v11.4.5 release notes

Co-authored-by: Jonas Winkler <jonas.winkler@jpwinkler.de>

.coveragerc

@@ -1,24 +0,0 @@
-[paths]
-source =
-	src
-	*/site-packages
-
-[run]
-branch = true
-parallel = true
-concurrency =
-	thread
-	multiprocessing
-source =
-	src/ocrmypdf
-
-[report]
-exclude_lines =
-	pragma: no cover
-	def __repr__
-	raise AssertionError
-	raise NotImplementedError
-	if 0:
-	if False:
-	if __name__ == .__main__.:
-	if TYPE_CHECKING:

docs/api.rst

@@ -56,6 +56,11 @@ Programs that call ``ocrmypdf.ocr()`` should also install a SIGBUS signal
 handler (except on Windows), to raise an exception if access to a memory
 mapped file fails. OCRmyPDF may use memory mapping.
 
+``ocrmypdf.ocr()`` will take a threading lock to prevent multiple runs of itself
+in the same Python interpreter process. This is not thread-safe, because of how
+OCRmyPDF's plugins and Python's library import system work. If you need to parallelize
+OCRmyPDF, use processes.
+
 .. warning::
 
     On Windows and macOS, the script that calls ``ocrmypdf.ocr()`` must be

docs/release_notes.rst

@@ -12,6 +12,21 @@ may be unreliable. Use the API to depend on precise behavior.
 The public API may be useful in scripts that launch OCRmyPDF processes or that
 wish to use some of its features for working with PDFs.
 
+v11.4.5
+=======
+
+-  Fixed an issue where files may not be closed when the API is used.
+-  Improved ``setup.cfg`` with better settings for test coverage.
+
+v11.4.4
+=======
+
+-  Fixed ``AttributeError: 'NoneType' object has no attribute 'userunit'``, issue #700,
+   related to OCRmyPDF not properly forwarded an error message from pdfminer.six.
+-  Adjusted typing of some arguments.
+-  ``ocrmypdf.ocr`` now takes a ``threading.Lock`` for reasons outlined in the
+   documentation.
+
 v11.4.3
 =======
 

setup.cfg

@@ -15,6 +15,8 @@ filterwarnings =
 	ignore:.*XMLParser.*:DeprecationWarning
 markers =
 	slow
+addopts =
+	-n auto
 
 [isort]
 multi_line_output=3
@@ -27,3 +29,30 @@ known_third_party = PIL,_cffi_backend,cffi,flask,img2pdf,pdfminer,pikepdf,pkg_re
 
 [metadata]
 license_file = LICENSE
+
+[coverage:paths]
+source =
+	src/
+
+[coverage:run]
+branch = true
+parallel = true
+concurrency = multiprocessing
+source =
+	src/ocrmypdf
+
+[coverage:report]
+# Regexes for lines to exclude from consideration
+exclude_lines =
+    # Have to re-enable the standard pragma
+    pragma: no cover
+
+    # Don't complain if tests don't hit defensive assertion code:
+    raise AssertionError
+    raise NotImplementedError
+
+    # Don't complain if non-runnable code isn't run:
+    if 0:
+	if False:
+    if __name__ == .__main__.:
+	if TYPE_CHECKING:

src/ocrmypdf/_concurrent.py

@@ -109,15 +109,11 @@ def exec_progress_pool(
         )
         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
+            for result in results:
+                if task_finished:
+                    task_finished(result, pbar)
+                else:
+                    pbar.update()
         except KeyboardInterrupt:
             # Terminate pool so we exit instantly
             pool.terminate()

src/ocrmypdf/_exec/tesseract.py

@@ -14,7 +14,7 @@ from collections import namedtuple
 from os import fspath
 from pathlib import Path
 from subprocess import PIPE, STDOUT, CalledProcessError, TimeoutExpired
-from typing import List
+from typing import List, Optional
 
 from PIL import Image
 
@@ -118,7 +118,7 @@ def get_languages():
     return set(lang.strip() for lang in rest)
 
 
-def tess_base_args(langs: List[str], engine_mode: int) -> List[str]:
+def tess_base_args(langs: List[str], engine_mode: Optional[int]) -> List[str]:
     args = ['tesseract']
     if langs:
         args.extend(['-l', '+'.join(langs)])
@@ -127,7 +127,7 @@ def tess_base_args(langs: List[str], engine_mode: int) -> List[str]:
     return args
 
 
-def get_orientation(input_file: Path, engine_mode: int, timeout: float):
+def get_orientation(input_file: Path, engine_mode: Optional[int], timeout: float):
     args_tesseract = tess_base_args(['osd'], engine_mode) + [
         '--psm',
         '0',

src/ocrmypdf/_sync.py

@@ -338,6 +338,8 @@ def run_pipeline(options, *, plugin_manager, api=False):
         and not api
     ):
         # Debug log for command line interface only with verbose output
+        # See https://github.com/pytest-dev/pytest/issues/5502 for why we skip this
+        # when pytest is running
         debug_log_handler = configure_debug_logging(Path(work_folder) / "debug.log")
 
     pikepdf_enable_mmap()

src/ocrmypdf/api.py

@@ -8,6 +8,7 @@
 import logging
 import os
 import sys
+import threading
 from enum import IntEnum
 from io import IOBase
 from pathlib import Path
@@ -30,6 +31,8 @@ except ModuleNotFoundError:
 StrPath = Union[os.PathLike, AnyStr]
 PathOrIO = Union[BinaryIO, StrPath]
 
+_api_lock = threading.Lock()
+
 
 class Verbosity(IntEnum):
     """Verbosity level for configure_logging."""
@@ -306,12 +309,18 @@ def ocr(  # pylint: disable=unused-argument
 
     parser = get_parser()
     create_options_kwargs['parser'] = parser
-    plugin_manager = get_plugin_manager(plugins)
-    plugin_manager.hook.add_options(parser=parser)  # pylint: disable=no-member
 
-    if 'verbose' in kwargs:
-        warn("ocrmypdf.ocr(verbose=) is ignored. Use ocrmypdf.configure_logging().")
+    with _api_lock:
+        # We can't allow multiple ocrmypdf.ocr() threads to run in parallel, because
+        # they might install different plugins, and generally speaking we have areas
+        # of code that use global state.
 
-    options = create_options(**create_options_kwargs)
-    check_options(options, plugin_manager)
-    return run_pipeline(options=options, plugin_manager=plugin_manager, api=True)
+        plugin_manager = get_plugin_manager(plugins)
+        plugin_manager.hook.add_options(parser=parser)  # pylint: disable=no-member
+
+        if 'verbose' in kwargs:
+            warn("ocrmypdf.ocr(verbose=) is ignored. Use ocrmypdf.configure_logging().")
+
+        options = create_options(**create_options_kwargs)
+        check_options(options, plugin_manager)
+        return run_pipeline(options=options, plugin_manager=plugin_manager, api=True)

src/ocrmypdf/optimize.py

@@ -498,7 +498,7 @@ def transcode_pngs(
 
 
 @deprecated
-def rewrite_png_as_g4(pike: Pdf, im_obj: Object, compdata) -> None:
+def rewrite_png_as_g4(pike: Pdf, im_obj: Object, compdata) -> None:  # pragma: no cover
     im_obj.BitsPerComponent = 1
     im_obj.Width = compdata.w
     im_obj.Height = compdata.h
@@ -519,7 +519,7 @@ def rewrite_png_as_g4(pike: Pdf, im_obj: Object, compdata) -> None:
 
 
 @deprecated
-def rewrite_png(pike: Pdf, im_obj: Object, compdata) -> None:
+def rewrite_png(pike: Pdf, im_obj: Object, compdata) -> None:  # pragma: no cover
     # 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

src/ocrmypdf/pdfinfo/info.py

@@ -22,7 +22,7 @@ import pikepdf
 from pikepdf import Object, Pdf, PdfMatrix
 
 from ocrmypdf._concurrent import exec_progress_pool
-from ocrmypdf.exceptions import EncryptedPdfError
+from ocrmypdf.exceptions import EncryptedPdfError, InputFileError
 from ocrmypdf.helpers import Resolution, available_cpu_count, pikepdf_enable_mmap
 from ocrmypdf.pdfinfo.layout import get_page_analysis, get_text_boxes
 
@@ -598,6 +598,8 @@ def _pdf_pageinfo_concurrent(
 
     def update_pageinfo(result, pbar):
         page = result
+        if not page:
+            raise InputFileError("Could read a page in the PDF")
         pages[page.pageno] = page
         pbar.update()
 

src/ocrmypdf/pdfinfo/layout.py

@@ -21,7 +21,7 @@ from pdfminer.pdffont import PDFSimpleFont, PDFUnicodeNotDefined
 from pdfminer.pdfpage import PDFPage
 from pdfminer.utils import bbox2str, matrix2str
 
-from ocrmypdf.exceptions import EncryptedPdfError
+from ocrmypdf.exceptions import EncryptedPdfError, InputFileError
 
 STRIP_NAME = re.compile(r'[0-9]+')
 
@@ -236,8 +236,13 @@ def get_page_analysis(infile, pageno, pscript5_mode):
 
     try:
         with Path(infile).open('rb') as f:
-            page = PDFPage.get_pages(f, pagenos=[pageno], maxpages=0)
-            interp.process_page(next(page))
+            page_iter = PDFPage.get_pages(f, pagenos=[pageno], maxpages=0)
+            page = next(page_iter, None)
+            if page is None:
+                raise InputFileError(
+                    f"pdfminer could not process page {pageno} (counting from 0)."
+                )
+            interp.process_page(page)
     except PDFTextExtractionNotAllowed as e:
         raise EncryptedPdfError() from e
     finally:

src/ocrmypdf/subprocess/__init__.py

@@ -77,20 +77,19 @@ def run_polling_stderr(args, *, callback, check=False, env=None, **kwargs):
     args, env, process_log, text = _fix_process_args(args, env, kwargs)
     assert text, "Must use text=True"
 
-    proc = Popen(args, env=env, **kwargs)
+    with Popen(args, env=env, **kwargs) as proc:
+        lines = []
+        while proc.poll() is None:
+            for msg in iter(proc.stderr.readline, ''):
+                if process_log.isEnabledFor(logging.DEBUG):
+                    process_log.debug(msg.strip())
+                callback(msg)
+                lines.append(msg)
+        stderr = ''.join(lines)
 
-    lines = []
-    while proc.poll() is None:
-        for msg in iter(proc.stderr.readline, ''):
-            if process_log.isEnabledFor(logging.DEBUG):
-                process_log.debug(msg.strip())
-            callback(msg)
-            lines.append(msg)
-    stderr = ''.join(lines)
-
-    if check and proc.returncode != 0:
-        raise CalledProcessError(proc.returncode, args, output=None, stderr=stderr)
-    return CompletedProcess(args, proc.returncode, None, stderr=stderr)
+        if check and proc.returncode != 0:
+            raise CalledProcessError(proc.returncode, args, output=None, stderr=stderr)
+        return CompletedProcess(args, proc.returncode, None, stderr=stderr)
 
 
 def _fix_process_args(args, env, kwargs):

tests/conftest.py

@@ -137,16 +137,7 @@ def run_ocrmypdf(input_file, output_file, *args, text=True):
         + [str(input_file), str(output_file)]
     )
 
-    # Tell subprocess where to find coverage.py configuration
-    # This has no unless except when coverage is running
-    # Details: https://coverage.readthedocs.io/en/coverage-5.0/subprocess.html
-    coverage_rc = Path(__file__).parent.parent / '.coveragerc'
     env = os.environ.copy()
-    if coverage_rc.exists():
-        env['COVERAGE_PROCESS_START'] = os.fspath(coverage_rc)
-    elif not running_in_docker():
-        assert False, "could not find .coveragerc"
-
     p = run(
         p_args,
         stdout=PIPE,

tests/test_pdfinfo.py

@@ -15,7 +15,11 @@ from PIL import Image
 from reportlab.pdfgen.canvas import Canvas
 
 from ocrmypdf import pdfinfo
+from ocrmypdf.exceptions import InputFileError
 from ocrmypdf.pdfinfo import Colorspace, Encoding
+from ocrmypdf.pdfinfo.layout import PDFPage
+
+run_ocrmypdf_api = pytest.helpers.run_ocrmypdf_api
 
 # pylint: disable=protected-access
 
@@ -179,3 +183,18 @@ def test_stack_abuse():
     with pytest.warns(None):
         with pytest.raises(RuntimeError):
             pdfinfo.info._interpret_contents(stream)
+
+
+def test_pages_issue700(monkeypatch, resources):
+    def get_no_pages(*args, **kwargs):
+        return iter([])
+
+    monkeypatch.setattr(PDFPage, 'get_pages', get_no_pages)
+
+    with pytest.raises(InputFileError, match="pdfminer"):
+        pdfinfo.PdfInfo(
+            resources / 'cardinal.pdf',
+            detailed_analysis=True,
+            progbar=False,
+            max_workers=1,
+        )