add support for cache construction from an index (#33)

PR introduces the ability to load an LLMCache from an underlying index.
Also makes some adjustments to the index class to support this.

- Add support for `SemanticCache.from_index(...)`
- Update implementation of the `SearchIndex`:
    - move `key_field` as a `load` method arg
- add `_get_key` method to create the redis key for an object and
associated test
- Update `SearchIndex.from_existing(...)` to accept a `redis_url` and
`connection_args`

Author: tylerhutcherson

README.md

@@ -67,7 +67,6 @@ index:
   name: user_index
   storage_type: hash
   prefix: users
-  key_field: user
 
 fields:
   # define tag fields

conftest.py

@@ -11,6 +11,9 @@
 aredis = get_async_redis_connection(REDIS_ADDRESS)
 redis = get_redis_connection(REDIS_ADDRESS)
 
+@pytest.fixture()
+def redis_url():
+    return REDIS_ADDRESS
 
 @pytest.fixture
 def async_client():

docs/examples/openai_qna.ipynb

@@ -748,8 +748,6 @@
     "index:\n",
     "    name: wiki\n",
     "    prefix: oaiWiki\n",
-    "    key_field: id\n",
-    "    storage_type: hash\n",
     "\n",
     "fields:\n",
     "    text:\n",

docs/user_guide/getting_started_01.ipynb

@@ -122,7 +122,6 @@
     "The index specification determines how data will be stored in Redis. This includes\n",
     "- ``name``: the name of the index\n",
     "- (Optional) ``prefix``: key prefix for each loaded entry\n",
-    "- (Optional) ``key_field``: field within the dataset to use as unique keys. If not specified, a unique key will be generated for each entry.\n",
     "\n",
     "The field specification determines what fields within the dataset will be available for queries. Each field corresponds to the name of a **column** within the dataset. The values within each specified column are arguments for the creation of that index that correspond directly to ``redis-py`` arguments.\n",
     "\n",
@@ -134,8 +133,6 @@
     "index:\n",
     "  name: user_index\n",
     "  prefix: v1\n",
-    "  key_field: user\n",
-    "  storage_type: hash\n",
     "\n",
     "fields:\n",
     "    # define tag fields\n",
@@ -168,8 +165,6 @@
     "    \"index\": {\n",
     "        \"name\": \"user_index\",\n",
     "        \"prefix\": \"v1\",\n",
-    "        \"key_field\": \"user\",\n",
-    "        \"storage_type\": \"hash\",\n",
     "    },\n",
     "    \"fields\": {\n",
     "        \"tag\": [{\"name\": \"credit_score\"}],\n",

docs/user_guide/hybrid_queries_02.ipynb

@@ -109,8 +109,6 @@
     "    \"index\": {\n",
     "        \"name\": \"user_index\",\n",
     "        \"prefix\": \"v1\",\n",
-    "        \"key_field\": \"user\",\n",
-    "        \"storage_type\": \"hash\",\n",
     "    },\n",
     "    \"fields\": {\n",
     "        \"tag\": [{\"name\": \"credit_score\"}],\n",

docs/user_guide/providers_03.ipynb

@@ -124,7 +124,6 @@
     "index:\n",
     "    name: providers\n",
     "    prefix: rvl\n",
-    "    storage_type: hash\n",
     "\n",
     "fields:\n",
     "    text:\n",

redisvl/cli/index.py

@@ -124,7 +124,7 @@ def _connect_to_index(self, args: Namespace) -> SearchIndex:
             exit(0)
 
         if args.index:
-            index = SearchIndex.from_existing(conn, args.index)
+            index = SearchIndex.from_existing(name=args.index, url=url)
         elif args.schema:
             index = SearchIndex.from_yaml(args.schema)
             index.set_client(conn)