/dev/naviparse.py |
---|
0,0 → 1,92 |
# -*- 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", u"t<0><1><2>ìng" ] |
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,fne,lenited? |
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"",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"",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"",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"",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""] |
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") |
print parsefix(u"tìtusìng") |
/webapp/main.py |
---|
9,13 → 9,12 |
import os |
import re |
from tsimapiak import parsenum |
from tsimapiak import dbconnector |
from tsimapiak import parse |
from tsimapiak.parsenum import parse |
from tsimapiak.dbconnector import getnavilist |
class Index(tornado.web.RequestHandler): |
def get(self): |
self.render("templates/index.html") |
self.redirect("/number") |
class Number(tornado.web.RequestHandler): |
def get(self): |
26,12 → 25,10 |
num = self.get_argument("num").strip() |
except: |
self.redirect("/number") |
numout = parsenum.parse(num.replace(" ","")) |
numout = parse(num.replace(" ","")) |
if numout == None: |
numoutt = -1 |
else: |
numoutt = (numout["dec"], numout["oct"]) |
self.render("templates/number.html", last=num, numout=numoutt) |
numout = -1 |
self.render("templates/number.html", last=num, numout=numout) |
class Restart(tornado.web.RequestHandler): |
def get(self): |
40,33 → 37,20 |
class TestDB(tornado.web.RequestHandler): |
def get(self): |
lis = dbconnector.getnavilist() |
lis = getnavilist() |
text = u"id | navi | infix | partofspeech<br />" |
text += u"<br />".join(u" | ".join(unicode(y) for y in x) for x in lis) |
self.write(text) |
class Parse(tornado.web.RequestHandler): |
def get(self): |
self.render("templates/parse.html", last="", out=None) |
def post(self): |
try: |
word = self.get_argument("word") |
except: |
self.redirect("/parse") |
out = parse.parsesent(word) |
self.render("templates/parse.html", last=word, out=out) |
application = tornado.web.Application([ |
("/", Index), |
("/number", Number), |
("/restart", Restart), |
("/testdb", TestDB), |
("/parse", Parse) |
("/testdb", TestDB) |
]) |
if __name__ == "__main__": |
http_server = tornado.httpserver.HTTPServer(application) |
http_server.listen(1337) |
#tornado.autoreload.start() |
tornado.ioloop.IOLoop.instance().start() |
tornado.autoreload.start() |
tornado.ioloop.IOLoop.instance().start() |
/webapp/templates/index.html |
---|
File deleted |
/webapp/templates/parse.html |
---|
File deleted |
/webapp/templates/number.html |
---|
17,4 → 17,4 |
<script type="text/javascript"> |
document.getElementById("num").focus(); |
</script> |
{% end %} |
{% end %} |
/tsimapiak/parse.py |
---|
File deleted |
\ No newline at end of file |
/tsimapiak/dbconnector.py |
---|
9,14 → 9,33 |
current = u"" |
db = tornado.database.Connection("127.0.0.1", "navi", user="navi", password="navi") |
for row in db.query(""" |
SELECT * |
SELECT *, CHAR_LENGTH(navi) AS NL |
FROM `metaWords` |
WHERE partOfSpeech <> 'num.' AND partOfSpeech <> "prefix" |
ORDER BY CHAR_LENGTH(navi) DESC"""): |
if row["infixes"]: |
ret.append({"id": row["id"], "navi": row["navi"], "infix": row["infixes"].lower(), "type": row["partOfSpeech"]}) |
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$") |
if len(current) == 3: |
current = current[0] + u"<0><1>" + current[1] + u"<2>" + current[2] |
elif len(current) == 2: |
current = current[0] + u"<0><1><2>" + current[1] |
else: |
current = u"<0><1><2>" + current[0] |
else: |
ret.append({"id": row["id"], "navi": row["navi"], "infix": row["navi"].lower(), "type": row["partOfSpeech"]}) |
current = unicode(row["navi"]) |
ret.append([row["id"], row["navi"], current, row["partOfSpeech"]]) |
db.close() |
return ret |
27,9 → 46,7 |
SELECT * |
FROM `metaWords` |
WHERE navi = ?""",word): |
if row["infixes"]: |
ret.append({"id": row["id"], "navi": row["navi"], "infix": row["infixes"].lower(), "type": row["partOfSpeech"]}) |
else: |
ret.append({"id": row["id"], "navi": row["navi"], "infix": row["navi"].lower(), "type": row["partOfSpeech"]}) |
ret.append([row["id"],row["navi"], row["infix"], row["partOfSpeech"]]) |
db.close() |
return ret |
return ret |
/tsimapiak/parsenum.py |
---|
30,17 → 30,20 |
numre = \ |
u"^(a?)(?:(" + "|".join(base) + u")zazam??)?" + \ |
u"^(?:(" + "|".join(base) + u")zazam??)?" + \ |
u"(?:(" + "|".join(base) + u")vozam??)?" + \ |
u"(?:(" + "|".join(base) + u")zam??)?" + \ |
u"(?:(" + "|".join(base) + u")vo(?:l(?=a|))?)?" + \ |
u"((?:" + "|".join(rem) + u")|" + \ |
u"(?:" + "|".join(num) + u"))?((?:ve)?)(a?)$" |
u"(?:" + "|".join(num) + u"))?$" |
numre = re.compile(numre) |
def parse(numin): |
if numin in (u"a", u"aa", u"ave", u"avea", u"ve", u"vea"): |
if type(numin) != unicode: |
return None |
if numin == u"": |
return None |
numin = numin.replace(u"í",u"ì").replace(u"á",u"ä") |
try: |
mat = numre.match(numin).groups() |
except: |
48,43 → 51,31 |
numout = 0 |
numoct = 0 |
try: |
numout += rem.index(mat[5]) + 1 |
numoct += rem.index(mat[5]) + 1 |
numout += rem.index(mat[4]) + 1 |
numoct += rem.index(mat[4]) + 1 |
except: |
try: |
numout += num.index(mat[5]) |
numoct += num.index(mat[5]) |
numout += num.index(mat[4]) |
numoct += num.index(mat[4]) |
except: pass |
try: |
numout += (base.index(mat[4]) + 1) * 8 |
numoct += (base.index(mat[4]) + 1) * 10 |
numout += (base.index(mat[3]) + 1) * 8 |
numoct += (base.index(mat[3]) + 1) * 10 |
except: pass |
try: |
numout += (base.index(mat[3]) + 1) * 8**2 |
numoct += (base.index(mat[3]) + 1) * 10**2 |
numout += (base.index(mat[2]) + 1) * 8**2 |
numoct += (base.index(mat[2]) + 1) * 10**2 |
except: pass |
try: |
numout += (base.index(mat[2]) + 1) * 8**3 |
numoct += (base.index(mat[2]) + 1) * 10**3 |
numout += (base.index(mat[1]) + 1) * 8**3 |
numoct += (base.index(mat[1]) + 1) * 10**3 |
except: pass |
try: |
numout += (base.index(mat[1]) + 1) * 8**4 |
numoct += (base.index(mat[1]) + 1) * 10**4 |
numout += (base.index(mat[0]) + 1) * 8**4 |
numoct += (base.index(mat[0]) + 1) * 10**4 |
except: pass |
retnum = unicode(numout) |
if mat[6] != u"": |
retnum += u"." |
prefs = [] |
posts = [] |
if mat[0] != u"": |
prefs.append(mat[0]) |
if mat[6] != u"": |
posts.append(mat[6]) |
if mat[7] != u"": |
posts.append(mat[7]) |
return {"word": {"id": 0, "navi": retnum, "infix": u"", "type": u""}, "pref": [prefs], "post": [posts], "inf": [u"", u"", u""], "len": False, "dec": numout, "oct": numoct} |
#return numout, numoct |
return numout, numoct |
if __name__ == "__main__": |
print parse(u"mrrvolawvea") |
print parse(u"mrrvolaw") |