WebSVN - navi - Path Comparison - / Rev 48 and / Rev 88

Ignore whitespace Rev 48 → Rev 88

 /tsimapiak/parse2.py
 ,0 → 1,84
+#!/usr/bin/python
+# -*- coding: utf-8 -*-
+import re
+import dbconnector
+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"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"]
+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"]
+#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 )?)"
+def parseword(wordin):
+    ret = {"word": {"id": 0, "navi": u"[" + u" ".join(wordin) + u"]", "infix": u"", "type": u""}}
+    for word in wordlist:
+        foundit = True
+        foundprefs = []
+        foundposts = []
+        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"":
+                foundit = False
+                break
+            pref, posf = wordin[wor].split(center)
+            for pre in prefixes:
+                if pref.endswith(pre):
+                    foundprefs[wor].append(pre)
+                    pref = pref[:-len(pre)]
+            if pref != u"":
+                foundit = False
+                break
+            for pos in postfixes:
+                if posf.startswith(pos):
+                    foundposts[wor].append(pos)
+                    posf = posf[len(pos):]
+            if posf != u"":
+                foundit = False
+                break
+        if foundit == True:
+            foundword = word
+            break
+    ret["pref"] = foundprefs
+    ret["post"] = foundposts
+    ret["inf"] = foundins
+    if foundit == True:
+        ret["word"] = foundword
+    return ret
+def parsesent(sent):
+    ret = []
+    left = len(sent)
+    while left:
+        word = parseword(sent[-left:])
+        left -= len(word["word"]["navi"].split(" "))
+        ret.append(word)
+    return ret

 /tsimapiak/parse.py
 ,0 → 1,86
+#!/usr/bin/python
+# -*- coding: utf-8 -*-
+import re
+from dbconnector import getnavilist
+wordlist = getnavilist()
+infixes0 = [ u"awn", u"eyk", u"us", u"äp" ]
+infixes1 = [ 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" ]
+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 wordlist:
+    regex = re.sub(u" ",u"[^ ]* [^ ]*",eachword["infix"])
+    regex = re.sub(u"^",u"[^ ]*",regex)
+    regex = re.sub(u"$",u"[^ ]*",regex)
+    regex = re.sub(u"<1><2>",u"[^ ]*",regex)
+    regex = re.sub(u"<3>",u"[^ ]*",regex)
+    if re.match(regex,original):
+      realword = eachword["infix"]
+      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"<1><2>.*",u"",realword)
+      middle = re.sub(u".*<1><2>(.*)<3>.*",ur"\1",realword)
+      end = re.sub(u".*<3>",u"",realword)
+      infix01 = re.sub(u".*?" + re.sub(u"<1><2>",u"([^ ]*)",re.sub(u"<3>",u"[^ ]*",realword)) + u".*?",ur"\1",original)
+      infix_2 = re.sub(u".*?" + re.sub(u"<3>",u"([^ ]*)",re.sub(u"<1><2>",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"<1><2><3>",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"<1><2><3>",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[:len(eachinfix)]
+          infix_2 = infix_2[len(eachinfix) - 1:]
+          gotinfix2 = True
+          break
+      if gotinfix2 == False or infix_2 != u"":
+        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""]

 /tsimapiak/dbconnector.py
 ,30 → 12,10
     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]
+        if row["infixes"]:
+            ret.append({"id": row["id"], "navi": row["navi"], "infix": row["infixes"], "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"], "type": row["partOfSpeech"]})
     db.close()
     return ret
 ,7 → 26,23
     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"], "type": row["partOfSpeech"]})
+        else:
+            ret.append({"id": row["id"], "navi": row["navi"], "infix": row["navi"], "type": row["partOfSpeech"]})
     db.close()
     return ret
+#def gettrans(id, cod):
+    #ret = []
+    #if cod not in (u"est",u"ptbr",u"de",u"eng",u"all"):
+        #return ret
+    #db = tornado.database.Connection("127.0.0.1", "navi", user="navi", password="navi")
+    #if cod == "all":
+            #for row in db.query("""
+            #SELECT *
+            #FROM `metaWords`
+            #WHERE id = ?""",idd):
+                #infix = makeinfix(row)
+            #ret.append([row["id"],row["navi"], infix, row["partOfSpeech"]])
+    #db.close()

 /webapp/templates/parse2.html
 ,0 → 1,38
+{% extends "base.html" %}
+{% block title %}Word parser{% end %}
+{% block body %}
+<b>Na'vi word:</b><br />
+<form action="/parse2" 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>Parts</th>
+    <th>Data</th>
+</tr>
+{% for wor in out %}
+<tr>
+    <td rowspan="3">{{ wor["word"]["navi"] }}</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>
+{% end %}
+</table>
+{% end %}
+<script type="text/javascript">
+document.getElementById("word").focus();
+</script>
+{% end %}

 /webapp/templates/parse.html
 ,0 → 1,20
+{% extends "base.html" %}
+{% block title %}Word parser{% end %}
+{% block body %}
+<b>Na'vi word:</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 type(out) == list %}
+{{ out[0]}} <br />
+{{ out[1]}} <br />
+{{ out[2]}} <br />
+{{ out[3]}}
+{% end %}
+<script type="text/javascript">
+document.getElementById("word").focus();
+</script>
+{% end %}

 /webapp/main.py
 ,8 → 9,10
 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
+from tsimapiak import parse2
 class Index(tornado.web.RequestHandler):
     def get(self):
 ,7 → 27,7
             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)
 ,20 → 39,46
 class TestDB(tornado.web.RequestHandler):
     def get(self):
-        lis = getnavilist()
+        lis = dbconnector.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").strip()
+        except:
+            self.redirect("/parse")
+        out = parse.parsefix(word)
+        self.render("templates/parse.html", last=word, out=out)
+class Parse2(tornado.web.RequestHandler):
+    def get(self):
+        self.render("templates/parse2.html", last="", out=None)
+    def post(self):
+        try:
+            word = self.get_argument("word").strip()
+        except:
+            self.redirect("/parse2")
+        out = parse2.parsesent(word.split(u" "))
+        self.render("templates/parse2.html", last=word, out=out)
 application = tornado.web.Application([
     ("/", Index),
     ("/number", Number),
     ("/restart", Restart),
-    ("/testdb", TestDB)
+    ("/testdb", TestDB),
+    ("/parse", Parse),
+    ("/parse2", Parse2)
 ])
 if __name__ == "__main__":
     http_server = tornado.httpserver.HTTPServer(application)
     http_server.listen(1337)
-    tornado.autoreload.start()
+    #tornado.autoreload.start()
     tornado.ioloop.IOLoop.instance().start()

Subversion Repositories navi

Compare Revisions

Ignore whitespace Rev 48 → Rev 88