def arabic_to_latin(input_file_path, output_file_path, sep_type, column_n):
print 'Transliterate from Arabic to Latin'
out=codecs.open(output_file_path, 'w', 'utf-8');
if args.preper==True:
status_tuple=commands.getstatusoutput('ruby1.8 ./pre_per2.rb {0} > {1}'.format(input_file_path, './tmp_file_1'));
if status_tuple[0]!=0:
sys.exit(status_tuple);
status_tuple=commands.getstatusoutput('nkf -w ./tmp_file_1 > conved');
if status_tuple[0]!=0:
sys.exit(status_tuple);
with codecs.open('./tmp_file_1', 'r', 'utf-8') as input_line:
for line in input_line:
items=line.split(sep_type);
Ins=translitr.transliter(line.split(sep_type)[column_n]);
if column_n==0:
out.write(Ins.arabic_to_unicode()+sep_type+sep_type.join(items[column_n+1:-1])+u'\n' );
elif column_n==-1:
out.write(sep_type.join(items[0:len(items)-1])+sep_type+Ins.arabic_to_unicode()+u'\n' );
#TODO あと，ここに中間のindexに対する記述をしていく
os.remove('./conved');
os.remove('./tmp_file_1');
else:
with codecs.open(input_file_path, 'r', 'utf-8') as lines:
for line in lines:
items=line.split(sep_type);
Ins=translitr.transliter(line.split(sep_type)[column_n]);
if column_n==0:
out.write(Ins.arabic_to_unicode()+sep_type+sep_type.join(items[column_n+1:-1])+u'\n' );
elif column_n==-1:
out.write(sep_type.join(items[0:len(items)-1])+sep_type+Ins.arabic_to_unicode()+u'\n' );
out.close();
def arabic_to_latin(input_file_path, output_file_path, sep_type, column_n):
print 'Transliterate from Arabic to Latin'
out = codecs.open(output_file_path, 'w', 'utf-8')
if args.preper == True:
status_tuple = commands.getstatusoutput(
'ruby1.8 ./pre_per2.rb {0} > {1}'.format(input_file_path,
'./tmp_file_1'))
if status_tuple[0] != 0:
sys.exit(status_tuple)
status_tuple = commands.getstatusoutput('nkf -w ./tmp_file_1 > conved')
if status_tuple[0] != 0:
sys.exit(status_tuple)
with codecs.open('./tmp_file_1', 'r', 'utf-8') as input_line:
for line in input_line:
items = line.split(sep_type)
Ins = translitr.transliter(line.split(sep_type)[column_n])
if column_n == 0:
out.write(Ins.arabic_to_unicode() + sep_type +
sep_type.join(items[column_n + 1:-1]) + u'\n')
elif column_n == -1:
out.write(
sep_type.join(items[0:len(items) - 1]) + sep_type +
Ins.arabic_to_unicode() + u'\n')
#TODO あと，ここに中間のindexに対する記述をしていく
os.remove('./conved')
os.remove('./tmp_file_1')
else:
with codecs.open(input_file_path, 'r', 'utf-8') as lines:
for line in lines:
items = line.split(sep_type)
Ins = translitr.transliter(line.split(sep_type)[column_n])
if column_n == 0:
out.write(Ins.arabic_to_unicode() + sep_type +
sep_type.join(items[column_n + 1:-1]) + u'\n')
elif column_n == -1:
out.write(
sep_type.join(items[0:len(items) - 1]) + sep_type +
Ins.arabic_to_unicode() + u'\n')
out.close()
def transliteration_whole(input_file_path, output_file_path, args):
print 'Whole transliteration'
print 'Mode is {}'.format(args.mode)
out = codecs.open(output_file_path, 'w', 'utf-8')
if args.mode == 'a_l':
if args.preper == True:
status_tuple = commands.getstatusoutput(
'ruby1.8 ./pre_per2.rb {0} > {1}'.format(
input_file_path, './tmp_file_1'))
if status_tuple[0] != 0:
sys.exit(status_tuple)
status_tuple = commands.getstatusoutput(
'nkf -w ./tmp_file_1 > conved')
if status_tuple[0] != 0:
sys.exit(status_tuple)
with codecs.open('./conved', 'r', 'utf-8') as input_line:
for line in input_line:
Ins = translitr.transliter(line)
out.write(Ins.arabic_to_unicode())
else:
try:
file_obj = codecs.open(input_file_path, 'rb', 'utf-8')
for i, line in enumerate(file_obj.readlines()):
Ins = translitr.transliter(line)
out.write(Ins.arabic_to_unicode())
except UnicodeDecodeError:
print u'Different Chara set is found.'
print u'Original sentence is :{} line at:{}'.format(line, i)
os.remove('./tmp_file_1')
os.remove('./conved')
out.close()
elif args.mode == 'l_a':
with codecs.open(input_file_path, 'r', 'utf-8') as input_line:
for line in input_line:
Ins = transliter.transliter(line)
out.write(Ins.unicode_to_arabic())
out.close()
def latin_to_arabic(input_file_path, output_file_path, sep_type, column_n):
print 'Transliterate from Latin to Arabic'
out=codecs.open(output_file_path, 'w', 'utf-8');
with codecs.open(input_file_path, 'r', 'utf-8') as input_line:
for line in input_line:
items=line.split(sep_type);
Ins=translitr.transliter(line.split(sep_type)[column_n]);
if column_n==0:
out.write(Ins.unicode_to_arabic()+sep_type+sep_type.join(items[column_n+1:-1])+u'\n' );
elif column_n==-1:
out.write(sep_type.join(items[0:len(items)-1])+sep_type+Ins.unicode_to_arabic()+u'\n' );
#TODO あと，ここに中間のindexに対する記述をしていく
out.close();
def transliteration_whole(input_file_path, output_file_path, args):
print 'Whole transliteration'
print 'Mode is {}'.format(args.mode);
out=codecs.open(output_file_path, 'w', 'utf-8');
if args.mode=='a_l':
if args.preper==True:
status_tuple=commands.getstatusoutput('ruby1.8 ./pre_per2.rb {0} > {1}'.format(input_file_path, './tmp_file_1'));
if status_tuple[0]!=0:
sys.exit(status_tuple);
status_tuple=commands.getstatusoutput('nkf -w ./tmp_file_1 > conved');
if status_tuple[0]!=0:
sys.exit(status_tuple);
with codecs.open('./conved', 'r', 'utf-8') as input_line:
for line in input_line:
Ins=translitr.transliter(line);
out.write(Ins.arabic_to_unicode());
else:
try:
file_obj=codecs.open(input_file_path, 'rb', 'utf-8');
for i, line in enumerate(file_obj.readlines()):
Ins=translitr.transliter(line);
out.write(Ins.arabic_to_unicode());
except UnicodeDecodeError:
print u'Different Chara set is found.'
print u'Original sentence is :{} line at:{}'.format(line, i);
os.remove('./tmp_file_1');
os.remove('./conved');
out.close();
elif args.mode=='l_a':
with codecs.open(input_file_path, 'r', 'utf-8') as input_line:
for line in input_line:
Ins=transliter.transliter(line);
out.write(Ins.unicode_to_arabic());
out.close();
def latin_to_arabic(input_file_path, output_file_path, sep_type, column_n):
print 'Transliterate from Latin to Arabic'
out = codecs.open(output_file_path, 'w', 'utf-8')
with codecs.open(input_file_path, 'r', 'utf-8') as input_line:
for line in input_line:
items = line.split(sep_type)
Ins = translitr.transliter(line.split(sep_type)[column_n])
if column_n == 0:
out.write(Ins.unicode_to_arabic() + sep_type +
sep_type.join(items[column_n + 1:-1]) + u'\n')
elif column_n == -1:
out.write(
sep_type.join(items[0:len(items) - 1]) + sep_type +
Ins.unicode_to_arabic() + u'\n')
#TODO あと，ここに中間のindexに対する記述をしていく
out.close()
