#
# [sgm]
# the .sgm file
#
# [apf]
# the .apf file
#
# Output: Written into the standard output.
# The line number of the selected sentence and the sentence separated by a space.
import sys
import re
import bisect
from xml.dom.minidom import parse
import serif
import datetime
if __name__ == "__main__":
if len(sys.argv) != 4:
print "Usage: pick_recent_date.py [date] [text] [apf]"
sys.exit(1)
date = serif.convert_to_date(sys.argv[1])
text = serif.read_sgm(sys.argv[2])
data = serif.read_apf(text, sys.argv[3])
btimex = serif.find_best_timex(date, text, data)
if btimex:
start, end = text.expand(btimex.start, btimex.end)
line = serif.resolveCoref(text, data, start, end)
print text.find(start) - 1, line.encode("utf-8")
return clusters
if __name__=='__main__':
if len(sys.argv) != 4:
print "Usage: pick_recent_date.py [YYYY-MM-DD] [/path/to/cluster/file] [/path/to/apf/directory/]"
sys.exit(1)
date = serif.convert_to_date(sys.argv[1])
clusters = read_clusters(open(sys.argv[2], 'r'))
path = sys.argv[3]
for cluster in clusters:
btimex = None
for article in cluster:
article = article.decode('utf-8')
try:
text = serif.read_sgm(os.path.join(path, article + '.sentences.sgm'))
data = serif.read_apf(os.path.join(path, article + '.sentences.sgm.apf'))
old_btimex = btimex
btimex = serif.find_best_timex(date, text, data, btimex)
if btimex != old_btimex:
start, end = text.expand(btimex.start, btimex.end)
line = serif.resolveCoref(text, data, start, end)
print article.encode('utf-8')
except IOError:
print article.encode('utf-8') + ' # exempted due to error. check if the apf file exists.'
if btimex and line:
print line.encode('utf-8')
print
