menekis/OCRmyPDF
Public Access
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Merge branch 'windows'

Browse Source
This commit is contained in:
James R. Barlow
2019-12-06 15:09:09 -08:00
parent fbf271a3ec 66bda3420a
commit 6b745d892f
34 changed files with 562 additions and 409 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/ocrmypdf/exec/__init__.py
+33 -2
View File
@@ -21,14 +21,35 @@ import logging
import os
import re
import sys
import shutil
from collections.abc import Mapping
from subprocess import PIPE, STDOUT, CalledProcessError, run
from subprocess import PIPE, STDOUT, CalledProcessError, run as subprocess_run
from ..exceptions import ExitCode, MissingDependencyError
log = logging.Logger(__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):
if not env:
env = os.environ
program = _get_program(args, env)
if os.name == 'nt' and program.lower().endswith('.py'):
args = [sys.executable, program] + args[1:]
else:
args = [program] + args[1:]
log.debug(args)
return subprocess_run(args, env=env, **kwargs)
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]
@@ -111,23 +132,33 @@ On RPM-based systems (Red Hat, Fedora), search for instructions on
installing the RPM for {program}.
'''
windows_install_advice = '''
If not already installed, install the Chocolatey package manager. Then use
a command prompt to install the missing package:
choco install {package}
'''
def _get_platform():
if sys.platform.startswith('freebsd'):
return 'freebsd'
elif sys.platform.startswith('linux'):
return 'linux'
elif sys.platform.startswith('win'):
return 'windows'
return sys.platform
def _error_trailer(program, package, **kwargs):
if isinstance(package, Mapping):
package = package[_get_platform()]
package = package.get(_get_platform(), program)
if _get_platform() == 'darwin':
log.info(osx_install_advice.format(**locals()))
elif _get_platform() == 'linux':
log.info(linux_install_advice.format(**locals()))
elif _get_platform() == 'windows':
log.info(windows_install_advice.format(**locals()))
def _error_missing_program(program, package, required_for, recommended):
src/ocrmypdf/exec/ghostscript.py
+118 -97
View File
@@ -19,24 +19,36 @@
import logging
import re
import os
import warnings
from contextlib import suppress
from functools import lru_cache
from io import BytesIO
from os import fspath
from shutil import copy
from subprocess import PIPE, STDOUT, run
from tempfile import NamedTemporaryFile
from pathlib import Path
from subprocess import PIPE, CalledProcessError
from shutil import which
from PIL import Image
from ..exceptions import SubprocessOutputError
from . import get_version
from ..exceptions import SubprocessOutputError, MissingDependencyError
from . import get_version, run
gslog = logging.getLogger()
GS = 'gs'
if os.name == 'nt':
GS = which('gswin64c')
if not GS:
GS = which('gswin32c')
if not GS:
raise MissingDependencyError("Ghostscript (gswin64c or gswin32c)")
GS = Path(GS).stem
@lru_cache(maxsize=1)
def version():
return get_version('gs')
return get_version(GS)
def jpeg_passthrough_available():
@@ -83,7 +95,7 @@ def extract_text(input_file, pageno=1):
args_gs = (
[
'gs',
GS,
'-dQUIET',
'-dSAFER',
'-dBATCH',
@@ -92,14 +104,15 @@ def extract_text(input_file, pageno=1):
'-dTextFormat=0',
]
+ pages
+ ['-o', '-', fspath(input_file)]
+ ['-o', '-', fspath(input_file), "-sstdout=%stderr"]
)
p = run(args_gs, stdout=PIPE, stderr=PIPE)
if p.returncode != 0:
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\n%s'
% (input_file, p.stdout.decode(), p.stderr.decode())
'Ghostscript text extraction failed\n%s\n%s'
% (input_file, e.stderr.decode(errors='replace'))
)
return p.stdout
@@ -141,54 +154,58 @@ def rasterize_pdf(
if not log:
log = gslog
with NamedTemporaryFile(delete=True) as tmp:
args_gs = (
[
'gs',
'-dQUIET',
'-dSAFER',
'-dBATCH',
'-dNOPAUSE',
f'-sDEVICE={raster_device}',
f'-dFirstPage={pageno}',
f'-dLastPage={pageno}',
f'-r{res[0]:f}x{res[1]:f}',
]
+ (['-dFILTERVECTOR'] if filter_vector else [])
+ [
'-o',
tmp.name,
'-dAutoRotatePages=/None', # Probably has no effect on raster
'-f',
fspath(input_file),
]
)
args_gs = (
[
GS,
'-dQUIET',
'-dSAFER',
'-dBATCH',
'-dNOPAUSE',
f'-sDEVICE={raster_device}',
f'-dFirstPage={pageno}',
f'-dLastPage={pageno}',
f'-r{res[0]:f}x{res[1]:f}',
]
+ (['-dFILTERVECTOR'] if filter_vector else [])
+ [
'-o',
'-',
'-sstdout=%stderr',
'-dAutoRotatePages=/None', # Probably has no effect on raster
'-f',
fspath(input_file),
]
)
log.debug(args_gs)
p = run(args_gs, stdout=PIPE, stderr=STDOUT, universal_newlines=True)
if _gs_error_reported(p.stdout):
log.error(p.stdout)
elif p.stdout:
log.debug(p.stdout)
log.debug(args_gs)
try:
p = run(args_gs, stdout=PIPE, stderr=PIPE, check=True)
except CalledProcessError as e:
with suppress(OSError):
Path(output_file).unlink() # no unfinished files
log.error(e.stderr.decode(errors='replace'))
raise SubprocessOutputError('Ghostscript rasterizing failed')
else:
stderr = p.stderr.decode(errors='replace')
if _gs_error_reported(stderr):
log.error(stderr)
elif stderr:
log.debug(stderr)
if p.returncode != 0:
raise SubprocessOutputError('Ghostscript rasterizing failed')
tmp.seek(0)
with Image.open(tmp) as im:
if rotation is not None:
log.debug("Rotating output by %i", rotation)
# rotation is a clockwise angle and Image.ROTATE_* is
# counterclockwise so this cancels out the rotation
if rotation == 90:
im = im.transpose(Image.ROTATE_90)
elif rotation == 180:
im = im.transpose(Image.ROTATE_180)
elif rotation == 270:
im = im.transpose(Image.ROTATE_270)
if rotation % 180 == 90:
page_dpi = page_dpi[1], page_dpi[0]
im.save(fspath(output_file), dpi=page_dpi)
with Image.open(BytesIO(p.stdout)) as im:
if rotation is not None:
log.debug("Rotating output by %i", rotation)
# rotation is a clockwise angle and Image.ROTATE_* is
# counterclockwise so this cancels out the rotation
if rotation == 90:
im = im.transpose(Image.ROTATE_90)
elif rotation == 180:
im = im.transpose(Image.ROTATE_180)
elif rotation == 270:
im = im.transpose(Image.ROTATE_270)
if rotation % 180 == 90:
page_dpi = page_dpi[1], page_dpi[0]
im.save(fspath(output_file), dpi=page_dpi)
def generate_pdfa(
@@ -256,37 +273,48 @@ def generate_pdfa(
# https://bugs.ghostscript.com/show_bug.cgi?id=699216
compression_args.append('-dPassThroughJPEGImages=false')
with NamedTemporaryFile(delete=True) as gs_pdf:
# nb no need to specify ProcessColorModel when ColorConversionStrategy
# is set; see:
# https://bugs.ghostscript.com/show_bug.cgi?id=699392
args_gs = (
[
"gs",
"-dQUIET",
"-dBATCH",
"-dNOPAUSE",
"-dSAFER",
"-dCompatibilityLevel=" + str(pdf_version),
"-sDEVICE=pdfwrite",
"-dAutoRotatePages=/None",
"-sColorConversionStrategy=" + strategy,
]
+ compression_args
+ [
"-dJPEGQ=95",
"-dPDFA=" + pdfa_part,
"-dPDFACompatibilityPolicy=1",
"-sOutputFile=" + gs_pdf.name,
]
)
args_gs.extend(fspath(s) for s in pdf_pages) # Stringify Path objs
log.debug(args_gs)
p = run(args_gs, stdout=PIPE, stderr=STDOUT, universal_newlines=True)
if _gs_error_reported(p.stdout):
log.error(p.stdout)
elif 'overprint mode not set' in p.stdout:
# nb no need to specify ProcessColorModel when ColorConversionStrategy
# is set; see:
# https://bugs.ghostscript.com/show_bug.cgi?id=699392
args_gs = (
[
GS,
"-dQUIET",
"-dBATCH",
"-dNOPAUSE",
"-dSAFER",
"-dCompatibilityLevel=" + str(pdf_version),
"-sDEVICE=pdfwrite",
"-dAutoRotatePages=/None",
"-sColorConversionStrategy=" + strategy,
]
+ compression_args
+ [
"-dJPEGQ=95",
"-dPDFA=" + pdfa_part,
"-dPDFACompatibilityPolicy=1",
"-o",
"-",
"-sstdout=%stderr",
]
)
args_gs.extend(fspath(s) for s in pdf_pages) # Stringify Path objs
log.debug(args_gs)
try:
with Path(output_file).open('wb') as output:
p = run(args_gs, stdout=output, stderr=PIPE, check=True)
except CalledProcessError as e:
# Ghostscript does not change return code when it fails to create
# PDF/A - check PDF/A status elsewhere
with suppress(OSError):
Path(output_file).unlink()
log.error(e.stderr.decode(errors='replace'))
raise SubprocessOutputError('Ghostscript PDF/A rendering failed')
else:
stderr = p.stderr.decode('utf-8', errors='replace')
if _gs_error_reported(stderr):
log.error(stderr)
elif 'overprint mode not set' in stderr:
# Unless someone is going to print PDF/A documents on a
# magical sRGB printer I can't see the removal of overprinting
# being a problem....
@@ -295,11 +323,4 @@ def generate_pdfa(
"input file to complete PDF/A conversion. "
)
else:
log.debug(p.stdout)
if p.returncode == 0:
# Ghostscript does not change return code when it fails to create
# PDF/A - check PDF/A status elsewhere
copy(gs_pdf.name, fspath(output_file))
else:
raise SubprocessOutputError('Ghostscript PDF/A rendering failed')
log.debug(stderr)
src/ocrmypdf/exec/jbig2enc.py
+2 -2
View File
@@ -18,10 +18,10 @@
"""Interface to jbig2 executable"""
from functools import lru_cache
from subprocess import PIPE, run
from subprocess import PIPE
from ..exceptions import MissingDependencyError
from . import get_version
from . import get_version, run
@lru_cache(maxsize=1)
src/ocrmypdf/exec/qpdf.py
+2 -2
View File
@@ -19,9 +19,9 @@
from functools import lru_cache
from os import fspath
from subprocess import PIPE, STDOUT, CalledProcessError, run
from subprocess import PIPE, STDOUT, CalledProcessError
from . import get_version
from . import get_version, run
@lru_cache(maxsize=1)
src/ocrmypdf/exec/tesseract.py
+8 -12
View File
@@ -23,15 +23,15 @@ from collections import namedtuple
from contextlib import suppress
import logging
from os import fspath
from subprocess import PIPE, STDOUT, CalledProcessError, TimeoutExpired, run
from subprocess import PIPE, STDOUT, CalledProcessError, TimeoutExpired
from ..exceptions import (
MissingDependencyError,
SubprocessOutputError,
TesseractConfigError,
)
from ..helpers import page_number
from . import get_version
from ..helpers import page_number, safe_symlink
from . import get_version, run
OrientationConfidence = namedtuple('OrientationConfidence', ('angle', 'confidence'))
@@ -128,9 +128,10 @@ def languages(tesseract_env=None):
except CalledProcessError as e:
raise MissingDependencyError(lang_error(e.output)) from e
for line in output.splitlines():
if line.startswith('Error'):
raise MissingDependencyError(lang_error(output))
header, *rest = output.splitlines()
if not header.startswith('List of available languages'):
raise MissingDependencyError(lang_error(output))
return set(lang.strip() for lang in rest)
@@ -194,12 +195,7 @@ def tesseract_log_output(mainlog, stdout, input_file):
try:
text = stdout.decode()
except UnicodeDecodeError:
log.error(
"Tesseract's output was not utf-8. "
"This usually means Tesseract's language packs do not match "
"the installed version of Tesseract."
)
text = stdout.decode('utf-8', 'backslashreplace')
text = stdout.decode('utf-8', 'ignore')
lines = text.splitlines()
for line in lines:
@@ -325,7 +321,7 @@ def use_skip_page(text_only, skip_pdf, output_pdf, output_text):
# Substitute a "skipped page"
with suppress(FileNotFoundError):
os.remove(output_pdf) # In case it was partially created
os.symlink(skip_pdf, output_pdf)
safe_symlink(skip_pdf, output_pdf)
return
# Or normally, just write a 0 byte file to the output to indicate a skip
src/ocrmypdf/exec/unpaper.py
+2 -3
View File
@@ -22,7 +22,6 @@
import os
import shlex
import subprocess
from functools import lru_cache
from subprocess import PIPE, STDOUT, CalledProcessError
from tempfile import TemporaryDirectory
@@ -30,7 +29,7 @@ from tempfile import TemporaryDirectory
from PIL import Image
from ..exceptions import MissingDependencyError, SubprocessOutputError
from . import get_version
from . import get_version, run as external_run
@lru_cache(maxsize=1)
@@ -77,7 +76,7 @@ def run(input_file, output_file, dpi, log, mode_args):
# their unpaper arguments (whether intentionally or otherwise)
args_unpaper.extend([input_pnm, output_pnm])
try:
proc = subprocess.run(
proc = external_run(
args_unpaper,
check=True,
close_fds=True,