50 lines
1005 B
Python
50 lines
1005 B
Python
|
# Bryan Schumaker (8/10/2010)
|
||
|
|
||
|
ttable = None
|
||
|
translate = None
|
||
|
|
||
|
class Index:
|
||
|
def __init__(self):
|
||
|
self.tokens = dict()
|
||
|
|
||
|
def setup(self):
|
||
|
import string
|
||
|
space = ord(" ")
|
||
|
strip = u"\"#$%&'*+<=>@[]^`{|}~.?!"
|
||
|
split = u"-\/,:;()_~+"
|
||
|
upper = string.uppercase
|
||
|
lower = string.lowercase
|
||
|
translate = string.translate
|
||
|
|
||
|
ttable = dict((ord(c),None) for c in strip)
|
||
|
splitt = dict((ord(c),space) for c in split)
|
||
|
lowert = dict((ord(c),ord(lower[i])) for i,c in enumerate(upper))
|
||
|
for t in (splitt, lowert):
|
||
|
for c in t:
|
||
|
ttable[c] = t[c]
|
||
|
|
||
|
def insert(self, id, tags):
|
||
|
global ttable
|
||
|
idset = set([id])
|
||
|
tokens = self.tokens
|
||
|
|
||
|
if ttable == None:
|
||
|
self.setup()
|
||
|
|
||
|
for tag in tags:
|
||
|
words = translate(tag,ttable).split()
|
||
|
for word in words:
|
||
|
set = tokens.get(word,None)
|
||
|
if set == None:
|
||
|
tokens[word] = idset
|
||
|
else:
|
||
|
set.update(idset)
|
||
|
for l in word:
|
||
|
set = tokens.get(l, None)
|
||
|
if set == None:
|
||
|
tokens[word] = idset
|
||
|
else:
|
||
|
set.update(idset)
|
||
|
|
||
|
|