langchain/docs/versioned_docs/version-0.2.x/integrations/stores/astradb.ipynb

{
 "cells": [
  {
   "cell_type": "raw",
   "metadata": {},
   "source": [
    "---\n",
    "sidebar_label: Astra DB\n",
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Astra DB\n",
    "\n",
    "DataStax [Astra DB](https://docs.datastax.com/en/astra/home/astra.html) is a serverless vector-capable database built on Cassandra and made conveniently available through an easy-to-use JSON API.\n",
    "\n",
    "`AstraDBStore` and `AstraDBByteStore` need the `astrapy` package to be installed:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "plaintext"
    }
   },
   "outputs": [],
   "source": [
    "%pip install --upgrade --quiet  astrapy"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The Store takes the following parameters:\n",
    "\n",
    "* `api_endpoint`: Astra DB API endpoint. Looks like `https://01234567-89ab-cdef-0123-456789abcdef-us-east1.apps.astra.datastax.com`\n",
    "* `token`: Astra DB token. Looks like `AstraCS:6gBhNmsk135....`\n",
    "* `collection_name` : Astra DB collection name\n",
    "* `namespace`: (Optional) Astra DB namespace"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## AstraDBStore\n",
    "\n",
    "The `AstraDBStore` is an implementation of `BaseStore` that stores everything in your DataStax Astra DB instance.\n",
    "The store keys must be strings and will be mapped to the `_id` field of the Astra DB document.\n",
    "The store values can be any object that can be serialized by `json.dumps`.\n",
    "In the database, entries will have the form:\n",
    "\n",
    "```json\n",
    "{\n",
    "  \"_id\": \"<key>\",\n",
    "  \"value\": <value>\n",
    "}\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from langchain_community.storage import AstraDBStore"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from getpass import getpass\n",
    "\n",
    "ASTRA_DB_API_ENDPOINT = input(\"ASTRA_DB_API_ENDPOINT = \")\n",
    "ASTRA_DB_APPLICATION_TOKEN = getpass(\"ASTRA_DB_APPLICATION_TOKEN = \")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "store = AstraDBStore(\n",
    "    api_endpoint=ASTRA_DB_API_ENDPOINT,\n",
    "    token=ASTRA_DB_APPLICATION_TOKEN,\n",
    "    collection_name=\"my_store\",\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['v1', [0.1, 0.2, 0.3]]\n"
     ]
    }
   ],
   "source": [
    "store.mset([(\"k1\", \"v1\"), (\"k2\", [0.1, 0.2, 0.3])])\n",
    "print(store.mget([\"k1\", \"k2\"]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Usage with CacheBackedEmbeddings\n",
    "\n",
    "You may use the `AstraDBStore` in conjunction with a [`CacheBackedEmbeddings`](/docs/modules/data_connection/text_embedding/caching_embeddings) to cache the result of embeddings computations.\n",
    "Note that `AstraDBStore` stores the embeddings as a list of floats without converting them first to bytes so we don't use `fromByteStore` there."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from langchain.embeddings import CacheBackedEmbeddings, OpenAIEmbeddings\n",
    "\n",
    "embeddings = CacheBackedEmbeddings(\n",
    "    underlying_embeddings=OpenAIEmbeddings(), document_embedding_store=store\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## AstraDBByteStore\n",
    "\n",
    "The `AstraDBByteStore` is an implementation of `ByteStore` that stores everything in your DataStax Astra DB instance.\n",
    "The store keys must be strings and will be mapped to the `_id` field of the Astra DB document.\n",
    "The store `bytes` values are converted to base64 strings for storage into Astra DB.\n",
    "In the database, entries will have the form:\n",
    "\n",
    "```json\n",
    "{\n",
    "  \"_id\": \"<key>\",\n",
    "  \"value\": \"bytes encoded in base 64\"\n",
    "}\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from langchain_community.storage import AstraDBByteStore"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from getpass import getpass\n",
    "\n",
    "ASTRA_DB_API_ENDPOINT = input(\"ASTRA_DB_API_ENDPOINT = \")\n",
    "ASTRA_DB_APPLICATION_TOKEN = getpass(\"ASTRA_DB_APPLICATION_TOKEN = \")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "store = AstraDBByteStore(\n",
    "    api_endpoint=ASTRA_DB_API_ENDPOINT,\n",
    "    token=ASTRA_DB_APPLICATION_TOKEN,\n",
    "    collection_name=\"my_store\",\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[b'v1', b'v2']\n"
     ]
    }
   ],
   "source": [
    "store.mset([(\"k1\", b\"v1\"), (\"k2\", b\"v2\")])\n",
    "print(store.mget([\"k1\", \"k2\"]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": ".venv",
   "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": "3.11.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}