def _pages_from_ranges(ranges):
if is_iterable_notstr(ranges):
return set(ranges)
pages = []
page_groups = ranges.replace(' ', '').split(',')
for g in page_groups:
if not g:
continue
try:
start, end = g.split('-')
except ValueError:
pages.append(int(g) - 1)
else:
try:
pages.extend(range(int(start) - 1, int(end)))
except ValueError:
raise BadArgsError("invalid page range")
if not monotonic(pages):
log.warning(
"List of pages to process contains duplicate pages, or pages that are "
"out of order")
if any(page < 0 for page in pages):
raise BadArgsError("pages refers to a page number less than 1")
log.debug("OCRing only these pages: %s", pages)
return set(pages)
def create_options(*, input_file: PathOrIO, output_file: PathOrIO,
parser: ArgumentParser, **kwargs):
cmdline = []
deferred = []
for arg, val in kwargs.items():
if val is None:
continue
# These arguments with special handling for which we bypass
# argparse
if arg in {'progress_bar', 'plugins'}:
deferred.append((arg, val))
continue
cmd_style_arg = arg.replace('_', '-')
# Booleans are special: add only if True, omit for False
if isinstance(val, bool):
if val:
cmdline.append(f"--{cmd_style_arg}")
continue
if is_iterable_notstr(val):
for elem in val:
cmdline.append(f"--{cmd_style_arg}")
cmdline.append(elem)
continue
# We have a parameter
cmdline.append(f"--{cmd_style_arg}")
if isinstance(val, (int, float)):
cmdline.append(str(val))
elif isinstance(val, str):
cmdline.append(val)
elif isinstance(val, Path):
cmdline.append(str(val))
else:
raise TypeError(f"{arg}: {val} ({type(val)})")
try:
cmdline.append(os.fspath(input_file))
except TypeError:
cmdline.append('stream://input_file')
try:
cmdline.append(os.fspath(output_file))
except TypeError:
cmdline.append('stream://output_file')
parser._api_mode = True
options = parser.parse_args(cmdline)
for keyword, val in deferred:
setattr(options, keyword, val)
if options.input_file == 'stream://input_file':
options.input_file = input_file
if options.output_file == 'stream://output_file':
options.output_file = output_file
return options
def _pages_from_ranges(ranges: str) -> Set[int]:
if is_iterable_notstr(ranges):
return set(ranges)
pages: List[int] = []
page_groups = ranges.replace(' ', '').split(',')
for g in page_groups:
if not g:
continue
try:
start, end = g.split('-')
except ValueError:
pages.append(int(g) - 1)
else:
try:
new_pages = list(range(int(start) - 1, int(end)))
if not new_pages:
raise BadArgsError(f"invalid page subrange '{start}-{end}'")
pages.extend(new_pages)
except ValueError:
raise BadArgsError("invalid page range") from None
if not pages:
raise BadArgsError(
f"The string of page ranges '{ranges}' did not contain any recognizable "
f"page ranges."
)
if not monotonic(pages):
log.warning(
"List of pages to process contains duplicate pages, or pages that are "
"out of order"
)
if any(page < 0 for page in pages):
raise BadArgsError("pages refers to a page number less than 1")
log.debug("OCRing only these pages: %s", pages)
return set(pages)