def add_arguments(title):
parser = fill_arg_for_processor(title)
parser.add_argument('--ignore_mess', default=None)
parser.add_argument('--show_mess', default=None)
parser.add_argument('--show_files', default=None)
parser.add_argument('--scheme', default=None)
parser.add_argument('--media', default=False, action='store_true')
return parser
fout.write('"{0}" - {1} \n'.format(val_k,val_v[2]))
# self.put_counter('changes_zh.txt', self.changes_zh)
# self.put_counter('changes_zh1.txt', self.changes_zh1)
'''
if self.feats_loader.wrong:
if '' in self.feats_loader.wrong:
print("Empty!")
print("Wrong feats: {0}".format(','.join(sorted(self.feats_loader.wrong))))
with open(Path('~/Documents/china_gr.txt').expanduser(), 'w') as fout:
fout.write("china feats:\n\n")
fout.write('\n'.join(sorted(self.feats))+'\n')
fout.write("china feat sets:\n\n")
fout.write('\n'.join(sorted(self.feat_grs))+'\n')
'''
if __name__ == '__main__':
parser = fill_arg_for_processor('feat checker')
parser_args = parser.parse_args()
feats_checker = FeatsChecker(parser_args)
feats_checker.process()
feats_checker.put_info()
class VersesCutter(ProcessorBasic):
def process_lxml_tree(self, tree):
handler = MyContentHandler()
sax.saxify(tree, handler)
if handler.max_br > 0:
print(self.inpname, handler.max_br, "строф")
return handler.etree
class VersesCutter2(ProcessorBasic):
def __init__(self, args):
super().__init__(args)
self.res = open(args.found, "w")
def process_lxml_tree(self, tree):
verses = tree.xpath('//p[@class="verse"]')
for verse in verses:
lines = verse.xpath("count(//br)")
if lines > 1000:
print(self.inpname, lines, "строк в строфе")
self.res.write(self.inpname)
return None
if __name__ == '__main__':
parser = fill_arg_for_processor('verses cutter')
parser.add_argument("--found", required=True)
parser_args = parser.parse_args()
cutter = VersesCutter2(parser_args)
cutter.process()
self.outfile_write(attr)
start = False
self.outfile_write('}')
# self.outfile_write('['+to_str[info[TEXT_POS]]+', '+to_str[info[TAIL_POS]]+']')
self.outfile_write(
'\t' + os.path.relpath(info[FIRST_DOC_POS], self.common_part) +
'\t\n')
for value in info[ELEM_POS].values():
self.put_info(value, shift + info[TAG_POS] + '/')
def process(self):
inp_paths = self.inppath.split('|')
for self.inppath in inp_paths:
if not super(Schema, self).process():
return
self.common_part = os.sep.join(self.common_part)
with open(self.schema, 'w') as self.outfile:
self.outfile_write('tags sequence\texample\tcomment\n')
for value in self.glob_info.values():
self.put_info(value, '')
self.outfile_write('\n(common path of examples - ' +
self.common_part + ')')
if __name__ == '__main__':
parser = fill_arg_for_processor('schema processing')
parser.add_argument('--schema', required=True)
parser_args = parser.parse_args()
processor = Schema(parser_args)
processor.process()
for lang, l_counter in self.counters.items():
with open(root_dir/Path(lang+".txt"), 'w') as f_count:
for stat in l_counter.most_common():
sym = stat[0]
if len(sym) > 1:
raise Exception('logic error')
cat = unicodedata.category(sym)
try:
sym_name = unicodedata.name(sym)
except ValueError:
sym = '#{0}'.format(ord(sym))
sym_name = 'UNKNOWN NAME'
cat_gr = cat[0]
count = stat[1]
if cat_gr == 'L' or cat == 'Cc' or cat == 'Nd':
continue
f_count.write('"{0}" {1}, {2}: {3} time(s)\n'.format(sym, cat, sym_name, count))
except (OSError, IOError) as e:
self.fatal_error("can't write statistics into {0}: {1}".format(self.sym_stat, e.message))
if __name__ == '__main__':
parser = fill_arg_for_processor('symbol counter')
parser.add_argument('--sym_stat', required=True)
parser_args = parser.parse_args()
counter = SymbolCounter(parser_args)
counter.process()
counter.report()
self.line = -1
"""
def change_accent(self, text):
if text is None:
return text
n = text.count(COMBINING_GRAVE_ACCENT)
if n == 0:
return text
self.count_mess(
"COMBINING_GRAVE_ACCENT changed to COMBINING_ACUTE_ACCENT", n)
return text.replace(COMBINING_GRAVE_ACCENT, COMBINING_ACUTE_ACCENT)
def process_lxml_tree(self, tree):
root = tree.getroot()
body = root.find('body')
if body is None:
return None
for elem in body.iter():
if self.nostructured(elem):
continue
elem.text = self.change_accent(elem.text)
elem.tail = self.change_accent(elem.tail)
return tree
if __name__ == '__main__':
parser = fill_arg_for_processor('speech converter', True)
parser_args = parser.parse_args()
converter = ConverterStihi(parser_args)
converter.process()
