Skip to content

Commit

Permalink
Example: Add per user retriever with auth using APIHandler (#370)
Browse files Browse the repository at this point in the history
This PR adds an example that shows how to implement per user logic with sample Auth using the APIHandler primitive.

Another example for:
#354
  • Loading branch information
eyurtsev authored Jan 5, 2024
1 parent 39ec0b4 commit b45a082
Show file tree
Hide file tree
Showing 4 changed files with 511 additions and 7 deletions.
212 changes: 212 additions & 0 deletions examples/auth/api_handler/client.ipynb
Original file line number Diff line number Diff line change
@@ -0,0 +1,212 @@
{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Client\n",
"\n",
"This is an example client that interacts with the server that has \"auth\".\n",
"\n",
"Please reference appropriate documentation in the server code and in FastAPI to actually make this secure.\n",
"\n",
"\n",
"**ATTENTION** Only the invoke endpoint has been defined by the server! \n",
"So batch/stream won't work. If you want to add stream and batch, you can do so as well on the server side.\n",
"The server is implemented using the APIHandler, it's more flexible, but does require a bit more code. :)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Login as Alice"
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"import requests\n",
"\n",
"response = requests.post(\"http://localhost:8000/token\", data={\"username\": \"alice\", \"password\": \"secret1\"})\n",
"result = response.json()"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"token = result['access_token']"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"inputs = {\"input\": \"hello\"}\n",
"response = requests.post(\"http://localhost:8000/my_runnable/invoke\", \n",
" json={\n",
" 'input': 'hello',\n",
" },\n",
" headers={\n",
" 'Authorization': f\"Bearer {token}\"\n",
" }\n",
")\n"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {
"tags": []
},
"outputs": [
{
"data": {
"text/plain": [
"{'output': [{'page_content': 'cats like mice',\n",
" 'metadata': {'owner_id': 'alice'},\n",
" 'type': 'Document'},\n",
" {'page_content': 'cats like cheese',\n",
" 'metadata': {'owner_id': 'alice'},\n",
" 'type': 'Document'}],\n",
" 'callback_events': [],\n",
" 'metadata': {'run_id': '1732c9aa-c6d3-4736-b8ca-01265fa8ba06'}}"
]
},
"execution_count": 4,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"response.json()"
]
},
{
"cell_type": "markdown",
"metadata": {
"tags": []
},
"source": [
"You can also interact with this via the RemoteRunnable interface (to use in other chains)"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"from langserve import RemoteRunnable\n",
"\n",
"remote_runnable = RemoteRunnable(\"http://localhost:8000/my_runnable\", headers={\"Authorization\": f\"Bearer {token}\"})"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {
"tags": []
},
"outputs": [
{
"data": {
"text/plain": [
"[Document(page_content='cats like mice', metadata={'owner_id': 'alice'}),\n",
" Document(page_content='cats like cheese', metadata={'owner_id': 'alice'})]"
]
},
"execution_count": 6,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"await remote_runnable.ainvoke(\"cat\")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Login as John"
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {
"tags": []
},
"outputs": [],
"source": [
"import requests\n",
"\n",
"response = requests.post(\"http://localhost:8000/token\", data={\"username\": \"john\", \"password\": \"secret2\"})\n",
"token = response.json()['access_token']\n",
"remote_runnable = RemoteRunnable(\"http://localhost:8000/my_runnable\", headers={\"Authorization\": f\"Bearer {token}\"})"
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {
"tags": []
},
"outputs": [
{
"data": {
"text/plain": [
"[Document(page_content='i like walks by the ocean', metadata={'owner_id': 'john'}),\n",
" Document(page_content='dogs like sticks', metadata={'owner_id': 'john'}),\n",
" Document(page_content='my favorite food is cheese', metadata={'owner_id': 'john'})]"
]
},
"execution_count": 8,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"await remote_runnable.ainvoke(\"water\")"
]
}
],
"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": "3.9.6"
}
},
"nbformat": 4,
"nbformat_minor": 4
}
Loading

0 comments on commit b45a082

Please sign in to comment.