def parse_transcript(self, filename):
meeting = PlenumMeeting()
meeting.save()
transcript = Transcript()
transcript.meeting = meeting
transcript.save()
matcher = Matcher()
text_lines = []
ordinal = count()
header = None
speaker = None
state = "cover"
for line in codecs.open(filename, "r", "utf-8"):
if state == "cover":
if line.strip() == u"תוכן עניינים":
cover = "".join(text_lines)
block = TranscriptBlock(transcript = transcript, ordinal = ordinal.next(), body = cover)
block.save()
text_lines = []
state = "toc"
elif re.search(ur"^<(?!הצע|החלט)", line):
text_lines = []
state = "speaker"
elif re.search(r"^ *<", line):
text_lines = []
state = "header"
elif matcher.search(ur"(הישיבה.*?)\s+$", line):
meeting.title = matcher.group(1)
print meeting.title
elif matcher.search(ur"(הישיבה.*?)\s+$", line):
meeting.title = matcher.group(1)
print meeting.title
elif state == "toc":
if re.search(ur"^<(?!הצע|החלט)", line):
text_lines = []
state = "speaker"
elif re.search(r"^ *<", line):
text_lines = []
state = "header"
elif state == "header":
if re.search(ur"^<(?!הצע|החלט)", line):
header = "".join(text_lines)
block = TranscriptBlock(transcript = transcript, ordinal = ordinal.next(), header = header)
block.save()
header = None
text_lines = []
state = "speaker"
elif state == "speaker":
speaker_line = "".join(text_lines).replace("\n", "")
if matcher.search(r'^<(.*?):?>$', speaker_line):
header = matcher.group(1)
try:
speaker = Person.objects.get(name__iexact=header)
except Person.DoesNotExist:
speaker = Person(name=header)
speaker.save()
text_lines = []
