chatlas 0.2.0__tar.gz → 0.4.0__tar.gz

{chatlas-0.2.0 → chatlas-0.4.0}/.github/workflows/check-update-types.yml

@@ -36,7 +36,12 @@ jobs:
       # Fail if there are changes under chatlas/types and throw a useful error message
       - name: Check for changes
         run: |
+          echo "Checking for changes in types..."
+          git status chatlas/types
+          git diff --stat chatlas/types
           if [[ -n $(git diff chatlas/types) ]]; then
+            echo "Changes detected:"
+            git diff chatlas/types
             echo "::error::Types are out of date. Please run 'make update-types' and commit the changes."
             exit 1
           fi

{chatlas-0.2.0 → chatlas-0.4.0}/.github/workflows/release.yml

@@ -5,8 +5,7 @@ on:
     types: [published]
 
 env:
-  UV_VERSION: "0.4.x"
-  PYTHON_VERSION: 3.13
+  PYTHON_VERSION: 3.12
 
 jobs:
   pypi-release:
@@ -27,8 +26,6 @@ jobs:
 
       - name: 🚀 Install uv
         uses: astral-sh/setup-uv@v3
-        with:
-          version: ${{ env.UV_VERSION }}
 
       - name: 🐍 Set up Python ${{ env.PYTHON_VERSION }}
         run: uv python install ${{ env.PYTHON_VERSION }}

{chatlas-0.2.0 → chatlas-0.4.0}/.github/workflows/test.yml

@@ -50,6 +50,7 @@ jobs:
         run: make check-tests
 
       - name: 📝 Check types
+        if: ${{ matrix.config.python-version != '3.9' }}
         run: make check-types
 
       - name: 📐 Check formatting

chatlas-0.4.0/CHANGELOG.md

@@ -0,0 +1,43 @@
+# Changelog
+
+<!--
+All notable changes to this project will be documented in this file.
+
+The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
+and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
+-->
+
+## [0.4.0] - 2025-02-19
+
+### New features
+
+* Added a `ChatVertex()` class to interact with Google Cloud's Vertex AI. (#50)
+* Added `.app(*, echo=)` support. This allows for chatlas to change the echo behavior when running the Shiny app. (#31)
+
+### Improvements
+
+* Migrated `ChatGoogle()`'s underlying python SDK from `google-generative` to `google-genai`. As a result, streaming tools are now working properly. (#50)
+
+### Bug fixes
+
+* Fixed a bug where synchronous chat tools would not work properly when used in a `_async()` context. (#56)
+* Fix broken `Chat`'s Shiny app when `.app(*, stream=True)` by using async chat tools. (#31)
+* Update formatting of exported markdown to use `repr()` instead of `str()` when exporting tool call results. (#30)
+
+## [0.3.0] - 2024-12-20
+
+### New features
+
+* `Chat`'s `.tokens()` method gains a `values` argument. Set it to `"discrete"` to get a result that can be summed to determine the token cost of submitting the current turns. The default (`"cumulative"`), remains the same (the result can be summed to determine the overall token cost of the conversation).
+* `Chat` gains a `.token_count()` method to help estimate token cost of new input. (#23)
+
+### Bug fixes
+
+* `ChatOllama` no longer fails when a `OPENAI_API_KEY` environment variable is not set.
+* `ChatOpenAI` now correctly includes the relevant `detail` on `ContentImageRemote()` input.
+* `ChatGoogle` now correctly logs its `token_usage()`. (#23)
+
+
+## [0.2.0] - 2024-12-11
+
+First stable release of `chatlas`, see the website to learn more <https://posit-dev.github.io/chatlas/>

{chatlas-0.2.0 → chatlas-0.4.0}/PKG-INFO

@@ -1,6 +1,6 @@
-Metadata-Version: 2.3
+Metadata-Version: 2.4
 Name: chatlas
-Version: 0.2.0
+Version: 0.4.0
 Summary: A simple and consistent interface for chatting with LLMs
 Project-URL: Homepage, https://posit-dev.github.io/chatlas
 Project-URL: Documentation, https://posit-dev.github.io/chatlas
@@ -22,7 +22,7 @@ Requires-Dist: pydantic>=2.0
 Requires-Dist: rich
 Provides-Extra: dev
 Requires-Dist: anthropic[bedrock]; extra == 'dev'
-Requires-Dist: google-generativeai>=0.8.3; extra == 'dev'
+Requires-Dist: google-genai>=1.2.0; extra == 'dev'
 Requires-Dist: matplotlib; extra == 'dev'
 Requires-Dist: numpy>1.24.4; extra == 'dev'
 Requires-Dist: openai; extra == 'dev'
@@ -30,15 +30,18 @@ Requires-Dist: pillow; extra == 'dev'
 Requires-Dist: python-dotenv; extra == 'dev'
 Requires-Dist: ruff>=0.6.5; extra == 'dev'
 Requires-Dist: shiny; extra == 'dev'
+Requires-Dist: tiktoken; extra == 'dev'
 Provides-Extra: docs
 Requires-Dist: griffe>=1; extra == 'docs'
 Requires-Dist: ipykernel; extra == 'docs'
 Requires-Dist: ipywidgets; extra == 'docs'
 Requires-Dist: nbclient; extra == 'docs'
 Requires-Dist: nbformat; extra == 'docs'
+Requires-Dist: numpy; extra == 'docs'
 Requires-Dist: pandas; extra == 'docs'
 Requires-Dist: pyyaml; extra == 'docs'
 Requires-Dist: quartodoc>=0.7; extra == 'docs'
+Requires-Dist: sentence-transformers; extra == 'docs'
 Provides-Extra: test
 Requires-Dist: pyright>=1.1.379; extra == 'test'
 Requires-Dist: pytest-asyncio; extra == 'test'
@@ -46,14 +49,24 @@ Requires-Dist: pytest>=8.3.2; extra == 'test'
 Requires-Dist: syrupy>=4; extra == 'test'
 Description-Content-Type: text/markdown
 
-# chatlas
+<h1 class="unnumbered unlisted"> chatlas <a href="https://posit-dev.github.io/chatlas"><img src="docs/images/logo.png" align="right" height="138" alt="chatlas website" /></a> </h1> 
+
+
+
+<p>
+<!-- badges start -->
+<a href="https://pypi.org/project/chatlas/"><img alt="PyPI" src="https://img.shields.io/pypi/v/chatlas?logo=python&logoColor=white&color=orange"></a>
+<a href="https://choosealicense.com/licenses/mit/"><img src="https://img.shields.io/badge/License-MIT-blue.svg" alt="MIT License"></a>
+<a href="https://github.com/posit-dev/chatlas"><img src="https://github.com/posit-dev/chatlas/actions/workflows/test.yml/badge.svg?branch=main" alt="Python Tests"></a>
+<!-- badges end -->
+</p>
 
 chatlas provides a simple and unified interface across large language model (llm) providers in Python. 
-It abstracts away complexity from common tasks like streaming chat interfaces, tool calling, structured output, and much more.
-chatlas helps you prototype faster without painting you into a corner; for example, switching providers is as easy as changing one line of code, but provider specific features are still accessible when needed.
-Developer experience is also a key focus of chatlas: typing support, rich console output, and built-in tooling are all included.
+It helps you prototype faster by abstracting away complexity from common tasks like streaming chat interfaces, tool calling, structured output, and much more.
+Switching providers is also as easy as changing one line of code, but you can also reach for provider-specific features when you need them.
+Developer experience is also a key focus of chatlas: typing support, rich console output, and extension points are all included.
 
-(Looking for something similar to chatlas, but in R? Check out [elmer](https://elmer.tidyverse.org/)!)
+(Looking for something similar to chatlas, but in R? Check out [ellmer](https://ellmer.tidyverse.org/)!)
 
 ## Install
 
@@ -85,6 +98,7 @@ It also supports the following enterprise cloud providers:
 
 * AWS Bedrock: [`ChatBedrockAnthropic()`](https://posit-dev.github.io/chatlas/reference/ChatBedrockAnthropic.html).
 * Azure OpenAI: [`ChatAzureOpenAI()`](https://posit-dev.github.io/chatlas/reference/ChatAzureOpenAI.html).
+* Vertex AI: [`ChatVertex()`](https://posit-dev.github.io/chatlas/reference/ChatVertex.html).
 
 To use a model provider that isn't listed here, you have two options:
 
@@ -123,7 +137,7 @@ From a `chat` instance, it's simple to start a web-based or terminal-based chat
 chat.app()
 ```
 
-<div style="display:flex;justify-content:center;">
+<div align="center">
 <img width="500" alt="A web app for chatting with an LLM via chatlas" src="https://github.com/user-attachments/assets/e43f60cb-3686-435a-bd11-8215cb024d2e" class="border rounded">
 </div>
 
@@ -279,7 +293,7 @@ asyncio.run(main())
 
 `chatlas` has full typing support, meaning that, among other things, autocompletion just works in your favorite editor:
 
-<div style="display:flex;justify-content:center;">
+<div align="center">
 <img width="500" alt="Autocompleting model options in ChatOpenAI" src="https://github.com/user-attachments/assets/163d6d8a-7d58-422d-b3af-cc9f2adee759" class="rounded">
 </div>
 
@@ -299,7 +313,7 @@ This shows important information like tool call results, finish reasons, and mor
 If the problem isn't self-evident, you can also reach into the `.get_last_turn()`, which contains the full response object, with full details about the completion.
 
 
-<div style="display:flex;justify-content:center;">
+<div align="center">
   <img width="500" alt="Turn completion details with typing support" src="https://github.com/user-attachments/assets/eaea338d-e44a-4e23-84a7-2e998d8af3ba" class="rounded">
 </div>
 

{chatlas-0.2.0 → chatlas-0.4.0}/README.md

@@ -1,11 +1,21 @@
-# chatlas
+<h1 class="unnumbered unlisted"> chatlas <a href="https://posit-dev.github.io/chatlas"><img src="docs/images/logo.png" align="right" height="138" alt="chatlas website" /></a> </h1> 
+
+
+
+<p>
+<!-- badges start -->
+<a href="https://pypi.org/project/chatlas/"><img alt="PyPI" src="https://img.shields.io/pypi/v/chatlas?logo=python&logoColor=white&color=orange"></a>
+<a href="https://choosealicense.com/licenses/mit/"><img src="https://img.shields.io/badge/License-MIT-blue.svg" alt="MIT License"></a>
+<a href="https://github.com/posit-dev/chatlas"><img src="https://github.com/posit-dev/chatlas/actions/workflows/test.yml/badge.svg?branch=main" alt="Python Tests"></a>
+<!-- badges end -->
+</p>
 
 chatlas provides a simple and unified interface across large language model (llm) providers in Python. 
-It abstracts away complexity from common tasks like streaming chat interfaces, tool calling, structured output, and much more.
-chatlas helps you prototype faster without painting you into a corner; for example, switching providers is as easy as changing one line of code, but provider specific features are still accessible when needed.
-Developer experience is also a key focus of chatlas: typing support, rich console output, and built-in tooling are all included.
+It helps you prototype faster by abstracting away complexity from common tasks like streaming chat interfaces, tool calling, structured output, and much more.
+Switching providers is also as easy as changing one line of code, but you can also reach for provider-specific features when you need them.
+Developer experience is also a key focus of chatlas: typing support, rich console output, and extension points are all included.
 
-(Looking for something similar to chatlas, but in R? Check out [elmer](https://elmer.tidyverse.org/)!)
+(Looking for something similar to chatlas, but in R? Check out [ellmer](https://ellmer.tidyverse.org/)!)
 
 ## Install
 
@@ -37,6 +47,7 @@ It also supports the following enterprise cloud providers:
 
 * AWS Bedrock: [`ChatBedrockAnthropic()`](https://posit-dev.github.io/chatlas/reference/ChatBedrockAnthropic.html).
 * Azure OpenAI: [`ChatAzureOpenAI()`](https://posit-dev.github.io/chatlas/reference/ChatAzureOpenAI.html).
+* Vertex AI: [`ChatVertex()`](https://posit-dev.github.io/chatlas/reference/ChatVertex.html).
 
 To use a model provider that isn't listed here, you have two options:
 
@@ -75,7 +86,7 @@ From a `chat` instance, it's simple to start a web-based or terminal-based chat
 chat.app()
 ```
 
-<div style="display:flex;justify-content:center;">
+<div align="center">
 <img width="500" alt="A web app for chatting with an LLM via chatlas" src="https://github.com/user-attachments/assets/e43f60cb-3686-435a-bd11-8215cb024d2e" class="border rounded">
 </div>
 
@@ -231,7 +242,7 @@ asyncio.run(main())
 
 `chatlas` has full typing support, meaning that, among other things, autocompletion just works in your favorite editor:
 
-<div style="display:flex;justify-content:center;">
+<div align="center">
 <img width="500" alt="Autocompleting model options in ChatOpenAI" src="https://github.com/user-attachments/assets/163d6d8a-7d58-422d-b3af-cc9f2adee759" class="rounded">
 </div>
 
@@ -251,7 +262,7 @@ This shows important information like tool call results, finish reasons, and mor
 If the problem isn't self-evident, you can also reach into the `.get_last_turn()`, which contains the full response object, with full details about the completion.
 
 
-<div style="display:flex;justify-content:center;">
+<div align="center">
   <img width="500" alt="Turn completion details with typing support" src="https://github.com/user-attachments/assets/eaea338d-e44a-4e23-84a7-2e998d8af3ba" class="rounded">
 </div>
 

{chatlas-0.2.0 → chatlas-0.4.0}/chatlas/__init__.py

@@ -3,7 +3,7 @@ from ._anthropic import ChatAnthropic, ChatBedrockAnthropic
 from ._chat import Chat
 from ._content_image import content_image_file, content_image_plot, content_image_url
 from ._github import ChatGithub
-from ._google import ChatGoogle
+from ._google import ChatGoogle, ChatVertex
 from ._groq import ChatGroq
 from ._interpolate import interpolate, interpolate_file
 from ._ollama import ChatOllama
@@ -24,6 +24,7 @@ __all__ = (
     "ChatOpenAI",
     "ChatAzureOpenAI",
     "ChatPerplexity",
+    "ChatVertex",
     "Chat",
     "content_image_file",
     "content_image_plot",

{chatlas-0.2.0 → chatlas-0.4.0}/chatlas/_anthropic.py

@@ -20,7 +20,7 @@ from ._logging import log_model_default
 from ._provider import Provider
 from ._tokens import tokens_log
 from ._tools import Tool, basemodel_to_param_schema
-from ._turn import Turn, normalize_turns
+from ._turn import Turn, normalize_turns, user_turn
 
 if TYPE_CHECKING:
     from anthropic.types import (
@@ -311,7 +311,8 @@ class AnthropicProvider(Provider[Message, RawMessageStreamEvent, Message]):
             if stream:
                 stream = False
                 warnings.warn(
-                    "Anthropic does not support structured data extraction in streaming mode."
+                    "Anthropic does not support structured data extraction in streaming mode.",
+                    stacklevel=2,
                 )
 
         kwargs_full: "SubmitInputArgs" = {
@@ -371,15 +372,65 @@ class AnthropicProvider(Provider[Message, RawMessageStreamEvent, Message]):
 
         return completion
 
-    def stream_turn(self, completion, has_data_model, stream) -> Turn:
-        return self._as_turn(completion, has_data_model)
-
-    async def stream_turn_async(self, completion, has_data_model, stream) -> Turn:
+    def stream_turn(self, completion, has_data_model) -> Turn:
         return self._as_turn(completion, has_data_model)
 
     def value_turn(self, completion, has_data_model) -> Turn:
         return self._as_turn(completion, has_data_model)
 
+    def token_count(
+        self,
+        *args: Content | str,
+        tools: dict[str, Tool],
+        data_model: Optional[type[BaseModel]],
+    ) -> int:
+        kwargs = self._token_count_args(
+            *args,
+            tools=tools,
+            data_model=data_model,
+        )
+        res = self._client.messages.count_tokens(**kwargs)
+        return res.input_tokens
+
+    async def token_count_async(
+        self,
+        *args: Content | str,
+        tools: dict[str, Tool],
+        data_model: Optional[type[BaseModel]],
+    ) -> int:
+        kwargs = self._token_count_args(
+            *args,
+            tools=tools,
+            data_model=data_model,
+        )
+        res = await self._async_client.messages.count_tokens(**kwargs)
+        return res.input_tokens
+
+    def _token_count_args(
+        self,
+        *args: Content | str,
+        tools: dict[str, Tool],
+        data_model: Optional[type[BaseModel]],
+    ) -> dict[str, Any]:
+        turn = user_turn(*args)
+
+        kwargs = self._chat_perform_args(
+            stream=False,
+            turns=[turn],
+            tools=tools,
+            data_model=data_model,
+        )
+
+        args_to_keep = [
+            "messages",
+            "model",
+            "system",
+            "tools",
+            "tool_choice",
+        ]
+
+        return {arg: kwargs[arg] for arg in args_to_keep if arg in kwargs}
+
     def _as_message_params(self, turns: list[Turn]) -> list["MessageParam"]:
         messages: list["MessageParam"] = []
         for turn in turns:
@@ -575,6 +626,53 @@ def ChatBedrockAnthropic(
         Additional arguments to pass to the `anthropic.AnthropicBedrock()`
         client constructor.
 
+    Troubleshooting
+    ---------------
+
+    If you encounter 400 or 403 errors when trying to use the model, keep the
+    following in mind:
+
+    ::: {.callout-note}
+    #### Incorrect model name
+
+    If the model name is completely incorrect, you'll see an error like
+    `Error code: 400 - {'message': 'The provided model identifier is invalid.'}`
+
+    Make sure the model name is correct and active in the specified region.
+    :::
+
+    ::: {.callout-note}
+    #### Models are region specific
+
+    If you encounter errors similar to `Error code: 403 - {'message': "You don't
+    have access to the model with the specified model ID."}`, make sure your
+    model is active in the relevant `aws_region`.
+
+    Keep in mind, if `aws_region` is not specified, and AWS_REGION is not set,
+    the region defaults to us-east-1, which may not match to your AWS config's
+    default region.
+    :::
+
+    ::: {.callout-note}
+    #### Cross region inference ID
+
+    In some cases, even if you have the right model and the right region, you
+    may still encounter an error like  `Error code: 400 - {'message':
+    'Invocation of model ID anthropic.claude-3-5-sonnet-20240620-v1:0 with
+    on-demand throughput isn't supported. Retry your request with the ID or ARN
+    of an inference profile that contains this model.'}`
+
+    In this case, you'll need to look up the 'cross region inference ID' for
+    your model. This might required opening your `aws-console` and navigating to
+    the 'Anthropic Bedrock' service page. From there, go to the 'cross region
+    inference' tab and copy the relevant ID.
+
+    For example, if the desired model ID is
+    `anthropic.claude-3-5-sonnet-20240620-v1:0`, the cross region ID might look
+    something like `us.anthropic.claude-3-5-sonnet-20240620-v1:0`.
+    :::
+
+
     Returns
     -------
     Chat