Rev 77 |
Rev 79 |
Go to most recent revision |
Blame |
Compare with Previous |
Last modification |
View Log
| RSS feed
#!/usr/bin/python
# -*- coding: utf-8 -*-
import re
import dbconnector
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""]
#prefixesn = ur"(?P<npr>(?:(?:fì|tsa)?(?:me|pxe|ay|fra)?|(?:fay)?|(?:tsay)?)(?:fne)?(?:tì|sä)?"
prefixes = [u"ay", u"me", u"pxe", u"pe", u"a", u"le", u"nì", u"sä", u"tì", u"ke", u"fì", u"fay", u"tsa", u"tsay", u"fra"]
#prefixesv = ur"(?P<vpr>(?:nì|sä|tì|rä'ä |ke )?)"
def parseword(wordin):
ret = {"word": {"id": 0, "navi": u" ".join(wordin), "infix": u"", "type": u""}}
for word in wordlist:
foundit = True
foundprefs = []
foundposts = []
splitword = word["infix"].split(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""
foundins = [u"", u"", u""]
pre = []
post = []
if u"<1>" in splitword[wor]:
for in1 in infixes1:
for in2 in infixes2:
for in3 in infixes3:
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"":
foundit = False
break
pref, posf = wordin[wor].split(center)
for pre in prefixes:
if pref.endswith(pre):
foundprefs[wor].append(pre)
pref = pref[:len(pref)-len(pre)]
if pref != u"":
foundit = False
break
foundposts[wor] = posf
if foundit == True:
foundword = word
break
if foundit == True:
ret["pref"] = foundprefs
ret["post"] = foundposts
ret["inf"] = foundins
ret["word"] = foundword
return ret
def parsesent(sent):
ret = []
left = len(sent)
while left:
word = parseword(sent)
left -= len(word["word"]["navi"].split(" "))
ret.append(word)
return ret