autocomplete-redis is a quora like automatic autocompletion based on redis.

• install pip (if you haven't yet) easy_install pip

• install pymmseg (support for Chinese Characters)： pip install pymmseg

• install autocomplete-redis： pip install -e git+https://github.com/fengli/autocomplete-redis.git#egg=autocomplete-dev

• Assume you have few items to index.

items=[{"uid":'1', "score":9, "term": u"hello world, that's great"},
       {"uid":'2', "score":10, "term": u"what the hell or yell"},
       {"uid":'3', "score":8.5, "term":u"World is like a box of chocolate"},
      ]

The code for build the index and search is simple:

from autocomplete import Autocomplete

#build index
au = Autocomplete ("scope")
for item in items:
  au.add_item (item)
#search
restuls = au.search_query (u'hel')

print results
[{'term': 'what the hell or yell', 'score': 10, 'uid': '2'}, {'term': "hello world, that's great", 'score': 9, 'uid': '1'}]

• Convention: the item you pass to autocomplete should have at least "uid" and "term", "score" is optional, but it's important if you want to return based on ranking. And you could have other fields as you like.

{"uid":'1', "score":9, "term": u"hello world, that's great", 'meta':"1992"}

• uid: the unique identifier for your item

• score: the returned items sorted by this value.

• term: the string to be indexed.

• def __init__ (self, scope, redisaddr="localhost", limits=5, cached=True)

  • scope: Scope allows you to index multiple independent indexes.
  • redisaddr: your redis address
  • limits: How many results you want to get.
  • cached: Cache multiple keys combination?

• def del_index (self)

Delete all the indexes. Warning: all data will be deleted.

• def add_item (self,item)

Add item to index.

• def del_item (self,item)

Delete item from index.

• def update_item (self, item)

Update item indexed with item['uid'] with the new version.

• def search_query (self,prefix)

Search in database for all items that item['term'] included PREFIX

• http://readpi.com