Rev 88 | Rev 90 | 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 -*- |
||
3 | |||
4 | import re |
||
66 | szabot | 5 | import dbconnector |
56 | szabot | 6 | |
65 | szabot | 7 | wordlist = dbconnector.getnavilist() |
8 | |||
9 | infixes1 = [u"awn", u"eyk", u"us", u"äp", u""] |
||
10 | 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""] |
||
11 | infixes3 = [u"äng", u"ats", u"eiy", u"ei", u"uy", u""] |
||
83 | szabot | 12 | prefixes = [u"tsay", u"fay", u"tsa", u"fra", u"pxe", u"ay", u"me", u"pe", u"le", u"nì", u"sä", u"tì", u"ke", u"fì", u"a"] |
13 | postfixes = [u"eyä", u"ìri", u"an", u"ìl", u"it", u"lo", u"ng", u"ri", u"ru", u"ti", u"ur", u"ve", u"yä", u"ya", u"tu", u"vi", u"yu", u"e", u"l", u"o", u"t", u"y", u"a", u"ä"] |
||
62 | szabot | 14 | #prefixesn = ur"(?P<npr>(?:(?:fì|tsa)?(?:me|pxe|ay|fra)?|(?:fay)?|(?:tsay)?)(?:fne)?(?:tì|sä)?" |
74 | szabot | 15 | #prefixesv = ur"(?P<vpr>(?:nì|sä|tì|rä'ä |ke )?)" |
56 | szabot | 16 | |
17 | def parseword(wordin): |
||
89 | szabot | 18 | ret = {"word": {"id": 0, "navi": u"[" + u" ".wordin[0] + u"]", "infix": u"", "type": u""}} |
65 | szabot | 19 | for word in wordlist: |
20 | foundit = True |
||
21 | foundprefs = [] |
||
22 | foundposts = [] |
||
74 | szabot | 23 | splitword = word["infix"].split(u" ") |
24 | if len(wordin) < len(splitword): |
||
68 | szabot | 25 | foundit = False |
26 | next |
||
65 | szabot | 27 | for wor in range(len(splitword)): |
76 | szabot | 28 | if not foundit: |
29 | break |
||
65 | szabot | 30 | foundprefs.append([]) |
31 | foundposts.append([]) |
||
32 | center = u"" |
||
33 | foundins = [u"", u"", u""] |
||
34 | pre = [] |
||
35 | post = [] |
||
36 | if u"<1>" in splitword[wor]: |
||
37 | for in1 in infixes1: |
||
38 | for in2 in infixes2: |
||
39 | for in3 in infixes3: |
||
40 | if splitword[wor].replace(u"<1><2>",in1+in2).replace(u"<3>",in3) in wordin[wor]: |
||
41 | center = splitword[wor].replace(u"<1><2>",in1+in2).replace(u"<3>",in3) |
||
42 | foundins = [in1, in2, in3] |
||
43 | break |
||
75 | szabot | 44 | if center != u"": break |
45 | if center != u"": break |
||
65 | szabot | 46 | else: |
47 | if splitword[wor] in wordin[wor]: |
||
48 | center = splitword[wor] |
||
74 | szabot | 49 | if center == u"": |
65 | szabot | 50 | foundit = False |
51 | break |
||
69 | szabot | 52 | pref, posf = wordin[wor].split(center) |
65 | szabot | 53 | for pre in prefixes: |
54 | if pref.endswith(pre): |
||
55 | foundprefs[wor].append(pre) |
||
80 | szabot | 56 | pref = pref[:-len(pre)] |
65 | szabot | 57 | if pref != u"": |
58 | foundit = False |
||
59 | break |
||
80 | szabot | 60 | for pos in postfixes: |
81 | szabot | 61 | if posf.startswith(pos): |
80 | szabot | 62 | foundposts[wor].append(pos) |
63 | posf = posf[len(pos):] |
||
82 | szabot | 64 | if posf != u"": |
80 | szabot | 65 | foundit = False |
66 | break |
||
65 | szabot | 67 | if foundit == True: |
68 | foundword = word |
||
56 | szabot | 69 | break |
87 | szabot | 70 | ret["pref"] = foundprefs |
71 | ret["post"] = foundposts |
||
72 | ret["inf"] = foundins |
||
65 | szabot | 73 | if foundit == True: |
71 | szabot | 74 | ret["word"] = foundword |
77 | szabot | 75 | return ret |
76 | |||
77 | def parsesent(sent): |
||
89 | szabot | 78 | sent = sent.strip().lower() |
79 | sent = re.sub(ur"[^\w ]",u"",sent) |
||
80 | sent = re.sub(ur"\ +",u" ",sent) |
||
81 | sent = sent.split(u" ") |
||
77 | szabot | 82 | ret = [] |
83 | left = len(sent) |
||
84 | while left: |
||
79 | szabot | 85 | word = parseword(sent[-left:]) |
78 | szabot | 86 | left -= len(word["word"]["navi"].split(" ")) |
77 | szabot | 87 | ret.append(word) |
70 | szabot | 88 | return ret |