Sort imports

misc/webservice.py

@@ -23,21 +23,22 @@ to emphasize that SaaS deployments should make sure they comply with
 Ghostscript's license as well as OCRmyPDF's.
 """
 
+import os
+import shlex
+from subprocess import PIPE, run
+from tempfile import TemporaryDirectory
+
 from flask import (
     Flask,
     Response,
-    flash,
-    request,
-    redirect,
-    url_for,
     abort,
+    flash,
+    redirect,
+    request,
     send_from_directory,
+    url_for,
 )
-from subprocess import run, PIPE
-from tempfile import TemporaryDirectory
 from werkzeug.utils import secure_filename
-import os
-import shlex
 
 app = Flask(__name__)
 app.secret_key = "secret"

src/ocrmypdf/__init__.py

@@ -24,24 +24,19 @@ __version__ = pkg_resources.get_distribution('ocrmypdf').version
 
 VERSION = __version__
 
+from . import helpers, hocrtransform, leptonica, pdfa, pdfinfo
+from .api import Verbosity, configure_logging, ocr
 from .exceptions import (
-    ExitCode,
     BadArgsError,
-    PdfMergeFailedError,
-    MissingDependencyError,
-    UnsupportedImageFormatError,
     DpiError,
-    OutputFileAccessError,
-    PriorOcrFoundError,
-    InputFileError,
-    SubprocessOutputError,
     EncryptedPdfError,
+    ExitCode,
+    InputFileError,
+    MissingDependencyError,
+    OutputFileAccessError,
+    PdfMergeFailedError,
+    PriorOcrFoundError,
+    SubprocessOutputError,
     TesseractConfigError,
+    UnsupportedImageFormatError,
 )
-
-from . import helpers
-from . import hocrtransform
-from . import leptonica
-from . import pdfa
-from . import pdfinfo
-from .api import ocr, configure_logging, Verbosity

src/ocrmypdf/__main__.py

@@ -21,12 +21,12 @@ import os
 import sys
 
 from . import __version__
-from .cli import parser
-from .api import configure_logging, Verbosity
 from ._jobcontext import make_logger
 from ._sync import run_pipeline
 from ._validation import check_closed_streams, check_options
-from .exceptions import ExitCode, BadArgsError, MissingDependencyError
+from .api import Verbosity, configure_logging
+from .cli import parser
+from .exceptions import BadArgsError, ExitCode, MissingDependencyError
 
 
 def run(args=None):

src/ocrmypdf/_jobcontext.py

@@ -16,9 +16,9 @@
 # along with OCRmyPDF.  If not, see <http://www.gnu.org/licenses/>.
 
 import logging
+import os
 import shutil
 import sys
-import os
 
 
 class PicklableLoggerMixin:

src/ocrmypdf/_pipeline.py

@@ -22,18 +22,17 @@ from datetime import datetime, timezone
 from shutil import copyfileobj
 
 import img2pdf
-from PIL import Image
-
 import pikepdf
 from pikepdf.models.metadata import encode_pdf_date
+from PIL import Image
 
 from . import PROGRAM_NAME, VERSION, leptonica
 from .exceptions import (
     DpiError,
     EncryptedPdfError,
     InputFileError,
-    UnsupportedImageFormatError,
     PriorOcrFoundError,
+    UnsupportedImageFormatError,
 )
 from .exec import ghostscript, tesseract
 from .helpers import re_symlink

src/ocrmypdf/_sync.py

@@ -28,6 +28,7 @@ from tempfile import mkdtemp
 from tqdm import tqdm
 
 from . import __version__
+from ._graft import OcrGrafter
 from ._jobcontext import PDFContext, cleanup_working_files, make_logger
 from ._pipeline import (
     convert_to_pdfa,
@@ -59,7 +60,6 @@ from ._validation import (
     create_input_file,
     report_output_file_size,
 )
-from ._graft import OcrGrafter
 from .exceptions import ExitCode, ExitCodeException
 from .exec import qpdf
 from .helpers import available_cpu_count

src/ocrmypdf/_validation.py

@@ -41,7 +41,7 @@ from .exec import (
     tesseract,
     unpaper,
 )
-from .helpers import is_file_writable, re_symlink, is_iterable_notstr, monotonic
+from .helpers import is_file_writable, is_iterable_notstr, monotonic, re_symlink
 
 # -------------
 # External dependencies

src/ocrmypdf/api.py

@@ -23,9 +23,9 @@ from pathlib import Path
 
 from tqdm import tqdm
 
-from .cli import parser
 from ._sync import run_pipeline
 from ._validation import check_options
+from .cli import parser
 
 
 class TqdmConsole:

src/ocrmypdf/exec/__init__.py

@@ -21,9 +21,10 @@ import logging
 import os
 import re
 import sys
-from subprocess import run, STDOUT, PIPE, CalledProcessError
-from ..exceptions import MissingDependencyError, ExitCode
 from collections.abc import Mapping
+from subprocess import PIPE, STDOUT, CalledProcessError, run
+
+from ..exceptions import ExitCode, MissingDependencyError
 
 log = logging.Logger(__name__)
 

src/ocrmypdf/exec/ghostscript.py

@@ -28,7 +28,6 @@ from PIL import Image
 from ..exceptions import SubprocessOutputError
 from . import get_version
 
-
 gslog = logging.getLogger()
 
 

src/ocrmypdf/exec/jbig2enc.py

@@ -18,8 +18,8 @@
 from functools import lru_cache
 from subprocess import PIPE, run
 
-from . import get_version
 from ..exceptions import MissingDependencyError
+from . import get_version
 
 
 @lru_cache(maxsize=1)

src/ocrmypdf/exec/pngquant.py

@@ -21,8 +21,8 @@ from tempfile import NamedTemporaryFile
 
 from PIL import Image
 
-from . import get_version
 from ..exceptions import MissingDependencyError
+from . import get_version
 
 
 @lru_cache(maxsize=1)

src/ocrmypdf/exec/tesseract.py

@@ -24,13 +24,13 @@ from functools import lru_cache
 from os import fspath
 from subprocess import PIPE, STDOUT, CalledProcessError, TimeoutExpired, run
 
-from . import get_version
 from ..exceptions import (
     MissingDependencyError,
-    TesseractConfigError,
     SubprocessOutputError,
+    TesseractConfigError,
 )
 from ..helpers import page_number
+from . import get_version
 
 OrientationConfidence = namedtuple('OrientationConfidence', ('angle', 'confidence'))
 

src/ocrmypdf/exec/unpaper.py

@@ -26,11 +26,11 @@ from functools import lru_cache
 from subprocess import PIPE, STDOUT, CalledProcessError
 from tempfile import TemporaryDirectory
 
-from . import get_version
-from ..exceptions import MissingDependencyError, SubprocessOutputError
-
 from PIL import Image
 
+from ..exceptions import MissingDependencyError, SubprocessOutputError
+from . import get_version
+
 
 @lru_cache(maxsize=1)
 def version():

src/ocrmypdf/filters.py

@@ -1,5 +1,5 @@
-from PIL import Image
 import PIL.ImageOps
+from PIL import Image
 
 
 def invert(im):

src/ocrmypdf/optimize.py

@@ -22,15 +22,15 @@ from collections import defaultdict
 from os import fspath
 from pathlib import Path
 
+import pikepdf
+from pikepdf import Dictionary, Name
 from PIL import Image
 from tqdm import tqdm
-import pikepdf
-from pikepdf import Name, Dictionary
 
 from . import leptonica
 from ._jobcontext import PDFContext
-from .exec import jbig2enc, pngquant
 from .exceptions import OutputFileAccessError
+from .exec import jbig2enc, pngquant
 from .helpers import re_symlink
 
 DEFAULT_JPEG_QUALITY = 75

src/ocrmypdf/pdfa.py

@@ -36,9 +36,8 @@ from binascii import hexlify
 from pathlib import Path
 from string import Template
 
-import pkg_resources
-
 import pikepdf
+import pkg_resources
 
 ICC_PROFILE_RELPATH = 'data/sRGB.icc'
 

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 <http://www.gnu.org/licenses/>.
 
-from .info import PdfInfo, Colorspace, Encoding
+from .info import Colorspace, Encoding, PdfInfo

src/ocrmypdf/pdfinfo/info.py

@@ -16,23 +16,24 @@
 # You should have received a copy of the GNU General Public License
 # along with OCRmyPDF.  If not, see <http://www.gnu.org/licenses/>.
 
+import logging
+import re
 from collections import namedtuple
 from decimal import Decimal
 from enum import Enum
-import logging
 from math import hypot, isclose
 from os import fspath
 from pathlib import Path
 from warnings import warn
-import re
 
-from pikepdf import PdfMatrix
 import pikepdf
+from pikepdf import PdfMatrix
 from tqdm import tqdm
 
+from ocrmypdf.exceptions import EncryptedPdfError, MissingDependencyError
+
 from . import ghosttext
 from .layout import get_page_analysis, get_text_boxes
-from ocrmypdf.exceptions import EncryptedPdfError, MissingDependencyError
 
 logger = logging.getLogger()
 

tests/cache/graph_ocred/__-l__deu__000001_ocr.png__000001_ocr_tess__pdf__txt/stderr.bin

@@ -0,0 +1 @@
+Tesseract Open Source OCR Engine v4.0.0 with Leptonica

tests/cache/graph_ocred/__-l__deu__000001_ocr.png__000001_ocr_tess__pdf__txt/txt.bin

@@ -0,0 +1,13 @@
+Portez ce vieux whisky au juge
+blond qui fume sur son Ile
+interieure, a cöte de l'alcöve
+ovoide, oU les büches se
+consument dans l'ätre, ce qui
+lui permet de penser & la
+caenogenese de |'etre dont il
+est question dans la cause
+ambigu& entendue a MoY, dans
+un capharnaüm qui, pense-t-il,
+diminue ca et la la qualite de son
+ceuvre.
+

tests/cache/manifest.jsonl

@@ -65,3 +65,4 @@
 {"tesseract_version": "tesseract 4.0.0  leptonica-1.78.0   libgif 5.1.4 : libjpeg 9c : libpng 1.6.37 : libtiff 4.0.10 : zlib 1.2.11 : libwebp 1.0.2 : libopenjp2 2.3.1  Found AVX2  Found AVX  Found SSE ", "platform": "Darwin-18.5.0-x86_64-i386-64bit", "python": "3.7.3", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/poster.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}
 {"tesseract_version": "tesseract 4.0.0  leptonica-1.78.0   libgif 5.1.4 : libjpeg 9c : libpng 1.6.37 : libtiff 4.0.10 : zlib 1.2.11 : libwebp 1.0.2 : libopenjp2 2.3.1  Found AVX2  Found AVX  Found SSE ", "platform": "Darwin-18.6.0-x86_64-i386-64bit", "python": "3.7.3", "argv_slug": "__-l__deu__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/cmyk.pdf", "args": ["-l", "deu", "-c", "textonly_pdf=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}
 {"tesseract_version": "tesseract 4.0.0  leptonica-1.78.0   libgif 5.1.4 : libjpeg 9c : libpng 1.6.37 : libtiff 4.0.10 : zlib 1.2.11 : libwebp 1.0.2 : libopenjp2 2.3.1  Found AVX2  Found AVX  Found SSE ", "platform": "Darwin-18.6.0-x86_64-i386-64bit", "python": "3.7.3", "argv_slug": "__-l__deu__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/lichtenstein.pdf", "args": ["-l", "deu", "-c", "textonly_pdf=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}
+{"tesseract_version": "tesseract 4.0.0  leptonica-1.78.0   libgif 5.1.4 : libjpeg 9c : libpng 1.6.37 : libtiff 4.0.10 : zlib 1.2.11 : libwebp 1.0.3 : libopenjp2 2.3.1  Found AVX2  Found AVX  Found SSE ", "platform": "Darwin-18.6.0-x86_64-i386-64bit", "python": "3.7.4", "argv_slug": "__-l__deu__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/graph_ocred.pdf", "args": ["-l", "deu", "-c", "textonly_pdf=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}