def process(patnum, doc, ref):
prepare_pdf_pages(patnum, doc)
locations = parser.parse_locations(ref)
basedir = config.data_dir + patnum + "/" + doc + "/"
basefile = config.data_dir + patnum + "/" + doc + "/" + doc
for type in locations:
if type in dir(processor):
for reference in locations[type]:
text = getattr(processor, type)(basefile, reference)
if text:
_store_reference_file(basedir, reference, text)
else:
print 'Could not evaluate ' + ref + ' in ' + basefile
print 'Processor: ' + str(processor) + ', ' + type
else:
'Processor ' + type + ' not implemented.'
def _scan_lines_for_references(indices, found, lines):
""" Scan lines for references.
"""
for pos, index in enumerate(indices):
try:
next_index = indices[pos + 1]
sublines = lines[index:next_index]
except:
sublines = lines[index:]
subtext = ""
for subline in sublines:
subtext += subline.lower()
for word in cfgdict.words:
subtext = subtext.replace(word, cfgdict.words[word])
refs = parser.parse_locations(subtext)
found[index]["references"] = refs
found[index]["abstract"] = parser.parse_for_word(subtext, cfgdict.abstract)
found[index]["figures"] = parser.parse_for(subtext, config.figures_parser, 'figures')
found[index]["claims"] = parser.parse_for(subtext, config.claims_parser, 'claims')
found[index]["paragraphs"] = parser.parse_for(subtext, config.paragraphs_parser, 'paragraphs')
found[index]["whole"] = parser.parse_for_word(subtext, cfgdict.whole_document)
found[index]["rawtext"] = subtext
return found