Feature/project crud (sprint 1) (#1)

* modyfiy dependency for langchain, add conditions to waith for mongo, add mongo-expers web client

* Move toy_files to examples, add jupyter notebook add env templates

* Project2 Internal Server Error fixed in template.
The error was caused by a missing message field in the request and response in the object stored in the database. It is possible that this depends on the AI model used.

* Transaction view

* Redirection from project to transaction view.
More specific data in other in transaction view

* Start of CRUD projects

* [PS-5] merge front

* add default pass to mongoexpress

* small changes, clear output

* installed python-multipart-0.0.6 to make forms run.
added simple form for project adding

* deleting projects

* api's things separated to other file (dunno if we'll need that)
crud for projects via ui done

* python-multipart added to poetry

* update poetry.lock

* testy api

* temp fix for timeout

* add openai lib example

* add langchain openai

* fix poetry version in docker

* add group for exmaples dependencies

* add langchain openai

* add base url

* edited project crud

* hotfix precreated projects

* 'TransactionContext' object has no attribute 'app' fix

* api get project hotfix

* trying to make transaction-tags

* tests, and api project fix

* new use cases for saving transaction , saving "as is"

* unpacking request and response error fix,
updated endpoints and what they returns,
model_name changed to ai_model_name in ProjectAIProviderSchema,
updated api tests

* webui endpoint hotfix

* pre-merge fixes

* pre-merge fixes after 06.12.23 call

* files not pushed in commit before

* last commit after call 08.12.2023

---------

Co-authored-by: ksopyla <[email protected]>
Co-authored-by: Przemysław Górecki <[email protected]>
Co-authored-by: unknown <[email protected]>

Dockerfile

@@ -7,11 +7,12 @@ WORKDIR /src
 
 # Install curl and Poetry in a single step, and clean up the apt cache to keep the image small
 RUN apt-get update && apt-get install -y --no-install-recommends curl \
-    && curl -sSL https://install.python-poetry.org | python3 - \
+    && curl -sSL https://install.python-poetry.org | python3 - --version 1.7.1 \
     && ln -s /root/.local/bin/poetry /usr/local/bin/poetry \
     && apt-get purge -y --auto-remove curl \
     && rm -rf /var/lib/apt/lists/*
 
+
 # Copy the Python dependency files into the image
 COPY pyproject.toml poetry.lock ./
 

README.md

@@ -101,7 +101,7 @@ file with the following lines:
 2. Build and run the docker image:
 
 ```bash
-docker-docker-compose up --build
+docker-compose up --build
 ```
 
 

docker-compose.yml

@@ -10,10 +10,24 @@ services:
       MONGO_INITDB_ROOT_USERNAME: root
       MONGO_INITDB_ROOT_PASSWORD: password
     healthcheck:
-      test: ["CMD", "curl", "http://localhost:21017"]
-      interval: 15s
+      test: echo 'db.runCommand("ping").ok' | mongosh localhost:27017/test --quiet
+      interval: 10s
       timeout: 10s
       retries: 5
+      start_period: 40s
+  mongoexperss:
+    image: mongo-express:latest
+    ports:
+      - "8081:8081"
+    environment: 
+      ME_CONFIG_MONGODB_URL: "mongodb://root:password@mongodb:27017"
+      ME_CONFIG_OPTIONS_EDITORTHEME: "ambiance"
+      ME_CONFIG_BASICAUTH_USERNAME: "admin" 
+      ME_CONFIG_BASICAUTH_PASSWORD: "pass"
+    depends_on:
+      mongodb:
+        condition: service_healthy
+
   app:
     build:
       context: .
@@ -26,8 +40,9 @@ services:
       BASE_URL: "http://promptsail.local"
       STATIC_DIRECTORY: "/static"
       MONGO_URL: "mongodb://root:password@mongodb:27017"
-    depends_on: 
-      - mongodb
+    depends_on:
+      mongodb:
+        condition: service_healthy
 
 networks:
   internal-network:

examples/.env.template

@@ -0,0 +1,2 @@
+OPENAI_API_KEY=sk-xxx
+OPENAI_ORG_ID=org-xxx

examples/langchain_openai.ipynb

@@ -0,0 +1,132 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# Prompt sail proxy for OpenAI with LangChain"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "OpenAI api loaded=sk-...e9Uzy\n"
+     ]
+    }
+   ],
+   "source": [
+    "\n",
+    "import os\n",
+    "from dotenv import load_dotenv\n",
+    "load_dotenv()\n",
+    "\n",
+    "\n",
+    "openai_key = os.getenv(\"OPENAI_API_KEY\")\n",
+    "openai_org_id = os.getenv(\"OPENAI_ORG_ID\")\n",
+    "print(\n",
+    "    f\"OpenAI api loaded={os.getenv('OPENAI_API_KEY')[0:3]}...{os.getenv('OPENAI_API_KEY')[-5:]}\"\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "At the application start we created two test projects with OpenAI API mappings. \n",
+    "There are visible on main dashboard http://promptsail.local/\n",
+    "\n",
+    "\n",
+    "* [project1](http://promptsail.local/ui/project/project1) -> https://api.openai.com/v1\n",
+    "* [project2](http://promptsail.local/ui/project/project2) -> https://api.openai.com/v1\n",
+    "\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from langchain.chat_models import ChatOpenAI\n",
+    "from langchain.prompts.chat import (\n",
+    "    ChatPromptTemplate,\n",
+    "    HumanMessagePromptTemplate,\n",
+    "    SystemMessagePromptTemplate,\n",
+    ")\n",
+    "from langchain.schema import HumanMessage, SystemMessage"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "api_base = \"http://project1.promptsail.local\"\n",
+    "chat = ChatOpenAI(\n",
+    "    temperature=0,\n",
+    "    openai_api_key=openai_key,\n",
+    "    openai_organization=openai_org_id,\n",
+    "    model=\"gpt-3.5-turbo\",\n",
+    "    base_url=api_base,\n",
+    ")\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "AIMessage(content='Bug: Out of Memory Error when Allocating CUDA Memory (GPU X)')"
+      ]
+     },
+     "execution_count": 12,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "\n",
+    "messages = [\n",
+    "    SystemMessage(\n",
+    "        content=\"You are a helpful assistant that help rewirte an jira ticket.\"\n",
+    "    ),\n",
+    "    HumanMessage(\n",
+    "        content=\"Give meaningful title to this bug, RuntimeError: CUDA out of memory. Tried to allocate X MiB (GPU X; X GiB total capacity; X GiB already allocated; X MiB free; X cached)\"\n",
+    "    ),\n",
+    "]\n",
+    "chat(messages)"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "prompt-sail-hDSOLtZB-py3.10",
+   "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": "3.10.6"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}

src/toy_langchain.py → examples/toy_langchain.py

@@ -11,7 +11,7 @@
 
 llm = OpenAI(
     model_name="text-davinci-003",
-    openai_api_base="http://project2.promptsail.local:8000",
+    openai_api_base="http://project2.promptsail.local",
 )
 output = llm("Explaining the meaning of life in one sentence")
 print(output)