def note_names(note):
from content_analysis import count_regex_matches
import content_analysis
global _names
global name_stop_list
if _names is None:
_names = list(set([x.lower() for x in names.read() if len(x) > 2 and (x.lower() not in name_stop_list)]))
rnames = [ "(^|\W+)%s($|\W+)" % nhit for nhit in [name for name in _names if name in note["contents"]]]
if len(rnames) > 0:
hits = [(n,count_regex_matches(n,note["contents"])) for n in rnames]
#print hits
hits = content_analysis.make_feature("names", reduce(lambda x,y: x + y, [count_regex_matches(n,note["contents"]) for n in rnames]))
return hits
return content_analysis.make_feature("names", 0)
def note_names(note):
from content_analysis import count_regex_matches
import content_analysis
global _names
global name_stop_list
if _names is None:
_names = list(
set([
x.lower() for x in names.read()
if len(x) > 2 and (x.lower() not in name_stop_list)
]))
rnames = [
"(^|\W+)%s($|\W+)" % nhit
for nhit in [name for name in _names if name in note["contents"]]
]
if len(rnames) > 0:
hits = [(n, count_regex_matches(n, note["contents"])) for n in rnames]
#print hits
hits = content_analysis.make_feature(
"names",
reduce(lambda x, y: x + y,
[count_regex_matches(n, note["contents"]) for n in rnames]))
return hits
return content_analysis.make_feature("names", 0)
def note_date_count(s):
import content_analysis
return content_analysis.make_feature("note_date_count" , len(date_matches(s["contents"]))) ##reduce(plus, [ count_regex_matches(f,s) for f in all_ ])
def note_date_count(s):
import content_analysis
return content_analysis.make_feature(
"note_date_count", len(date_matches(s["contents"]))
) ##reduce(plus, [ count_regex_matches(f,s) for f in all_ ])