fixes conflicts on poetry.lock

.github/workflows/publish.yml

@@ -3,6 +3,7 @@ name: Publish packages (PyPI, npm)
 on:
   release:
     types: [published]
+  workflow_dispatch:
 
 jobs:
   publish-pypi:
@@ -71,6 +72,6 @@ jobs:
       - name: Publish package distributions to npm
         working-directory: ./frontend
         run: |
-          pnpm publish
+          pnpm publish --no-git-checks
         env:
           NODE_AUTH_TOKEN: ${{ secrets.NPM_API_TOKEN }}

.python-version

@@ -1 +1 @@
-3.12.3
+3.12.4

CHANGELOG.md

@@ -8,6 +8,14 @@ frontend TypeScript client, `django-ai-assistant-client`.
     The backend and the frontend are versioned together, that is, they have the same version number.
     When you update the backend, you should also update the frontend to the same version.
 
+## 0.0.4 <small>July 5, 2024</small> {id="0.0.4"}
+
+- Fix frontend README.
+
+## 0.0.3 <small>July 5, 2024</small> {id="0.0.3"}
+
+- Less restrictive Python version in pyproject.toml. Support future Python versions.
+
 ## 0.0.2 <small>June 28, 2024</small> {id="0.0.2"}
 
 - Add support for Django 4.2 LTS

CONTRIBUTING.md

@@ -12,22 +12,37 @@ Please follow this guide to learn more about how to develop and test the project
 git clone [email protected]:vintasoftware/django-ai-assistant.git
 ```
 
-### Set up a virtualenv, optionally set up nvm, and activate your environment(s)
+### Install development tools
 
-You can use [pyenv](https://github.com/pyenv/pyenv), [pipenv](https://github.com/pypa/pipenv/blob/main/docs/installation.md), vanilla venvs or the tool of your choice.
+This project uses [Poetry](https://python-poetry.org/docs/) for dependency and virtual environment management.
+
+If you need to install the version of Python recommended for the project, you can use [Pyenv](https://github.com/pyenv/pyenv).
 
 For installing Node, we recommend [NVM](https://github.com/nvm-sh/nvm).
 
 ### Install dependencies
 
 #### Backend
 
-Go to the project root and install the Python dependencies:
+Go to the project root. To instantiate the virtual environment, run
+
+```bash
+poetry shell
+```
+
+Install the Python dependencies:
 
 ```bash
 poetry install
 ```
 
+If you encounter an error regarding the Python version required for the project, you can use pyenv to install the appropriate version based on [.python-version](.python-version):
+
+```bash
+pyenv install
+```
+
+
 #### Frontend
 
 Go to the frontend directory and install the Node dependencies:
@@ -68,6 +83,12 @@ Then follow the instructions in the [example README](https://github.com/vintasof
 
 ## Tests
 
+Before running tests copy the `.env.example` file to `.env.tests`.
+
+```bash
+cp .env.example .env.tests
+```
+
 Run tests with:
 
 ```bash
@@ -76,7 +97,8 @@ poetry run pytest
 
 The tests use `pytest-vcr` to record and replay HTTP requests to AI models.
 
-If you're implementing a new test that needs to call a real AI model, you need to set the `OPENAI_API_KEY` environment variable on root `.env` file.
+If you're implementing a new test that needs to call a real AI model, you need to set the `OPENAI_API_KEY` environment variable with a real API key in the `.env.tests` file.
+
 Then, you will run the tests in record mode:
 
 ```bash
@@ -102,10 +124,19 @@ poetry run mkdocs serve
 To release and publish a new version, follow these steps:
 
 1. Update the version in `pyproject.toml` and `frontend/package.json`.
-2. In the project root, run `poetry run python manage.py generate_openapi_schema --output frontend/openapi_schema.json` to update the OpenAPI schema.
-3. In the frontend directory, run `pnpm run generate-client` to update the TypeScript client with the new OpenAPI schema.
-4. Update the changelog in `CHANGELOG.md`.
-5. Open a PR with the changes.
-6. Once the PR is merged, run the [Release GitHub Action](https://github.com/vintasoftware/django-ai-assistant/actions/workflows/release.yml) to create a draft release.
-7. Review the draft release, ensure the description has at least the associated changelog entry, and publish it.
-8. Once the review is publish, the [Publish GitHub Action](https://github.com/vintasoftware/django-ai-assistant/actions/workflows/publish.yml) will automatically run to publish the new version to [PyPI](https://pypi.org/project/django-ai-assistant) and [npm](https://www.npmjs.com/package/django-ai-assistant-client). Check the logs to ensure the publication was successful.
+2. Re-install the local version of the Python project: `poetry install`
+3. In the project root, run `poetry run python manage.py generate_openapi_schema --output frontend/openapi_schema.json` to update the OpenAPI schema.
+4. Re-install the local version of the frontend project:
+
+```bash
+cd frontend
+pnpm install
+pnpm run build
+```
+
+5. In the frontend directory, run `pnpm run generate-client` to update the TypeScript client with the new OpenAPI schema.
+6. Update the changelog in `CHANGELOG.md`.
+7. Open a PR with the changes.
+8. Once the PR is merged, run the [Release GitHub Action](https://github.com/vintasoftware/django-ai-assistant/actions/workflows/release.yml) to create a draft release.
+9. Review the draft release, ensure the description has at least the associated changelog entry, and publish it.
+10. Once the review is published, the [Publish GitHub Action](https://github.com/vintasoftware/django-ai-assistant/actions/workflows/publish.yml) will automatically run to publish the new version to [PyPI](https://pypi.org/project/django-ai-assistant) and [npm](https://www.npmjs.com/package/django-ai-assistant-client). Check the logs to ensure the publication was successful.

README.md

@@ -1,17 +1,26 @@
 [![CI](https://github.com/vintasoftware/django-ai-assistant/actions/workflows/ci.yml/badge.svg)](https://github.com/vintasoftware/django-ai-assistant/actions/workflows/ci.yml)
 [![Coverage Status](https://coveralls.io/repos/github/vintasoftware/django-ai-assistant/badge.svg?branch=main)](https://coveralls.io/github/vintasoftware/django-ai-assistant?branch=main)
+[![Discord Server](https://img.shields.io/discord/1260577482122203206)](https://discord.gg/mqdubnPb)
 
-# django-ai-assistant <img src="https://raw.githubusercontent.com/vintasoftware/django-ai-assistant/main/docs/images/robot-happy-outline.svg" width="20" alt="robot logo">
+# django-ai-assistant
+
+<img align="left" src="https://raw.githubusercontent.com/vintasoftware/django-ai-assistant/main/docs/images/favicon.svg" height="120" alt="robot logo">
 
 Combine the power of LLMs with Django's productivity to build intelligent applications.
 Let AI Assistants call methods from Django's side and do anything your users need!
 
 Use AI Tool Calling and RAG with Django to easily build state of the art AI Assistants.
 
-⚠️ Under heavy development. Not publicly launched yet!
-
 Please check the documentation: [https://vintasoftware.github.io/django-ai-assistant/](https://vintasoftware.github.io/django-ai-assistant/)
 
+## 5 minute demo
+
+https://github.com/vintasoftware/django-ai-assistant/assets/397989/715d7608-e657-4dce-a7be-2909679f814b
+
+## Community
+
+Feel free to ask questions and share your work on our [Discord Server](https://discord.gg/fQfH9PkJM6)!
+
 ## Contributing
 
 You're welcome to contribute with Django AI Assistant! Please feel free to tackle existing [issues](https://github.com/vintasoftware/django-ai-assistant/issues). If you have a new idea, please create a thread on [Discussions](https://github.com/vintasoftware/django-ai-assistant/discussions). 

django_ai_assistant/admin.py

@@ -1,19 +1,58 @@
+from typing import ClassVar, List, Type
+
 from django.contrib import admin
+from django.contrib.admin.options import InlineModelAdmin
+from django.urls import reverse
+from django.utils.safestring import mark_safe
 
 from django_ai_assistant.models import Message, Thread
 
 
+class MessageInline(admin.TabularInline):
+    model = Message
+    extra = 0
+    fields = ("pk", "message_type", "content", "created_at")
+    readonly_fields = fields
+    ordering = ("created_at",)
+    show_change_link = True
+
+    def pk(self, obj):
+        display_text = "<a href={}>{}</a>".format(
+            reverse(
+                f"admin:{Message._meta.app_label}_{Message._meta.model_name}_change", args=(obj.pk,)
+            ),
+            obj.pk,
+        )
+        return mark_safe(display_text)  # noqa: S308
+
+    def message_type(self, obj):
+        return obj.message.get("type") if obj.message else None
+
+    def content(self, obj):
+        return obj.message.get("data", {}).get("content") if obj.message else None
+
+    def has_add_permission(self, request, obj=None):
+        return False
+
+    def has_delete_permission(self, request, obj=None):
+        return False
+
+    def has_change_permission(self, request, obj=None):
+        return False
+
+
 @admin.register(Thread)
 class ThreadAdmin(admin.ModelAdmin):
     list_display = ("name", "created_at", "created_by", "updated_at")
     search_fields = ("name",)
     list_filter = ("created_at", "updated_at")
     raw_id_fields = ("created_by",)
+    inlines: ClassVar[List[Type[InlineModelAdmin]]] = [MessageInline]
 
 
 @admin.register(Message)
 class MessageAdmin(admin.ModelAdmin):
-    list_display = ("id", "thread", "created_at", "created_at")
+    list_display = ("id", "thread", "created_at")
     search_fields = ("thread__name", "message")
     list_filter = ("created_at",)
     raw_id_fields = ("thread",)

django_ai_assistant/helpers/assistants.py

@@ -12,7 +12,10 @@
     DEFAULT_DOCUMENT_PROMPT,
     DEFAULT_DOCUMENT_SEPARATOR,
 )
-from langchain_core.chat_history import BaseChatMessageHistory, InMemoryChatMessageHistory
+from langchain_core.chat_history import (
+    BaseChatMessageHistory,
+    InMemoryChatMessageHistory,
+)
 from langchain_core.language_models import BaseChatModel
 from langchain_core.output_parsers import StrOutputParser
 from langchain_core.prompts import (
@@ -295,7 +298,9 @@ def get_message_history(self, thread_id: Any | None) -> BaseChatMessageHistory:
         """
 
         # DjangoChatMessageHistory must be here because Django may not be loaded yet elsewhere:
-        from django_ai_assistant.langchain.chat_message_histories import DjangoChatMessageHistory
+        from django_ai_assistant.langchain.chat_message_histories import (
+            DjangoChatMessageHistory,
+        )
 
         if thread_id is None:
             return InMemoryChatMessageHistory()
@@ -538,7 +543,7 @@ def invoke(self, *args: Any, thread_id: Any | None, **kwargs: Any) -> dict:
         return chain.invoke(*args, **kwargs)
 
     @with_cast_id
-    def run(self, message: str, thread_id: Any | None, **kwargs: Any) -> str:
+    def run(self, message: str, thread_id: Any | None = None, **kwargs: Any) -> str:
         """Run the assistant with the given message and thread ID.\n
         This is the higher-level method to run the assistant.\n