/tsimapiak/parse2.py |
---|
3,7 → 3,6 |
import re |
import dbconnector |
import parsenum |
wordlist = dbconnector.getnavilist() |
12,7 → 11,7 |
infixes3 = (u"äng", u"ats", u"eiy", u"ei", u"uy", u"") |
prefixes = (u"a", u"pe", u"le", u"nì", u"sä", u"tì", u"fne", u"tsay", u"fay", u"fra", u"pxe", u"ay", u"me", u"tsa", u"fì", u"ke") |
adpositions = (u"kxamlä", u"mungwrr", u"nemfa", u"pximaw", u"pxisre", u"tafkip", u"takip", u"teri", u"mìkam", u"ìla", u"fkip", u"fpi", u"ftu", u"kip", u"lok", u"luke", u"maw", u"pxel", u"pxaw", u"rofa", 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") |
postfixes = (u"an", u"ng", u"eyä", u"e", u"tsyìp", u"o", 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"l", u"t", u"y", u"a", u"ä") + adpositions |
postfixes = (u"an", u"eyä", u"e", u"tsyìp", u"o", u"ìri", 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"l", u"t", u"y", u"a", u"ä") + adpositions |
#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 )?)" |
64,10 → 63,6 |
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 |
104,16 → 99,14 |
return ret |
def parsesent(sent): |
sent = sent.strip().lower().replace(u"’", u"'") |
sent = re.sub(ur"[^\wìä' ]",u"",sent) |
sent = sent.strip().lower() |
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[0]) |
if word == None: |
word = parseword(sent[-left:]) |
word = parseword(sent[-left:]) |
left -= len(word["word"]["navi"].split(" ")) |
ret.append(word) |
return ret |
/tsimapiak/dbconnector.py |
---|
9,10 → 9,9 |
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"""): |
ORDER BY NL DESC"""): |
if row["infixes"]: |
ret.append({"id": row["id"], "navi": row["navi"], "infix": row["infixes"].lower(), "type": row["partOfSpeech"]}) |
else: |
/tsimapiak/parsenum.py |
---|
30,12 → 30,12 |
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): |
51,34 → 51,30 |
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"." |
return {"word": {"id": 0, "navi": retnum, "infix": u"", "type": u""}, "pref": [mat[0]], "post": [mat[6], mat[7]], "inf": [u"", u"", u""], "len": False, "dec": numout, "oct": numdec} |
#return numout, numoct |
return numout, numoct |
if __name__ == "__main__": |
/webapp/main.py |
---|
29,10 → 29,8 |
self.redirect("/number") |
numout = parsenum.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): |