def pass1():
global output
outfile = '__h1output.txt'
with open(outfile, 'w') as f:
count = 0
for t in tests:
nabcc = False
if t.has_key('mode') and t['mode'] == 'NABCC':
nabcc = True
if t.has_key('output'):
result, inpos1 = translator1.translateWithInPos(t['input'], nabcc=nabcc)
if t.has_key('inpos1'):
correct_inpos1 = ','.join(['%d' % n for n in t['inpos1'] ])
else:
correct_inpos1 = None
result_inpos1 = ','.join(['%d' % n for n in inpos1])
if result != t['output'] or \
(correct_inpos1 and result_inpos1 != correct_inpos1) or \
(len(result) != len(inpos1)):
count+=1
f.write("input: " + t['input'].encode('utf-8') + "\n")
f.write("result: " + result.encode('utf-8') + "\n")
f.write("correct: " + t['output'].encode('utf-8') + "\n")
if correct_inpos1:
f.write("correct_inpos1: " + correct_inpos1 + "\n")
f.write("result_inpos1: " + result_inpos1 + "\n")
if 'comment' in t:
f.write("comment: " + t['comment'].encode('utf-8') + "\n")
f.write("\n")
print 'h1: %d error(s). see %s' % (count, outfile)
return (count, outfile)
def translateWithInPos2(inbuf, logwrite=_logwrite, nabcc=False):
if not mecab_initialized:
initialize()
# do not translate if string is unicode braille
if all((0x2800 <= ord(c) <= 0x28ff or c == ' ') for c in inbuf):
outbuf = inbuf
inpos2 = [n for n in xrange(len(inbuf))]
else:
outbuf, inpos2 = japanese_braille_separate(inbuf, logwrite, nabcc=nabcc)
result, inpos1 = translator1.translateWithInPos(outbuf, nabcc=nabcc)
result = result.replace('□', ' ')
return (outbuf, result, inpos1, inpos2)
