Python Text.from_file方法代码示例

# 需要导入模块: from Text import Text [as 别名]
# 或者: from Text.Text import from_file [as 别名]
def extract_bisents(file1, lang1, file2, lang2, alignment_file):
    assert isinstance(lang1, (unicode, str)) and len(lang1)==2
    assert isinstance(lang2, (unicode, str)) and len(lang2)==2
    t1 = Text.from_file(file1, lang1)
    t2 = Text.from_file(file2, lang2)
    alignment = Alignment.from_file(alignment_file)
    bisents = alignment.as_pairs(t1.as_sentences_flat(),
                                 t2.as_sentences_flat())
    return bisents

# 需要导入模块: from Text import Text [as 别名]
# 或者: from Text.Text import from_file [as 别名]
def fetch_sentences(basename, lang):
    assert lang in ('pl', 'plm',
                    'cu', 'cum', 'cut', 'cue',
                    'el', 'elm', 'elt'), "invalid lang " + lang

    real_lang = lang[:2]
    transformation = lang[2:]

    basename_with_lang = ("%s/%s" % (basename, real_lang))

    try:
        #TODO maybe open ready metaphone files?
        with file("%s.sentences" % basename_with_lang) as f:
            t = [line.decode('utf-8').strip() for line in f.readlines()]
    except IOError:
        t = Text.from_file("%s.txt" % basename_with_lang,
                           lang=real_lang).as_sentences_flat()

    if transformation:
        if transformation == 'm':
            return [metaphone_text(s, lang=real_lang) for s in t]
        elif transformation == 't':
            return [translit_pl(s, real_lang) for s in t]
        elif transformation == 'e':
            return [expand_cu(s, numbers=True) for s in t]
    return t

# 需要导入模块: from Text import Text [as 别名]
# 或者: from Text.Text import from_file [as 别名]
def export_sentences(input_file, lang, export_type):
    from translit.metaphone import metaphone
    t = Text.from_file(input_file, lang)
    for s in t.as_sentences(paragraph_separator='¶'):
        if export_type == 'hunalign':
            if s == '¶':
                s = '<p>'
            else:
                s = ' '.join(metaphone(w) for w in s.split())
        print s.encode('utf-8')

# 需要导入模块: from Text import Text [as 别名]
# 或者: from Text.Text import from_file [as 别名]
if __name__ == '__main__':
    import sys
    import re
    from Text import Text

    try:
        alignment_filename = sys.argv[1]
        if sys.argv[2:]:
            [fn1, fn2] = sys.argv[2:]
            assert not fn1.endswith('.txt')
            assert not fn2.endswith('.txt')
            with open(fn1) as f:
                seq1 = [l.decode('utf-8').strip() for l in f.readlines()]
                with open(fn2) as f:
                    seq2 = [l.decode('utf-8').strip() for l in f.readlines()]
        else:
            m = re.match(r'(.*/)?(..)-(..)\.(.*)$', alignment_filename)
            t1 = Text.from_file("%s%s.txt" % (m.group(1), m.group(2)),
                                lang=m.group(2))
            t2 = Text.from_file("%s%s.txt" % (m.group(1), m.group(3)),
                                lang=m.group(3))
            seq1 = t1.as_sentences_flat()
            seq2 = t2.as_sentences_flat()
        a = Alignment.from_file(alignment_filename)
        a.pretty_print(seq1, seq2)
        print "Total cost: " + str(sum(c for (_, _, c) in a.data))
    except IndexError, ValueError:
        print >> sys.stderr, __doc__
    except IOError, e:
        print >> sys.stderr, e

# 需要导入模块: from Text import Text [as 别名]
# 或者: from Text.Text import from_file [as 别名]
 def get_text(self, lang):
     return Text.from_file(self._p(str(lang) + ".txt"), lang=str(lang))

# 需要导入模块: from Text import Text [as 别名]
# 或者: from Text.Text import from_file [as 别名]
    def gen():
        yield (0, 0, 0)
        prev_i2 = 0
        for (i1, i2) in al12:
            for _i2 in range(prev_i2+1, i2+1):
                try:
                    i3s = map23[_i2]
                    for i3 in i3s:
                        # if map31[i3] == i1:
                        yield (i1, _i2, i3)
                except KeyError:
                    pass
            prev_i2 = i2
    return Alignment(list(gen()), no_costs=True)

if __name__ == '__main__':
    import sys
    name = sys.argv[1]

    a1 = Alignment.from_file(name + '/pl-cu.my').as_ladder()
    a2 = Alignment.from_file(name + '/cu-el.my').as_ladder()
    a3 = Alignment.from_file(name + '/pl-el.my').as_ladder()
    a3 = [(b, a) for (a, b) in a3]

    ma = merge_3_alignments(a1, a2, a3)

    ma.pretty_print(Text.from_file(name + '/pl.txt', lang='pl').as_sentences_flat(),
                    Text.from_file(name + '/cu.txt', lang='cu').as_sentences_flat(),
                    Text.from_file(name + '/el.txt', lang='el').as_sentences_flat())