def pdfrefs(in_fp):
"""Extract references from a pdf file"""
xml = pdf2xml(in_fp)
ref_re = re.compile('\[(\d+)\] (.*)')
in_ref = False
refs = dict()
key = None
for t in xml.iter('text'):
if not t.text:
continue
if 'references' in t.text.lower():
in_ref = True
continue
if not in_ref:
continue
ref_m = ref_re.match(t.text)
if ref_m:
key, text = int(ref_m.group(1)), ref_m.group(2)
refs[key] = [text]
else:
refs[key].append(t.text)
refs = {k: ' '.join(v) for k, v in refs.items()}
return refs
def pdftitle(in_fp):
xml = pdf2xml(in_fp)
fontspecs = xml.findall("./page[@number='1']/fontspec")
max_size = -1
max_i = None
for f in fontspecs:
i = int(f.attrib['id'])
size = int(f.attrib['size'])
if size >= max_size:
max_size = size
max_i = i
titles = xml.xpath(".//text[@font='{0}']/descendant-or-self::text()".format(max_i))
title = ' '.join(titles).strip()
title = re.sub('\s+', ' ', title)
return title
