truetype_font_nomapping.pdf', 'type3_font_nomapping.pdf') ) -@pytest.mark.xfail( - ghostscript.version() in ('9.52',), reason="gs 9.52 txtwrite doesn't work" -) def test_corrupt_font_detection(resources, testfile): filename = resources / testfile - with pytest.raises(NotImplementedError): - pdf = pdfinfo.PdfInfo(filename) - pdf[0].has_corrupt_text - - pdf = pdfinfo.PdfInfo(filename, detailed_page_analysis=True) + pdf = pdfinfo.PdfInfo(filename) assert pdf[0].has_corrupt_text diff --git a/tests/test_preprocessing.py b/tests/test_preprocessing.py index 4054e4e4..7cabe827 100644 --- a/tests/test_preprocessing.py +++ b/tests/test_preprocessing.py @@ -15,13 +15,13 @@ # You should have received a copy of the GNU General Public License # along with OCRmyPDF. If not, see . -import logging from math import isclose import pytest from PIL import Image -from ocrmypdf.exec import ghostscript +from ocrmypdf._exec import ghostscript +from ocrmypdf.helpers import Resolution from ocrmypdf.leptonica import Pix from ocrmypdf.pdfinfo import PdfInfo @@ -31,31 +31,30 @@ from ocrmypdf.pdfinfo import PdfInfo check_ocrmypdf = pytest.helpers.check_ocrmypdf run_ocrmypdf = pytest.helpers.run_ocrmypdf run_ocrmypdf_api = pytest.helpers.run_ocrmypdf_api -spoof = pytest.helpers.spoof RENDERERS = ['hocr', 'sandwich'] -def test_deskew(spoof_tesseract_noop, resources, outdir): +def test_deskew(resources, outdir): # Run with deskew deskewed_pdf = check_ocrmypdf( - resources / 'skew.pdf', outdir / 'skew.pdf', '-d', env=spoof_tesseract_noop + resources / 'skew.pdf', + outdir / 'skew.pdf', + '-d', + '--plugin', + 'tests/plugins/tesseract_noop.py', ) # Now render as an image again and use Leptonica to find the skew angle # to confirm that it was deskewed - log = logging.getLogger() - deskewed_png = outdir / 'deskewed.png' ghostscript.rasterize_pdf( deskewed_pdf, deskewed_png, - xres=150, - yres=150, raster_device='pngmono', - log=log, + raster_dpi=Resolution(150, 150), pageno=1, ) @@ -66,7 +65,7 @@ def test_deskew(spoof_tesseract_noop, resources, outdir): assert -0.5 < skew_angle < 0.5, "Deskewing failed" -def test_remove_background(spoof_tesseract_noop, resources, outdir): +def test_remove_background(resources, outdir): # Ensure the input image does not contain pure white/black with Image.open(resources / 'congress.jpg') as im: assert im.getextrema() != ((0, 255), (0, 255), (0, 255)) @@ -77,20 +76,17 @@ def test_remove_background(spoof_tesseract_noop, resources, outdir): '--remove-background', '--image-dpi', '150', - env=spoof_tesseract_noop, + '--plugin', + 'tests/plugins/tesseract_noop.py', ) - log = logging.getLogger() - output_png = outdir / 'remove_bg.png' ghostscript.rasterize_pdf( output_pdf, output_png, - xres=100, - yres=100, raster_device='png16m', - log=log, + raster_dpi=Resolution(100, 100), pageno=1, ) @@ -105,9 +101,7 @@ def test_remove_background(spoof_tesseract_noop, resources, outdir): ) @pytest.mark.parametrize("renderer", ['sandwich', 'hocr']) @pytest.mark.parametrize("output_type", ['pdf', 'pdfa']) -def test_exotic_image( - spoof_tesseract_cache, pdf, renderer, output_type, resources, outdir -): +def test_exotic_image(pdf, renderer, output_type, resources, outdir): outfile = outdir / f'test_{pdf}_{renderer}.pdf' check_ocrmypdf( resources / pdf, @@ -121,40 +115,39 @@ def test_exotic_image( '--skip-text', '--pdf-renderer', renderer, - env=spoof_tesseract_cache, + '--plugin', + 'tests/plugins/tesseract_cache.py', ) assert outfile.with_suffix('.pdf.txt').exists() @pytest.mark.parametrize('renderer', RENDERERS) -def test_non_square_resolution(renderer, spoof_tesseract_cache, resources, outpdf): +def test_non_square_resolution(renderer, resources, outpdf): # Confirm input image is non-square resolution in_pageinfo = PdfInfo(resources / 'aspect.pdf') - assert in_pageinfo[0].xres != in_pageinfo[0].yres + assert in_pageinfo[0].dpi.x != in_pageinfo[0].dpi.y check_ocrmypdf( resources / 'aspect.pdf', outpdf, '--pdf-renderer', renderer, - env=spoof_tesseract_cache, + '--plugin', + 'tests/plugins/tesseract_cache.py', ) out_pageinfo = PdfInfo(outpdf) # Confirm resolution was kept the same - assert in_pageinfo[0].xres == out_pageinfo[0].xres - assert in_pageinfo[0].yres == out_pageinfo[0].yres + assert in_pageinfo[0].dpi == out_pageinfo[0].dpi @pytest.mark.parametrize('renderer', RENDERERS) -def test_convert_to_square_resolution( - renderer, spoof_tesseract_cache, resources, outpdf -): +def test_convert_to_square_resolution(renderer, resources, outpdf): # Confirm input image is non-square resolution in_pageinfo = PdfInfo(resources / 'aspect.pdf') - assert in_pageinfo[0].xres != in_pageinfo[0].yres + assert in_pageinfo[0].dpi.x != in_pageinfo[0].dpi.y # --force-ocr requires means forced conversion to square resolution check_ocrmypdf( @@ -163,7 +156,8 @@ def test_convert_to_square_resolution( '--force-ocr', '--pdf-renderer', renderer, - env=spoof_tesseract_cache, + '--plugin', + 'tests/plugins/tesseract_cache.py', ) out_pageinfo = PdfInfo(outpdf) @@ -171,7 +165,7 @@ def test_convert_to_square_resolution( in_p0, out_p0 = in_pageinfo[0], out_pageinfo[0] # Resolution show now be equal - assert out_p0.xres == out_p0.yres + assert out_p0.dpi.x == out_p0.dpi.y # Page size should match input page size assert isclose(in_p0.width_inches, out_p0.width_inches) @@ -179,7 +173,7 @@ def test_convert_to_square_resolution( # Because we rasterized the page to produce a new image, it should occupy # the entire page - out_im_w = out_p0.images[0].width / out_p0.images[0].xres - out_im_h = out_p0.images[0].height / out_p0.images[0].yres + out_im_w = out_p0.images[0].width / out_p0.images[0].dpi.x + out_im_h = out_p0.images[0].height / out_p0.images[0].dpi.y assert isclose(out_p0.width_inches, out_im_w) assert isclose(out_p0.height_inches, out_im_h) diff --git a/tests/test_rotation.py b/tests/test_rotation.py index 05d42300..f0e7fcfd 100644 --- a/tests/test_rotation.py +++ b/tests/test_rotation.py @@ -26,7 +26,8 @@ import pytest from PIL import Image from ocrmypdf import leptonica -from ocrmypdf.exec import ghostscript, tesseract +from ocrmypdf._exec import ghostscript, tesseract +from ocrmypdf.helpers import Resolution from ocrmypdf.pdfinfo import PdfInfo # pytest.helpers is dynamic @@ -48,8 +49,6 @@ RENDERERS = ['hocr', 'sandwich'] def check_monochrome_correlation( outdir, reference_pdf, reference_pageno, test_pdf, test_pageno ): - gslog = logging.getLogger() - reference_png = outdir / f'{reference_pdf.name}.ref{reference_pageno:04d}.png' test_png = outdir / f'{test_pdf.name}.test{test_pageno:04d}.png' @@ -60,10 +59,8 @@ def check_monochrome_correlation( ghostscript.rasterize_pdf( pdf, png, - xres=100, - yres=100, raster_device='pngmono', - log=gslog, + raster_dpi=Resolution(100, 100), pageno=pageno, rotation=0, ) @@ -100,7 +97,7 @@ def test_monochrome_correlation(resources, outdir): @pytest.mark.slow @pytest.mark.parametrize('renderer', RENDERERS) -def test_autorotate(spoof_tesseract_cache, renderer, resources, outdir): +def test_autorotate(renderer, resources, outdir): # cardinal.pdf contains four copies of an image rotated in each cardinal # direction - these ones are "burned in" not tagged with /Rotate out = check_ocrmypdf( @@ -111,7 +108,8 @@ def test_autorotate(spoof_tesseract_cache, renderer, resources, outdir): '1', '--pdf-renderer', renderer, - env=spoof_tesseract_cache, + '--plugin', + 'tests/plugins/tesseract_cache.py', ) for n in range(1, 4 + 1): correlation = check_monochrome_correlation( @@ -131,9 +129,7 @@ def test_autorotate(spoof_tesseract_cache, renderer, resources, outdir): ('99', 'correlation < 0.10'), # High thres -> never rotate -> low corr ], ) -def test_autorotate_threshold( - spoof_tesseract_cache, threshold, correlation_test, resources, outdir -): +def test_autorotate_threshold(threshold, correlation_test, resources, outdir): out = check_ocrmypdf( resources / 'cardinal.pdf', outdir / 'out.pdf', @@ -142,7 +138,8 @@ def test_autorotate_threshold( '-r', # '-v', # '1', - env=spoof_tesseract_cache, + '--plugin', + 'tests/plugins/tesseract_cache.py', ) correlation = check_monochrome_correlation( @@ -268,7 +265,6 @@ def test_tesseract_orientation(resources, tmp_path): pix_rotated = pix.rotate_orth(2) # 180 degrees clockwise pix_rotated.write_implied_format(tmp_path / '000001.png') - log = logging.getLogger() tesseract.get_orientation( # Test results of this are unreliable - tmp_path / '000001.png', engine_mode='3', timeout=10, log=log + tmp_path / '000001.png', engine_mode='3', timeout=10 ) diff --git a/tests/test_stdio.py b/tests/test_stdio.py index 21112468..478661c2 100644 --- a/tests/test_stdio.py +++ b/tests/test_stdio.py @@ -23,35 +23,33 @@ from subprocess import DEVNULL, PIPE, CalledProcessError, Popen, run import pytest from ocrmypdf.exceptions import ExitCode -from ocrmypdf.exec import qpdf +from ocrmypdf.helpers import check_pdf # pytest.helpers is dynamic # pylint: disable=no-member,redefined-outer-name run_ocrmypdf = pytest.helpers.run_ocrmypdf run_ocrmypdf_api = pytest.helpers.run_ocrmypdf -spoof = pytest.helpers.spoof -def test_stdin(spoof_tesseract_noop, ocrmypdf_exec, resources, outpdf): +def test_stdin(ocrmypdf_exec, resources, outpdf): input_file = str(resources / 'francais.pdf') output_file = str(outpdf) # Runs: ocrmypdf - output.pdf < testfile.pdf with open(input_file, 'rb') as input_stream: - p_args = ocrmypdf_exec + ['-', output_file] - p = run( - p_args, - stdout=PIPE, - stderr=PIPE, - stdin=input_stream, - env=spoof_tesseract_noop, - ) + p_args = ocrmypdf_exec + [ + '-', + output_file, + '--plugin', + 'tests/plugins/tesseract_noop.py', + ] + p = run(p_args, stdout=PIPE, stderr=PIPE, stdin=input_stream) assert p.returncode == ExitCode.ok -def test_stdout(spoof_tesseract_noop, ocrmypdf_exec, resources, outpdf): - if 'COV_CORE_DATAFILE' in spoof_tesseract_noop: +def test_stdout(ocrmypdf_exec, resources, outpdf): + if 'COV_CORE_DATAFILE' in os.environ: pytest.skip(msg="Coverage uses stdout") input_file = str(resources / 'francais.pdf') @@ -59,24 +57,23 @@ def test_stdout(spoof_tesseract_noop, ocrmypdf_exec, resources, outpdf): # Runs: ocrmypdf francais.pdf - > test_stdout.pdf with open(output_file, 'wb') as output_stream: - p_args = ocrmypdf_exec + [input_file, '-'] - p = run( - p_args, - stdout=output_stream, - stderr=PIPE, - stdin=DEVNULL, - env=spoof_tesseract_noop, - ) + p_args = ocrmypdf_exec + [ + input_file, + '-', + '--plugin', + 'tests/plugins/tesseract_noop.py', + ] + p = run(p_args, stdout=output_stream, stderr=PIPE, stdin=DEVNULL) assert p.returncode == ExitCode.ok - assert qpdf.check(output_file, log=None) + assert check_pdf(output_file) @pytest.mark.skipif( sys.version_info[0:3] >= (3, 6, 4), reason="issue fixed in Python 3.6.4" ) @pytest.mark.skipif(os.name == 'nt', reason="POSIX problem") -def test_closed_streams(spoof_tesseract_noop, ocrmypdf_exec, resources, outpdf): +def test_closed_streams(ocrmypdf_exec, resources, outpdf): input_file = str(resources / 'francais.pdf') output_file = str(outpdf) @@ -84,14 +81,18 @@ def test_closed_streams(spoof_tesseract_noop, ocrmypdf_exec, resources, outpdf): os.close(0) os.close(1) - p_args = ocrmypdf_exec + [input_file, output_file] + p_args = ocrmypdf_exec + [ + input_file, + output_file, + '--plugin', + 'tests/plugins/tesseract_noop.py', + ] p = Popen( # pylint: disable=subprocess-popen-preexec-fn p_args, close_fds=True, stdout=None, stderr=PIPE, stdin=None, - env=spoof_tesseract_noop, preexec_fn=evil_closer, ) out, err = p.communicate() @@ -104,11 +105,9 @@ def test_closed_streams(spoof_tesseract_noop, ocrmypdf_exec, resources, outpdf): Path('/etc/alpine-release').exists(), reason="invalid test on alpine" ) @pytest.mark.skipif(os.name == 'nt', reason="invalid test on Windows") -def test_bad_locale(): - env = os.environ.copy() - env['LC_ALL'] = 'C' - - p, out, err = run_ocrmypdf('a', 'b', env=env) +def test_bad_locale(monkeypatch): + monkeypatch.setenv('LC_ALL', 'C') + p, out, err = run_ocrmypdf('a', 'b') assert out == '', "stdout not clean" assert p.returncode != 0 assert 'configured to use ASCII as encoding' in err, "should whine" @@ -118,12 +117,16 @@ def test_bad_locale(): os.name == 'nt' and sys.version_info < (3, 8), reason="Windows does not like this; not sure how to fix", ) -def test_dev_null(spoof_tesseract_noop, resources): - if 'COV_CORE_DATAFILE' in spoof_tesseract_noop: +def test_dev_null(resources): + if 'COV_CORE_DATAFILE' in os.environ: pytest.skip(msg="Coverage uses stdout") p, out, err = run_ocrmypdf( - resources / 'trivial.pdf', os.devnull, '--force-ocr', env=spoof_tesseract_noop + resources / 'trivial.pdf', + os.devnull, + '--force-ocr', + '--plugin', + 'tests/plugins/tesseract_noop.py', ) assert p.returncode == 0, "could not send output to /dev/null" assert len(out) == 0, "wrote to stdout" diff --git a/tests/test_tess4.py b/tests/test_tesseract.py similarity index 75% rename from tests/test_tess4.py rename to tests/test_tesseract.py index a66f2186..0db09110 100644 --- a/tests/test_tess4.py +++ b/tests/test_tesseract.py @@ -18,25 +18,19 @@ import logging import os import subprocess -from contextlib import contextmanager from os import fspath from pathlib import Path import pytest from ocrmypdf import pdfinfo +from ocrmypdf._exec import tesseract from ocrmypdf.exceptions import MissingDependencyError -from ocrmypdf.exec import tesseract -# pylint: disable=no-member,w0621 +# pylint: disable=no-member,redefined-outer-name check_ocrmypdf = pytest.helpers.check_ocrmypdf run_ocrmypdf = pytest.helpers.run_ocrmypdf -spoof = pytest.helpers.spoof - - -def test_tesseract_v4(): - assert tesseract.v4() @pytest.mark.parametrize('basename', ['graph_ocred.pdf', 'cardinal.pdf']) @@ -60,8 +54,8 @@ def test_skip_pages_does_not_replicate(resources, basename, outdir): for page in info: assert len(page.images) == 1, "skipped page was replicated" - for n in range(len(info_in)): - assert info[n].width_inches == info_in[n].width_inches + for n, info_out_n in enumerate(info): + assert info_out_n.width_inches == info_in[n].width_inches def test_content_preservation(resources, outpdf): @@ -76,70 +70,57 @@ def test_content_preservation(resources, outpdf): assert len(page.images) > 1, "masks were rasterized" -def test_no_languages(tmp_path): - env = os.environ.copy() +def test_no_languages(tmp_path, monkeypatch): (tmp_path / 'tessdata').mkdir() - env['TESSDATA_PREFIX'] = fspath(tmp_path) - + monkeypatch.setenv('TESSDATA_PREFIX', fspath(tmp_path)) with pytest.raises(MissingDependencyError): - tesseract.languages(tesseract_env=env) + tesseract.get_languages() def test_image_too_large_hocr(monkeypatch, resources, outdir): - log = logging.getLogger('test_image_too_large_hocr') - def dummy_run(args, *, env=None, **kwargs): raise subprocess.CalledProcessError(1, 'tesseract', output=b'Image too large') monkeypatch.setattr(tesseract, 'run', dummy_run) tesseract.generate_hocr( input_file=resources / 'crom.png', - output_files=[outdir / 'out.hocr', outdir / 'out.txt'], - language=['eng'], + output_hocr=outdir / 'out.hocr', + output_text=outdir / 'out.txt', + languages=['eng'], engine_mode=None, tessconfig=[], timeout=180.0, pagesegmode=None, - log=log, user_words=None, user_patterns=None, - tesseract_env=None, ) assert "name='ocr-capabilities'" in Path(outdir / 'out.hocr').read_text() def test_image_too_large_pdf(monkeypatch, resources, outdir): - log = logging.getLogger('test_image_too_large_pdf') - def dummy_run(args, *, env=None, **kwargs): raise subprocess.CalledProcessError(1, 'tesseract', output=b'Image too large') monkeypatch.setattr(tesseract, 'run', dummy_run) tesseract.generate_pdf( - input_image=resources / 'crom.png', - skip_pdf=resources / 'blank.pdf', + input_file=resources / 'crom.png', output_pdf=outdir / 'pdf.pdf', output_text=outdir / 'txt.txt', - language=['eng'], + languages=['eng'], engine_mode=None, - text_only=False, tessconfig=[], timeout=180.0, pagesegmode=None, - log=log, user_words=None, user_patterns=None, - tesseract_env=None, ) assert Path(outdir / 'txt.txt').read_text() == '[skipped page]' if os.name != 'nt': # different semantics - assert Path(outdir / 'pdf.pdf').samefile(resources / 'blank.pdf') + assert Path(outdir / 'pdf.pdf').stat().st_size == 0 def test_timeout(caplog): - log = logging.getLogger('test_timeout') - tesseract.page_timedout(log, '123456.png', 5) - assert "123456" in caplog.text + tesseract.page_timedout(5) assert "took too long" in caplog.text @@ -160,10 +141,8 @@ def test_timeout(caplog): ], ) def test_tesseract_log_output(caplog, in_, logged): - log = logging.getLogger('tesseract_log_output') - log.setLevel(logging.INFO) - - tesseract.tesseract_log_output(log, in_, 'dummy') + caplog.set_level(logging.INFO) + tesseract.tesseract_log_output(in_) if logged == '': assert caplog.text == '' else: @@ -171,7 +150,6 @@ def test_tesseract_log_output(caplog, in_, logged): def test_tesseract_log_output_raises(caplog): - log = logging.getLogger('tesseract_log_output') with pytest.raises(tesseract.TesseractConfigError): - tesseract.tesseract_log_output(log, b'parameter not found: moo', 'dummy') + tesseract.tesseract_log_output(b'parameter not found: moo') assert 'not found' in caplog.text diff --git a/tests/test_unpaper.py b/tests/test_unpaper.py index f0e90b80..6e28235a 100644 --- a/tests/test_unpaper.py +++ b/tests/test_unpaper.py @@ -20,92 +20,93 @@ from unittest.mock import patch import pytest +from ocrmypdf._plugin_manager import get_parser_options_plugins from ocrmypdf._validation import check_options -from ocrmypdf.cli import parser +from ocrmypdf.cli import get_parser from ocrmypdf.exceptions import ExitCode, MissingDependencyError -from ocrmypdf.exec import unpaper # pytest.helpers is dynamic -# pylint: disable=no-member +# pylint: disable=no-member,redefined-outer-name # pylint: disable=w0612 check_ocrmypdf = pytest.helpers.check_ocrmypdf run_ocrmypdf = pytest.helpers.run_ocrmypdf -spoof = pytest.helpers.spoof - - -def have_unpaper(): - try: - unpaper.version() - except Exception: - return False - else: - return True - - -@pytest.fixture -def spoof_unpaper_oldversion(tmp_path_factory): - return spoof(tmp_path_factory, unpaper="unpaper_oldversion.py") +have_unpaper = pytest.helpers.have_unpaper def test_no_unpaper(resources, no_outpdf): input_ = fspath(resources / "c02-22.pdf") output = fspath(no_outpdf) - options = parser.parse_args(args=["--clean", input_, output]) - with patch("ocrmypdf.exec.unpaper.version") as mock_unpaper_version: + _parser, options, pm = get_parser_options_plugins(["--clean", input_, output]) + with patch("ocrmypdf._exec.unpaper.version") as mock_unpaper_version: mock_unpaper_version.side_effect = FileNotFoundError("unpaper") + with pytest.raises(MissingDependencyError): - check_options(options) + check_options(options, pm) -def test_old_unpaper(spoof_unpaper_oldversion, resources, no_outpdf): - p, out, err = run_ocrmypdf( - resources / "c02-22.pdf", no_outpdf, "--clean", env=spoof_unpaper_oldversion +def test_old_unpaper(resources, no_outpdf): + input_ = fspath(resources / "c02-22.pdf") + output = fspath(no_outpdf) + + _parser, options, pm = get_parser_options_plugins(["--clean", input_, output]) + with patch("ocrmypdf._exec.unpaper.version") as mock_unpaper_version: + mock_unpaper_version.return_value = '0.5' + + with pytest.raises(MissingDependencyError): + check_options(options, pm) + + +@pytest.mark.skipif(not have_unpaper(), reason="requires unpaper") +def test_clean(resources, outpdf): + check_ocrmypdf( + resources / "skew.pdf", + outpdf, + "-c", + '--plugin', + 'tests/plugins/tesseract_noop.py', ) - assert p.returncode == ExitCode.missing_dependency @pytest.mark.skipif(not have_unpaper(), reason="requires unpaper") -def test_clean(spoof_tesseract_noop, resources, outpdf): - check_ocrmypdf(resources / "skew.pdf", outpdf, "-c", env=spoof_tesseract_noop) - - -@pytest.mark.skipif(not have_unpaper(), reason="requires unpaper") -def test_unpaper_args_valid(spoof_tesseract_noop, resources, outpdf): +def test_unpaper_args_valid(resources, outpdf): check_ocrmypdf( resources / "skew.pdf", outpdf, "-c", "--unpaper-args", "--layout double", # Spaces required here - env=spoof_tesseract_noop, + '--plugin', + 'tests/plugins/tesseract_noop.py', ) @pytest.mark.skipif(not have_unpaper(), reason="requires unpaper") -def test_unpaper_args_invalid_filename(spoof_tesseract_noop, resources, outpdf): +def test_unpaper_args_invalid_filename(resources, outpdf): p, out, err = run_ocrmypdf( resources / "skew.pdf", outpdf, "-c", "--unpaper-args", "/etc/passwd", - env=spoof_tesseract_noop, + '--plugin', + 'tests/plugins/tesseract_noop.py', ) assert "No filenames allowed" in err assert p.returncode == ExitCode.bad_args @pytest.mark.skipif(not have_unpaper(), reason="requires unpaper") -def test_unpaper_args_invalid(spoof_tesseract_noop, resources, outpdf): +def test_unpaper_args_invalid(resources, outpdf): p, out, err = run_ocrmypdf( resources / "skew.pdf", outpdf, "-c", "--unpaper-args", "unpaper is not going to like these arguments", - env=spoof_tesseract_noop, + '--plugin', + 'tests/plugins/tesseract_noop.py', ) # Can't tell difference between unpaper choking on bad arguments or some # other unpaper failure diff --git a/tests/test_userunit.py b/tests/test_userunit.py index 83ad01d4..60f97d08 100644 --- a/tests/test_userunit.py +++ b/tests/test_userunit.py @@ -25,7 +25,6 @@ from ocrmypdf.pdfinfo import PdfInfo check_ocrmypdf = pytest.helpers.check_ocrmypdf run_ocrmypdf = pytest.helpers.run_ocrmypdf run_ocrmypdf_api = pytest.helpers.run_ocrmypdf_api -spoof = pytest.helpers.spoof @pytest.fixture @@ -39,15 +38,26 @@ def test_userunit_ghostscript_fails(poster, no_outpdf, caplog): assert 'not supported by Ghostscript' in caplog.text -def test_userunit_qpdf_passes(spoof_tesseract_cache, poster, outpdf): +def test_userunit_pdf_passes(poster, outpdf): before = PdfInfo(poster) - check_ocrmypdf(poster, outpdf, '--output-type=pdf', env=spoof_tesseract_cache) + check_ocrmypdf( + poster, + outpdf, + '--output-type=pdf', + '--plugin', + 'tests/plugins/tesseract_cache.py', + ) after = PdfInfo(outpdf) assert isclose(before[0].width_inches, after[0].width_inches) -def test_rotate_interaction(spoof_tesseract_cache, poster, outpdf): +def test_rotate_interaction(poster, outpdf): check_ocrmypdf( - poster, outpdf, '--output-type=pdf', '--rotate-pages', env=spoof_tesseract_cache + poster, + outpdf, + '--output-type=pdf', + '--rotate-pages', + '--plugin', + 'tests/plugins/tesseract_cache.py', ) diff --git a/tests/test_validation.py b/tests/test_validation.py index 01f393ba..52dfb74b 100644 --- a/tests/test_validation.py +++ b/tests/test_validation.py @@ -23,47 +23,65 @@ import pikepdf import pytest import ocrmypdf._validation as vd +from ocrmypdf._plugin_manager import get_plugin_manager from ocrmypdf.api import create_options +from ocrmypdf.cli import get_parser from ocrmypdf.exceptions import BadArgsError, MissingDependencyError from ocrmypdf.pdfinfo import PdfInfo -def make_opts(input_file='a.pdf', output_file='b.pdf', language='eng', **kwargs): +def make_opts_pm(input_file='a.pdf', output_file='b.pdf', language='eng', **kwargs): if language is not None: kwargs['language'] = language - return create_options(input_file=input_file, output_file=output_file, **kwargs) + parser = get_parser() + pm = get_plugin_manager(kwargs.get('plugins', [])) + pm.hook.add_options(parser=parser) + return ( + create_options( + input_file=input_file, output_file=output_file, parser=parser, **kwargs + ), + pm, + ) + + +def make_opts(*args, **kwargs): + opts, _pm = make_opts_pm(*args, **kwargs) + return opts def test_hocr_notlatin_warning(caplog): - vd.check_options_output(make_opts(language='chi_sim', pdf_renderer='hocr')) + vd.check_options( + *make_opts_pm(language='chi_sim', pdf_renderer='hocr', output_type='pdfa') + ) assert 'PDF renderer is known to cause' in caplog.text def test_old_ghostscript(caplog): - with patch('ocrmypdf.exec.ghostscript.version', return_value='9.19'), patch( - 'ocrmypdf.exec.tesseract.has_textonly_pdf', return_value=True + with patch('ocrmypdf._exec.ghostscript.version', return_value='9.19'), patch( + 'ocrmypdf._exec.tesseract.has_textonly_pdf', return_value=True ): - vd.check_options_output(make_opts(language='chi_sim', output_type='pdfa')) + vd.check_options(*make_opts_pm(language='chi_sim', output_type='pdfa')) assert 'Ghostscript does not work correctly' in caplog.text - with patch('ocrmypdf.exec.ghostscript.version', return_value='9.18'), patch( - 'ocrmypdf.exec.tesseract.has_textonly_pdf', return_value=True + with patch('ocrmypdf._exec.ghostscript.version', return_value='9.18'), patch( + 'ocrmypdf._exec.tesseract.has_textonly_pdf', return_value=True ): with pytest.raises(MissingDependencyError): - vd.check_options_output(make_opts(output_type='pdfa-3')) + vd.check_options(*make_opts_pm(output_type='pdfa-3')) - with patch('ocrmypdf.exec.ghostscript.version', return_value='9.24'), patch( - 'ocrmypdf.exec.tesseract.has_textonly_pdf', return_value=True + with patch('ocrmypdf._exec.ghostscript.version', return_value='9.24'), patch( + 'ocrmypdf._exec.tesseract.has_textonly_pdf', return_value=True ): with pytest.raises(MissingDependencyError): - vd.check_dependency_versions(make_opts()) + vd.check_options(*make_opts_pm()) def test_old_tesseract_error(): - with patch('ocrmypdf.exec.tesseract.has_textonly_pdf', return_value=False): + with patch('ocrmypdf._exec.tesseract.has_textonly_pdf', return_value=False): with pytest.raises(MissingDependencyError): opts = make_opts(pdf_renderer='sandwich', language='eng') - vd.check_options_output(opts) + plugin_manager = get_plugin_manager(opts.plugins) + vd.check_options(opts, plugin_manager) def test_lossless_redo(): @@ -90,12 +108,16 @@ def test_optimizing(caplog): def test_user_words(caplog): - with patch('ocrmypdf.exec.tesseract.version', return_value='4.0.0'): - vd.check_options_advanced(make_opts(user_words='foo')) + with patch('ocrmypdf._exec.tesseract.has_user_words', return_value=False): + opts = make_opts(user_words='foo') + plugin_manager = get_plugin_manager(opts.plugins) + vd.check_options(opts, plugin_manager) assert '4.0 ignores --user-words' in caplog.text caplog.clear() - with patch('ocrmypdf.exec.tesseract.version', return_value='4.1.0'): - vd.check_options_advanced(make_opts(user_patterns='foo')) + with patch('ocrmypdf._exec.tesseract.has_user_words', return_value=True): + opts = make_opts(user_patterns='foo') + plugin_manager = get_plugin_manager(opts.plugins) + vd.check_options(opts, plugin_manager) assert '4.0 ignores --user-words' not in caplog.text @@ -148,16 +170,17 @@ def test_report_file_size(tmp_path, caplog): def test_false_action_store_true(): opts = make_opts(keep_temporary_files=True) - assert opts.keep_temporary_files == True + assert opts.keep_temporary_files opts = make_opts(keep_temporary_files=False) - assert opts.keep_temporary_files == False + assert not opts.keep_temporary_files @pytest.mark.parametrize('progress_bar', [True, False]) def test_no_progress_bar(progress_bar, resources): opts = make_opts(progress_bar=progress_bar, input_file=(resources / 'trivial.pdf')) - with patch('ocrmypdf.pdfinfo.info.tqdm', autospec=True) as tqdmpatch: - vd.check_options(opts) + plugin_manager = get_plugin_manager(opts.plugins) + with patch('ocrmypdf._concurrent.tqdm', autospec=True) as tqdmpatch: + vd.check_options(opts, plugin_manager) pdfinfo = PdfInfo(opts.input_file, progbar=opts.progress_bar) assert pdfinfo is not None assert tqdmpatch.called @@ -167,20 +190,21 @@ def test_no_progress_bar(progress_bar, resources): def test_language_warning(caplog): opts = make_opts(language=None) + plugin_manager = get_plugin_manager(opts.plugins) caplog.set_level(logging.DEBUG) with patch( 'ocrmypdf._validation.locale.getlocale', return_value=('en_US', 'UTF-8') ): - vd.check_options_languages(opts) - assert opts.language == ['eng'] + vd.check_options_languages(opts, plugin_manager) + assert opts.languages == {'eng'} assert '' in caplog.text opts = make_opts(language=None) with patch( 'ocrmypdf._validation.locale.getlocale', return_value=('fr_FR', 'UTF-8') ): - vd.check_options_languages(opts) - assert opts.language == ['eng'] + vd.check_options_languages(opts, plugin_manager) + assert opts.languages == {'eng'} assert 'assuming --language' in caplog.text @@ -237,3 +261,10 @@ def test_optional_program_recommended(caplog): (loglevel == logging.WARNING and "recommended" in msg) for _logger_name, loglevel, msg in caplog.record_tuples ) + + +def test_pagesegmode_warning(caplog): + opts = make_opts(tesseract_pagesegmode='0') + plugin_manager = get_plugin_manager(opts.plugins) + vd.check_options(opts, plugin_manager) + assert 'disable OCR' in caplog.text