Subversion Repositories navi

Compare Revisions

Ignore whitespace Rev 68 → Rev 69

/tsimapiak/parse2.py
6,13 → 6,6
 
wordlist = dbconnector.getnavilist()
 
#wordlist = [{"id": 1, "navi": u"si", "infix": u"s<1><2><3>i", "type": u"v."},
#{"id": 2, "navi": u"wou", "infix": u"w<1><2>o<3>u", "type": u"v."},
#{"id": 3, "navi": u"kame", "infix": u"k<1><2>am<3>e", "type": u"v."},
#{"id": 4, "navi": u"tìng", "infix": u"t<1><2><3>ìng", "type": u"v."},
#{"id": 5, "navi": u"yerik", "infix": u"yerik", "type": u"n."},
#{"id": 6, "navi": u"nari", "infix": u"nari", "type": u"n."}]
 
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""]
22,18 → 15,18
 
def parseword(wordin):
wordin = wordin.split(" ")
ret = {"word": {"id": 0, "navi": u" ".join(wordin), "infix": u"", "type": u""}}
ret = [{"word": {"id": 0, "navi": u" ".join(wordin), "infix": u"", "type": u""}}]
foundword = u""
for word in wordlist:
foundinf = False
foundit = True
foundprefs = []
foundposts = []
splitword = word["infix"].split(" ")
if len(wordin) != len(splitword):
if len(wordin) < len(splitword):
foundit = False
next
for wor in range(len(splitword)):
foundinf = False
foundprefs.append([])
foundposts.append([])
center = u""
59,7 → 52,7
if not foundinf:
foundit = False
break
pref, posf = wordin[wor].replace(center,u" ").split(" ")
pref, posf = wordin[wor].split(center)
for pre in prefixes:
if pref.endswith(pre):
foundprefs[wor].append(pre)
72,17 → 65,11
foundword = word
break
if foundit == True:
ret["pref"] = foundprefs
ret["post"] = foundposts
ret["inf"] = foundins
ret["word"] = foundword
return ret
 
#print(parseword([u"tìtìng"]))
#print(parseword([u"keykame"]))
#print(parseword([u"kameie"]))
#print(parseword([u"tìsusiti"]))
#print(parseword([u"weykayoeiu"]))
#print(parseword([u"yerikit"]))
#print(parseword([u"menari"]))
#print(parseword([u"menariri"]))
ret[0]["pref"] = foundprefs
ret[0]["post"] = foundposts
ret[0]["inf"] = foundins
ret[0]["word"] = foundword
if len(wordin) > len(splitword):
return ret + parseword(u" ".join(wordin[len(splitword):]))
else:
return ret