def domatch(self):
"""
Do the matching and counting based on lines in pdf's txt content file,
use tuple to store matches
use list to store matches within a line a file
each match consists of line number and pattern with context
The structure of match result looks like:
[[(pageno, lineno, location, context),...],...]
[[(1,1,location1, ...match...)],[(1,20,location2, ...match...),(2,32,location3, ...match...)]]
"""
import re
pattern = normalizer.normalize_pattern(self.pattern)
if self.icase:
pattern = re.compile(pattern, re.I)
else:
pattern = re.compile(pattern)
count = 0
results = []
for line in self.lines:
# line: (pageno, lineno, cotent)
pageno = line[0]
lineno = line[1]
linecontent = line[2]
if pattern.search(linecontent):
res = []
for match in pattern.finditer(linecontent):
count += 1
# context
if self.context > 0:
s = self.startindex(match.start(), self.context)
e = match.end() + self.context
else:
s = self.startposition(match.start(), linecontent)
e = self.endposition(match.end(), linecontent)
# location
if self.location:
# toc
if self.dictionary:
toc = TOC(self.file, path=self.dictionary)
else:
toc = TOC(self.file)
toc_dictionary = toc.gettoc_filter_by_dictionary(self.lines)
location = self.matchlocation(lineno, toc_dictionary)
else:
location = 'All'
res.append((pageno,lineno,location,linecontent[s:e]))
#print count
results.append(res)
self.count = count
return results
