Subversion Repositories navi

Rev

Rev 176 | Rev 181 | Go to most recent revision | Details | Compare with Previous | Last modification | View Log | RSS feed

Rev Author Line No. Line
56 szabot 1
#!/usr/bin/python
2
# -*- coding: utf-8 -*-
176 muzer 3
#    This file is part of Tsim Apiak.
4
#
5
#    Tsim Apiak is free software: you can redistribute it and/or modify
6
#    it under the terms of the GNU General Public Licence as published by
7
#    the Free Software Foundation, either version 3 of the Licence, or
8
#    (at your option) any later version. 
9
# 
10
#    In addition to this, you must also comply with clause 4 of the
11
#    Apache Licence, version 2.0, concerning attribution. Where there
12
#    is a contradiction between the two licences, the GPL
13
#    takes preference.
14
#
15
#    Foobar is distributed in the hope that it will be useful,
16
#    but WITHOUT ANY WARRANTY; without even the implied warranty of
17
#    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
18
#    GNU General Public License for more details.
19
#
20
#    You should have received a copy of the GNU General Public License
21
#    along with Tsim Apiak.  If not, see <http://www.gnu.org/licenses/>.
56 szabot 22
 
23
import re
66 szabot 24
import dbconnector
103 szabot 25
import parsenum
56 szabot 26
 
65 szabot 27
wordlist = dbconnector.getnavilist()
28
 
94 szabot 29
infixes1 = (u"awn", u"eyk", u"us", u"äp", u"")
30
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"")
31
infixes3 = (u"äng", u"ats", u"eiy", u"ei", u"uy", u"")
156 muzer 32
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")
171 muzer 33
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")
121 szabot 34
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")
62 szabot 35
#prefixesn = ur"(?P<npr>(?:(?:fì|tsa)?(?:me|pxe|ay|fra)?|(?:fay)?|(?:tsay)?)(?:fne)?(?:tì|sä)?"
74 szabot 36
#prefixesv = ur"(?P<vpr>(?:nì|sä|tì|rä'ä |ke )?)"
56 szabot 37
 
91 szabot 38
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""))
39
 
56 szabot 40
def parseword(wordin):
90 szabot 41
    ret = {"word": {"id": 0, "navi": u"[" + wordin[0] + u"]", "infix": u"", "type": u""}}
65 szabot 42
    for word in wordlist:
43
        foundit = True
44
        foundprefs = []
45
        foundposts = []
99 szabot 46
        lenited = False
74 szabot 47
        splitword = word["infix"].split(u" ")
172 muzer 48
        foundins = [u"", u"", u""]
74 szabot 49
        if len(wordin) < len(splitword):
68 szabot 50
            foundit = False
51
            next
65 szabot 52
        for wor in range(len(splitword)):
76 szabot 53
            if not foundit:
54
                break
65 szabot 55
            foundprefs.append([])
56
            foundposts.append([])
57
            center = u""
58
            pre = []
59
            post = []
60
            if u"<1>" in splitword[wor]:
61
                for in1 in infixes1:
180 szabot 62
                    if in1 in splitword[wor]:
63
                        tempin1.append(in1)
64
                for in2 in infixes2:
65
                    if in2 in splitword[wor]:
66
                        tempin2.append(in2)
67
                for in3 in infixes3:
68
                    if in3 in splitword[wor]:
69
                        tempin3.append(in3)
70
                for in1 in tempin1:
71
                    for in2 in tempin2:
72
                        for in3 in tempin3:
65 szabot 73
                            if splitword[wor].replace(u"<1><2>",in1+in2).replace(u"<3>",in3) in wordin[wor]:
74
                                center = splitword[wor].replace(u"<1><2>",in1+in2).replace(u"<3>",in3)
75
                                foundins = [in1, in2, in3]
76
                                break
75 szabot 77
                        if center != u"": break
78
                    if center != u"": break
65 szabot 79
            else:
80
                if splitword[wor] in wordin[wor]:
81
                    center = splitword[wor]
95 szabot 82
                if center == u"":
91 szabot 83
                    for i in lenit:
92 szabot 84
                        temp = u""
91 szabot 85
                        if splitword[wor].startswith(i[0]):
92 szabot 86
                            temp = i[1] + splitword[wor][len(i[0]):]
87
                            if temp in wordin[wor]:
99 szabot 88
                                lenited = True
92 szabot 89
                                center = temp
95 szabot 90
                if center == u"":
91
                    if splitword[wor].endswith(u"nga"):
97 szabot 92
                        temp = splitword[wor][:-3] + u"ng"
95 szabot 93
                        if temp in wordin[wor]:
94
                            center = temp
103 szabot 95
                    if splitword[wor].endswith(u"po"):
174 muzer 96
                        temp = splitword[wor][:-2] + u"p"
103 szabot 97
                        if temp in wordin[wor]:
98
                            center = temp
74 szabot 99
            if center == u"":
65 szabot 100
                foundit = False
101
                break
91 szabot 102
            temp = wordin[wor].split(center)
103
            if len(temp) != 2:
104
                foundit = False
105
                break
106
            pref, posf = temp
119 szabot 107
            last = u""
108
            while last != pref:
109
                last = pref
110
                for pre in prefixes:
111
                    if pref != u"":
112
                        if pref.endswith(pre):
113
                            if pre in foundprefs[wor]:
114
                                break
115
                            foundprefs[wor].append(pre)
116
                            pref = pref[:-len(pre)]
120 szabot 117
                            break
65 szabot 118
            if pref != u"":
119
                foundit = False
120
                break
119 szabot 121
            last = u""
122
            while last != posf:
123
                last = posf
124
                for pos in postfixes:
125
                    if posf != u"":
126
                        if posf.startswith(pos):
127
                            if pos in foundposts[wor]:
128
                                break
129
                            foundposts[wor].append(pos)
130
                            posf = posf[len(pos):]
120 szabot 131
                            break
82 szabot 132
            if posf != u"":
80 szabot 133
                foundit = False
134
                break
65 szabot 135
        if foundit == True:
136
            foundword = word
56 szabot 137
            break
87 szabot 138
    ret["pref"] = foundprefs
139
    ret["post"] = foundposts
140
    ret["inf"] = foundins
99 szabot 141
    ret["len"] = lenited
65 szabot 142
    if foundit == True:
71 szabot 143
        ret["word"] = foundword
77 szabot 144
    return ret
145
 
146
def parsesent(sent):
101 szabot 147
    sent = sent.strip().lower().replace(u"’", u"'")
100 szabot 148
    sent = re.sub(ur"[^\wìä' ]",u"",sent)
89 szabot 149
    sent = re.sub(ur"\ +",u" ",sent)
150
    sent = sent.split(u" ")
77 szabot 151
    ret = []
152
    left = len(sent)
153
    while left:
108 szabot 154
        word = parsenum.parse(sent[len(sent)-left])
103 szabot 155
        if word == None:
156
            word = parseword(sent[-left:])
78 szabot 157
        left -= len(word["word"]["navi"].split(" "))
77 szabot 158
        ret.append(word)
136 muzer 159
    return ret