Rev 185 |
Rev 187 |
Go to most recent revision |
Blame |
Compare with Previous |
Last modification |
View Log
| RSS feed
#!/usr/bin/python
# -*- coding: utf-8 -*-
# This file is part of Tsim Apiak.
#
# Tsim Apiak is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public Licence as published by
# the Free Software Foundation, either version 3 of the Licence, or
# (at your option) any later version.
#
# In addition to this, you must also comply with clause 4 of the
# Apache Licence, version 2.0, concerning attribution. Where there
# is a contradiction between the two licences, the GPL
# takes preference.
#
# Tsim Apiak is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with Tsim Apiak. If not, see <http://www.gnu.org/licenses/>.
import re
import dbconnector
import parsenum
wordlist = dbconnector.getnavilist()
infixes1 = (u"awn", u"eyk", u"us", u"äp", u"")
infixes2 = (u"ìyev", u"iyev", u"ìmìy", u"arm", u"asy", u"ilv", u"ìmv", u"imv", u"ìrm", u"irv", u"ìsy", u"aly", u"ary", u"ìly", u"ìry", u"ìlm", u"alm", u"am", u"ay", u"er", u"ìm", u"iv", u"ìy", u"ol", u"")
infixes3 = (u"äng", u"ats", u"eiy", u"ei", u"uy", u"")
prefixes = (u"tsay", u"say", u"fay", u"fra", u"pxe", u"fne", u"tsa", u"sa", u"pe", u"fe", u"le", u"nì", u"sä", u"tì", u"sì", u"ay", u"me", u"fì", u"ke", u"he", u"a")
adpositions = (u"mungwrr", u"kxamlä", u"pximaw", u"pxisre", u"tafkip", u"nemfa", u"takip", u"mìkam", u"teri", u"fkip", u"luke", u"pxel", u"pxaw", u"rofa", u"ìlä", u"fpi", u"ftu", u"kip", u"lok", u"maw", u"sre", u"sìn", u"vay", u"eo", u"fa", u"hu", u"io", u"ka", u"mì", u"na", u"ne", u"ro", u"ta", u"uo", u"wä", u"äo", u"to", u"sì")
postfixes = adpositions + (u"tsyìp", u"eyä", u"ìri", u"ìl", u"it", u"lo", u"ri", u"ru", u"ti", u"ur", u"ve", u"yä", u"ya", u"tu", u"vi", u"yu", u"an", u"ng", u"ke", u"e", u"o", u"l", u"t", u"y", u"a", u"ä", u"r")
#prefixesn = ur"(?P<npr>(?:(?:fì|tsa)?(?:me|pxe|ay|fra)?|(?:fay)?|(?:tsay)?)(?:fne)?(?:tì|sä)?"
#prefixesv = ur"(?P<vpr>(?:nì|sä|tì|rä'ä |ke )?)"
lenit = ((u"px", u"p"), (u"tx", u"t"), (u"kx", u"k"), (u"ts", u"s"), (u"t", u"s"), (u"p", u"f"), (u"k", u"h"), (u"'", u""))
def parseword(wordin):
ret = {"word": {"id": 0, "navi": u"[" + wordin[0] + u"]", "infix": u"", "type": u""}}
for word in wordlist:
foundit = True
foundprefs = []
foundposts = []
lenited = False
splitword = word["infix"].split(u" ")
foundins = [u"", u"", u""]
if len(wordin) < len(splitword):
foundit = False
next
for wor in range(len(splitword)):
if not foundit:
break
foundprefs.append([])
foundposts.append([])
center = u""
pre = []
post = []
if u"<1>" in splitword[wor]:
tempin1 = []
tempin2 = []
tempin3 = []
for in1 in infixes1:
if in1 in wordin[wor]:
tempin1.append(in1)
for in2 in infixes2:
if in2 in wordin[wor]:
tempin2.append(in2)
for in3 in infixes3:
if in3 in wordin[wor]:
tempin3.append(in3)
for in1 in tempin1:
for in2 in tempin2:
for in3 in tempin3:
if splitword[wor].replace(u"<1><2>",in1+in2).replace(u"<3>",in3) in wordin[wor]:
center = splitword[wor].replace(u"<1><2>",in1+in2).replace(u"<3>",in3)
foundins = [in1, in2, in3]
break
if center != u"": break
if center != u"": break
else:
if splitword[wor] in wordin[wor]:
center = splitword[wor]
if center == u"":
for i in lenit:
temp = u""
if splitword[wor].startswith(i[0]):
temp = i[1] + splitword[wor][len(i[0]):]
if temp in wordin[wor]:
lenited = True
center = temp
if center == u"":
if splitword[wor].endswith(u"nga"):
temp = splitword[wor][:-3] + u"ng"
if temp in wordin[wor]:
center = temp
if splitword[wor].endswith(u"po"):
temp = splitword[wor][:-2] + u"p"
if temp in wordin[wor]:
center = temp
if center == u"":
foundit = False
break
temp = wordin[wor].split(center)
if len(temp) != 2:
foundit = False
break
pref, posf = temp
last = u""
while last != pref:
last = pref
for pre in prefixes:
if pref != u"":
if pref.endswith(pre):
if pre in foundprefs[wor]:
break
foundprefs[wor].append(pre)
pref = pref[:-len(pre)]
break
if pref != u"":
foundit = False
break
last = u""
while last != posf:
last = posf
for pos in postfixes:
if posf != u"":
if posf.startswith(pos):
if pos in foundposts[wor]:
break
foundposts[wor].append(pos)
posf = posf[len(pos):]
break
if posf != u"":
foundit = False
break
if foundit == True:
foundword = word
break
ret["pref"] = foundprefs
ret["post"] = foundposts
ret["inf"] = foundins
ret["len"] = lenited
if foundit == True:
ret["word"] = foundword
return ret
def parsesent(sent):
sent = sent.strip().lower().replace(u"’", u"'")
sent = re.sub(ur"[^\wìä' ]",u"",sent)
sent = re.sub(ur"\ +",u" ",sent)
sent = sent.split(u" ")
ret = []
left = len(sent)
while left:
word = parsenum.parse(sent[len(sent)-left])
if word == None:
word = parseword(sent[-left:])
left -= len(word["word"]["navi"].split(" "))
ret.append(word)
return ret