| /tsimapiak/dbconnector.py |
|---|
| 0,0 → 1,52 |
| #!/usr/bin/python |
| # -*- coding: utf-8 -*- |
| import tornado.database |
| import re |
| def getnavilist(): |
| ret = [] |
| current = u"" |
| db = tornado.database.Connection("127.0.0.1", "navi", user="navi", password="navi") |
| for row in db.query(""" |
| SELECT *, CHAR_LENGTH(navi) AS NL |
| FROM `metaWords` |
| ORDER BY NL DESC"""): |
| if row["partOfSpeech"] in (u"v.", u"vin.", u"vtr."): |
| current = unicode(row["ipa"]) |
| current = current.replace(ur"ɛ",ur"e").replace(ur".",ur"").replace(ur"ɾ",ur"r") \ |
| .replace(ur"ɪ",ur"ì").replace(ur"ˈ",ur"").replace(ur"'",ur"x") \ |
| .replace(ur"ŋ",ur"ng").replace(ur"j",ur"y").replace(ur"ʔ",ur"'") \ |
| .replace(ur"æ",ur"ä").replace(ur"ˌ",ur"").replace(ur"\t{ts}",ur"ts") \ |
| .replace(ur"ṛ",ur"rr").replace(ur"ḷ",ur"ll").replace(ur"k̚",ur"k ") \ |
| .replace(ur"p̚",ur"p ").replace(ur"t̚",ur"t ").replace(ur"'̚",ur"' ") \ |
| .replace(u"\\",ur"").replace(ur"(",ur"").replace(ur")",ur"") \ |
| .replace(ur"[",ur"").replace(ur"]",ur"").replace(ur" "," ") \ |
| .strip() |
| current = re.sub(ur" or.*","",current) |
| current = re.sub(ur"z(.*)engk(.*)e",ur"z\1enk\2e",current) |
| current = re.sub(ur"t(.*)ì(m|n)\ ",ur"t\1ìng ",current) |
| current = current.split(ur"$cdot$") |
| if len(current) == 3: |
| current = current[0] + u"<0><1>" + current[1] + u"<2>" + current[2] |
| elif len(current) == 2: |
| current = current[0] + u"<0><1><2>" + current[1] |
| else: |
| current = u"<0><1><2>" + current[0] |
| else: |
| current = unicode(row["navi"]) |
| ret.append([row["id"], row["navi"], current, row["partOfSpeech"]]) |
| db.close() |
| return ret |
| def getnavi(word): |
| ret = [] |
| db = tornado.database.Connection("127.0.0.1", "navi", user="navi", password="navi") |
| for row in db.query(""" |
| SELECT * |
| FROM `metaWords` |
| WHERE navi = ?""",word): |
| ret.append([row["id"],row["navi"], row["infix"], row["partOfSpeech"]]) |
| db.close() |
| return ret |
| /webapp/main.py |
|---|
| 5,12 → 5,12 |
| import tornado.ioloop |
| import tornado.web |
| import tornado.autoreload |
| import tornado.database |
| import os |
| import re |
| from tsimapiak.parsenum import parse |
| from tsimapiak.dbconnector import getnavilist |
| class Index(tornado.web.RequestHandler): |
| def get(self): |
| 34,39 → 34,6 |
| def get(self): |
| os.system("/usr/bin/restartnavi") |
| def getwords(): |
| ret = [] |
| current = u"" |
| db = tornado.database.Connection("127.0.0.1", "navi", user="navi", password="navi") |
| for row in db.query(""" |
| SELECT *, CHAR_LENGTH(navi) AS NL |
| FROM `metaWords` |
| ORDER BY NL DESC"""): |
| if row["partOfSpeech"] in (u"v.", u"vin.", u"vtr."): |
| current = unicode(row["ipa"]) |
| current = current.replace(ur"ɛ",ur"e").replace(ur".",ur"").replace(ur"ɾ",ur"r") \ |
| .replace(ur"ɪ",ur"ì").replace(ur"ˈ",ur"").replace(ur"'",ur"x") \ |
| .replace(ur"ŋ",ur"ng").replace(ur"j",ur"y").replace(ur"ʔ",ur"'") \ |
| .replace(ur"æ",ur"ä").replace(ur"ˌ",ur"").replace(ur"\t{ts}",ur"ts") \ |
| .replace(ur"ṛ",ur"rr").replace(ur"ḷ",ur"ll").replace(ur"k̚",ur"k ") \ |
| .replace(ur"p̚",ur"p ").replace(ur"t̚",ur"t ").replace(ur"'̚",ur"' ") \ |
| .replace(u"\\",ur"").replace(ur"(",ur"").replace(ur")",ur"") \ |
| .replace(ur"[",ur"").replace(ur"]",ur"").replace(ur" "," ") \ |
| .strip() |
| current = re.sub(ur" or.*","",current) |
| current = re.sub(ur"z(.*)engk(.*)e",ur"z\1enk\2e",current) |
| current = re.sub(ur"t(.*)ì(m|n)\ ",ur"t\1ìng ",current) |
| current = current.split(ur"$cdot$") |
| if len(current) == 3: |
| current = current[0] + u"<0><1>" + current[1] + u"<2>" + current[2] |
| elif len(current) == 2: |
| current = current[0] + u"<0><1><2>" + current[1] |
| else: |
| current = u"<0><1><2>" + current[0] |
| else: |
| current = unicode(row["navi"]) |
| ret.append([row["id"], row["navi"], current, row["partOfSpeech"]]) |
| return ret |
| class TestDB(tornado.web.RequestHandler): |
| def get(self): |
| /dev/naviparse.py |
|---|
| 0,0 → 1,91 |
| # -*- coding: utf-8 -*- |
| import re |
| strings = [ u"oe", u"nga", u"k<0><1>am<2>e", u"t<0><1><2>ìng nari", u"s<0><1><2>i" ] |
| infixes0 = [ u"awn", u"eyk", u"us", u"äp" ] |
| infixes1 = [ u"ìyev", u"iyev", 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" ] |
| # 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 |
| def parsefix(original): |
| realword = u"" |
| infix0 = u"" |
| infix1 = u"" |
| infix2 = u"" |
| infix01 = u"" |
| infix_1 = u"" |
| infix_2 = u"" |
| for eachword in strings: |
| regex = re.sub(u" ",u"[^ ]* [^ ]*",eachword) |
| regex = re.sub(u"^",u"[^ ]*",regex) |
| regex = re.sub(u"$",u"[^ ]*",regex) |
| regex = re.sub(u"<0><1>",u"[^ ]*",regex) |
| regex = re.sub(u"<2>",u"[^ ]*",regex) |
| if re.match(regex,original): |
| realword = eachword |
| 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""] |
| else: |
| if re.search(u"<",realword): |
| beginning = re.sub(u"<0><1>.*",u"",realword) |
| middle = re.sub(u".*<0><1>(.*)<2>.*",ur"\1",realword) |
| end = re.sub(u".*<2>",u"",realword) |
| infix01 = re.sub(u".*?" + re.sub(u"<0><1>",u"([^ ]*)",re.sub(u"<2>",u"[^ ]*",realword)) + u".*?",ur"\1",original) |
| infix_2 = re.sub(u".*?" + re.sub(u"<2>",u"([^ ]*)",re.sub(u"<0><1>",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"<0><1><2>",realword): |
| infix_2 = infix_1 |
| else: |
| return [u"[" + original + 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"<0><1><2>",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""] |
| gotinfix2 = False |
| for eachinfix in infixes2: |
| if infix_2.startswith(eachinfix): |
| infix2 = infix_2 |
| gotinfix2 = True |
| break |
| if gotinfix2 == False: |
| 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""] |
| # 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""] |
| else: |
| return [realword,u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u"",u""] |
| print parsefix(u"oel") |
| print parsefix(u"ngati") |
| print parsefix(u"kameie") |
| print parsefix(u"kìyevame") |
| print parsefix(u"English") |
| print parsefix(u"keykivame") |
| print parsefix(u"tìsusiti") |
| print parsefix(u"tayìng nari") |