type="choice",
choices=["sdrt", "pdtb", "minsdrt", "subord_coord"],
help="merge classes according to given relation grouping (cf doc)")
(options, args) = parser.parse_args()
allfuncs = {}
if options.timex:
allfuncs["timex"] = add_timex
if options.anaphor:
allfuncs["anaph"] = add_anaph_feats
if options.verb_class:
allfuncs["verb"] = add_verbclass
if options.voisins and LEX:
#allfuncs["voisins"] = add_voisins
db_filename, table = init_voisins(options.voisins)
allfuncs["voisins"] = edu_similarity
if options.relations:
allfuncs["relations"] = add_relations
if (options.simple or options.strand_orphans) and allfuncs == {}:
allfuncs["dummy"] = add_nothing
print >> sys.stderr, "Requested additions:", allfuncs.keys()
# basename = args[0]
# doc = annodisAnnot(basename+".xml")
# feats = FeatureMap(basename+".features")
# prep = Preprocess( basename+".txt.prep.xml")
if options.merge:
if __name__=="__main__":
test=sys.argv[1]
try:
doc=annodisAnnot(test)
#print doc.text().encode("utf8")
#print doc.get_edu("2").attrib
#print doc.get_edu_text("1")
except:
print >> sys.stderr, "pb reading", test
try:
prep=Preprocess(test.split(".xml")[0]+".txt.prep.xml")
doc.add_preprocess(prep)
voisins = sys.argv[2]
db_filename,table=init_voisins(voisins)
corpus.add_lexical_relations(table)
# for pid,p in doc.parses().items():
# print p.dependencies()
# for i in range(10):
# print "Mentions for EDU %s: %s" %(i,doc.get_edu_mentions( str(i) ))
mentions = doc.mentions().values()
#mentions.sort()
# PM: commented out for speed
#for m in mentions:
# print m
for i in range(1,10):
try:
print i, doc.get_edu_text( str(i) ).encode("utf8")
print i, doc.get_edu_mentions( str(i) )
if __name__ == "__main__":
test = sys.argv[1]
try:
doc = annodisAnnot(test)
#print doc.text().encode("utf8")
#print doc.get_edu("2").attrib
#print doc.get_edu_text("1")
except:
print >> sys.stderr, "pb reading", test
try:
prep = Preprocess(test.split(".xml")[0] + ".txt.prep.xml")
doc.add_preprocess(prep)
voisins = sys.argv[2]
db_filename, table = init_voisins(voisins)
corpus.add_lexical_relations(table)
# for pid,p in doc.parses().items():
# print p.dependencies()
# for i in range(10):
# print "Mentions for EDU %s: %s" %(i,doc.get_edu_mentions( str(i) ))
mentions = doc.mentions().values()
#mentions.sort()
# PM: commented out for speed
#for m in mentions:
# print m
for i in range(1, 10):
try:
print i, doc.get_edu_text(str(i)).encode("utf8")
print i, doc.get_edu_mentions(str(i))
(options, args) = parser.parse_args()
allfuncs = {}
if options.timex:
allfuncs["timex"] = add_timex
if options.anaphor:
allfuncs["anaph"] = add_anaph_feats
if options.verb_class:
allfuncs["verb"] = add_verbclass
if options.voisins and LEX:
#allfuncs["voisins"] = add_voisins
db_filename, table = init_voisins(options.voisins)
allfuncs["voisins"] = edu_similarity
if options.relations:
allfuncs["relations"] = add_relations
if (options.simple or options.strand_orphans) and allfuncs == {}:
allfuncs["dummy"] = add_nothing
print >> sys.stderr, "Requested additions:", allfuncs.keys()
# basename = args[0]
# doc = annodisAnnot(basename+".xml")
# feats = FeatureMap(basename+".features")
# prep = Preprocess( basename+".txt.prep.xml")
if options.merge: