def context_morphs(all_sentence, word_list):
num = re.compile(r'[0-9]')
for word in word_list:
chunk_list = module(word)
for sent in chunk_list:
for j in range(len(sent)):
context = []
for k in sent[j].morph:
if k.pos == '名詞':
context.append(k.surface)
else:
if len(context) >= 2:
a = ' '.join(context)
if a in nouns_100:
for b in sent[int(sent[j].dst)].morph:
if b.pos == '名詞' or b.pos == '動詞' or b.pos == '形容詞':
print a + '\t -> ' + b.base
for c in sent[j].srcs:
for d in sent[int(c)].morph:
if b.pos == '名詞' or b.pos == '動詞' or b.pos == '形容詞':
print a + '\t <- ' + d.base
context = []
if len(context) >= 2:
a = ' '.join(context)
if a in nouns_100:
for i in sent[int(sent[j].dst)].morph:
if b.pos == '名詞' or b.pos == '動詞' or b.pos == '形容詞':
print a + '\t -> ' + b.base
def noun_verb(FILE):
chunk_list = module(FILE)
for chunk in chunk_list:
for i in range(len(chunk)):
if chunk[i].dst != '-1':
noun, verb = '', ''
noun_check, verb_check = 0, 0
for j1 in chunk[i].morphs:
if j1.surface != '、 ' and j1.surface != '。 ':
noun += j1.surface
if j1.pos == '名詞':
noun_check = 1
if noun_check == 0:
continue
for j2 in chunk[int(chunk[i].dst)].morphs:
if j2.surface != '、 ' and j1.surface != '。 ':
verb += j2.surface
if j1.pos == '動詞':
verb_check = 1
if verb_check == 0:
continue
print noun + '\t' + verb + '\n'
def main(): print module.__dict__ print "**********start *************" m=module() print m # print "general method in python other package" # m.details() print "*********Eastablish data connection *******" conn=module.connection() print conn
def tab_extract(FILE):
chunk_list = module(FILE)
for chunk in chunk_list:
for i in range(len(chunk)):
if chunk[i].dst != '-1':
moto, saki = '', ''
for j1 in chunk[i].morphs:
if j1.pos != '記号':
moto += j1.surface
for j2 in chunk[int(chunk[i].dst)].morphs:
if j2.pos != '記号':
saki += j2.surface
print moto + '\t' + saki + '\n'
def more_than_2_phrase(FILE):
chunk_list = module(FILE)
for chunk in chunk_list:
for i in range(len(chunk)):
if len(chunk[i].srcs) >= 2:
print chunk[i].srcs
saki = ''
for j in chunk[i].morphs:
saki += j.surface
print saki + '\n'
for src in chunk[i].srcs:
moto = ''
for k in chunk[int(src)].morphs:
moto += k.surface
print moto
def phrase_expression(FILE):
chunk_list = module(FILE)
chunk1, chunk2 = '', ''
for chunk1 in chunk_list:
for chunk2 in chunk_list:
for i in range(len(chunk1)):
for j in range(len(chunk2)):
if chunk1[i].dst == chunk2[j].num:
for j1 in chunk1[i].morphs:
for j2 in chunk2[j].morphs:
if j1.pos == '名詞' and j2.pos == '名詞':
print j1.surface + '\t' + j2.surface + '\n'
print chunk2[j].num + '->' + chunk1[i].num
def load_shelve (file_name, progress_bar=None):
'''
Load a module from disk.
@see: dump_shelve()
@type name: String
@param name: File name to import from
@rtype: pida.module
@return: Imported module
'''
import shelve
sh = shelve.open(file_name, flag='r', protocol=2)
mod = module()
# attributes from pida.module
mod.name = sh["name"]
mod.base = sh["base"]
mod.depth = sh["depth"]
mod.analysis = sh["analysis"]
mod.signature = sh["signature"]
mod.version = sh["version"]
mod.ext = sh["ext"]
# attributes inherited from pgraph.graph
mod.id = sh["id"]
mod.clusters = sh["clusters"]
mod.edges = sh["edges"]
mod.nodes = {}
# we restore the node dictionary piece by piece to avoid out of memory conditions.
for key, val in sh["nodes"].items():
mod.nodes[key] = val
return
pre_w = 'None'
pre_pos = 'None'
for one_chunk in one_sentence:
if one_chunk.chunk == 'B-NP':
if word.startswith('#'):
make_feature(word, features, one_chunk.w, one_chunk.pos, pre_pos)
head_w = find_head(one_chunk.w.lower())
features = feature(pre_w, pre_pos, head_w, one_chunk.pos)
word = '# ' + one_chunk.w
elif one_chunk.chunk == 'I-NP' and word != ' ':
if features.head_w != 'NONE':
features.f_pos = one_chunk.pos
word += ' ' + one_chunk.w
elif word != ' ':
make_feature(word, features, one_chunk.w, one_chunk.pos, pre_pos)
word = ' '
else: pass
pre_w = one_chunk.w
pre_pos = one_chunk.pos
if __name__ == '__main__':
for name in glob.glob('8_071_GENIA_tagger_?.txt'):
result = open('8_076_output' + name[-5] + '.f', 'w')
11feature(module(name))
result.close()
for k in sent[j].morph:
if k.pos == '名詞':
context.append(k.surface)
else:
if len(context) >= 2:
a = ' '.join(context)
if a in nouns_100:
for b in sent[int(sent[j].dst)].morph:
if b.pos == '名詞' or b.pos == '動詞' or b.pos == '形容詞':
print a + '\t -> ' + b.base
for c in sent[j].srcs:
for d in sent[int(c)].morph:
if b.pos == '名詞' or b.pos == '動詞' or b.pos == '形容詞':
print a + '\t <- ' + d.base
context = []
if len(context) >= 2:
a = ' '.join(context)
if a in nouns_100:
for i in sent[int(sent[j].dst)].morph:
if b.pos == '名詞' or b.pos == '動詞' or b.pos == '形容詞':
print a + '\t -> ' + b.base
if __name__ == '__main__':
word_list = make_word_list()
for num in glob.glob('7_061_output_?.txt'):
all_sentence = module(num)
context_morphs(all_sentence, word_list)
def a_the_none(all_sentence):
for one_sentence in all_sentence:
word = ''
pre_w = 'None'
pre_pos = 'None'
for one_chunk in one_sentence:
if one_chunk.chunk == 'B-NP':
if word.startswith('#'):
make_feature(word, features, one_chunk.w, one_chunk.pos, pre_pos)
head_w = find_head(one_chunk.w.lower())
features = feature(pre_w, pre_pos, head_w, one_chunk.pos)
word = '# ' + one_chunk.w
elif one_chunk.chunk == 'I-NP' and word != '':
if features.head_w != 'NONE':
features.f_pos = one_chunk.pos
word += ' ' + one_chunk.w
elif word != ' ':
make_feature(word, features, one_chunk.w, one_chunk.pos, pre_pos)
word = ' '
else: pass
pre_w = one_chunk.w
pre_pos = one_chunk.pos
if __name__ == '__main__':
a_the_none(module(sys.argv[1]))
def _start(module, args=None):
module(args)
from module import *
m = module(".bashrc")
print(m)
print m.commentCharacter
print m.activeFile
print m.homePath
m.set_file(".vimrc")
print m.activeFile
m.set_file(".bashrc")
print (m.activeFile)
#print(m.check_for_file())
#m.append_to_file("DELETEME")
#m.uncomment_line("ben=")
m.backup_file(".origin")
#!/usr/bin/env python
# coding: utf-8
from module import *
import sys
def np(all_sentence):
for one_sentence in all_sentence:
word = ''
for one_chunk in one_sentence:
if one_chunk.chunk == 'B-NP':
if word.startswith('#'):
print word
word = '# ' + one_chunk.w
elif one_chunk.chunk == 'I-NP' and word != ' ':
word += ' ' + one_chunk.w
elif word != ' ':
print word
word = ' '
else: pass
if __name__ == '__main__':
np(module(sys.argv[1]))
def labels():
text = module('Cristiano',10)
return str(text)
app = Flask(__name__)
@app.route('/')
def hello_world():
return """
<!DOCTYPE html>
<head>
<title> API Exercise pt. 2 </title>
</head>
<body>
<p> Andre's Code is being tested with the twitter handle 'Cristiano' and a number of 10 tweets </p>
<p> To view the labels returned by the Google Cloud Vision API, type '/labels' after the current URL </p>
<p> OUTPUT VIDEO as a .mp4 file: </p>
<video controls>
<source src="/output.mp4" type="video/mp4">
</video>
</body>
"""
#only way i could think of printing a result from a module
@app.route("/labels")
def labels():
text = module('Cristiano',10)
return str(text)
if __name__ == '__main__':
text = module('Cristiano',10)
app.run(debug = True, use_reloader = True)