/tsimapiak/parsenum.py |
---|
1,81 → 1,121 |
#!/usr/bin/python |
# -*- coding: utf-8 -*- |
import re |
num = [u"kew", |
u"'aw", |
u"mune", |
u"pxey", |
u"tsìng", |
u"mrr", |
u"pukap", |
u"kinä"] |
rem = [u"aw", |
u"mun", |
u"pey", |
u"sìng", |
u"mrr", |
u"fu", |
u"hin"] |
base = [u"", |
u"me", |
u"pxe", |
u"tsì", |
u"mrr", |
u"pu", |
u"ki"] |
numre = \ |
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"))?$" |
numre = re.compile(numre) |
def parse(numin): |
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: |
return None |
numout = 0 |
numoct = 0 |
try: |
numout += rem.index(mat[4]) + 1 |
numoct += rem.index(mat[4]) + 1 |
except: |
try: |
numout += num.index(mat[4]) |
numoct += num.index(mat[4]) |
except: pass |
try: |
numout += (base.index(mat[3]) + 1) * 8 |
numoct += (base.index(mat[3]) + 1) * 10 |
except: pass |
try: |
numout += (base.index(mat[2]) + 1) * 8**2 |
numoct += (base.index(mat[2]) + 1) * 10**2 |
except: pass |
try: |
numout += (base.index(mat[1]) + 1) * 8**3 |
numoct += (base.index(mat[1]) + 1) * 10**3 |
except: pass |
try: |
numout += (base.index(mat[0]) + 1) * 8**4 |
numoct += (base.index(mat[0]) + 1) * 10**4 |
except: pass |
return numout, numoct |
if __name__ == "__main__": |
print parse(u"mrrvolaw") |
#!/usr/bin/python |
# -*- coding: utf-8 -*- |
num = [u"kew", |
u"'aw", |
u"mune", |
u"pxey", |
u"tsìng", |
u"mrr", |
u"pukap", |
u"kinä"] |
numord = [u"kew", |
u"'aw", |
u"mu", |
u"pxey", |
u"tsì", |
u"mrr", |
u"pu", |
u"ki"] |
rem = [u"aw", |
u"mun", |
u"pey", |
u"sìng", |
u"mrr", |
u"fu", |
u"hin"] |
remord = [u"aw", |
u"mu", |
u"pey", |
u"sì", |
u"mrr", |
u"fu", |
u"hi"] |
base = [u"", |
u"me", |
u"pxe", |
u"tsì", |
u"mrr", |
u"pu", |
u"ki"] |
def parse(numin): |
if u"mm" in numin: |
return None |
if (numin[0] == u"a") and (numin[len(numin)-1] == u"a"): |
return None |
prefs = [] |
posts = [] |
outoct = 0 |
outdec = 0 |
ret = {"word": {"id": 0, "navi": u"", "infix": u"", "type": u""}, "pref": [prefs], "post": [posts], "inf": [u"", u"", u""], "len": False, "dec": outdec, "oct": outoct} |
if numin[0] == u"a": |
prefs.append(u"a") |
numin = numin[1:] |
if numin[-2:] == u"ve": |
posts.append(u"ve") |
numin = numin[:-2] |
if numin[len(numin)-1] == u"a": |
posts.append(u"a") |
numin = numin[:-1] |
#base numbers |
for n in range(len(num)): |
if (numin == num[n]) or (numin == numord[n]): |
outoct = n |
outdec = n |
ret["word"]["navi"] = unicode(outdec) if not u"ve" in posts else unicode(outdec) + u"." |
ret["dec"] = outdec |
ret["oct"] = outoct |
return ret |
#other numbers |
for n in range(len(base)): |
if numin.startswith(base[n] + u"zazam"): |
outoct += (n+1) * (10**4) |
outdec += (n+1) * (8**4) |
numin = numin[len(base[n]) + 5:] |
for n in range(len(base)): |
if numin.startswith(base[n] + u"vozam"): |
outoct += (n+1) * (10**3) |
outdec += (n+1) * (8**3) |
numin = numin[len(base[n]) + 5:] |
for n in range(len(base)): |
if numin.startswith(base[n] + u"zam"): |
outoct += (n+1) * (10**2) |
outdec += (n+1) * (8**2) |
numin = numin[len(base[n]) + 3:] |
for n in range(len(base)): |
if numin.startswith(base[n] + u"vol"): |
outoct += (n+1) * 10 |
outdec += (n+1) * 8 |
numin = numin[len(base[n]) + 3:] |
if numin.startswith(base[n] + u"vo"): |
outoct += (n+1) * 10 |
outdec += (n+1) * 8 |
numin = numin[len(base[n]) + 2:] |
for n in range(len(rem)): |
if u"ve" in posts: |
if numin == remord[n]: |
outoct += n + 1 |
outdec += n + 1 |
numin = u"" |
else: |
if numin == rem[n]: |
outoct += n + 1 |
outdec += n + 1 |
numin = u"" |
if numin == u"": |
ret["word"]["navi"] = unicode(outdec) if not u"ve" in posts else unicode(outdec) + u"." |
ret["dec"] = outdec |
ret["oct"] = outoct |
return ret |
else: |
return None |
if __name__ == "__main__": |
print parse(u"mevolawve") |
/tsimapiak/dbconnector.py |
---|
9,33 → 9,14 |
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 |
SELECT * |
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] |
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"]}) |
else: |
current = unicode(row["navi"]) |
ret.append([row["id"], row["navi"], current, row["partOfSpeech"]]) |
ret.append({"id": row["id"], "navi": row["navi"], "infix": row["navi"].lower(), "type": row["partOfSpeech"]}) |
db.close() |
return ret |
46,7 → 27,20 |
SELECT * |
FROM `metaWords` |
WHERE navi = ?""",word): |
ret.append([row["id"],row["navi"], row["infix"], row["partOfSpeech"]]) |
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"]}) |
db.close() |
return ret |
def translate(id,language): |
db = tornado.database.Connection("127.0.0.1", "navi", user="navi", password="navi") |
for row in db.query(""" |
SELECT * |
FROM `localizedWords` |
WHERE id = ? AND languageCode = ?""",id,language): |
ret = row["localized"] |
break |
db.close() |
return ret |
/tsimapiak/parse.py |
---|
0,0 → 1,135 |
#!/usr/bin/python |
# -*- coding: utf-8 -*- |
import re |
import dbconnector |
import parsenum |
wordlist = dbconnector.getnavilist() |
infixes1 = (u"awn", u"eyk", u"us", u"äp", u"") |
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"") |
infixes3 = (u"äng", u"ats", u"eiy", u"ei", u"uy", u"") |
prefixes = (u"tsay", u"fay", u"fra", u"pxe", u"fne", u"tsa", u"pe", u"le", u"nì", u"sä", u"tì", u"ay", u"me", u"fì", u"ke", u"a") |
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"ìla", 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") |
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") |
#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 )?)" |
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"")) |
def parseword(wordin): |
ret = {"word": {"id": 0, "navi": u"[" + wordin[0] + u"]", "infix": u"", "type": u""}} |
for word in wordlist: |
foundit = True |
foundprefs = [] |
foundposts = [] |
lenited = False |
splitword = word["infix"].split(u" ") |
if len(wordin) < len(splitword): |
foundit = False |
next |
for wor in range(len(splitword)): |
if not foundit: |
break |
foundprefs.append([]) |
foundposts.append([]) |
center = u"" |
foundins = [u"", u"", u""] |
pre = [] |
post = [] |
if u"<1>" in splitword[wor]: |
for in1 in infixes1: |
for in2 in infixes2: |
for in3 in infixes3: |
if splitword[wor].replace(u"<1><2>",in1+in2).replace(u"<3>",in3) in wordin[wor]: |
center = splitword[wor].replace(u"<1><2>",in1+in2).replace(u"<3>",in3) |
foundins = [in1, in2, in3] |
break |
if center != u"": break |
if center != u"": break |
else: |
if splitword[wor] in wordin[wor]: |
center = splitword[wor] |
if center == u"": |
for i in lenit: |
temp = u"" |
if splitword[wor].startswith(i[0]): |
temp = i[1] + splitword[wor][len(i[0]):] |
if temp in wordin[wor]: |
lenited = True |
center = temp |
if center == u"": |
if splitword[wor].endswith(u"nga"): |
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 |
temp = wordin[wor].split(center) |
if len(temp) != 2: |
foundit = False |
break |
pref, posf = temp |
last = u"" |
while last != pref: |
last = pref |
for pre in prefixes: |
if pref != u"": |
if pref.endswith(pre): |
if pre in foundprefs[wor]: |
break |
foundprefs[wor].append(pre) |
pref = pref[:-len(pre)] |
break |
if pref != u"": |
foundit = False |
break |
last = u"" |
while last != posf: |
last = posf |
for pos in postfixes: |
if posf != u"": |
if posf.startswith(pos): |
if pos in foundposts[wor]: |
break |
foundposts[wor].append(pos) |
posf = posf[len(pos):] |
break |
if posf != u"": |
foundit = False |
break |
if foundit == True: |
foundword = word |
break |
ret["pref"] = foundprefs |
ret["post"] = foundposts |
ret["inf"] = foundins |
ret["len"] = lenited |
if foundit == True: |
ret["word"] = foundword |
return ret |
def parsesent(sent): |
sent = sent.strip().lower().replace(u"’", u"'") |
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[len(sent)-left]) |
if word == None: |
word = parseword(sent[-left:]) |
left -= len(word["word"]["navi"].split(" ")) |
if word["word"]["id"] != 0: |
word["translated"] = dbconnector.translate(word["word"]["id"],"eng") |
else: |
word["translated"] = word["word"]["navi"] |
ret.append(word) |
return ret |
/webapp/templates/parse.html |
---|
0,0 → 1,45 |
{% extends "base.html" %} |
{% block title %}Sentence parser{% end %} |
{% block body %} |
<b>Na'vi sentence:</b><br /> |
<form action="/parse" method="post"> |
<input id="word" name="word" type="text" value="{{last}}" style="width: 100%;" /> |
<input name="btn" type="submit" value="Parse!" /> |
</form> |
{% if out %} |
<table border="1"> |
<tr> |
<th>Words</th> |
<th>English</th> |
<th>Parts</th> |
<th>Data</th> |
</tr> |
{% for wor in out %} |
<tr> |
<td rowspan="4">{{ wor["word"]["navi"] }}</td> |
<td rowspan="4">{{ wor["translated"] }}</td> |
<td>Infixes:</td> |
<td>{{ u", ".join(wor["inf"]) }}</td> |
</tr> |
<tr> |
<td>Prefixes:</td> |
<td>{{ u"; ".join(u", ".join(x) for x in wor["pref"]) }}</td> |
</tr> |
<tr> |
<td>Postfixes:</td> |
<td>{{ u"; ".join(u", ".join(x) for x in wor["post"]) }}</td> |
</tr> |
<tr> |
<td>Lenited:</td> |
<td>{{ str(wor["len"]) }}</td> |
</tr> |
{% end %} |
</table> |
{% end %} |
<p>This program uses Eana Eltu for the list of words and infix positions (but nothing else), created by Tuiq and Taronyu. Thanks also go to the rest of the Learn Na'vi community!</p> |
<script type="text/javascript"> |
document.getElementById("word").focus(); |
</script> |
{% end %} |
/webapp/templates/base.html |
---|
1,6 → 1,7 |
<html> |
<head> |
<title>Tsim Apiak - {% block title %}Title{% end %}</title> |
<link rel="shortcut icon" type="image/x-icon" href="static/favicon.ico" /> |
<style type="text/css"> |
body { |
background: #145179; |
/webapp/templates/index.html |
---|
0,0 → 1,8 |
{% extends "base.html" %} |
{% block title %}Home{% end %} |
{% block body %} |
<a href="/number"><b>Number translator</b></a> - this webapp allows you to translate written-out Na'vi numbers into decimal and octal.<br /> |
<a href="/parse"><b>Parser</b></a> - this webapp can parse Na'vi sentences into the base words, prefixes, infixes and suffixes. It does not currently translate the words, but that will come. |
{% end %} |
/webapp/templates/number.html |
---|
17,4 → 17,4 |
<script type="text/javascript"> |
document.getElementById("num").focus(); |
</script> |
{% end %} |
{% end %} |
/webapp/static/favicon.ico |
---|
Cannot display: file marked as a binary type. |
svn:mime-type = application/octet-stream |
/webapp/static/favicon.ico |
---|
Property changes: |
Added: svn:mime-type |
## -0,0 +1 ## |
+application/octet-stream |
\ No newline at end of property |
Index: webapp/main.py |
=================================================================== |
--- webapp/main.py (revision 48) |
+++ webapp/main.py (revision 141) |
@@ -9,12 +9,13 @@ |
import os |
import re |
-from tsimapiak.parsenum import parse |
-from tsimapiak.dbconnector import getnavilist |
+from tsimapiak import parsenum |
+from tsimapiak import dbconnector |
+from tsimapiak import parse |
class Index(tornado.web.RequestHandler): |
def get(self): |
- self.redirect("/number") |
+ self.render("templates/index.html") |
class Number(tornado.web.RequestHandler): |
def get(self): |
@@ -25,10 +26,12 @@ |
num = self.get_argument("num").strip() |
except: |
self.redirect("/number") |
- numout = parse(num.replace(" ","")) |
+ numout = parsenum.parse(num.replace(" ","")) |
if numout == None: |
- numout = -1 |
- self.render("templates/number.html", last=num, numout=numout) |
+ numoutt = -1 |
+ else: |
+ numoutt = (numout["dec"], numout["oct"]) |
+ self.render("templates/number.html", last=num, numout=numoutt) |
class Restart(tornado.web.RequestHandler): |
def get(self): |
@@ -37,20 +40,37 @@ |
class TestDB(tornado.web.RequestHandler): |
def get(self): |
- lis = getnavilist() |
+ lis = dbconnector.getnavilist() |
text = u"id | navi | infix | partofspeech " |
text += u" ".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) |
+ |
+settings = { |
+ "static_path": os.path.join(os.path.dirname(__file__), "static") |
+} |
+ |
application = tornado.web.Application([ |
("/", Index), |
("/number", Number), |
("/restart", Restart), |
- ("/testdb", TestDB) |
-]) |
+ ("/testdb", TestDB), |
+ ("/parse", Parse) |
+], **settings) |
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() |
\ No newline at end of file |
/dev/naviparse.py |
---|
File deleted |