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

Move Tesseract options validation into plugin

Browse Source
This commit is contained in:
James R. Barlow
2020-05-25 01:31:46 -07:00
parent 9bccff4f88
commit a0f9ca3a30
5 changed files with 55 additions and 40 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/ocrmypdf/_sync.py
-2
View File
@@ -323,8 +323,6 @@ def run_pipeline(options, *, plugin_manager, api=False):
original_filename, start_input_file, work_folder / 'origin.pdf', options
)
plugin_manager.hook.prepare(options=options)
# Gather pdfinfo and create context
pdfinfo = get_pdfinfo(
origin_pdf,
src/ocrmypdf/_validation.py
+1 -30
View File
@@ -123,18 +123,6 @@ def check_options_output(options):
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'
@@ -277,18 +265,6 @@ 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."
)
if options.tesseract_pagesegmode in (0, 2):
log.warning(
"The --tesseract-pagesegmode argument you select will disable OCR. "
"This may cause processing to fail."
)
def check_options_metadata(options):
@@ -322,6 +298,7 @@ def check_options(options, plugin_manager):
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
@@ -464,12 +441,6 @@ def report_output_file_size(options, input_file, output_file):
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',
src/ocrmypdf/builtin_plugins/tesseract_ocr.py
+40 -1
View File
@@ -15,11 +15,16 @@
# 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
from ocrmypdf import hookimpl
from ocrmypdf.cli import numeric
from ocrmypdf.exec import tesseract
from ocrmypdf.exceptions import MissingDependencyError
from ocrmypdf.exec import check_external_program, tesseract
from ocrmypdf.pluginspec import OcrEngine
log = logging.getLogger(__name__)
@hookimpl
def add_options(parser):
@@ -76,6 +81,40 @@ def add_options(parser):
)
@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(
options.tesseract_env, set(options.language)
):
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(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."
)
if options.tesseract_pagesegmode in (0, 2):
log.warning(
"The --tesseract-pagesegmode argument you select will disable OCR. "
"This may cause processing to fail."
)
class TesseractOcrEngine(OcrEngine):
@staticmethod
def version():
src/ocrmypdf/pluginspec.py
+1 -1
View File
@@ -39,7 +39,7 @@ def add_options(parser: ArgumentParser) -> None:
@hookspec
def prepare(options: Namespace) -> None:
def check_options(options: Namespace) -> None:
"""Called to notify a plugin that a file will be processed.
The plugin may modify the *options*. All objects that are in options must
tests/test_validation.py
+13 -6
View File
@@ -69,7 +69,8 @@ def test_old_tesseract_error():
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():
@@ -96,12 +97,17 @@ 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
@@ -223,5 +229,6 @@ def test_version_comparison():
def test_pagesegmode_warning(caplog):
opts = make_opts(tesseract_pagesegmode='0')
vd.check_options_advanced(opts)
plugin_manager = get_plugin_manager(opts.plugins)
vd.check_options(opts, plugin_manager)
assert 'disable OCR' in caplog.text