def load_figures_json(filename: str) -> Dict[str, List[Figure]]:
d = file_util.read_json(filename)
res = {
page: [Figure.from_dict(dict_fig) for dict_fig in page_dicts]
for (page, page_dicts) in d.items()
}
return res
def extract(self, pdf_path, output_dir, use_cache=True):
"""Return results from extracting a PDF with pdffigures2.
:param str pdf_path: path to the PDF to extract.
:param str output_dir: path to the output directory.
:param bool use_cache: whether or not to use cached data from
disk if it's available.
:returns: results from running pdffigures2 on the PDF.
"""
pdffigures_dir = os.path.join(output_dir, 'pdffigures/')
if not os.path.exists(pdffigures_dir):
os.makedirs(pdffigures_dir)
success_file_path = os.path.join(pdffigures_dir, '_SUCCESS')
error_file_path = os.path.join(pdffigures_dir, '_ERROR')
pdffigures_jar_path = file_util.cache_file(
settings.PDFFIGURES_JAR_PATH)
if not os.path.exists(success_file_path) or not use_cache:
try:
subprocess.check_call(
'java'
' -jar {pdffigures_jar_path}'
' --figure-data-prefix {pdffigures_dir}'
' --save-regionless-captions'
' {pdf_path}'.format(
pdffigures_jar_path=pdffigures_jar_path,
pdf_path=pdf_path,
pdffigures_dir=pdffigures_dir),
shell=True)
except subprocess.CalledProcessError:
# Writes an error file to indicate that a problem occured
with open(error_file_path, 'w') as f_out:
f_out.write('')
# return nothing
return
# add a success file to verify that the operation completed
with open(success_file_path, 'w') as f_out:
f_out.write('')
return file_util.read_json(
os.path.join(pdffigures_dir,
os.path.basename(pdf_path)[:-4] + '.json'))
def _get_hypes(self) -> dict:
return file_util.read_json(os.path.join(self.save_dir, 'hypes.json'))
def _get_hypes(self) -> dict:
return file_util.read_json(self.save_dir + 'hypes.json')