def removeDocument(self, docid):
try:
wordids = self._doc2wid[docid]
except KeyError:
return # silently ignore
wordids = wordids.get() # unwrap _PS instance
tree = self._wid2doc
tree_has = tree.has_key
for wordid in decode(wordids):
if tree_has(wordid):
try:
tree[wordid].remove(docid)
except KeyError:
pass
if not tree[wordid]:
del tree[wordid]
del self._doc2wid[docid]
del self._docweight[docid]
self._length.change(-1)
def removeDocument(self, docid):
try:
wordids = self._doc2wid[docid]
except KeyError:
return # silently ignore
wordids = wordids.get() # unwrap _PS instance
tree = self._wid2doc
tree_has = tree.has_key
for wordid in decode(wordids):
if tree_has(wordid):
try:
tree[wordid].remove(docid)
except KeyError:
pass
if not tree[wordid]:
del tree[wordid]
del self._doc2wid[docid]
del self._docweight[docid]
self._length.change(-1)
def getPositions(self, docid, wordid):
""" return a sequence of positions of occurrences of wordid within
a document given by its docid.
"""
encoded_wid = encode((wordid, ))
encoded_document = self._doc2wid[docid].get()
positions = IITreeSet()
for pos, wid in enumerate(decode(encoded_document)):
if wid == wordid:
positions.insert(pos)
return positions
def getPositions(self, docid, wordid):
""" return a sequence of positions of occurrences of wordid within
a document given by its docid.
"""
encoded_wid = encode((wordid,))
encoded_document = self._doc2wid[docid].get()
positions = IITreeSet()
for pos, wid in enumerate(decode(encoded_document)):
if wid == wordid:
positions.insert(pos)
return positions
def insertDocument(self, docid, widlist):
if not self._doc2wid.has_key(docid):
self._length.change(1)
enc_widlist = encode(widlist)
old_enc_widlist = self._doc2wid.get(docid)
if old_enc_widlist is not None:
old_enc_widlist = old_enc_widlist.get() # unwrap _PS instance
removed_wordids = []
if old_enc_widlist != enc_widlist:
self._doc2wid[docid] = _PS(enc_widlist)
if old_enc_widlist is not None:
old_widlist = IISet(decode(old_enc_widlist))
removed_wordids = difference(old_widlist, IISet(widlist))
tree = self._wid2doc
tree_has = tree.has_key
count = 0
for wid in widlist:
count += 1
if not tree_has(wid):
tree[wid] = DocidList([docid])
else:
if not docid in tree[wid]:
tree[wid].insert(docid)
for wid in removed_wordids:
if tree_has(wid):
try:
tree[wid].remove(docid)
except KeyError:
pass
self._docweight[docid] = count
def insertDocument(self, docid, widlist):
if not self._doc2wid.has_key(docid):
self._length.change(1)
enc_widlist = encode(widlist)
old_enc_widlist = self._doc2wid.get(docid)
if old_enc_widlist is not None:
old_enc_widlist = old_enc_widlist.get() # unwrap _PS instance
removed_wordids = []
if old_enc_widlist != enc_widlist :
self._doc2wid[docid] = _PS(enc_widlist)
if old_enc_widlist is not None:
old_widlist = IISet(decode(old_enc_widlist))
removed_wordids = difference(old_widlist, IISet(widlist))
tree = self._wid2doc
tree_has = tree.has_key
count = 0
for wid in widlist:
count += 1
if not tree_has(wid):
tree[wid] = DocidList([docid])
else:
if not docid in tree[wid]:
tree[wid].insert(docid)
for wid in removed_wordids:
if tree_has(wid):
try:
tree[wid].remove(docid)
except KeyError:
pass
self._docweight[docid] = count
def getWordIdsForDocId(self, docid):
try:
ps_wrapper = self._doc2wid[docid]
return decode(ps_wrapper.get())
except KeyError:
raise StorageException('No such docid: %d' % docid)
def getWordIdsForDocId(self, docid):
try:
ps_wrapper = self._doc2wid[docid]
return decode(ps_wrapper.get())
except KeyError:
raise StorageException('No such docid: %d' % docid)