2,85 → 2,118 |
# -*- coding: utf-8 -*- |
|
import re |
from dbconnector import getnavilist |
import dbconnector |
import parsenum |
|
wordlist = getnavilist() |
wordlist = dbconnector.getnavilist() |
|
infixes0 = [ u"awn", u"eyk", u"us", u"äp" ] |
infixes1 = [ 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" ] |
infixes2 = [ u"äng", u"ats", u"eiy", u"ei", u"uy" ] |
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"fay", u"fra", u"pxe", u"fne", u"tsa", u"pe", u"le", u"nì", u"sä", u"tì", u"ay", u"me", u"fì", u"ke", 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"ìla", 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") |
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"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 )?)" |
|
# Returns array with Word,Infix 0,Infix 1,Infix 2,Case,Gender,Number suffixes,Inclusive,Indefinite,Vocative (suffix),Plural,Adposition,Adject pre,Adject suff,am/ay/tu/vi/yu,adverbial,nominalise,sä,fne,lenited? |
def parsefix(original): |
realword = u"" |
infix0 = u"" |
infix1 = u"" |
infix2 = u"" |
infix01 = u"" |
infix_1 = u"" |
infix_2 = u"" |
for eachword in wordlist: |
regex = re.sub(u" ",u"[^ ]* [^ ]*",eachword["infix"]) |
regex = re.sub(u"^",u"[^ ]*",regex) |
regex = re.sub(u"$",u"[^ ]*",regex) |
regex = re.sub(u"<1><2>",u"[^ ]*",regex) |
regex = re.sub(u"<3>",u"[^ ]*",regex) |
if re.match(regex,original): |
realword = eachword["infix"] |
break |
if realword == u"": |
return [u"[" + original + u"]",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u""] |
else: |
if re.search(u"<",realword): |
beginning = re.sub(u"<1><2>.*",u"",realword) |
middle = re.sub(u".*<1><2>(.*)<3>.*",ur"\1",realword) |
end = re.sub(u".*<3>",u"",realword) |
infix01 = re.sub(u".*?" + re.sub(u"<1><2>",u"([^ ]*)",re.sub(u"<3>",u"[^ ]*",realword)) + u".*?",ur"\1",original) |
infix_2 = re.sub(u".*?" + re.sub(u"<3>",u"([^ ]*)",re.sub(u"<1><2>",u"[^ ]*",realword)) + u".*?",ur"\1",original) |
for eachinfix in infixes0: |
if infix01.startswith(eachinfix): |
infix0 = eachinfix |
infix_1 = infix01[len(eachinfix):] |
break |
else: |
infix0 = u"" |
infix_1 = infix01 |
gotinfix1 = False |
for eachinfix in infixes1: |
if infix_1.startswith(eachinfix): |
infix1 = eachinfix |
infix_1 = infix_1[len(eachinfix):] |
if infix_1 != u"": |
if re.search(u"<1><2><3>",realword): |
infix_2 = infix_1 |
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" ") |
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: |
return [u"[" + original + u"]",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u""] |
gotinfix1 = True |
break |
if gotinfix1 == False: |
if re.search(u"<1><2><3>",realword): |
if infix_1 == u"": |
infix_2 = infix_1 |
infix1 = u"" |
elif infix_1 == u"": |
infix1 = u"" |
else: |
return [u"[" + original + u"]",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u""] |
gotinfix2 = False |
for eachinfix in infixes2: |
if infix_2.startswith(eachinfix): |
infix2 = infix_2[:len(eachinfix)] |
infix_2 = infix_2[len(eachinfix) - 1:] |
gotinfix2 = True |
break |
if gotinfix2 == False or infix_2 != u"": |
if infix_2.startswith(end): |
suffixes = infix2[len(end) - 1:] + end |
elif infix_2 == u"": |
infix2 = u"" |
else: |
return [u"[" + original + u"]",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u""] |
# print u"0" + unicode(infix0) + u" 1" + unicode(infix1) + u" 2" + unicode(infix2) |
return [realword,infix0,infix1,infix2,u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u""] |
else: |
return [realword,u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u""] |
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][:-3] + 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 |
for pre in prefixes: |
if pref != u"": |
if pref.endswith(pre): |
foundprefs[wor].append(pre) |
pref = pref[:-len(pre)] |
if pref != u"": |
foundit = False |
break |
for pos in postfixes: |
if posf != u"": |
if posf.startswith(pos): |
foundposts[wor].append(pos) |
posf = posf[len(pos):] |
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 |