Walenty / Slowal

# encoding=UTF-8

# Copyright © 2007, 2008, 2010, 2011 Jakub Wilk <jwilk@jwilk.net>
#
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the “Software”), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in
# all copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED “AS IS”, WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
# SOFTWARE.

'''
Bindings for Morfeusz_, a Polish morphological analyser.

.. _Morfeusz:
   http://sgjp.pl/morfeusz/
'''

from __future__ import with_statement

import collections
import ctypes
import sys

py3k = sys.version_info >= (3, 0)

if py3k:
    import _thread as thread
else:
    import thread
if not py3k:
    from itertools import izip as zip

if py3k:
    unicode = str

__author__ = 'Jakub Wilk <jwilk@jwilk.net>'
__version__ = '0.3300'
__all__ = ['analyse', 'about', 'expand_tags', 'ATTRIBUTES', 'VALUES']

ATTRIBUTES = '''
subst=number case gender
depr=number case gender
adj=number case gender degree
adja=
adjc=
adjp=
adv=degree
num=number case gender accommodability
numcol=number case gender accommodability
ppron12=number case gender person accentability
ppron3=number case gender person accentability post_prepositionality
siebie=case
fin=number person aspect
bedzie=number person aspect
aglt=number person aspect vocalicity
praet=number gender aspect agglutination
impt=number person aspect
imps=aspect
inf=aspect
pcon=aspect
pant=aspect
ger=number case gender aspect negation
pact=number case gender aspect negation
ppas=number case gender aspect negation
winien=number gender aspect
pred=
prep=case vocalicity
conj=
comp=
brev=fullstoppedness
burk=
interj=
qub=vocalicity
xxs=number case gender
xxx=
interp=
ign=
sp=
'''
ATTRIBUTES = \
dict(
    (key, tuple(values.split()))
    for line in ATTRIBUTES.splitlines() if line
    for (key, values) in (line.split('=', 1),)
)

VALUES = '''
number=sg pl
case=nom gen dat acc inst loc voc
gender=m1 m2 m3 f n1 n2 p1 p2 p3
person=pri sec ter
degree=pos comp sup
aspect=imperf perf
negation=aff neg
accentability=akc nakc
post_prepositionality=npraep praep
accommodability=congr rec
agglutination=agl nagl
vocalicity=nwok wok
fullstoppedness=pun npun
'''
VALUES = \
dict(
    (key, tuple(values.split()))
    for line in VALUES.splitlines() if line
    for (key, values) in (line.split('=', 1),)
)

libmorfeusz = ctypes.CDLL('libmorfeusz.so.0')

MORFOPT_ENCODING = 1
MORFEUSZ_UTF_8 = 8

MORFOPT_WHITESPACE = 2
MORFEUSZ_SKIP_WHITESPACE = 0
MORFEUSZ_KEEP_WHITESPACE = 2

libmorfeusz.morfeusz_set_option(MORFOPT_ENCODING, MORFEUSZ_UTF_8)
libmorfeusz_lock = thread.allocate_lock()

class InterpEdge(ctypes.Structure):
    _fields_ = \
    (
        ('i', ctypes.c_int),
        ('j', ctypes.c_int),
        ('_orth', ctypes.c_char_p),
        ('_base', ctypes.c_char_p),
        ('_tags', ctypes.c_char_p)
    )

    if py3k:
        @property
        def tags(self):
            if self._tags is not None:
                return self._tags.decode('UTF-8')
    else:
        @property
        def tags(self):
            return self._tags

    @property
    def orth(self):
        if self._orth is not None:
            return self._orth.decode('UTF-8')

    @property
    def base(self):
        if self._base is not None:
            return self._base.decode('UTF-8')

libmorfeusz_analyse = libmorfeusz.morfeusz_analyse
libmorfeusz_analyse.restype = ctypes.POINTER(InterpEdge)
libmorfeusz_about = libmorfeusz.morfeusz_about
libmorfeusz_about.restype = ctypes.c_char_p

def expand_tags(tags, expand_dot=True, expand_underscore=True):

    if tags is None:
        yield
        return
    tags = str(tags)
    for tag in tags.split('|'):
        tag = tag.split(':')
        pos = tag.pop(0)
        chunks = [(pos,)]
        chunks += \
        (
            VALUES[attribute] if chunk == '_' and expand_underscore
            else chunk.split('.')
            for chunk, attribute in zip(tag, ATTRIBUTES[pos])
        )

        if not expand_dot:
            yield ':'.join('.'.join(values) for values in chunks)
            continue

        def expand_chunks(i):
            if i >= len(chunks):
                yield ()
            else:
                tail = tuple(expand_chunks(i + 1))
                for chunk_variant in chunks[i]:
                    for tail_variant in tail:
                        yield (chunk_variant,) + tail_variant

        for x in expand_chunks(0):
            yield ':'.join(x)

_expand_tags = expand_tags

def _dont_expand_tags(s, **kwargs):
    return [s]

def analyse(text, expand_tags=True, expand_dot=True, expand_underscore=True, dag=False, keep_whitespace=False):
    '''
    Analyse the text.
    '''
    expand_tags = _expand_tags if expand_tags else _dont_expand_tags
    text = unicode(text)
    text = text.encode('UTF-8')
    analyse = _analyse_as_dag if dag else _analyse_as_list
    return analyse(
        text=text,
        expand_tags=expand_tags,
        expand_dot=expand_dot,
        expand_underscore=expand_underscore,
        keep_whitespace=keep_whitespace
    )

def _analyse_as_dag(text, expand_tags, expand_dot, expand_underscore, keep_whitespace):
    result = []
    with libmorfeusz_lock:
        if keep_whitespace:
            if libmorfeusz.morfeusz_set_option(MORFOPT_WHITESPACE, MORFEUSZ_KEEP_WHITESPACE) != 1:
                raise NotImplementedError("This version of Morfeusz doesn't support keep_whitespace")
        for edge in libmorfeusz_analyse(text):
            if edge.i == -1:
                break
            for tag in expand_tags(edge.tags, expand_dot=expand_dot, expand_underscore=expand_underscore):
                result += [(edge.i, edge.j, (edge.orth, edge.base, tag))]
        if keep_whitespace:
            libmorfeusz.morfeusz_set_option(MORFOPT_WHITESPACE, MORFEUSZ_SKIP_WHITESPACE)
    return result

def _analyse_as_list(text, expand_tags, expand_dot, expand_underscore, keep_whitespace):
    dag = collections.defaultdict(list)
    with libmorfeusz_lock:
        if keep_whitespace:
            if libmorfeusz.morfeusz_set_option(MORFOPT_WHITESPACE, MORFEUSZ_KEEP_WHITESPACE) != 1:
                raise NotImplementedError("This version of Morfeusz doesn't support keep_whitespace")
        for edge in libmorfeusz_analyse(text):
            if edge.i == -1:
                break
            for tag in expand_tags(edge.tags, expand_dot=expand_dot, expand_underscore=expand_underscore):
                dag[edge.i] += [((edge.orth, edge.base, tag), edge.j)]
        if keep_whitespace:
            libmorfeusz.morfeusz_set_option(MORFOPT_WHITESPACE, MORFEUSZ_SKIP_WHITESPACE)
    def expand_dag(i):
        nexts = dag[i]
        if not nexts:
            yield []
        else:
            for head, j in nexts:
                for tail in expand_dag(j):
                    yield [head] + tail
    return list(expand_dag(0))

def about():
    '''
    Return a string containing information on authors and version of the
    underlying library.
    '''
    about = libmorfeusz_about()
    try:
        return about.decode('UTF-8')
    except UnicodeError:
        return about.decode('ISO-8859-2')

# vim:ts=4 sw=4 et