github/langchain
1
0
Fork 0
mirror of https://github.com/hwchase17/langchain.git synced 2025-08-07 20:15:40 +00:00
Code Issues Packages Projects Releases Wiki Activity

couchbase: Add ttl support to caches & chat_message_history (#26214)

Browse Source 
**Description:** Add support to delete documents automatically from the
caches & chat message history by adding a new optional parameter, `ttl`.


- [x] **Add tests and docs**: If you're adding a new integration, please
include
1. a test for the integration, preferably unit tests that do not rely on
network access,
2. an example notebook showing its use. It lives in
`docs/docs/integrations` directory.


- [x] **Lint and test**: Run `make format`, `make lint` and `make test`
from the root of the package(s) you've modified. See contribution
guidelines for more: https://python.langchain.com/docs/contributing/

---------

Co-authored-by: Nithish Raghunandanan <nithishr@users.noreply.github.com>
Co-authored-by: Erick Friis <erick@langchain.dev>
This commit is contained in:
Nithish Raghunandanan 2024-09-21 01:44:29 +02:00 committed by GitHub
parent c6c508ee96
commit 2d21274bf6
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
10 changed files with 901 additions and 417 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

91
docs/docs/integrations/llm_caching.ipynb
View File

@ -14,7 +14,7 @@
},
{
"cell_type": "code",
"execution_count": null,
"execution_count": 2,
"id": "88486f6f",
"metadata": {},
"outputs": [],
@ -30,7 +30,7 @@
},
{
"cell_type": "code",
"execution_count": 2,
"execution_count": 3,
"id": "10ad9224",
"metadata": {
"ExecuteTime": {
@ -2400,7 +2400,7 @@
},
{
"cell_type": "code",
"execution_count": 2,
"execution_count": 4,
"id": "9b4764e4-c75f-4185-b326-524287a826be",
"metadata": {},
"outputs": [],
@ -2430,7 +2430,7 @@
},
{
"cell_type": "code",
"execution_count": 3,
"execution_count": 5,
"id": "4b5e73c5-92c1-4eab-84e2-77924ea9c123",
"metadata": {},
"outputs": [],
@ -2452,7 +2452,7 @@
},
{
"cell_type": "code",
"execution_count": 4,
"execution_count": 6,
"id": "db8d28cc-8d93-47b4-8326-57a29a06fb3c",
"metadata": {},
"outputs": [
@ -2483,7 +2483,7 @@
},
{
"cell_type": "code",
"execution_count": 5,
"execution_count": 7,
"id": "b470dc81-2e7f-4743-9435-ce9071394eea",
"metadata": {},
"outputs": [
@ -2491,17 +2491,17 @@
"name": "stdout",
"output_type": "stream",
"text": [
"CPU times: user 53 ms, sys: 29 ms, total: 82 ms\n",
"Wall time: 84.2 ms\n"
"CPU times: user 25.9 ms, sys: 15.3 ms, total: 41.3 ms\n",
"Wall time: 144 ms\n"
]
},
{
"data": {
"text/plain": [
"\"\\n\\nWhy couldn't the bicycle stand up by itself? Because it was two-tired!\""
"\"\\n\\nWhy don't scientists trust atoms?\\n\\nBecause they make up everything.\""
]
},
"execution_count": 5,
"execution_count": 7,
"metadata": {},
"output_type": "execute_result"
}
@ -2512,6 +2512,35 @@
"llm.invoke(\"Tell me a joke\")"
]
},
{
"cell_type": "markdown",
"id": "1dca39d8-233a-45ba-ad7d-0920dfbc4a50",
"metadata": {},
"source": [
"### Specifying a Time to Live (TTL) for the Cached entries\n",
"The Cached documents can be deleted after a specified time automatically by specifying a `ttl` parameter along with the initialization of the Cache."
]
},
{
"cell_type": "code",
"execution_count": 8,
"id": "a3edcc6f-2ccd-45ba-8ca4-f65b5e51b461",
"metadata": {},
"outputs": [],
"source": [
"from datetime import timedelta\n",
"\n",
"set_llm_cache(\n",
" CouchbaseCache(\n",
" cluster=cluster,\n",
" bucket_name=BUCKET_NAME,\n",
" scope_name=SCOPE_NAME,\n",
" collection_name=COLLECTION_NAME,\n",
" ttl=timedelta(minutes=5),\n",
" )\n",
")"
]
},
{
"cell_type": "markdown",
"id": "43626f33-d184-4260-b641-c9341cef5842",
@ -2523,7 +2552,7 @@
},
{
"cell_type": "code",
"execution_count": 6,
"execution_count": 9,
"id": "6b470c03-d7fe-4270-89e1-638251619a53",
"metadata": {},
"outputs": [],
@ -2653,7 +2682,7 @@
},
{
"cell_type": "code",
"execution_count": 7,
"execution_count": 10,
"id": "ae0766c8-ea34-4604-b0dc-cf2bbe8077f4",
"metadata": {},
"outputs": [],
@ -2679,7 +2708,7 @@
},
{
"cell_type": "code",
"execution_count": 8,
"execution_count": 11,
"id": "a2e82743-10ea-4319-b43e-193475ae5449",
"metadata": {},
"outputs": [
@ -2703,7 +2732,7 @@
},
{
"cell_type": "code",
"execution_count": 9,
"execution_count": 12,
"id": "c36f4e29-d872-4334-a1f1-0e6d10c5d9f2",
"metadata": {},
"outputs": [
@ -2725,6 +2754,38 @@
"print(llm.invoke(\"What is the expected lifespan of a dog?\"))"
]
},
{
"cell_type": "markdown",
"id": "f6f674fa-70b5-4cf9-a208-992aad2c3c89",
"metadata": {},
"source": [
"### Specifying a Time to Live (TTL) for the Cached entries\n",
"The Cached documents can be deleted after a specified time automatically by specifying a `ttl` parameter along with the initialization of the Cache."
]
},
{
"cell_type": "code",
"execution_count": 13,
"id": "7e127d0a-5049-47e0-abf4-6424ad7d9fec",
"metadata": {},
"outputs": [],
"source": [
"from datetime import timedelta\n",
"\n",
"set_llm_cache(\n",
" CouchbaseSemanticCache(\n",
" cluster=cluster,\n",
" embedding=embeddings,\n",
" bucket_name=BUCKET_NAME,\n",
" scope_name=SCOPE_NAME,\n",
" collection_name=COLLECTION_NAME,\n",
" index_name=INDEX_NAME,\n",
" score_threshold=0.8,\n",
" ttl=timedelta(minutes=5),\n",
" )\n",
")"
]
},
{
"cell_type": "markdown",
"id": "ae1f5e1c-085e-4998-9f2d-b5867d2c3d5b",
@ -2802,7 +2863,7 @@
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.12.3"
"version": "3.10.13"
}
},
"nbformat": 4,

677
docs/docs/integrations/memory/couchbase_chat_message_history.ipynb
View File

@ -1,325 +1,354 @@
{
"cells": [
{
"cell_type": "markdown",
"id": "a283d2fd-e26e-4811-a486-d3cf0ecf6749",
"metadata": {},
"source": [
"# Couchbase\n",
"> Couchbase is an award-winning distributed NoSQL cloud database that delivers unmatched versatility, performance, scalability, and financial value for all of your cloud, mobile, AI, and edge computing applications. Couchbase embraces AI with coding assistance for developers and vector search for their applications.\n",
"\n",
"This notebook goes over how to use the `CouchbaseChatMessageHistory` class to store the chat message history in a Couchbase cluster\n"
]
},
{
"cell_type": "markdown",
"id": "ff868a6c-3e17-4c3d-8d32-67b01f4d7bcc",
"metadata": {},
"source": [
"## Set Up Couchbase Cluster\n",
"To run this demo, you need a Couchbase Cluster. \n",
"\n",
"You can work with both [Couchbase Capella](https://www.couchbase.com/products/capella/) and your self-managed Couchbase Server."
]
},
{
"cell_type": "markdown",
"id": "41fa85e7-6968-45e4-a445-de305d80f332",
"metadata": {},
"source": [
"## Install Dependencies\n",
"`CouchbaseChatMessageHistory` lives inside the `langchain-couchbase` package. "
]
},
{
"cell_type": "code",
"execution_count": 1,
"id": "b744ca05-b8c6-458c-91df-f50ca2c20b3c",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Note: you may need to restart the kernel to use updated packages.\n"
]
}
],
"source": [
"%pip install --upgrade --quiet langchain-couchbase"
]
},
{
"cell_type": "markdown",
"id": "41f29205-6452-493b-ba18-8a3b006bcca4",
"metadata": {},
"source": [
"## Create Couchbase Connection Object\n",
"We create a connection to the Couchbase cluster initially and then pass the cluster object to the Vector Store. \n",
"\n",
"Here, we are connecting using the username and password. You can also connect using any other supported way to your cluster. \n",
"\n",
"For more information on connecting to the Couchbase cluster, please check the [Python SDK documentation](https://docs.couchbase.com/python-sdk/current/hello-world/start-using-sdk.html#connect)."
]
},
{
"cell_type": "code",
"execution_count": 2,
"id": "f394908e-f5fe-408a-84d7-b97fdebcfa26",
"metadata": {},
"outputs": [],
"source": [
"COUCHBASE_CONNECTION_STRING = (\n",
" \"couchbase://localhost\" # or \"couchbases://localhost\" if using TLS\n",
")\n",
"DB_USERNAME = \"Administrator\"\n",
"DB_PASSWORD = \"Password\""
]
},
{
"cell_type": "code",
"execution_count": 3,
"id": "ad4dce21-d80c-465a-b709-fd366ba5ce35",
"metadata": {},
"outputs": [],
"source": [
"from datetime import timedelta\n",
"\n",
"from couchbase.auth import PasswordAuthenticator\n",
"from couchbase.cluster import Cluster\n",
"from couchbase.options import ClusterOptions\n",
"\n",
"auth = PasswordAuthenticator(DB_USERNAME, DB_PASSWORD)\n",
"options = ClusterOptions(auth)\n",
"cluster = Cluster(COUCHBASE_CONNECTION_STRING, options)\n",
"\n",
"# Wait until the cluster is ready for use.\n",
"cluster.wait_until_ready(timedelta(seconds=5))"
]
},
{
"cell_type": "markdown",
"id": "e3d0210c-e2e6-437a-86f3-7397a1899fef",
"metadata": {},
"source": [
"We will now set the bucket, scope, and collection names in the Couchbase cluster that we want to use for storing the message history.\n",
"\n",
"Note that the bucket, scope, and collection need to exist before using them to store the message history."
]
},
{
"cell_type": "code",
"execution_count": 4,
"id": "e8c7f846-a5c4-4465-a40e-4a9a23ac71bd",
"metadata": {},
"outputs": [],
"source": [
"BUCKET_NAME = \"langchain-testing\"\n",
"SCOPE_NAME = \"_default\"\n",
"COLLECTION_NAME = \"conversational_cache\""
]
},
{
"cell_type": "markdown",
"id": "283959e1-6af7-4768-9211-5b0facc6ef65",
"metadata": {},
"source": [
"## Usage\n",
"In order to store the messages, you need the following:\n",
"- Couchbase Cluster object: Valid connection to the Couchbase cluster\n",
"- bucket_name: Bucket in cluster to store the chat message history\n",
"- scope_name: Scope in bucket to store the message history\n",
"- collection_name: Collection in scope to store the message history\n",
"- session_id: Unique identifier for the session\n",
"\n",
"Optionally you can configure the following:\n",
"- session_id_key: Field in the chat message documents to store the `session_id`\n",
"- message_key: Field in the chat message documents to store the message content\n",
"- create_index: Used to specify if the index needs to be created on the collection. By default, an index is created on the `message_key` and the `session_id_key` of the documents"
]
},
{
"cell_type": "code",
"execution_count": 5,
"id": "43c3b2d5-aae2-44a9-9e9f-f10adf054cfa",
"metadata": {},
"outputs": [],
"source": [
"from langchain_couchbase.chat_message_histories import CouchbaseChatMessageHistory\n",
"\n",
"message_history = CouchbaseChatMessageHistory(\n",
" cluster=cluster,\n",
" bucket_name=BUCKET_NAME,\n",
" scope_name=SCOPE_NAME,\n",
" collection_name=COLLECTION_NAME,\n",
" session_id=\"test-session\",\n",
")\n",
"\n",
"message_history.add_user_message(\"hi!\")\n",
"\n",
"message_history.add_ai_message(\"how are you doing?\")"
]
},
{
"cell_type": "code",
"execution_count": 6,
"id": "e7e348ef-79e9-481c-aeef-969ae03dea6a",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"[HumanMessage(content='hi!'), AIMessage(content='how are you doing?')]"
]
},
"execution_count": 6,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"message_history.messages"
]
},
{
"cell_type": "markdown",
"id": "c8b942a7-93fa-4cd9-8414-d047135c2733",
"metadata": {},
"source": [
"## Chaining\n",
"The chat message history class can be used with [LCEL Runnables](https://python.langchain.com/docs/how_to/message_history/)"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "8a9f0d91-d1d6-481d-8137-ea11229f485a",
"metadata": {},
"outputs": [],
"source": [
"import getpass\n",
"import os\n",
"\n",
"from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder\n",
"from langchain_core.runnables.history import RunnableWithMessageHistory\n",
"from langchain_openai import ChatOpenAI\n",
"\n",
"os.environ[\"OPENAI_API_KEY\"] = getpass.getpass()"
]
},
{
"cell_type": "code",
"execution_count": 8,
"id": "946d45aa-5a61-49ae-816b-1c3949c56d9a",
"metadata": {},
"outputs": [],
"source": [
"prompt = ChatPromptTemplate.from_messages(\n",
" [\n",
" (\"system\", \"You are a helpful assistant.\"),\n",
" MessagesPlaceholder(variable_name=\"history\"),\n",
" (\"human\", \"{question}\"),\n",
" ]\n",
")\n",
"\n",
"# Create the LCEL runnable\n",
"chain = prompt | ChatOpenAI()"
]
},
{
"cell_type": "code",
"execution_count": 9,
"id": "20dfd838-b549-42ed-b3ba-ac005f7e024c",
"metadata": {},
"outputs": [],
"source": [
"chain_with_history = RunnableWithMessageHistory(\n",
" chain,\n",
" lambda session_id: CouchbaseChatMessageHistory(\n",
" cluster=cluster,\n",
" bucket_name=BUCKET_NAME,\n",
" scope_name=SCOPE_NAME,\n",
" collection_name=COLLECTION_NAME,\n",
" session_id=session_id,\n",
" ),\n",
" input_messages_key=\"question\",\n",
" history_messages_key=\"history\",\n",
")"
]
},
{
"cell_type": "code",
"execution_count": 10,
"id": "17bd09f4-896d-433d-bb9a-369a06e7aa8a",
"metadata": {},
"outputs": [],
"source": [
"# This is where we configure the session id\n",
"config = {\"configurable\": {\"session_id\": \"testing\"}}"
]
},
{
"cell_type": "code",
"execution_count": 11,
"id": "4bda1096-2fc2-40d7-a046-0d5d8e3a8f75",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"AIMessage(content='Hello Bob! How can I assist you today?', response_metadata={'token_usage': {'completion_tokens': 10, 'prompt_tokens': 22, 'total_tokens': 32}, 'model_name': 'gpt-4o-mini', 'system_fingerprint': None, 'finish_reason': 'stop', 'logprobs': None}, id='run-a0f8a29e-ddf4-4e06-a1fe-cf8c325a2b72-0', usage_metadata={'input_tokens': 22, 'output_tokens': 10, 'total_tokens': 32})"
]
},
"execution_count": 11,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"chain_with_history.invoke({\"question\": \"Hi! I'm bob\"}, config=config)"
]
},
{
"cell_type": "code",
"execution_count": 12,
"id": "1cfb31da-51bb-4c5f-909a-b7118b0ae08d",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"AIMessage(content='Your name is Bob.', response_metadata={'token_usage': {'completion_tokens': 5, 'prompt_tokens': 43, 'total_tokens': 48}, 'model_name': 'gpt-4o-mini', 'system_fingerprint': None, 'finish_reason': 'stop', 'logprobs': None}, id='run-f764a9eb-999e-4042-96b6-fe47b7ae4779-0', usage_metadata={'input_tokens': 43, 'output_tokens': 5, 'total_tokens': 48})"
]
},
"execution_count": 12,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"chain_with_history.invoke({\"question\": \"Whats my name\"}, config=config)"
]
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3 (ipykernel)",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.10.13"
}
},
"nbformat": 4,
"nbformat_minor": 5
}
"cells": [
{
"cell_type": "markdown",
"id": "a283d2fd-e26e-4811-a486-d3cf0ecf6749",
"metadata": {},
"source": [
"# Couchbase\n",
"> Couchbase is an award-winning distributed NoSQL cloud database that delivers unmatched versatility, performance, scalability, and financial value for all of your cloud, mobile, AI, and edge computing applications. Couchbase embraces AI with coding assistance for developers and vector search for their applications.\n",
"\n",
"This notebook goes over how to use the `CouchbaseChatMessageHistory` class to store the chat message history in a Couchbase cluster\n"
]
},
{
"cell_type": "markdown",
"id": "ff868a6c-3e17-4c3d-8d32-67b01f4d7bcc",
"metadata": {},
"source": [
"## Set Up Couchbase Cluster\n",
"To run this demo, you need a Couchbase Cluster. \n",
"\n",
"You can work with both [Couchbase Capella](https://www.couchbase.com/products/capella/) and your self-managed Couchbase Server."
]
},
{
"cell_type": "markdown",
"id": "41fa85e7-6968-45e4-a445-de305d80f332",
"metadata": {},
"source": [
"## Install Dependencies\n",
"`CouchbaseChatMessageHistory` lives inside the `langchain-couchbase` package. "
]
},
{
"cell_type": "code",
"execution_count": 1,
"id": "b744ca05-b8c6-458c-91df-f50ca2c20b3c",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Note: you may need to restart the kernel to use updated packages.\n"
]
}
],
"source": [
"%pip install --upgrade --quiet langchain-couchbase"
]
},
{
"cell_type": "markdown",
"id": "41f29205-6452-493b-ba18-8a3b006bcca4",
"metadata": {},
"source": [
"## Create Couchbase Connection Object\n",
"We create a connection to the Couchbase cluster initially and then pass the cluster object to the Vector Store. \n",
"\n",
"Here, we are connecting using the username and password. You can also connect using any other supported way to your cluster. \n",
"\n",
"For more information on connecting to the Couchbase cluster, please check the [Python SDK documentation](https://docs.couchbase.com/python-sdk/current/hello-world/start-using-sdk.html#connect)."
]
},
{
"cell_type": "code",
"execution_count": 2,
"id": "f394908e-f5fe-408a-84d7-b97fdebcfa26",
"metadata": {},
"outputs": [],
"source": [
"COUCHBASE_CONNECTION_STRING = (\n",
" \"couchbase://localhost\" # or \"couchbases://localhost\" if using TLS\n",
")\n",
"DB_USERNAME = \"Administrator\"\n",
"DB_PASSWORD = \"Password\""
]
},
{
"cell_type": "code",
"execution_count": 3,
"id": "ad4dce21-d80c-465a-b709-fd366ba5ce35",
"metadata": {},
"outputs": [],
"source": [
"from datetime import timedelta\n",
"\n",
"from couchbase.auth import PasswordAuthenticator\n",
"from couchbase.cluster import Cluster\n",
"from couchbase.options import ClusterOptions\n",
"\n",
"auth = PasswordAuthenticator(DB_USERNAME, DB_PASSWORD)\n",
"options = ClusterOptions(auth)\n",
"cluster = Cluster(COUCHBASE_CONNECTION_STRING, options)\n",
"\n",
"# Wait until the cluster is ready for use.\n",
"cluster.wait_until_ready(timedelta(seconds=5))"
]
},
{
"cell_type": "markdown",
"id": "e3d0210c-e2e6-437a-86f3-7397a1899fef",
"metadata": {},
"source": [
"We will now set the bucket, scope, and collection names in the Couchbase cluster that we want to use for storing the message history.\n",
"\n",
"Note that the bucket, scope, and collection need to exist before using them to store the message history."
]
},
{
"cell_type": "code",
"execution_count": 4,
"id": "e8c7f846-a5c4-4465-a40e-4a9a23ac71bd",
"metadata": {},
"outputs": [],
"source": [
"BUCKET_NAME = \"langchain-testing\"\n",
"SCOPE_NAME = \"_default\"\n",
"COLLECTION_NAME = \"conversational_cache\""
]
},
{
"cell_type": "markdown",
"id": "283959e1-6af7-4768-9211-5b0facc6ef65",
"metadata": {},
"source": [
"## Usage\n",
"In order to store the messages, you need the following:\n",
"- Couchbase Cluster object: Valid connection to the Couchbase cluster\n",
"- bucket_name: Bucket in cluster to store the chat message history\n",
"- scope_name: Scope in bucket to store the message history\n",
"- collection_name: Collection in scope to store the message history\n",
"- session_id: Unique identifier for the session\n",
"\n",
"Optionally you can configure the following:\n",
"- session_id_key: Field in the chat message documents to store the `session_id`\n",
"- message_key: Field in the chat message documents to store the message content\n",
"- create_index: Used to specify if the index needs to be created on the collection. By default, an index is created on the `message_key` and the `session_id_key` of the documents\n",
"- ttl: Used to specify a time in `timedelta` to live for the documents after which they will get deleted automatically from the storage."
]
},
{
"cell_type": "code",
"execution_count": 5,
"id": "43c3b2d5-aae2-44a9-9e9f-f10adf054cfa",
"metadata": {},
"outputs": [],
"source": [
"from langchain_couchbase.chat_message_histories import CouchbaseChatMessageHistory\n",
"\n",
"message_history = CouchbaseChatMessageHistory(\n",
" cluster=cluster,\n",
" bucket_name=BUCKET_NAME,\n",
" scope_name=SCOPE_NAME,\n",
" collection_name=COLLECTION_NAME,\n",
" session_id=\"test-session\",\n",
")\n",
"\n",
"message_history.add_user_message(\"hi!\")\n",
"\n",
"message_history.add_ai_message(\"how are you doing?\")"
]
},
{
"cell_type": "code",
"execution_count": 6,
"id": "e7e348ef-79e9-481c-aeef-969ae03dea6a",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"[HumanMessage(content='hi!'), AIMessage(content='how are you doing?')]"
]
},
"execution_count": 6,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"message_history.messages"
]
},
{
"cell_type": "markdown",
"id": "b993fe69-4462-4cb4-ad0a-eb31a1a4d7d9",
"metadata": {},
"source": [
"## Specifying a Time to Live (TTL) for the Chat Messages\n",
"The stored messages can be deleted after a specified time automatically by specifying a `ttl` parameter along with the initialization of the chat message history store."
]
},
{
"cell_type": "code",
"execution_count": 7,
"id": "d32d9302-de97-4319-a484-c83530bab508",
"metadata": {},
"outputs": [],
"source": [
"from langchain_couchbase.chat_message_histories import CouchbaseChatMessageHistory\n",
"\n",
"message_history = CouchbaseChatMessageHistory(\n",
" cluster=cluster,\n",
" bucket_name=BUCKET_NAME,\n",
" scope_name=SCOPE_NAME,\n",
" collection_name=COLLECTION_NAME,\n",
" session_id=\"test-session\",\n",
" ttl=timedelta(hours=24),\n",
")"
]
},
{
"cell_type": "markdown",
"id": "c8b942a7-93fa-4cd9-8414-d047135c2733",
"metadata": {},
"source": [
"## Chaining\n",
"The chat message history class can be used with [LCEL Runnables](https://python.langchain.com/docs/how_to/message_history/)"
]
},
{
"cell_type": "code",
"execution_count": 8,
"id": "8a9f0d91-d1d6-481d-8137-ea11229f485a",
"metadata": {},
"outputs": [],
"source": [
"import getpass\n",
"import os\n",
"\n",
"from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder\n",
"from langchain_core.runnables.history import RunnableWithMessageHistory\n",
"from langchain_openai import ChatOpenAI\n",
"\n",
"os.environ[\"OPENAI_API_KEY\"] = getpass.getpass()"
]
},
{
"cell_type": "code",
"execution_count": 9,
"id": "946d45aa-5a61-49ae-816b-1c3949c56d9a",
"metadata": {},
"outputs": [],
"source": [
"prompt = ChatPromptTemplate.from_messages(\n",
" [\n",
" (\"system\", \"You are a helpful assistant.\"),\n",
" MessagesPlaceholder(variable_name=\"history\"),\n",
" (\"human\", \"{question}\"),\n",
" ]\n",
")\n",
"\n",
"# Create the LCEL runnable\n",
"chain = prompt | ChatOpenAI()"
]
},
{
"cell_type": "code",
"execution_count": 10,
"id": "20dfd838-b549-42ed-b3ba-ac005f7e024c",
"metadata": {},
"outputs": [],
"source": [
"chain_with_history = RunnableWithMessageHistory(\n",
" chain,\n",
" lambda session_id: CouchbaseChatMessageHistory(\n",
" cluster=cluster,\n",
" bucket_name=BUCKET_NAME,\n",
" scope_name=SCOPE_NAME,\n",
" collection_name=COLLECTION_NAME,\n",
" session_id=session_id,\n",
" ),\n",
" input_messages_key=\"question\",\n",
" history_messages_key=\"history\",\n",
")"
]
},
{
"cell_type": "code",
"execution_count": 11,
"id": "17bd09f4-896d-433d-bb9a-369a06e7aa8a",
"metadata": {},
"outputs": [],
"source": [
"# This is where we configure the session id\n",
"config = {\"configurable\": {\"session_id\": \"testing\"}}"
]
},
{
"cell_type": "code",
"execution_count": 12,
"id": "4bda1096-2fc2-40d7-a046-0d5d8e3a8f75",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"AIMessage(content='Hello, Bob! How can I assist you today?', additional_kwargs={'refusal': None}, response_metadata={'token_usage': {'completion_tokens': 11, 'prompt_tokens': 22, 'total_tokens': 33}, 'model_name': 'gpt-3.5-turbo-0125', 'system_fingerprint': None, 'finish_reason': 'stop', 'logprobs': None}, id='run-62e54e3d-db70-429d-9ee0-e5e8eb2489a1-0', usage_metadata={'input_tokens': 22, 'output_tokens': 11, 'total_tokens': 33})"
]
},
"execution_count": 12,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"chain_with_history.invoke({\"question\": \"Hi! I'm bob\"}, config=config)"
]
},
{
"cell_type": "code",
"execution_count": 13,
"id": "1cfb31da-51bb-4c5f-909a-b7118b0ae08d",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"AIMessage(content='Your name is Bob.', additional_kwargs={'refusal': None}, response_metadata={'token_usage': {'completion_tokens': 5, 'prompt_tokens': 44, 'total_tokens': 49}, 'model_name': 'gpt-3.5-turbo-0125', 'system_fingerprint': None, 'finish_reason': 'stop', 'logprobs': None}, id='run-d84a570a-45f3-4931-814a-078761170bca-0', usage_metadata={'input_tokens': 44, 'output_tokens': 5, 'total_tokens': 49})"
]
},
"execution_count": 13,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"chain_with_history.invoke({\"question\": \"Whats my name\"}, config=config)"
]
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3 (ipykernel)",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.10.13"
}
},
"nbformat": 4,
"nbformat_minor": 5
}

42
libs/partners/couchbase/langchain_couchbase/cache.py
View File

@ -9,6 +9,7 @@ are duplicated in this utility from modules:
import hashlib
import json
import logging
from datetime import timedelta
from typing import Any, Dict, Optional, Union
from couchbase.cluster import Cluster
@ -87,6 +88,16 @@ def _loads_generations(generations_str: str) -> Union[RETURN_VAL_TYPE, None]:
return None
def _validate_ttl(ttl: Optional[timedelta]) -> None:
"""Validate the time to live"""
if not isinstance(ttl, timedelta):
raise ValueError(f"ttl should be of type timedelta but was {type(ttl)}.")
if ttl <= timedelta(seconds=0):
raise ValueError(
f"ttl must be greater than 0 but was {ttl.total_seconds()} seconds."
)
class CouchbaseCache(BaseCache):
"""Couchbase LLM Cache
LLM Cache that uses Couchbase as the backend
@ -140,6 +151,7 @@ class CouchbaseCache(BaseCache):
bucket_name: str,
scope_name: str,
collection_name: str,
ttl: Optional[timedelta] = None,
**kwargs: Dict[str, Any],
) -> None:
"""Initialize the Couchbase LLM Cache
@ -149,6 +161,8 @@ class CouchbaseCache(BaseCache):
scope_name (str): name of the scope in bucket to store documents in.
collection_name (str): name of the collection in the scope to store
documents in.
ttl (Optional[timedelta]): TTL or time for the document to live in the cache
After this time, the document will get deleted from the cache.
"""
if not isinstance(cluster, Cluster):
raise ValueError(
@ -162,6 +176,8 @@ class CouchbaseCache(BaseCache):
self._scope_name = scope_name
self._collection_name = collection_name
self._ttl = None
# Check if the bucket exists
if not self._check_bucket_exists():
raise ValueError(
@ -185,6 +201,11 @@ class CouchbaseCache(BaseCache):
except Exception as e:
raise e
# Check if the time to live is provided and valid
if ttl is not None:
_validate_ttl(ttl)
self._ttl = ttl
def lookup(self, prompt: str, llm_string: str) -> Optional[RETURN_VAL_TYPE]:
"""Look up from cache based on prompt and llm_string."""
try:
@ -206,10 +227,16 @@ class CouchbaseCache(BaseCache):
self.LLM: llm_string,
self.RETURN_VAL: _dumps_generations(return_val),
}
document_key = self._generate_key(prompt, llm_string)
try:
self._collection.upsert(
key=self._generate_key(prompt, llm_string), value=doc
)
if self._ttl:
self._collection.upsert(
key=document_key,
value=doc,
expiry=self._ttl,
)
else:
self._collection.upsert(key=document_key, value=doc)
except Exception:
logger.error("Error updating cache")
@ -242,6 +269,7 @@ class CouchbaseSemanticCache(BaseCache, CouchbaseVectorStore):
collection_name: str,
index_name: str,
score_threshold: Optional[float] = None,
ttl: Optional[timedelta] = None,
) -> None:
"""Initialize the Couchbase LLM Cache
Args:
@ -253,6 +281,8 @@ class CouchbaseSemanticCache(BaseCache, CouchbaseVectorStore):
documents in.
index_name (str): name of the Search index to use.
score_threshold (float): score threshold to use for filtering results.
ttl (Optional[timedelta]): TTL or time for the document to live in the cache
After this time, the document will get deleted from the cache.
"""
if not isinstance(cluster, Cluster):
raise ValueError(
@ -265,6 +295,7 @@ class CouchbaseSemanticCache(BaseCache, CouchbaseVectorStore):
self._bucket_name = bucket_name
self._scope_name = scope_name
self._collection_name = collection_name
self._ttl = None
# Check if the bucket exists
if not self._check_bucket_exists():
@ -291,6 +322,10 @@ class CouchbaseSemanticCache(BaseCache, CouchbaseVectorStore):
self.score_threshold = score_threshold
if ttl is not None:
_validate_ttl(ttl)
self._ttl = ttl
# Initialize the vector store
super().__init__(
cluster=cluster,
@ -334,6 +369,7 @@ class CouchbaseSemanticCache(BaseCache, CouchbaseVectorStore):
self.RETURN_VAL: _dumps_generations(return_val),
}
],
ttl=self._ttl,
)
except Exception:
logger.error("Error updating cache")

54
libs/partners/couchbase/langchain_couchbase/chat_message_histories.py
View File

@ -3,7 +3,8 @@
import logging
import time
import uuid
from typing import Any, Dict, List, Sequence
from datetime import timedelta
from typing import Any, Dict, List, Optional, Sequence
from couchbase.cluster import Cluster
from langchain_core.chat_history import BaseChatMessageHistory
@ -22,6 +23,16 @@ DEFAULT_INDEX_NAME = "LANGCHAIN_CHAT_HISTORY"
DEFAULT_BATCH_SIZE = 100
def _validate_ttl(ttl: Optional[timedelta]) -> None:
"""Validate the time to live"""
if not isinstance(ttl, timedelta):
raise ValueError(f"ttl should be of type timedelta but was {type(ttl)}.")
if ttl <= timedelta(seconds=0):
raise ValueError(
f"ttl must be greater than 0 but was {ttl.total_seconds()} seconds."
)
class CouchbaseChatMessageHistory(BaseChatMessageHistory):
"""Couchbase Chat Message History
Chat message history that uses Couchbase as the storage
@ -77,6 +88,7 @@ class CouchbaseChatMessageHistory(BaseChatMessageHistory):
session_id_key: str = DEFAULT_SESSION_ID_KEY,
message_key: str = DEFAULT_MESSAGE_KEY,
create_index: bool = True,
ttl: Optional[timedelta] = None,
) -> None:
"""Initialize the Couchbase Chat Message History
Args:
@ -92,6 +104,8 @@ class CouchbaseChatMessageHistory(BaseChatMessageHistory):
message_key (str): name of the field to use for the messages
Set to "message" by default.
create_index (bool): create an index if True. Set to True by default.
ttl (timedelta): time to live for the documents in the collection.
When set, the documents are automatically deleted after the ttl expires.
"""
if not isinstance(cluster, Cluster):
raise ValueError(
@ -104,6 +118,7 @@ class CouchbaseChatMessageHistory(BaseChatMessageHistory):
self._bucket_name = bucket_name
self._scope_name = scope_name
self._collection_name = collection_name
self._ttl = None
# Check if the bucket exists
if not self._check_bucket_exists():
@ -134,6 +149,10 @@ class CouchbaseChatMessageHistory(BaseChatMessageHistory):
self._session_id = session_id
self._ts_key = DEFAULT_TS_KEY
if ttl is not None:
_validate_ttl(ttl)
self._ttl = ttl
# Create an index if it does not exist if requested
if create_index:
index_fields = (
@ -156,15 +175,27 @@ class CouchbaseChatMessageHistory(BaseChatMessageHistory):
# get utc timestamp for ordering the messages
timestamp = time.time()
message_content = message_to_dict(message)
try:
self._collection.insert(
document_key,
value={
self._message_key: message_content,
self._session_id_key: self._session_id,
self._ts_key: timestamp,
},
)
if self._ttl:
self._collection.insert(
document_key,
value={
self._message_key: message_content,
self._session_id_key: self._session_id,
self._ts_key: timestamp,
},
expiry=self._ttl,
)
else:
self._collection.insert(
document_key,
value={
self._message_key: message_content,
self._session_id_key: self._session_id,
self._ts_key: timestamp,
},
)
except Exception as e:
logger.error("Error adding message: ", e)
@ -192,7 +223,10 @@ class CouchbaseChatMessageHistory(BaseChatMessageHistory):
batch = messages_to_insert[i : i + batch_size]
# Convert list of dictionaries to a single dictionary to insert
insert_batch = {list(d.keys())[0]: list(d.values())[0] for d in batch}
self._collection.insert_multi(insert_batch)
if self._ttl:
self._collection.insert_multi(insert_batch, expiry=self._ttl)
else:
self._collection.insert_multi(insert_batch)
except Exception as e:
logger.error("Error adding messages: ", e)

9
libs/partners/couchbase/langchain_couchbase/vectorstores.py
View File

@ -377,6 +377,9 @@ class CouchbaseVectorStore(VectorStore):
if metadatas is None:
metadatas = [{} for _ in texts]
# Check if TTL is provided
ttl = kwargs.get("ttl", None)
embedded_texts = self._embedding_function.embed_documents(list(texts))
documents_to_insert = [
@ -396,7 +399,11 @@ class CouchbaseVectorStore(VectorStore):
for i in range(0, len(documents_to_insert), batch_size):
batch = documents_to_insert[i : i + batch_size]
try:
result = self._collection.upsert_multi(batch[0])
# Insert with TTL if provided
if ttl:
result = self._collection.upsert_multi(batch[0], expiry=ttl)
else:
result = self._collection.upsert_multi(batch[0])
if result.all_ok:
doc_ids.extend(batch[0].keys())
except DocumentExistsException as e:

170
libs/partners/couchbase/poetry.lock generated
View File

@ -121,8 +121,27 @@ files = [
{file = "annotated_types-0.7.0.tar.gz", hash = "sha256:aff07c09a53a08bc8cfccb9c85b05f1aa9a2a6f23728d790723543408344ce89"},
]
[[package]]
name = "anyio"
version = "4.4.0"
description = "High level compatibility layer for multiple asynchronous event loop implementations"
optional = false
python-versions = ">=3.8"
files = [
{file = "anyio-4.4.0-py3-none-any.whl", hash = "sha256:c1b2d8f46a8a812513012e1107cb0e68c17159a7a594208005a57dc776e1bdc7"},
{file = "anyio-4.4.0.tar.gz", hash = "sha256:5aadc6a1bbb7cdb0bede386cac5e2940f5e2ff3aa20277e991cf028e0585ce94"},
]
[package.dependencies]
typing-extensions = {version = ">=4.0.0", markers = "python_version < \"3.9\""}
exceptiongroup = {version = ">=1.0.2", markers = "python_version < \"3.11\""}
idna = ">=2.8"
sniffio = ">=1.1"
typing-extensions = {version = ">=4.1", markers = "python_version < \"3.11\""}
[package.extras]
doc = ["Sphinx (>=7)", "packaging", "sphinx-autodoc-typehints (>=1.2.0)", "sphinx-rtd-theme"]
test = ["anyio[trio]", "coverage[toml] (>=7)", "exceptiongroup (>=1.2.0)", "hypothesis (>=4.0)", "psutil (>=5.9)", "pytest (>=7.0)", "pytest-mock (>=3.6.1)", "trustme", "uvloop (>=0.17)"]
trio = ["trio (>=0.23)"]
[[package]]
name = "async-timeout"
@ -497,6 +516,63 @@ files = [
docs = ["Sphinx", "furo"]
test = ["objgraph", "psutil"]
[[package]]
name = "h11"
version = "0.14.0"
description = "A pure-Python, bring-your-own-I/O implementation of HTTP/1.1"
optional = false
python-versions = ">=3.7"
files = [
{file = "h11-0.14.0-py3-none-any.whl", hash = "sha256:e3fe4ac4b851c468cc8363d500db52c2ead036020723024a109d37346efaa761"},
{file = "h11-0.14.0.tar.gz", hash = "sha256:8f19fbbe99e72420ff35c00b27a34cb9937e902a8b810e2c88300c6f0a3b699d"},
]
[[package]]
name = "httpcore"
version = "1.0.5"
description = "A minimal low-level HTTP client."
optional = false
python-versions = ">=3.8"
files = [
{file = "httpcore-1.0.5-py3-none-any.whl", hash = "sha256:421f18bac248b25d310f3cacd198d55b8e6125c107797b609ff9b7a6ba7991b5"},
{file = "httpcore-1.0.5.tar.gz", hash = "sha256:34a38e2f9291467ee3b44e89dd52615370e152954ba21721378a87b2960f7a61"},
]
[package.dependencies]
certifi = "*"
h11 = ">=0.13,<0.15"
[package.extras]
asyncio = ["anyio (>=4.0,<5.0)"]
http2 = ["h2 (>=3,<5)"]
socks = ["socksio (==1.*)"]
trio = ["trio (>=0.22.0,<0.26.0)"]
[[package]]
name = "httpx"
version = "0.27.2"
description = "The next generation HTTP client."
optional = false
python-versions = ">=3.8"
files = [
{file = "httpx-0.27.2-py3-none-any.whl", hash = "sha256:7bb2708e112d8fdd7829cd4243970f0c223274051cb35ee80c03301ee29a3df0"},
{file = "httpx-0.27.2.tar.gz", hash = "sha256:f7c2be1d2f3c3c3160d441802406b206c2b76f5947b11115e6df10c6c65e66c2"},
]
[package.dependencies]
anyio = "*"
certifi = "*"
httpcore = "==1.*"
idna = "*"
sniffio = "*"
[package.extras]
brotli = ["brotli", "brotlicffi"]
cli = ["click (==8.*)", "pygments (==2.*)", "rich (>=10,<14)"]
http2 = ["h2 (>=3,<5)"]
socks = ["socksio (==1.*)"]
zstd = ["zstandard (>=0.18.0)"]
[[package]]
name = "idna"
version = "3.7"
@ -546,26 +622,26 @@ files = [
[[package]]
name = "langchain"
version = "0.2.9"
version = "0.3.0"
description = "Building applications with LLMs through composability"
optional = false
python-versions = "<4.0,>=3.8.1"
python-versions = "<4.0,>=3.9"
files = [
{file = "langchain-0.2.9-py3-none-any.whl", hash = "sha256:be23fcb29adbd5059944f1fed08fa575f0739d420b1c4127531e0fbf5663fcca"},
{file = "langchain-0.2.9.tar.gz", hash = "sha256:cc326a7f6347787a19882928c324433b1a79df629bba45604b2d26495ee5d69c"},
{file = "langchain-0.3.0-py3-none-any.whl", hash = "sha256:59a75a6a1eb7bfd2a6bf0c7a5816409a8fdc9046187b07af287b23b9899617af"},
{file = "langchain-0.3.0.tar.gz", hash = "sha256:a7c23892440bd1f5b9e029ff0dd709dd881ae927c4c0a3210ac64dba9bbf3f7f"},
]
[package.dependencies]
aiohttp = ">=3.8.3,<4.0.0"
async-timeout = {version = ">=4.0.0,<5.0.0", markers = "python_version < \"3.11\""}
langchain-core = ">=0.2.20,<0.3.0"
langchain-text-splitters = ">=0.2.0,<0.3.0"
langchain-core = ">=0.3.0,<0.4.0"
langchain-text-splitters = ">=0.3.0,<0.4.0"
langsmith = ">=0.1.17,<0.2.0"
numpy = [
{version = ">=1,<2", markers = "python_version < \"3.12\""},
{version = ">=1.26.0,<2.0.0", markers = "python_version >= \"3.12\""},
]
pydantic = ">=1,<3"
pydantic = ">=2.7.4,<3.0.0"
PyYAML = ">=5.3"
requests = ">=2,<3"
SQLAlchemy = ">=1.4,<3"
@ -573,23 +649,24 @@ tenacity = ">=8.1.0,<8.4.0 || >8.4.0,<9.0.0"
[[package]]
name = "langchain-core"
version = "0.2.20"
version = "0.3.0"
description = "Building applications with LLMs through composability"
optional = false
python-versions = ">=3.8.1,<4.0"
python-versions = ">=3.9,<4.0"
files = []
develop = true
[package.dependencies]
jsonpatch = "^1.33"
langsmith = "^0.1.75"
langsmith = "^0.1.117"
packaging = ">=23.2,<25"
pydantic = [
{version = ">=1,<3", markers = "python_full_version < \"3.12.4\""},
{version = ">=2.5.2,<3.0.0", markers = "python_full_version < \"3.12.4\""},
{version = ">=2.7.4,<3.0.0", markers = "python_full_version >= \"3.12.4\""},
]
PyYAML = ">=5.3"
tenacity = "^8.1.0,!=8.4.0"
typing-extensions = ">=4.7"
[package.source]
type = "directory"
@ -597,30 +674,31 @@ url = "../../core"
[[package]]
name = "langchain-text-splitters"
version = "0.2.2"
version = "0.3.0"
description = "LangChain text splitting utilities"
optional = false
python-versions = "<4.0,>=3.8.1"
python-versions = "<4.0,>=3.9"
files = [
{file = "langchain_text_splitters-0.2.2-py3-none-any.whl", hash = "sha256:1c80d4b11b55e2995f02d2a326c0323ee1eeff24507329bb22924e420c782dff"},
{file = "langchain_text_splitters-0.2.2.tar.gz", hash = "sha256:a1e45de10919fa6fb080ef0525deab56557e9552083600455cb9fa4238076140"},
{file = "langchain_text_splitters-0.3.0-py3-none-any.whl", hash = "sha256:e84243e45eaff16e5b776cd9c81b6d07c55c010ebcb1965deb3d1792b7358e83"},
{file = "langchain_text_splitters-0.3.0.tar.gz", hash = "sha256:f9fe0b4d244db1d6de211e7343d4abc4aa90295aa22e1f0c89e51f33c55cd7ce"},
]
[package.dependencies]
langchain-core = ">=0.2.10,<0.3.0"
langchain-core = ">=0.3.0,<0.4.0"
[[package]]
name = "langsmith"
version = "0.1.87"
version = "0.1.120"
description = "Client library to connect to the LangSmith LLM Tracing and Evaluation Platform."
optional = false
python-versions = "<4.0,>=3.8.1"
files = [
{file = "langsmith-0.1.87-py3-none-any.whl", hash = "sha256:4cd19539c29367f812667e43c0fb0b6d304a078246508df85c38c4ea3df2d0cf"},
{file = "langsmith-0.1.87.tar.gz", hash = "sha256:d2422099708af5717d01559731c1c62d49aebf05a420015e30f6dca5ed44227c"},
{file = "langsmith-0.1.120-py3-none-any.whl", hash = "sha256:54d2785e301646c0988e0a69ebe4d976488c87b41928b358cb153b6ddd8db62b"},
{file = "langsmith-0.1.120.tar.gz", hash = "sha256:25499ca187b41bd89d784b272b97a8d76f60e0e21bdf20336e8a2aa6a9b23ac9"},
]
[package.dependencies]
httpx = ">=0.23.0,<1"
orjson = ">=3.9.14,<4.0.0"
pydantic = [
{version = ">=1,<3", markers = "python_full_version < \"3.12.4\""},
@ -785,43 +863,6 @@ files = [
{file = "mypy_extensions-1.0.0.tar.gz", hash = "sha256:75dbf8955dc00442a438fc4d0666508a9a97b6bd41aa2f0ffe9d2f2725af0782"},
]
[[package]]
name = "numpy"
version = "1.24.4"
description = "Fundamental package for array computing in Python"
optional = false
python-versions = ">=3.8"
files = [
{file = "numpy-1.24.4-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:c0bfb52d2169d58c1cdb8cc1f16989101639b34c7d3ce60ed70b19c63eba0b64"},
{file = "numpy-1.24.4-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:ed094d4f0c177b1b8e7aa9cba7d6ceed51c0e569a5318ac0ca9a090680a6a1b1"},
{file = "numpy-1.24.4-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:79fc682a374c4a8ed08b331bef9c5f582585d1048fa6d80bc6c35bc384eee9b4"},
{file = "numpy-1.24.4-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:7ffe43c74893dbf38c2b0a1f5428760a1a9c98285553c89e12d70a96a7f3a4d6"},
{file = "numpy-1.24.4-cp310-cp310-win32.whl", hash = "sha256:4c21decb6ea94057331e111a5bed9a79d335658c27ce2adb580fb4d54f2ad9bc"},
{file = "numpy-1.24.4-cp310-cp310-win_amd64.whl", hash = "sha256:b4bea75e47d9586d31e892a7401f76e909712a0fd510f58f5337bea9572c571e"},
{file = "numpy-1.24.4-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:f136bab9c2cfd8da131132c2cf6cc27331dd6fae65f95f69dcd4ae3c3639c810"},
{file = "numpy-1.24.4-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:e2926dac25b313635e4d6cf4dc4e51c8c0ebfed60b801c799ffc4c32bf3d1254"},
{file = "numpy-1.24.4-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:222e40d0e2548690405b0b3c7b21d1169117391c2e82c378467ef9ab4c8f0da7"},
{file = "numpy-1.24.4-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:7215847ce88a85ce39baf9e89070cb860c98fdddacbaa6c0da3ffb31b3350bd5"},
{file = "numpy-1.24.4-cp311-cp311-win32.whl", hash = "sha256:4979217d7de511a8d57f4b4b5b2b965f707768440c17cb70fbf254c4b225238d"},
{file = "numpy-1.24.4-cp311-cp311-win_amd64.whl", hash = "sha256:b7b1fc9864d7d39e28f41d089bfd6353cb5f27ecd9905348c24187a768c79694"},
{file = "numpy-1.24.4-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:1452241c290f3e2a312c137a9999cdbf63f78864d63c79039bda65ee86943f61"},
{file = "numpy-1.24.4-cp38-cp38-macosx_11_0_arm64.whl", hash = "sha256:04640dab83f7c6c85abf9cd729c5b65f1ebd0ccf9de90b270cd61935eef0197f"},
{file = "numpy-1.24.4-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:a5425b114831d1e77e4b5d812b69d11d962e104095a5b9c3b641a218abcc050e"},
{file = "numpy-1.24.4-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:dd80e219fd4c71fc3699fc1dadac5dcf4fd882bfc6f7ec53d30fa197b8ee22dc"},
{file = "numpy-1.24.4-cp38-cp38-win32.whl", hash = "sha256:4602244f345453db537be5314d3983dbf5834a9701b7723ec28923e2889e0bb2"},
{file = "numpy-1.24.4-cp38-cp38-win_amd64.whl", hash = "sha256:692f2e0f55794943c5bfff12b3f56f99af76f902fc47487bdfe97856de51a706"},
{file = "numpy-1.24.4-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:2541312fbf09977f3b3ad449c4e5f4bb55d0dbf79226d7724211acc905049400"},
{file = "numpy-1.24.4-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:9667575fb6d13c95f1b36aca12c5ee3356bf001b714fc354eb5465ce1609e62f"},
{file = "numpy-1.24.4-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:f3a86ed21e4f87050382c7bc96571755193c4c1392490744ac73d660e8f564a9"},
{file = "numpy-1.24.4-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:d11efb4dbecbdf22508d55e48d9c8384db795e1b7b51ea735289ff96613ff74d"},
{file = "numpy-1.24.4-cp39-cp39-win32.whl", hash = "sha256:6620c0acd41dbcb368610bb2f4d83145674040025e5536954782467100aa8835"},
{file = "numpy-1.24.4-cp39-cp39-win_amd64.whl", hash = "sha256:befe2bf740fd8373cf56149a5c23a0f601e82869598d41f8e188a0e9869926f8"},
{file = "numpy-1.24.4-pp38-pypy38_pp73-macosx_10_9_x86_64.whl", hash = "sha256:31f13e25b4e304632a4619d0e0777662c2ffea99fcae2029556b17d8ff958aef"},
{file = "numpy-1.24.4-pp38-pypy38_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:95f7ac6540e95bc440ad77f56e520da5bf877f87dca58bd095288dce8940532a"},
{file = "numpy-1.24.4-pp38-pypy38_pp73-win_amd64.whl", hash = "sha256:e98f220aa76ca2a977fe435f5b04d7b3470c0a2e6312907b37ba6068f26787f2"},
{file = "numpy-1.24.4.tar.gz", hash = "sha256:80f5e3a4e498641401868df4208b74581206afbee7cf7b8329daae82676d9463"},
]
[[package]]
name = "numpy"
version = "1.26.4"
@ -1237,6 +1278,17 @@ files = [
{file = "ruff-0.5.2.tar.gz", hash = "sha256:2c0df2d2de685433794a14d8d2e240df619b748fbe3367346baa519d8e6f1ca2"},
]
[[package]]
name = "sniffio"
version = "1.3.1"
description = "Sniff out which async library your code is running under"
optional = false
python-versions = ">=3.7"
files = [
{file = "sniffio-1.3.1-py3-none-any.whl", hash = "sha256:2f6da418d1f1e0fddd844478f41680e794e6051915791a034ff65e5f100525a2"},
{file = "sniffio-1.3.1.tar.gz", hash = "sha256:f4324edc670a0f49750a81b895f35c3adb843cca46f0530f79fc1babb23789dc"},
]
[[package]]
name = "sqlalchemy"
version = "2.0.31"
@ -1497,5 +1549,5 @@ multidict = ">=4.0"
[metadata]
lock-version = "2.0"
python-versions = ">=3.8.1,<4.0"
content-hash = "0a2c4601a26691fd3bf061961c95a19a3869c491d4670dc38b8a4c46d2c01a6e"
python-versions = ">=3.9,<4.0"
content-hash = "0834872527b82aef5921c6b869a70c49ce36507d858e1371c05a079a77cfad86"

2
libs/partners/couchbase/pyproject.toml
View File

@ -4,7 +4,7 @@ build-backend = "poetry.core.masonry.api"
[tool.poetry]
name = "langchain-couchbase"
version = "0.1.1"
version = "0.2.0"
description = "An integration package connecting Couchbase and LangChain"
authors = []
readme = "README.md"

102
libs/partners/couchbase/tests/integration_tests/test_cache.py
View File

@ -1,7 +1,7 @@
"""Test Couchbase Cache functionality"""
import os
from datetime import timedelta
from datetime import datetime, timedelta
from typing import Any
import pytest
@ -12,7 +12,13 @@ from langchain_core.globals import get_llm_cache, set_llm_cache
from langchain_core.outputs import Generation
from langchain_couchbase.cache import CouchbaseCache, CouchbaseSemanticCache
from tests.utils import FakeEmbeddings, FakeLLM
from tests.utils import (
FakeEmbeddings,
FakeLLM,
cache_key_hash_function,
fetch_document_expiry_time,
get_document_keys,
)
CONNECTION_STRING = os.getenv("COUCHBASE_CONNECTION_STRING", "")
BUCKET_NAME = os.getenv("COUCHBASE_BUCKET_NAME", "")
@ -88,6 +94,39 @@ class TestCouchbaseCache:
output = get_llm_cache().lookup("bar", llm_string)
assert output != [Generation(text="fizz")]
def test_cache_with_ttl(self, cluster: Any) -> None:
"""Test standard LLM cache functionality with TTL"""
ttl = timedelta(minutes=10)
set_llm_cache(
CouchbaseCache(
cluster=cluster,
bucket_name=BUCKET_NAME,
scope_name=SCOPE_NAME,
collection_name=CACHE_COLLECTION_NAME,
ttl=ttl,
)
)
llm = FakeLLM()
params = llm.dict()
params["stop"] = None
llm_string = str(sorted([(k, v) for k, v in params.items()]))
get_llm_cache().update("foo", llm_string, [Generation(text="fizz")])
cache_output = get_llm_cache().lookup("foo", llm_string)
assert cache_output == [Generation(text="fizz")]
# Check the document's expiry time by fetching it from the database
document_key = cache_key_hash_function("foo" + llm_string)
document_expiry_time = fetch_document_expiry_time(
cluster, BUCKET_NAME, SCOPE_NAME, CACHE_COLLECTION_NAME, document_key
)
current_time = datetime.now()
time_to_expiry = document_expiry_time - current_time
assert time_to_expiry < ttl
def test_semantic_cache(self, cluster: Any) -> None:
"""Test semantic LLM cache functionality"""
set_llm_cache(
@ -118,3 +157,62 @@ class TestCouchbaseCache:
get_llm_cache().clear()
output = get_llm_cache().lookup("bar", llm_string)
assert output != [Generation(text="fizz"), Generation(text="Buzz")]
def test_semantic_cache_with_ttl(self, cluster: Any) -> None:
"""Test semantic LLM cache functionality with TTL"""
ttl = timedelta(minutes=10)
set_llm_cache(
CouchbaseSemanticCache(
cluster=cluster,
embedding=FakeEmbeddings(),
index_name=INDEX_NAME,
bucket_name=BUCKET_NAME,
scope_name=SCOPE_NAME,
collection_name=SEMANTIC_CACHE_COLLECTION_NAME,
ttl=ttl,
)
)
llm = FakeLLM()
params = llm.dict()
params["stop"] = None
llm_string = str(sorted([(k, v) for k, v in params.items()]))
# Add a document to the cache
seed_prompt = "foo"
get_llm_cache().update(
seed_prompt, llm_string, [Generation(text="fizz"), Generation(text="Buzz")]
)
# foo and bar will have the same embedding produced by FakeEmbeddings
cache_output = get_llm_cache().lookup("bar", llm_string)
assert cache_output == [Generation(text="fizz"), Generation(text="Buzz")]
# Check the document's expiry time by fetching it from the database
fetch_document_query = (
f"SELECT meta().id, * from `{SEMANTIC_CACHE_COLLECTION_NAME}` doc "
f"WHERE doc.text = '{seed_prompt}'"
)
document_keys = get_document_keys(
cluster=cluster,
bucket_name=BUCKET_NAME,
scope_name=SCOPE_NAME,
query=fetch_document_query,
)
assert len(document_keys) == 1
document_expiry_time = fetch_document_expiry_time(
cluster=cluster,
bucket_name=BUCKET_NAME,
scope_name=SCOPE_NAME,
collection_name=SEMANTIC_CACHE_COLLECTION_NAME,
document_key=document_keys[0],
)
current_time = datetime.now()
time_to_expiry = document_expiry_time - current_time
assert time_to_expiry < ttl

132
libs/partners/couchbase/tests/integration_tests/test_chat_message_history.py
View File

@ -2,7 +2,7 @@
import os
import time
from datetime import timedelta
from datetime import datetime, timedelta
from typing import Any
import pytest
@ -13,6 +13,7 @@ from langchain.memory import ConversationBufferMemory
from langchain_core.messages import AIMessage, HumanMessage
from langchain_couchbase.chat_message_histories import CouchbaseChatMessageHistory
from tests.utils import fetch_document_expiry_time, get_document_keys
CONNECTION_STRING = os.getenv("COUCHBASE_CONNECTION_STRING", "")
BUCKET_NAME = os.getenv("COUCHBASE_BUCKET_NAME", "")
@ -162,3 +163,132 @@ class TestCouchbaseCache:
memory_b.chat_memory.clear()
time.sleep(SLEEP_DURATION)
assert memory_b.chat_memory.messages == []
def test_memory_message_with_ttl(self, cluster: Any) -> None:
"""Test chat message history with a message being saved with a TTL"""
ttl = timedelta(minutes=5)
session_id = "test-session-ttl"
message_history = CouchbaseChatMessageHistory(
cluster=cluster,
bucket_name=BUCKET_NAME,
scope_name=SCOPE_NAME,
collection_name=MESSAGE_HISTORY_COLLECTION_NAME,
session_id=session_id,
ttl=ttl,
)
memory = ConversationBufferMemory(
memory_key="baz", chat_memory=message_history, return_messages=True
)
# clear the memory
memory.chat_memory.clear()
# wait for the messages to be cleared
time.sleep(SLEEP_DURATION)
assert memory.chat_memory.messages == []
# add some messages
ai_message = AIMessage(content="Hello, how are you doing ?")
memory.chat_memory.add_ai_message(ai_message)
# wait until the messages can be retrieved
time.sleep(SLEEP_DURATION)
# check that the messages are in the memory
messages = memory.chat_memory.messages
assert len(messages) == 1
# check that the messages are in the order of creation
assert messages == [ai_message]
# Check the document's expiry time by fetching it from the database
fetch_documents_query = (
f"SELECT meta().id, * from `{MESSAGE_HISTORY_COLLECTION_NAME}` doc"
f" WHERE doc.session_id = '{session_id}'"
)
document_keys = get_document_keys(
cluster=cluster,
bucket_name=BUCKET_NAME,
scope_name=SCOPE_NAME,
query=fetch_documents_query,
)
assert len(document_keys) == 1
# Ensure that the document will expire within the TTL
document_expiry_time = fetch_document_expiry_time(
cluster=cluster,
bucket_name=BUCKET_NAME,
scope_name=SCOPE_NAME,
collection_name=MESSAGE_HISTORY_COLLECTION_NAME,
document_key=document_keys[0],
)
current_time = datetime.now()
assert document_expiry_time - current_time < ttl
def test_memory_messages_with_ttl(self, cluster: Any) -> None:
"""Test chat message history with messages being stored with a TTL"""
ttl = timedelta(minutes=5)
session_id = "test-session-ttl"
message_history = CouchbaseChatMessageHistory(
cluster=cluster,
bucket_name=BUCKET_NAME,
scope_name=SCOPE_NAME,
collection_name=MESSAGE_HISTORY_COLLECTION_NAME,
session_id=session_id,
ttl=ttl,
)
memory = ConversationBufferMemory(
memory_key="baz", chat_memory=message_history, return_messages=True
)
# clear the memory
memory.chat_memory.clear()
# wait for the messages to be cleared
time.sleep(SLEEP_DURATION)
assert memory.chat_memory.messages == []
# add some messages
ai_message = AIMessage(content="Hello, how are you doing ?")
user_message = HumanMessage(content="I'm good, how are you?")
memory.chat_memory.add_messages([ai_message, user_message])
# wait until the messages can be retrieved
time.sleep(SLEEP_DURATION)
# check that the messages are in the memory
messages = memory.chat_memory.messages
assert len(messages) == 2
# check that the messages are in the order of creation
assert messages == [ai_message, user_message]
# Check the documents' expiry time by fetching the documents from the database
fetch_documents_query = (
f"SELECT meta().id, * from `{MESSAGE_HISTORY_COLLECTION_NAME}` doc"
f" WHERE doc.session_id = '{session_id}'"
)
document_keys = get_document_keys(
cluster=cluster,
bucket_name=BUCKET_NAME,
scope_name=SCOPE_NAME,
query=fetch_documents_query,
)
assert len(document_keys) == 2
# Ensure that each document will expire within the TTL
for document_key in document_keys:
document_expiry_time = fetch_document_expiry_time(
cluster=cluster,
bucket_name=BUCKET_NAME,
scope_name=SCOPE_NAME,
collection_name=MESSAGE_HISTORY_COLLECTION_NAME,
document_key=document_key,
)
current_time = datetime.now()
assert document_expiry_time - current_time < ttl

39
libs/partners/couchbase/tests/utils.py
View File

@ -1,7 +1,11 @@
"""Fake Embedding class for testing purposes."""
"""Utilities for testing purposes."""
import hashlib
from datetime import datetime
from typing import Any, Dict, List, Mapping, Optional, cast
from couchbase.cluster import Cluster
from couchbase.options import GetOptions
from langchain_core.callbacks import CallbackManagerForLLMRun
from langchain_core.embeddings import Embeddings
from langchain_core.language_models.llms import LLM
@ -97,3 +101,36 @@ class FakeLLM(LLM):
response = queries[list(queries.keys())[self.response_index]]
self.response_index = self.response_index + 1
return response
def cache_key_hash_function(_input: str) -> str:
"""Use a deterministic hashing approach."""
return hashlib.md5(_input.encode()).hexdigest()
def fetch_document_expiry_time(
cluster: Cluster,
bucket_name: str,
scope_name: str,
collection_name: str,
document_key: str,
) -> datetime:
"""Fetch the document's expiry time from the database."""
collection = (
cluster.bucket(bucket_name).scope(scope_name).collection(collection_name)
)
result = collection.get(document_key, GetOptions(with_expiry=True))
return result.expiryTime
def get_document_keys(
cluster: Cluster, bucket_name: str, scope_name: str, query: str
) -> List[str]:
"""Get the document key from the database based on the query using meta().id."""
scope = cluster.bucket(bucket_name).scope(scope_name)
result = scope.query(query).execute()
document_keys = [row["id"] for row in result]
return document_keys