"message": "As this PR has been waiting for the original user for a while but seems to be inactive, it's now going to be closed. But if there's anyone interested, feel free to create a new PR.",
"reminder": {
"before": "P3D",
"message": "Heads-up: this will be closed in 3 days unless there’s new activity."
"message": "Heads-up: this will be closed in 3 days unless there's new activity."
}
},
"invalid": {
"delay": 0,
"message": "This was marked as invalid and will be closed now. If this is an error, please provide additional details."
},
"maybe-ai": {
"delay": 0,
"message": "This was marked as potentially AI generated and will be closed now. If this is an error, please provide additional details, make sure to read the docs about contributing and AI."
* **Fast**: Very high performance, on par with **NodeJS** and **Go** (thanks to Starlette and Pydantic). [One of the fastest Python frameworks available](#performance).
* **Fast to code**: Increase the speed to develop features by about 200% to 300%. *
* **Fewer bugs**: Reduce about 40% of human (developer) induced errors. *
* **Intuitive**: Great editor support. <abbrtitle="also known as auto-complete, autocompletion, IntelliSense">Completion</abbr> everywhere. Less time debugging.
* **Intuitive**: Great editor support. <dfntitle="also known as auto-complete, autocompletion, IntelliSense">Completion</dfn> everywhere. Less time debugging.
* **Easy**: Designed to be easy to use and learn. Less time reading docs.
* **Short**: Minimize code duplication. Multiple features from each parameter declaration. Fewer bugs.
* **Robust**: Get production-ready code. With automatic interactive documentation.
* Automatic and clear errors when the data is invalid.
* Validation even for deeply nested JSON objects.
* <abbrtitle="also known as: serialization, parsing, marshalling">Conversion</abbr> of input data: coming from the network to Python data and types. Reading from:
* <dfntitle="also known as: serialization, parsing, marshalling">Conversion</dfn> of input data: coming from the network to Python data and types. Reading from:
* JSON.
* Path parameters.
* Query parameters.
@ -385,7 +379,7 @@ item: Item
* Headers.
* Forms.
* Files.
* <abbrtitle="also known as: serialization, parsing, marshalling">Conversion</abbr> of output data: converting from Python data and types to network data (as JSON):
* <dfntitle="also known as: serialization, parsing, marshalling">Conversion</dfn> of output data: converting from Python data and types to network data (as JSON):
@ -448,7 +442,7 @@ For a more complete example including more features, see the <a href="https://fa
* Declaration of **parameters** from other different places as: **headers**, **cookies**, **form fields** and **files**.
* How to set **validation constraints** as `maximum_length` or `regex`.
* A very powerful and easy to use **<abbrtitle="also known as components, resources, providers, services, injectables">Dependency Injection</abbr>** system.
* A very powerful and easy to use **<dfntitle="also known as components, resources, providers, services, injectables">Dependency Injection</dfn>** system.
* Security and authentication, including support for **OAuth2** with **JWT tokens** and **HTTP Basic** auth.
* More advanced (but equally easy) techniques for declaring **deeply nested JSON models** (thanks to Pydantic).
* **GraphQL** integration with <ahref="https://strawberry.rocks"class="external-link"target="_blank">Strawberry</a> and other libraries.
@ -533,7 +527,7 @@ Used by Starlette:
* <ahref="https://www.python-httpx.org"target="_blank"><code>httpx</code></a> - Required if you want to use the `TestClient`.
* <ahref="https://jinja.palletsprojects.com"target="_blank"><code>jinja2</code></a> - Required if you want to use the default template configuration.
* <ahref="https://github.com/Kludex/python-multipart"target="_blank"><code>python-multipart</code></a> - Required if you want to support form <abbrtitle="converting the string that comes from an HTTP request into Python data">"parsing"</abbr>, with `request.form()`.
* <ahref="https://github.com/Kludex/python-multipart"target="_blank"><code>python-multipart</code></a> - Required if you want to support form <dfntitle="converting the string that comes from an HTTP request into Python data">"parsing"</dfn>, with `request.form()`.
@ -35,7 +35,7 @@ Siehe Abschnitt `### Content of code snippets` im allgemeinen Prompt in `scripts
//// tab | Test
Gestern schrieb mein Freund: „Wenn man unkorrekt korrekt schreibt, hat man es unkorrekt geschrieben“. Worauf ich antwortete: „Korrekt, aber ‚unkorrekt‘ ist unkorrekterweise nicht ‚„unkorrekt“‘“.
Gestern schrieb mein Freund: „Wenn man ‚incorrectly‘ korrekt schreibt, hat man es falsch geschrieben“. Worauf ich antwortete: „Korrekt, aber ‚incorrectly‘ ist inkorrekterweise nicht ‚„incorrectly“‘“.
/// note | Hinweis
@ -189,7 +189,7 @@ Siehe Abschnitt `### Links` im allgemeinen Prompt in `scripts/translate.py`.
////
## HTML„abbr“-Elemente { #html-abbr-elements }
## HTML-„abbr“-Elemente { #html-abbr-elements }
//// tab | Test
@ -202,11 +202,6 @@ Hier einige Dinge, die in HTML-„abbr“-Elemente gepackt sind (einige sind erf
### Das abbr gibt eine Erklärung { #the-abbr-gives-an-explanation }
* <abbrtitle="Eine Gruppe von Maschinen, die so konfiguriert sind, dass sie verbunden sind und in irgendeiner Weise zusammenarbeiten.">Cluster</abbr>
* <abbrtitle="Eine Methode des Machine Learning, die künstliche neuronale Netze mit zahlreichen versteckten Schichten zwischen Eingabe- und Ausgabeschicht verwendet und so eine umfassende interne Struktur entwickelt">Deep Learning</abbr>
### Das abbr gibt eine vollständige Phrase und eine Erklärung { #the-abbr-gives-a-full-phrase-and-an-explanation }
* <abbrtitle="Mozilla Developer Network – Mozilla-Entwicklernetzwerk: Dokumentation für Entwickler, geschrieben von den Firefox-Leuten">MDN</abbr>
@ -224,6 +219,11 @@ Siehe Abschnitt `### HTML abbr elements` im allgemeinen Prompt in `scripts/trans
////
## HTML „dfn“-Elemente { #html-dfn-elements }
* <dfntitle="Eine Gruppe von Maschinen, die so konfiguriert sind, dass sie verbunden sind und in irgendeiner Weise zusammenarbeiten.">Cluster</dfn>
* <dfntitle="Eine Methode des Machine Learning, die künstliche neuronale Netze mit zahlreichen versteckten Schichten zwischen Eingabe- und Ausgabeschicht verwendet und so eine umfassende interne Struktur entwickelt">Deep Learning</dfn>
## Überschriften { #headings }
//// tab | Test
@ -248,7 +248,7 @@ Die einzige strenge Regel für Überschriften ist, dass das LLM den Hash-Teil in
Siehe Abschnitt `### Headings` im allgemeinen Prompt in `scripts/translate.py`.
Für einige sprachspezifische Anweisungen, siehe z. B. den Abschnitt `### Headings` in `docs/de/llm-prompt.md`.
Für einige sprachsspezifische Anweisungen, siehe z. B. den Abschnitt `### Headings` in `docs/de/llm-prompt.md`.
│ ├── __init__.py # diese Datei macht „app“ zu einem „Python-Package“
│ ├── main.py # „main“-Modul, z. B. import app.main
│ ├── dependencies.py # „dependencies“-Modul, z. B. import app.dependencies
│ └── routers # „routers“ ist ein „Python-Subpackage“
│ │ ├── __init__.py # macht „routers“ zu einem „Python-Subpackage“
│ │ ├── items.py # „items“-Submodul, z. B. import app.routers.items
│ │ └── users.py # „users“-Submodul, z. B. import app.routers.users
│ └── internal # „internal“ ist ein „Python-Subpackage“
│ ├── __init__.py # macht „internal“ zu einem „Python-Subpackage“
│ └── admin.py # „admin“-Submodul, z. B. import app.internal.admin
├── app # "app" ist ein Python-Package
│ ├── __init__.py # diese Datei macht "app" zu einem "Python-Package"
│ ├── main.py # "main"-Modul, z. B. import app.main
│ ├── dependencies.py # "dependencies"-Modul, z. B. import app.dependencies
│ └── routers # "routers" ist ein "Python-Subpackage"
│ │ ├── __init__.py # macht "routers" zu einem "Python-Subpackage"
│ │ ├── items.py # "items"-Submodul, z. B. import app.routers.items
│ │ └── users.py # "users"-Submodul, z. B. import app.routers.users
│ └── internal # "internal" ist ein "Python-Subpackage"
│ ├── __init__.py # macht "internal" zu einem "Python-Subpackage"
│ └── admin.py # "admin"-Submodul, z. B. import app.internal.admin
```
## `APIRouter` { #apirouter }
@ -479,7 +479,7 @@ $ fastapi dev app/main.py
</div>
und öffnen Sie die Dokumentation unter <ahref="http://127.0.0.1:8000/docs"class="external-link"target="_blank">http://127.0.0.1:8000/docs</a>.
Und öffnen Sie die Dokumentation unter <ahref="http://127.0.0.1:8000/docs"class="external-link"target="_blank">http://127.0.0.1:8000/docs</a>.
Sie sehen die automatische API-Dokumentation, einschließlich der Pfade aller Submodule, mit den richtigen Pfaden (und Präfixen) und den richtigen Tags:
# Body – Mehrere Parameter { #body-multiple-parameters }
Nun, da wir gesehen haben, wie `Path` und `Query` verwendet werden, schauen wir uns fortgeschrittenere Verwendungsmöglichkeiten von <abbrtitle="Anfragekörper">Requestbody</abbr>-Deklarationen an.
Nun, da wir gesehen haben, wie `Path` und `Query` verwendet werden, schauen wir uns fortgeschrittenere Verwendungsmöglichkeiten von <abbrtitle="Requestbody">Requestbody</abbr>-Deklarationen an.
## `Path`-, `Query`- und Body-Parameter vermischen { #mix-path-query-and-body-parameters }
@ -101,13 +101,13 @@ Natürlich können Sie auch, wann immer Sie das brauchen, weitere Query-Paramete
Da einfache Werte standardmäßig als Query-Parameter interpretiert werden, müssen Sie `Query` nicht explizit hinzufügen, Sie können einfach schreiben:
@ -202,11 +202,6 @@ Here some things wrapped in HTML "abbr" elements (Some are invented):
* <abbrtitle="XML Web Token">XWT</abbr>
* <abbrtitle="Parallel Server Gateway Interface">PSGI</abbr>
### The abbr gives an explanation { #the-abbr-gives-an-explanation }
* <abbrtitle="A group of machines that are configured to be connected and work together in some way.">cluster</abbr>
* <abbrtitle="A method of machine learning that uses artificial neural networks with numerous hidden layers between input and output layers, thereby developing a comprehensive internal structure">Deep Learning</abbr>
### The abbr gives a full phrase and an explanation { #the-abbr-gives-a-full-phrase-and-an-explanation }
* <abbrtitle="Mozilla Developer Network: documentation for developers, written by the Firefox people">MDN</abbr>
@ -224,6 +219,11 @@ See section `### HTML abbr elements` in the general prompt in `scripts/translate
////
## HTML "dfn" elements { #html-dfn-elements }
* <dfntitle="A group of machines that are configured to be connected and work together in some way.">cluster</dfn>
* <dfntitle="A method of machine learning that uses artificial neural networks with numerous hidden layers between input and output layers, thereby developing a comprehensive internal structure">Deep Learning</dfn>
In this case, this `__call__` is what **FastAPI** will use to check for additional parameters and sub-dependencies, and this is what will be called to pass a value to the parameter in your *path operation function* later.
@ -26,7 +26,7 @@ In this case, this `__call__` is what **FastAPI** will use to check for addition
And now, we can use `__init__` to declare the parameters of the instance that we can use to "parameterize" the dependency:
Here are some additional ideas that might be useful when working with Python types.
## Using `Union` or `Optional` { #using-union-or-optional }
If your code for some reason can't use `|`, for example if it's not in a type annotation but in something like `response_model=`, instead of using the vertical bar (`|`) you can use `Union` from `typing`.
For example, you could declare that something could be a `str` or `None`:
```python
from typing import Union
def say_hi(name: Union[str, None]):
print(f"Hi {name}!")
```
`typing` also has a shortcut to declare that something could be `None`, with `Optional`.
Here's a tip from my very **subjective** point of view:
* 🚨 Avoid using `Optional[SomeType]`
* Instead ✨ **use `Union[SomeType, None]`** ✨.
Both are equivalent and underneath they are the same, but I would recommend `Union` instead of `Optional` because the word "**optional**" would seem to imply that the value is optional, and it actually means "it can be `None`", even if it's not optional and is still required.
I think `Union[SomeType, None]` is more explicit about what it means.
It's just about the words and names. But those words can affect how you and your teammates think about the code.
As an example, let's take this function:
```python
from typing import Optional
def say_hi(name: Optional[str]):
print(f"Hey {name}!")
```
The parameter `name` is defined as `Optional[str]`, but it is **not optional**, you cannot call the function without the parameter:
```Python
say_hi() # Oh, no, this throws an error! 😱
```
The `name` parameter is **still required** (not *optional*) because it doesn't have a default value. Still, `name` accepts `None` as the value:
```Python
say_hi(name=None) # This works, None is valid 🎉
```
The good news is, in most cases, you will be able to simply use `|` to define unions of types:
```python
def say_hi(name: str | None):
print(f"Hey {name}!")
```
So, normally you don't have to worry about names like `Optional` and `Union`. 😎
And the proxy would be **"stripping"** the **path prefix** on the fly before transmitting the request to the app server (probably Uvicorn via FastAPI CLI), keeping your application convinced that it is being served at `/app`, so that you don't have to update all your code to include the prefix `/api/v1`.
@ -193,7 +193,7 @@ You can get the current `root_path` used by your application for each request, i
Here we are including it in the message just for demonstration purposes.
@ -220,7 +220,7 @@ The response would be something like:
Alternatively, if you don't have a way to provide a command line option like `--root-path` or equivalent, you can set the `root_path` parameter when creating your FastAPI app:
@ -30,7 +30,7 @@ This is because by default, FastAPI will inspect every item inside and make sure
But if you are certain that the content that you are returning is **serializable with JSON**, you can pass it directly to the response class and avoid the extra overhead that FastAPI would have by passing your return content through the `jsonable_encoder` before passing it to the response class.
In this example, the function `generate_html_response()` already generates and returns a `Response` instead of returning the HTML in a `str`.
@ -136,7 +136,7 @@ It accepts the following parameters:
FastAPI (actually Starlette) will automatically include a Content-Length header. It will also include a Content-Type header, based on the `media_type` and appending a charset for text types.
1. This is the generator function. It's a "generator function" because it contains `yield` statements inside.
2. By using a `with` block, we make sure that the file-like object is closed after the generator function is done. So, after it finishes sending the response.
@ -256,11 +256,11 @@ Takes a different set of arguments to instantiate than the other response types:
File responses will include appropriate `Content-Length`, `Last-Modified` and `ETag` headers.
In this case, you can return the file path directly from your *path operation* function.
@ -274,7 +274,7 @@ Let's say you want it to return indented and formatted JSON, so you want to use
You could create a `CustomORJSONResponse`. The main thing you have to do is create a `Response.render(content)` method that returns the content as `bytes`:
Here we are simulating the expensive *startup* operation of loading the model by putting the (fake) model function in the dictionary with machine learning models before the `yield`. This code will be executed **before** the application **starts taking requests**, during the *startup*.
@ -48,7 +48,7 @@ Maybe you need to start a new version, or you just got tired of running it. 🤷
The first thing to notice, is that we are defining an async function with `yield`. This is very similar to Dependencies with `yield`.
@ -32,7 +32,7 @@ Webhooks are available in OpenAPI 3.1.0 and above, supported by FastAPI `0.99.0`
When you create a **FastAPI** application, there is a `webhooks` attribute that you can use to define *webhooks*, the same way you would define *path operations*, for example with `@app.webhooks.post()`.
@ -40,7 +40,7 @@ Even if they are in different modules (Python files).
To exclude a *path operation* from the generated OpenAPI schema (and thus, from the automatic documentation systems), use the parameter `include_in_schema` and set it to `False`:
## Advanced description from docstring { #advanced-description-from-docstring }
@ -92,7 +92,7 @@ You can extend the OpenAPI schema for a *path operation* using the parameter `op
This `openapi_extra` can be helpful, for example, to declare [OpenAPI Extensions](https://github.com/OAI/OpenAPI-Specification/blob/main/versions/3.0.3.md#specificationExtensions):
In this example, we didn't declare any Pydantic model. In fact, the request body is not even <abbrtitle="converted from some plain format, like bytes, into Python objects">parsed</abbr> as JSON, it is read directly as `bytes`, and the function `magic_data_reader()` would be in charge of parsing it in some way.
In this example, we didn't declare any Pydantic model. In fact, the request body is not even <dfntitle="converted from some plain format, like bytes, into Python objects">parsed</dfn> as JSON, it is read directly as `bytes`, and the function `magic_data_reader()` would be in charge of parsing it in some way.
Nevertheless, we can declare the expected schema for the request body.
@ -153,7 +153,7 @@ And you could do this even if the data type in the request is not JSON.
For example, in this application we don't use FastAPI's integrated functionality to extract the JSON Schema from Pydantic models nor the automatic validation for JSON. In fact, we are declaring the request content type as YAML, not JSON:
Nevertheless, although we are not using the default integrated functionality, we are still using a Pydantic model to manually generate the JSON Schema for the data that we want to receive in YAML.
@ -161,7 +161,7 @@ Then we use the request directly, and extract the body as `bytes`. This means th
And then in our code, we parse that YAML content directly, and then we are again using the same Pydantic model to validate the YAML content:
And then you can return any object you need, as you normally would (a `dict`, a database model, etc).
@ -22,7 +22,7 @@ You can also add headers when you return a `Response` directly.
Create a response as described in [Return a Response Directly](response-directly.md){.internal-link target=_blank} and pass the headers as an additional parameter:
When you try to open the URL for the first time (or click the "Execute" button in the docs) the browser will ask you for your username and password:
@ -40,7 +40,7 @@ To handle that, we first convert the `username` and `password` to `bytes` encodi
Then we can use `secrets.compare_digest()` to ensure that `credentials.username` is `"stanleyjobson"`, and that `credentials.password` is `"swordfish"`.
@ -104,4 +104,4 @@ That way, using `secrets.compare_digest()` in your application code, it will be
After detecting that the credentials are incorrect, return an `HTTPException` with a status code 401 (the same returned when no credentials are provided) and add the header `WWW-Authenticate` to make the browser show the login prompt again:
@ -54,7 +54,7 @@ The same way as with Pydantic models, you declare class attributes with type ann
You can use all the same validation features and tools you use for Pydantic models, like different data types and additional validations with `Field()`.
Then for any subsequent call of `get_settings()` in the dependencies for the next requests, instead of executing the internal code of `get_settings()` and creating a new `Settings` object, it will return the same object that was returned on the first call, again and again.
* Declare a `Request` parameter in the *path operation* that will return a template.
* Use the `templates` you created to render and return a `TemplateResponse`, pass the name of the template, the request object, and a "context" dictionary with key-value pairs to be used inside of the Jinja2 template.
You can read more details about the ["Running lifespan in tests in the official Starlette documentation site."](https://www.starlette.dev/lifespan/#running-lifespan-in-tests)
For the deprecated `startup` and `shutdown` events, you can use the `TestClient` as follows:
You can receive and send binary, text, and JSON data.
@ -154,7 +154,7 @@ With that you can connect the WebSocket and then send and receive messages:
When a WebSocket connection is closed, the `await websocket.receive_text()` will raise a `WebSocketDisconnect` exception, which you can then catch and handle like in this example.
One of the main features needed by API systems is data "<abbrtitle="also called marshalling, conversion">serialization</abbr>" which is taking data from the code (Python) and converting it into something that can be sent through the network. For example, converting an object containing data from a database into a JSON object. Converting `datetime` objects into strings, etc.
One of the main features needed by API systems is data "<dfntitle="also called marshalling, conversion">serialization</dfn>" which is taking data from the code (Python) and converting it into something that can be sent through the network. For example, converting an object containing data from a database into a JSON object. Converting `datetime` objects into strings, etc.
Another big feature needed by APIs is data validation, making sure that the data is valid, given certain parameters. For example, that some field is an `int`, and not some random string. This is especially useful for incoming data.
@ -145,7 +145,7 @@ Without a data validation system, you would have to do all the checks by hand, i
These features are what Marshmallow was built to provide. It is a great library, and I have used it a lot before.
But it was created before there existed Python type hints. So, to define every <abbrtitle="the definition of how data should be formed">schema</abbr> you need to use specific utils and classes provided by Marshmallow.
But it was created before there existed Python type hints. So, to define every <dfntitle="the definition of how data should be formed">schema</dfn> you need to use specific utils and classes provided by Marshmallow.
/// check | Inspired **FastAPI** to
@ -155,7 +155,7 @@ Use code to define "schemas" that provide data types and validation, automatical
Starlette is a lightweight <abbrtitle="The new standard for building asynchronous Python web applications">ASGI</abbr> framework/toolkit, which is ideal for building high-performance asyncio services.
Starlette is a lightweight <dfntitle="The new standard for building asynchronous Python web applications">ASGI</dfn> framework/toolkit, which is ideal for building high-performance asyncio services.
It is very simple and intuitive. It's designed to be easily extensible, and have modular components.
@ -6,44 +6,20 @@ First, you might want to see the basic ways to [help FastAPI and get help](help-
If you already cloned the <ahref="https://github.com/fastapi/fastapi"class="external-link"target="_blank">fastapi repository</a> and you want to deep dive in the code, here are some guidelines to set up your environment.
### Virtual environment
Follow the instructions to create and activate a [virtual environment](virtual-environments.md){.internal-link target=_blank} for the internal code of `fastapi`.
### Install requirements
After activating the environment, install the required packages:
//// tab | `pip`
<divclass="termy">
```console
$ pip install -r requirements.txt
---> 100%
```
</div>
////
//// tab | `uv`
If you have <ahref="https://github.com/astral-sh/uv"class="external-link"target="_blank">`uv`</a>:
Create a virtual environment and install the required packages with <ahref="https://github.com/astral-sh/uv"class="external-link"target="_blank">`uv`</a>:
<divclass="termy">
```console
$ uv pip install -r requirements.txt
$ uv sync --extra all
---> 100%
```
</div>
////
It will install all the dependencies and your local FastAPI in your local environment.
### Using your local FastAPI
@ -56,9 +32,9 @@ That way, you don't have to "install" your local version to be able to test ever
/// note | Technical Details
This only happens when you install using this included `requirements.txt` instead of running `pip install fastapi` directly.
This only happens when you install using `uv sync --extra all` instead of running `pip install fastapi` directly.
That is because inside the `requirements.txt` file, the local version of FastAPI is marked to be installed in "editable" mode, with the `-e` option.
That is because `uv sync --extra all` will install the local version of FastAPI in "editable" mode by default.
///
@ -201,252 +177,85 @@ as Uvicorn by default will use the port `8000`, the documentation on port `8008`
### Translations
/// warning | Attention
**Update on Translations**
We're updating the way we handle documentation translations.
Until now, we invited community members to translate pages via pull requests, which were then reviewed by at least two native speakers. While this has helped bring FastAPI to many more users, we’ve also run into several challenges - some languages have only a few translated pages, others are outdated and hard to maintain over time.
To improve this, we’re working on automation tools 🤖 to manage translations more efficiently. Once ready, documentation will be machine-translated and still reviewed by at least two native speakers ✅ before publishing. This will allow us to keep translations up-to-date while reducing the review burden on maintainers.
What’s changing now:
* 🚫 We’re no longer accepting new community-submitted translation PRs.
* ⏳ Existing open PRs will be reviewed and can still be merged if completed within the next 3 weeks (since July 11 2025).
* 🌐 In the future, we will only support languages where at least three active native speakers are available to review and maintain translations.
This transition will help us keep translations more consistent and timely while better supporting our contributors 🙌. Thank you to everyone who has contributed so far — your help has been invaluable! 💖
///
Help with translations is VERY MUCH appreciated! And it can't be done without the help from the community. 🌎 🚀
Here are the steps to help with translations.
Translation pull requests are made by LLMs guided with prompts designed by the FastAPI team together with the community of native speakers for each supported language.
#### Tips and guidelines
#### LLM Prompt per Language
* Check the currently<ahref="https://github.com/fastapi/fastapi/pulls"class="external-link"target="_blank">existing pull requests</a> for your language. You can filter the pull requests by the ones with the label for your language. For example, for Spanish, the label is <ahref="https://github.com/fastapi/fastapi/pulls?q=is%3Aopen+sort%3Aupdated-desc+label%3Alang-es+label%3Aawaiting-review"class="external-link"target="_blank">`lang-es`</a>.
Each language has a directory: <ahref="https://github.com/fastapi/fastapi/tree/master/docs"class="external-link"target="_blank">https://github.com/fastapi/fastapi/tree/master/docs</a>, in it you can see a file `llm-prompt.md` with the prompt specific for that language.
* Review those pull requests, requesting changes or approving them. For the languages I don't speak, I'll wait for several others to review the translation before merging.
For example, for Spanish, the prompt is at: <ahref="https://github.com/fastapi/fastapi/blob/master/docs/es/llm-prompt.md"class="external-link"target="_blank">`docs/es/llm-prompt.md`</a>.
/// tip
You can <ahref="https://help.github.com/en/github/collaborating-with-issues-and-pull-requests/commenting-on-a-pull-request"class="external-link"target="_blank">add comments with change suggestions</a> to existing pull requests.
If you see mistakes in your language, you can make suggestions to the prompt in that file for your language, and request the specific pages you would like to re-generate after the changes.
Check the docs about <ahref="https://help.github.com/en/github/collaborating-with-issues-and-pull-requests/about-pull-request-reviews"class="external-link"target="_blank">adding a pull request review</a> to approve it or request changes.
#### Reviewing Translation PRs
///
You can also check the currently <ahref="https://github.com/fastapi/fastapi/pulls"class="external-link"target="_blank">existing pull requests</a> for your language. You can filter the pull requests by the ones with the label for your language. For example, for Spanish, the label is <ahref="https://github.com/fastapi/fastapi/pulls?q=is%3Aopen+sort%3Aupdated-desc+label%3Alang-es+label%3Aawaiting-review"class="external-link"target="_blank">`lang-es`</a>.
* Check if there's a <ahref="https://github.com/fastapi/fastapi/discussions/categories/translations"class="external-link"target="_blank">GitHub Discussion</a> to coordinate translations for your language. You can subscribe to it, and when there's a new pull request to review, an automatic comment will be added to the discussion.
When reviewing a pull request, it's better not to suggest changes in the same pull request, because it is LLM generated, and it won't be possible to make sure that small individual changes are replicated in other similar sections, or that they are preserved when translating the same content again.
* If you translate pages, add a single pull request per page translated. That will make it much easier for others to review it.
* To check the 2-letter code for the language you want to translate, you can use the table <ahref="https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes"class="external-link"target="_blank">List of ISO 639-1 codes</a>.
#### Existing language
Let's say you want to translate a page for a language that already has translations for some pages, like Spanish.
In the case of Spanish, the 2-letter code is `es`. So, the directory for Spanish translations is located at `docs/es/`.
Instead of adding suggestions to the translation PR, make the suggestions to the LLM prompt file for that language, in a new PR. For example, for Spanish, the LLM prompt file is at: <ahref="https://github.com/fastapi/fastapi/blob/master/docs/es/llm-prompt.md"class="external-link"target="_blank">`docs/es/llm-prompt.md`</a>.
/// tip
The main ("official") language is English, located at `docs/en/`.
///
Now run the live server for the docs in Spanish:
<divclass="termy">
```console
// Use the command "live" and pass the language code as a CLI argument
$ python ./scripts/docs.py live es
<spanstyle="color: green;">[INFO]</span> Serving on http://127.0.0.1:8008
Alternatively, you can perform the same steps that scripts does manually.
Go into the language directory, for the Spanish translations it's at `docs/es/`:
```console
$ cd docs/es/
```
Then run `mkdocs` in that directory:
```console
$ mkdocs serve --dev-addr 127.0.0.1:8008
```
///
Now you can go to <ahref="http://127.0.0.1:8008"class="external-link"target="_blank">http://127.0.0.1:8008</a> and see your changes live.
You will see that every language has all the pages. But some pages are not translated and have an info box at the top, about the missing translation.
Now let's say that you want to add a translation for the section [Features](features.md){.internal-link target=_blank}.
* Copy the file at:
```
docs/en/docs/features.md
```
* Paste it in exactly the same location but for the language you want to translate, e.g.:
```
docs/es/docs/features.md
```
/// tip
Notice that the only change in the path and file name is the language code, from `en` to `es`.
Check the docs about <ahref="https://help.github.com/en/github/collaborating-with-issues-and-pull-requests/about-pull-request-reviews"class="external-link"target="_blank">adding a pull request review</a> to approve it or request changes.
///
If you go to your browser you will see that now the docs show your new section (the info box at the top is gone). 🎉
Now you can translate it all and see how it looks as you save the file.
#### Don't Translate these Pages
#### Subscribe to Notifications for Your Language
🚨 Don't translate:
Check if there's a <ahref="https://github.com/fastapi/fastapi/discussions/categories/translations"class="external-link"target="_blank">GitHub Discussion</a> to coordinate translations for your language. You can subscribe to it, and when there's a new pull request to review, an automatic comment will be added to the discussion.
* Files under `reference/`
* `release-notes.md`
* `fastapi-people.md`
* `external-links.md`
* `newsletter.md`
* `management-tasks.md`
* `management.md`
* `contributing.md`
Some of these files are updated very frequently and a translation would always be behind, or they include the main content from English source files, etc.
To check the 2-letter code for the language you want to translate, you can use the table <ahref="https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes"class="external-link"target="_blank">List of ISO 639-1 codes</a>.
#### Request a New Language
Let's say that you want to request translations for a language that is not yet translated, not even some pages. For example, Latin.
If there is no discussion for that language, you can start by requesting the new language. For that, you can follow these steps:
* The first step would be for you to find other 2 people that would be willing to be reviewing translation PRs for that language with you.
* Once there are at least 3 people that would be willing to commit to help maintain that language, you can continue the next steps.
* Create a new discussion following the template.
* Get a few native speakers to comment on the discussion and commit to help review translations for that language.
* Tag the other 2 people that will help with the language, and ask them to confirm there they will help.
Once there are several people in the discussion, the FastAPI team can evaluate it and can make it an official translation.
Then the docs will be automatically translated using AI, and the team of native speakers can review the translation, and help tweak the AI prompts.
Then the docs will be automatically translated using LLMs, and the team of native speakers can review the translation, and help tweak the LLM prompts.
Once there's a new translation, for example if docs are updated or there's a new section, there will be a comment in the same discussion with the link to the new translation to review.
#### New Language
/// note
These steps will be performed by the FastAPI team.
///
Checking the link from above (List of ISO 639-1 codes), you can see that the 2-letter code for Latin is `la`.
Now you can create a new directory for the new language, running the following script:
<divclass="termy">
```console
// Use the command new-lang, pass the language code as a CLI argument
$ python ./scripts/docs.py new-lang la
Successfully initialized: docs/la
```
</div>
Now you can check in your code editor the newly created directory `docs/la/`.
That command created a file `docs/la/mkdocs.yml` with a simple config that inherits everything from the `en` version:
```yaml
INHERIT: ../en/mkdocs.yml
```
/// tip
You could also simply create that file with those contents manually.
## Automated Code and AI
///
That command also created a dummy file `docs/la/index.md` for the main page, you can start by translating that one.
You are encouraged to use all the tools you want to do your work and contribute as efficiently as possible, this includes AI (LLM) tools, etc. Nevertheless, contributions should have meaningful human intervention, judgement, context, etc.
You can continue with the previous instructions for an "Existing Language" for that process.
If the **human effort** put in a PR, e.g. writing LLM prompts, is **less** than the **effort we would need to put** to **review it**, please **don't** submit the PR.
You can make the first pull request with those two files, `docs/la/mkdocs.yml` and `docs/la/index.md`. 🎉
Think of it this way: we can already write LLM prompts or run automated tools ourselves, and that would be faster than reviewing external PRs.
#### Preview the result
### Closing Automated and AI PRs
As already mentioned above, you can use the `./scripts/docs.py` with the `live` command to preview the results (or `mkdocs serve`).
If we see PRs that seem AI generated or automated in similar ways, we'll flag them and close them.
Once you are done, you can also test it all as it would look online, including all the other languages.
The same applies to comments and descriptions, please don't copy paste the content generated by an LLM.
To do that, first build all the docs:
<divclass="termy">
### Human Effort Denial of Service
```console
// Use the command "build-all", this will take a bit
$ python ./scripts/docs.py build-all
Building docs for: en
Building docs for: es
Successfully built docs for: es
```
</div>
This builds all those independent MkDocs sites for each language, combines them, and generates the final output at `./site/`.
Then you can serve that with the command `serve`:
<divclass="termy">
```console
// Use the command "serve" after running "build-all"
$ python ./scripts/docs.py serve
Warning: this is a very simple server. For development, use mkdocs serve instead.
This is here only to preview a site with translations already built.
Make sure you run the build-all command first.
Serving at: http://127.0.0.1:8008
```
</div>
Using automated tools and AI to submit PRs or comments that we have to carefully review and handle would be the equivalent of a <ahref="https://en.wikipedia.org/wiki/Denial-of-service_attack"class="external-link"target="_blank">Denial-of-service attack</a> on our human effort.
#### Translation specific tips and guidelines
It would be very little effort from the person submitting the PR (an LLM prompt) that generates a large amount of effort on our side (carefully reviewing code).
* Translate only the Markdown documents (`.md`). Do not translate the code examples at `./docs_src`.
Please don't do that.
* In code blocks within the Markdown document, translate comments (`# a comment`), but leave the rest unchanged.
We'll need to block accounts that spam us with repeated automated PRs or comments.
* Do not change anything enclosed in "``" (inline code).
### Use Tools Wisely
* In lines starting with `///` translate only the text part after `|`. Leave the rest unchanged.
As Uncle Ben said:
* You can translate info boxes like `/// warning` with for example `/// warning | Achtung`. But do not change the word immediately after the `///`, it determines the color of the info box.
<blockquote>
With great <strike>power</strike><strong>tools</strong> comes great responsibility.
</blockquote>
* Do not change the paths in links to images, code files, Markdown documents.
Avoid inadvertently doing harm.
* However, when a Markdown document is translated, the `#hash-parts` in links to its headings may change. Update these links if possible.
* Search for such links in the translated document using the regex `#[^# ]`.
* Search in all documents already translated into your language for `your-translated-document.md`. For example VS Code has an option "Edit" -> "Find in Files".
* When translating a document, do not "pre-translate" `#hash-parts` that link to headings in untranslated documents.
You have amazing tools at hand, use them wisely to help effectively.
@ -168,7 +166,7 @@ Now in the same project directory create a file `Dockerfile` with:
```{ .dockerfile .annotate }
# (1)!
FROM python:3.9
FROM python:3.14
# (2)!
WORKDIR /code
@ -392,7 +390,7 @@ If your FastAPI is a single file, for example, `main.py` without an `./app` dire
Then you would just have to change the corresponding paths to copy the file inside the `Dockerfile`:
```{ .dockerfile .annotate hl_lines="10 13" }
FROM python:3.9
FROM python:3.14
WORKDIR /code
@ -456,7 +454,7 @@ Without using containers, making applications run on startup and with restarts c
## Replication - Number of Processes { #replication-number-of-processes }
If you have a <abbrtitle="A group of machines that are configured to be connected and work together in some way.">cluster</abbr> of machines with **Kubernetes**, Docker Swarm Mode, Nomad, or another similar complex system to manage distributed containers on multiple machines, then you will probably want to **handle replication** at the **cluster level** instead of using a **process manager** (like Uvicorn with workers) in each container.
If you have a <dfntitle="A group of machines that are configured to be connected and work together in some way.">cluster</dfn> of machines with **Kubernetes**, Docker Swarm Mode, Nomad, or another similar complex system to manage distributed containers on multiple machines, then you will probably want to **handle replication** at the **cluster level** instead of using a **process manager** (like Uvicorn with workers) in each container.
One of those distributed container management systems like Kubernetes normally has some integrated way of handling **replication of containers** while still supporting **load balancing** for the incoming requests. All at the **cluster level**.
@ -501,7 +499,7 @@ Of course, there are **special cases** where you could want to have **a containe
In those cases, you can use the `--workers` command line option to set the number of workers that you want to run:
@ -65,7 +65,7 @@ Here's an example of how an HTTPS API could look like, step by step, paying atte
It would probably all start by you **acquiring** some **domain name**. Then, you would configure it in a DNS server (possibly your same cloud provider).
You would probably get a cloud server (a virtual machine) or something similar, and it would have a <abbrtitle="That doesn't change">fixed</abbr>**public IP address**.
You would probably get a cloud server (a virtual machine) or something similar, and it would have a <dfntitle="Doesn't change over time. Not dynamic.">fixed</dfn>**public IP address**.
In the DNS server(s) you would configure a record (an "`A record`") to point **your domain** to the public **IP address of your server**.
@ -196,31 +196,11 @@ They have contributed source code, documentation, etc. 📦
There are hundreds of other contributors, you can see them all in the <ahref="https://github.com/fastapi/fastapi/graphs/contributors"class="external-link"target="_blank">FastAPI GitHub Contributors page</a>. 👷
## Top Translators
These are the **Top Translators**. 🌐
These users have created the most Pull Requests with [translations to other languages](contributing.md#translations){.internal-link target=_blank} that have been *merged*.
<divclass="user-list user-list-center">
{% for user in (translators.values() | list)[:50] %}
These users are the **Top Translation Reviewers**. 🕵️
I only speak a few languages (and not very well 😅). So, the reviewers are the ones that have the [**power to approve translations**](contributing.md#translations){.internal-link target=_blank} of the documentation. Without them, there wouldn't be documentation in several other languages.
Translation reviewers have the [**power to approve translations**](contributing.md#translations){.internal-link target=_blank} of the documentation. Without them, there wouldn't be documentation in several other languages.
<divclass="user-list user-list-center">
{% for user in (translation_reviewers.values() | list)[:50] %}
### Based on open standards { #based-on-open-standards }
* <ahref="https://github.com/OAI/OpenAPI-Specification"class="external-link"target="_blank"><strong>OpenAPI</strong></a> for API creation, including declarations of <abbrtitle="also known as: endpoints, routes">path</abbr><abbrtitle="also known as HTTP methods, as POST, GET, PUT, DELETE">operations</abbr>, parameters, request bodies, security, etc.
* <ahref="https://github.com/OAI/OpenAPI-Specification"class="external-link"target="_blank"><strong>OpenAPI</strong></a> for API creation, including declarations of <dfntitle="also known as: endpoints, routes">path</dfn><dfntitle="also known as HTTP methods, as POST, GET, PUT, DELETE">operations</dfn>, parameters, request bodies, security, etc.
* Automatic data model documentation with <ahref="https://json-schema.org/"class="external-link"target="_blank"><strong>JSON Schema</strong></a> (as OpenAPI itself is based on JSON Schema).
* Designed around these standards, after a meticulous study. Instead of an afterthought layer on top.
* This also allows using automatic **client code generation** in many languages.
@ -136,7 +136,7 @@ All built as reusable tools and components that are easy to integrate with your
FastAPI includes an extremely easy to use, but extremely powerful <abbrtitle='also known as "components", "resources", "services", "providers"'><strong>Dependency Injection</strong></abbr> system.
FastAPI includes an extremely easy to use, but extremely powerful <dfntitle='also known as "components", "resources", "services", "providers"'><strong>Dependency Injection</strong></dfn> system.
* Even dependencies can have dependencies, creating a hierarchy or **"graph" of dependencies**.
* All **automatically handled** by the framework.
@ -153,8 +153,8 @@ Any integration is designed to be so simple to use (with dependencies) that you
### Tested { #tested }
* 100% <abbrtitle="The amount of code that is automatically tested">test coverage</abbr>.
* 100% <abbrtitle="Python type annotations, with this your editor and external tools can give you better support">type annotated</abbr> code base.
* 100% <dfntitle="The amount of code that is automatically tested">test coverage</dfn>.
* 100% <dfntitle="Python type annotations, with this your editor and external tools can give you better support">type annotated</dfn> code base.
* Used in production applications.
## Starlette features { #starlette-features }
@ -190,7 +190,7 @@ With **FastAPI** you get all of **Pydantic**'s features (as FastAPI is based on
* **No brainfuck**:
* No new schema definition micro-language to learn.
* If you know Python types you know how to use Pydantic.
* Plays nicely with your **<abbrtitle="Integrated Development Environment: similar to a code editor">IDE</abbr>/<abbrtitle="A program that checks for code errors">linter</abbr>/brain**:
* Plays nicely with your **<abbrtitle="Integrated Development Environment: similar to a code editor">IDE</abbr>/<dfntitle="A program that checks for code errors">linter</dfn>/brain**:
* Because pydantic data structures are just instances of classes you define; auto-completion, linting, mypy and your intuition should all work properly with your validated data.
* Validate **complex structures**:
* Use of hierarchical Pydantic models, Python `typing`’s `List` and `Dict`, etc.
* **Fast**: Very high performance, on par with **NodeJS** and **Go** (thanks to Starlette and Pydantic). [One of the fastest Python frameworks available](#performance).
* **Fast to code**: Increase the speed to develop features by about 200% to 300%. *
* **Fewer bugs**: Reduce about 40% of human (developer) induced errors. *
* **Intuitive**: Great editor support. <abbrtitle="also known as auto-complete, autocompletion, IntelliSense">Completion</abbr> everywhere. Less time debugging.
* **Intuitive**: Great editor support. <dfntitle="also known as auto-complete, autocompletion, IntelliSense">Completion</dfn> everywhere. Less time debugging.
* **Easy**: Designed to be easy to use and learn. Less time reading docs.
* **Short**: Minimize code duplication. Multiple features from each parameter declaration. Fewer bugs.
* **Robust**: Get production-ready code. With automatic interactive documentation.
* Automatic and clear errors when the data is invalid.
* Validation even for deeply nested JSON objects.
* <abbrtitle="also known as: serialization, parsing, marshalling">Conversion</abbr> of input data: coming from the network to Python data and types. Reading from:
* <dfntitle="also known as: serialization, parsing, marshalling">Conversion</dfn> of input data: coming from the network to Python data and types. Reading from:
* JSON.
* Path parameters.
* Query parameters.
@ -382,7 +376,7 @@ item: Item
* Headers.
* Forms.
* Files.
* <abbrtitle="also known as: serialization, parsing, marshalling">Conversion</abbr> of output data: converting from Python data and types to network data (as JSON):
* <dfntitle="also known as: serialization, parsing, marshalling">Conversion</dfn> of output data: converting from Python data and types to network data (as JSON):
@ -445,7 +439,7 @@ For a more complete example including more features, see the <a href="https://fa
* Declaration of **parameters** from other different places as: **headers**, **cookies**, **form fields** and **files**.
* How to set **validation constraints** as `maximum_length` or `regex`.
* A very powerful and easy to use **<abbrtitle="also known as components, resources, providers, services, injectables">Dependency Injection</abbr>** system.
* A very powerful and easy to use **<dfntitle="also known as components, resources, providers, services, injectables">Dependency Injection</dfn>** system.
* Security and authentication, including support for **OAuth2** with **JWT tokens** and **HTTP Basic** auth.
* More advanced (but equally easy) techniques for declaring **deeply nested JSON models** (thanks to Pydantic).
* **GraphQL** integration with <ahref="https://strawberry.rocks"class="external-link"target="_blank">Strawberry</a> and other libraries.
@ -530,7 +524,7 @@ Used by Starlette:
* <ahref="https://www.python-httpx.org"target="_blank"><code>httpx</code></a> - Required if you want to use the `TestClient`.
* <ahref="https://jinja.palletsprojects.com"target="_blank"><code>jinja2</code></a> - Required if you want to use the default template configuration.
* <ahref="https://github.com/Kludex/python-multipart"target="_blank"><code>python-multipart</code></a> - Required if you want to support form <abbrtitle="converting the string that comes from an HTTP request into Python data">"parsing"</abbr>, with `request.form()`.
* <ahref="https://github.com/Kludex/python-multipart"target="_blank"><code>python-multipart</code></a> - Required if you want to support form <dfntitle="converting the string that comes from an HTTP request into Python data">"parsing"</dfn>, with `request.form()`.
@ -74,7 +74,7 @@ Make sure you use a supported label from the <a href="https://github.com/tiangol
* `refactor`: Refactors
* This is normally for changes to the internal code that don't change the behavior. Normally it improves maintainability, or enables future features, etc.
* `upgrade`: Upgrades
* This is for upgrades to direct dependencies from the project, or extra optional dependencies, normally in `pyproject.toml`. So, things that would affect final users, they would end up receiving the upgrade in their code base once they update. But this is not for upgrades to internal dependencies used for development, testing, docs, etc. Those internal dependencies, normally in `requirements.txt` files or GitHub Action versions should be marked as `internal`, not `upgrade`.
* This is for upgrades to direct dependencies from the project, or extra optional dependencies, normally in `pyproject.toml`. So, things that would affect final users, they would end up receiving the upgrade in their code base once they update. But this is not for upgrades to internal dependencies used for development, testing, docs, etc. Those internal dependencies or GitHub Action versions should be marked as `internal`, not `upgrade`.
* `docs`: Docs
* Changes in docs. This includes updating the docs, fixing typos. But it doesn't include changes to translations.
* You can normally quickly detect it by going to the "Files changed" tab in the PR and checking if the updated file(s) starts with `docs/en/docs`. The original version of the docs is always in English, so in `docs/en/docs`.
@ -106,135 +106,25 @@ This way, we can notice when there are new translations ready, because they have
## Merge Translation PRs
For Spanish, as I'm a native speaker and it's a language close to me, I will give it a final review myself and in most cases tweak the PR a bit before merging it.
Translations are generated automatically with LLMs and scripts.
For the other languages, confirm that:
There's one GitHub Action that can be manually run to add or update translations for a language: <ahref="https://github.com/fastapi/fastapi/actions/workflows/translate.yml"class="external-link"target="_blank">`translate.yml`</a>.
* The title is correct following the instructions above.
* It has the labels `lang-all` and `lang-{lang code}`.
* The PR changes only one Markdown file adding a translation.
* Or in some cases, at most two files, if they are small, for the same language, and people reviewed them.
* If it's the first translation for that language, it will have additional `mkdocs.yml` files, for those cases follow the instructions below.
* The PR doesn't add any additional or extraneous files.
* The translation seems to have a similar structure as the original English file.
* The translation doesn't seem to change the original content, for example with obvious additional documentation sections.
* The translation doesn't use different Markdown structures, for example adding HTML tags when the original didn't have them.
* The "admonition" sections, like `tip`, `info`, etc. are not changed or translated. For example:
```
/// tip
This is a tip.
///
```
looks like this:
/// tip
This is a tip.
///
...it could be translated as:
```
/// tip
Esto es un consejo.
///
```
...but needs to keep the exact `tip` keyword. If it was translated to `consejo`, like:
```
/// consejo
Esto es un consejo.
///
```
it would change the style to the default one, it would look like:
/// consejo
Esto es un consejo.
///
Those don't have to be translated, but if they are, they need to be written as:
For these language translation PRs, confirm that:
```
/// tip | consejo
Esto es un consejo.
///
```
Which looks like:
/// tip | consejo
Esto es un consejo.
///
## First Translation PR
When there's a first translation for a language, it will have a `docs/{lang code}/docs/index.md` translated file and a `docs/{lang code}/mkdocs.yml`.
For example, for Bosnian, it would be:
* `docs/bs/docs/index.md`
* `docs/bs/mkdocs.yml`
The `mkdocs.yml` file will have only the following content:
```YAML
INHERIT: ../en/mkdocs.yml
```
The language code would normally be in the <ahref="https://en.wikipedia.org/wiki/List_of_ISO_639_language_codes"class="external-link"target="_blank">ISO 639-1 list of language codes</a>.
In any case, the language code should be in the file <ahref="https://github.com/fastapi/fastapi/blob/master/docs/language_names.yml"class="external-link"target="_blank">docs/language_names.yml</a>.
There won't be yet a label for the language code, for example, if it was Bosnian, there wouldn't be a `lang-bs`. Before creating the label and adding it to the PR, create the GitHub Discussion:
* Go to the <ahref="https://github.com/fastapi/fastapi/discussions/categories/translations"class="external-link"target="_blank">Translations GitHub Discussions</a>
* Create a new discussion with the title `Bosnian Translations` (or the language name in English)
* A description of:
```Markdown
## Bosnian translations
This is the issue to track translations of the docs to Bosnian. 🚀
Here are the [PRs to review with the label `lang-bs`](https://github.com/fastapi/fastapi/pulls?q=is%3Apr+is%3Aopen+sort%3Aupdated-desc+label%3Alang-bs+label%3A%22awaiting-review%22). 🤓
```
Update "Bosnian" with the new language.
And update the search link to point to the new language label that will be created, like `lang-bs`.
Create and add the label to that new Discussion just created, like `lang-bs`.
* The PR was automated (authored by @tiangolo), not made by another user.
* It has the labels `lang-all` and `lang-{lang code}`.
* If the PR is approved by at least one native speaker, you can merge it.
Then go back to the PR, and add the label, like `lang-bs`, and `lang-all` and `awaiting-review`.
## Review PRs
Now the GitHub action will automatically detect the label `lang-bs` and will post in that Discussion that this PR is waiting to be reviewed.
* If a PR doesn't explain what it does or why, if it seems like it could be useful, ask for more information. Otherwise, feel free to close it.
## Review PRs
* If a PR seems to be spam, meaningless, only to change statistics (to appear as "contributor") or similar, you can simply mark it as `invalid`, and it will be automatically closed.
If a PR doesn't explain what it does or why, ask for more information.
* If a PR seems to be AI generated, and seems like reviewing it would take more time from you than the time it took to write the prompt, mark it as `maybe-ai`, and it will be automatically closed.
A PR should have a specific use case that it is solving.
* A PR should have a specific use case that it is solving.
* If the PR is for a feature, it should have docs.
* Unless it's a feature we want to discourage, like support for a corner case that we don't want users to use.
@ -254,27 +144,12 @@ Every month, a GitHub Action updates the FastAPI People data. Those PRs look lik
If the tests are passing, you can merge it right away.
## External Links PRs
When people add external links they edit this file <ahref="https://github.com/fastapi/fastapi/blob/master/docs/en/data/external_links.yml"class="external-link"target="_blank">external_links.yml</a>.
* Make sure the new link is in the correct category (e.g. "Podcasts") and language (e.g. "Japanese").
* A new link should be at the top of its list.
* The link URL should work (it should not return a 404).
* The content of the link should be about FastAPI.
* The new addition should have these fields:
* `author`: The name of the author.
* `link`: The URL with the content.
* `title`: The title of the link (the title of the article, podcast, etc).
After checking all these things and ensuring the PR has the right labels, you can merge it.
## Dependabot PRs
Dependabot will create PRs to update dependencies for several things, and those PRs all look similar, but some are way more delicate than others.
* If the PR is for a direct dependency, so, Dependabot is modifying `pyproject.toml`, **don't merge it**. 😱 Let me check it first. There's a good chance that some additional tweaks or updates are needed.
* If the PR updates one of the internal dependencies, for example it's modifying `requirements.txt` files, or GitHub Action versions, if the tests are passing, the release notes (shown in a summary in the PR) don't show any obvious potential breaking change, you can merge it. 😎
* If the PR is for a direct dependency, so, Dependabot is modifying `pyproject.toml` in the main dependencies, **don't merge it**. 😱 Let me check it first. There's a good chance that some additional tweaks or updates are needed.
* If the PR updates one of the internal dependencies, for example the group `dev` in `pyproject.toml`, or GitHub Action versions, if the tests are passing, the release notes (shown in a summary in the PR) don't show any obvious potential breaking change, you can merge it. 😎
Python has support for optional "type hints" (also called "type annotations").
These **"type hints"** or annotations are a special syntax that allow declaring the <abbrtitle="for example: str, int, float, bool">type</abbr> of a variable.
These **"type hints"** or annotations are a special syntax that allow declaring the <dfntitle="for example: str, int, float, bool">type</dfn> of a variable.
By declaring types for your variables, editors and tools can give you better support.
@ -22,7 +22,7 @@ If you are a Python expert, and you already know everything about type hints, sk
### Generic types with type parameters { #generic-types-with-type-parameters }
### `typing` module { #typing-module }
There are some data structures that can contain other values, like `dict`, `list`, `set` and `tuple`. And the internal values can have their own type too.
For some additional use cases, you might need to import some things from the standard library `typing` module, for example when you want to declare that something has "any type", you can use `Any` from `typing`:
These types that have internal types are called "**generic**" types. And it's possible to declare them, even with their internal types.
```python
from typing import Any
To declare those types and the internal types, you can use the standard Python module `typing`. It exists specifically to support these type hints.
#### Newer versions of Python { #newer-versions-of-python }
The syntax using `typing` is **compatible** with all versions, from Python 3.6 to the latest ones, including Python 3.9, Python 3.10, etc.
def some_function(data: Any):
print(data)
```
As Python advances, **newer versions** come with improved support for these type annotations and in many cases you won't even need to import and use the `typing` module to declare the type annotations.
### Generic types { #generic-types }
If you can choose a more recent version of Python for your project, you will be able to take advantage of that extra simplicity.
Some types can take "type parameters" in square brackets, to define their internal types, for example a "list of strings" would be declared `list[str]`.
In all the docs there are examples compatible with each version of Python (when there's a difference).
These types that can take type parameters are called **Generic types** or **Generics**.
For example "**Python 3.6+**" means it's compatible with Python 3.6 or above (including 3.7, 3.8, 3.9, 3.10, etc). And "**Python 3.9+**" means it's compatible with Python 3.9 or above (including 3.10, etc).
You can use the same builtin types as generics (with square brackets and types inside):
If you can use the **latest versions of Python**, use the examples for the latest version, those will have the **best and simplest syntax**, for example, "**Python 3.10+**".
* `list`
* `tuple`
* `set`
* `dict`
#### List { #list }
@ -167,7 +170,7 @@ As the type, put `list`.
As the list is a type that contains some internal types, you put them in square brackets:
You can declare that a variable can be any of **several types**, for example, an `int` or a `str`.
In Python 3.6 and above (including Python 3.10) you can use the `Union` type from `typing` and put inside the square brackets the possible types to accept.
In Python 3.10 there's also a **new syntax** where you can put the possible types separated by a <abbrtitle='also called "bitwise or operator", but that meaning is not relevant here'>vertical bar (`|`)</abbr>.
To define it you use the <dfntitle='also called "bitwise or operator", but that meaning is not relevant here'>vertical bar (`|`)</dfn> to separate both types.
//// tab | Python 3.10+
This is called a "union", because the variable can be anything in the union of those two sets of types.
Using `Optional[str]` instead of just `str` will let the editor help you detect errors where you could be assuming that a value is always a `str`, when it could actually be `None` too.
`Optional[Something]` is actually a shortcut for `Union[Something, None]`, they are equivalent.
This also means that in Python 3.10, you can use `Something | None`:
//// tab | Python 3.10+
```Python hl_lines="1"
@ -266,96 +245,7 @@ This also means that in Python 3.10, you can use `Something | None`:
#### Using `Union` or `Optional` { #using-union-or-optional }
If you are using a Python version below 3.10, here's a tip from my very **subjective** point of view:
* 🚨 Avoid using `Optional[SomeType]`
* Instead ✨ **use `Union[SomeType, None]`** ✨.
Both are equivalent and underneath they are the same, but I would recommend `Union` instead of `Optional` because the word "**optional**" would seem to imply that the value is optional, and it actually means "it can be `None`", even if it's not optional and is still required.
I think `Union[SomeType, None]` is more explicit about what it means.
It's just about the words and names. But those words can affect how you and your teammates think about the code.
And then you won't have to worry about names like `Optional` and `Union`. 😎
#### Generic types { #generic-types }
These types that take type parameters in square brackets are called **Generic types** or **Generics**, for example:
//// tab | Python 3.10+
You can use the same builtin types as generics (with square brackets and types inside):
* `list`
* `tuple`
* `set`
* `dict`
And the same as with previous Python versions, from the `typing` module:
* `Union`
* `Optional`
* ...and others.
In Python 3.10, as an alternative to using the generics `Union` and `Optional`, you can use the <abbrtitle='also called "bitwise or operator", but that meaning is not relevant here'>vertical bar (`|`)</abbr> to declare unions of types, that's a lot better and simpler.
////
//// tab | Python 3.9+
You can use the same builtin types as generics (with square brackets and types inside):
* `list`
* `tuple`
* `set`
* `dict`
And generics from the `typing` module:
* `Union`
* `Optional`
* ...and others.
////
Using `str | None` instead of just `str` will let the editor help you detect errors where you could be assuming that a value is always a `str`, when it could actually be `None` too.
### Classes as types { #classes-as-types }
@ -363,11 +253,11 @@ You can also declare a class as the type of a variable.
@ -403,19 +293,13 @@ To learn more about <a href="https://docs.pydantic.dev/" class="external-link" t
You will see a lot more of all this in practice in the [Tutorial - User Guide](tutorial/index.md){.internal-link target=_blank}.
/// tip
Pydantic has a special behavior when you use `Optional` or `Union[Something, None]` without a default value, you can read more about it in the Pydantic docs about <ahref="https://docs.pydantic.dev/2.3/usage/models/#required-fields"class="external-link"target="_blank">Required Optional fields</a>.
///
## Type Hints with Metadata Annotations { #type-hints-with-metadata-annotations }
Python also has a feature that allows putting **additional <abbr title="Data about the data, in this case, information about the type, e.g. a description.">metadata</abbr>** in these type hints using `Annotated`.
Python also has a feature that allows putting **additional <dfn title="Data about the data, in this case, information about the type, e.g. a description.">metadata</dfn>** in these type hints using `Annotated`.
Since Python 3.9, `Annotated` is a part of the standard library, so you can import it from `typing`.
You can declare a parameter in a *path operation function* or dependency to be of type `Request` and then you can access the raw request object directly, without any validation, etc.
Read more about it in the [FastAPI docs about using Request directly](https://fastapi.tiangolo.com/advanced/using-request-directly/)
@ -4,6 +4,8 @@ You can declare a parameter in a *path operation function* or dependency to be o
You can also use it directly to create an instance of it and return it from your *path operations*.
Read more about it in the [FastAPI docs about returning a custom Response](https://fastapi.tiangolo.com/advanced/response-directly/#returning-a-custom-response)
@ -56,6 +56,8 @@ There are a couple of custom FastAPI response classes, you can use them to optim
## Starlette Responses
You can read more about all of them in the [FastAPI docs for Custom Response](https://fastapi.tiangolo.com/advanced/custom-response/) and in the [Starlette docs about Responses](https://starlette.dev/responses/).
When defining WebSockets, you normally declare a parameter of type `WebSocket` and with it you can read data from the client and send data to it.
Read more about it in the [FastAPI docs for WebSockets](https://fastapi.tiangolo.com/advanced/websockets/)
It is provided directly by Starlette, but you can import it from `fastapi`:
```python
@ -44,16 +46,6 @@ When you want to define dependencies that should be compatible with both HTTP an
- send_json
- close
When a client disconnects, a `WebSocketDisconnect` exception is raised, you can catch it.
You can import it directly form `fastapi`:
```python
from fastapi import WebSocketDisconnect
```
::: fastapi.WebSocketDisconnect
## WebSockets - additional classes
Additional classes for handling WebSockets.
@ -66,4 +58,16 @@ from fastapi.websockets import WebSocketDisconnect, WebSocketState
::: fastapi.websockets.WebSocketDisconnect
When a client disconnects, a `WebSocketDisconnect` exception is raised, you can catch it.
You can import it directly form `fastapi`:
```python
from fastapi import WebSocketDisconnect
```
Read more about it in the [FastAPI docs for WebSockets](https://fastapi.tiangolo.com/advanced/websockets/#handling-disconnections-and-multiple-clients)
::: fastapi.websockets.WebSocketState
`WebSocketState` is an enumeration of the possible states of a WebSocket connection.
* ➖ Drop support for Python 3.9. PR [#14897](https://github.com/fastapi/fastapi/pull/14897) by [@tiangolo](https://github.com/tiangolo).
### Refactors
* 🎨 Update internal types for Python 3.10. PR [#14898](https://github.com/fastapi/fastapi/pull/14898) by [@tiangolo](https://github.com/tiangolo).
### Docs
* 📝 Update highlights in webhooks docs. PR [#14905](https://github.com/fastapi/fastapi/pull/14905) by [@tiangolo](https://github.com/tiangolo).
* 📝 Update source examples and docs from Python 3.9 to 3.10. PR [#14900](https://github.com/fastapi/fastapi/pull/14900) by [@tiangolo](https://github.com/tiangolo).
### Internal
* 🔨 Update docs.py scripts to migrate Python 3.9 to Python 3.10. PR [#14906](https://github.com/fastapi/fastapi/pull/14906) by [@tiangolo](https://github.com/tiangolo).
## 0.128.8
### Docs
* 📝 Fix grammar in `docs/en/docs/tutorial/first-steps.md`. PR [#14708](https://github.com/fastapi/fastapi/pull/14708) by [@SanjanaS10](https://github.com/SanjanaS10).
### Internal
* 🔨 Tweak PDM hook script. PR [#14895](https://github.com/fastapi/fastapi/pull/14895) by [@tiangolo](https://github.com/tiangolo).
* ♻️ Update build setup for `fastapi-slim`, deprecate it, and make it only depend on `fastapi`. PR [#14894](https://github.com/fastapi/fastapi/pull/14894) by [@tiangolo](https://github.com/tiangolo).
## 0.128.7
### Features
* ✨ Show a clear error on attempt to include router into itself. PR [#14258](https://github.com/fastapi/fastapi/pull/14258) by [@JavierSanchezCastro](https://github.com/JavierSanchezCastro).
* ✨ Replace `dict` by `Mapping` on `HTTPException.headers`. PR [#12997](https://github.com/fastapi/fastapi/pull/12997) by [@rijenkii](https://github.com/rijenkii).
### Refactors
* ♻️ Simplify reading files in memory, do it sequentially instead of (fake) parallel. PR [#14884](https://github.com/fastapi/fastapi/pull/14884) by [@tiangolo](https://github.com/tiangolo).
### Docs
* 📝 Use `dfn` tag for definitions instead of `abbr` in docs. PR [#14744](https://github.com/fastapi/fastapi/pull/14744) by [@YuriiMotov](https://github.com/YuriiMotov).
### Internal
* ✅ Tweak comment in test to reference PR. PR [#14885](https://github.com/fastapi/fastapi/pull/14885) by [@tiangolo](https://github.com/tiangolo).
* 🔧 Update LLM-prompt for `abbr` and `dfn` tags. PR [#14747](https://github.com/fastapi/fastapi/pull/14747) by [@YuriiMotov](https://github.com/YuriiMotov).
* ✅ Test order for the submitted byte Files. PR [#14828](https://github.com/fastapi/fastapi/pull/14828) by [@valentinDruzhinin](https://github.com/valentinDruzhinin).
* 🔧 Configure `test` workflow to run tests with `inline-snapshot=review`. PR [#14876](https://github.com/fastapi/fastapi/pull/14876) by [@YuriiMotov](https://github.com/YuriiMotov).
## 0.128.6
### Fixes
* 🐛 Fix `on_startup` and `on_shutdown` parameters of `APIRouter`. PR [#14873](https://github.com/fastapi/fastapi/pull/14873) by [@YuriiMotov](https://github.com/YuriiMotov).
### Translations
* 🌐 Update translations for zh (update-outdated). PR [#14843](https://github.com/fastapi/fastapi/pull/14843) by [@tiangolo](https://github.com/tiangolo).
### Internal
* ✅ Fix parameterized tests with snapshots. PR [#14875](https://github.com/fastapi/fastapi/pull/14875) by [@YuriiMotov](https://github.com/YuriiMotov).
## 0.128.5
### Refactors
* ♻️ Refactor and simplify Pydantic v2 (and v1) compatibility internal utils. PR [#14862](https://github.com/fastapi/fastapi/pull/14862) by [@tiangolo](https://github.com/tiangolo).
### Internal
* ✅ Add inline snapshot tests for OpenAPI before changes from Pydantic v2. PR [#14864](https://github.com/fastapi/fastapi/pull/14864) by [@tiangolo](https://github.com/tiangolo).
## 0.128.4
### Refactors
* ♻️ Refactor internals, simplify Pydantic v2/v1 utils, `create_model_field`, better types for `lenient_issubclass`. PR [#14860](https://github.com/fastapi/fastapi/pull/14860) by [@tiangolo](https://github.com/tiangolo).
* ♻️ Simplify internals, remove Pydantic v1 only logic, no longer needed. PR [#14857](https://github.com/fastapi/fastapi/pull/14857) by [@tiangolo](https://github.com/tiangolo).
* ♻️ Refactor internals, cleanup unneeded Pydantic v1 specific logic. PR [#14856](https://github.com/fastapi/fastapi/pull/14856) by [@tiangolo](https://github.com/tiangolo).
### Translations
* 🌐 Update translations for fr (outdated pages). PR [#14839](https://github.com/fastapi/fastapi/pull/14839) by [@YuriiMotov](https://github.com/YuriiMotov).
* 🌐 Update translations for tr (outdated and missing). PR [#14838](https://github.com/fastapi/fastapi/pull/14838) by [@YuriiMotov](https://github.com/YuriiMotov).
### Internal
* ⬆️ Upgrade development dependencies. PR [#14854](https://github.com/fastapi/fastapi/pull/14854) by [@tiangolo](https://github.com/tiangolo).
## 0.128.3
### Refactors
* ♻️ Re-implement `on_event` in FastAPI for compatibility with the next Starlette, while keeping backwards compatibility. PR [#14851](https://github.com/fastapi/fastapi/pull/14851) by [@tiangolo](https://github.com/tiangolo).
### Upgrades
* ⬆️ Upgrade Starlette supported version range to `starlette>=0.40.0,<1.0.0`. PR [#14853](https://github.com/fastapi/fastapi/pull/14853) by [@tiangolo](https://github.com/tiangolo).
### Translations
* 🌐 Update translations for ru (update-outdated). PR [#14834](https://github.com/fastapi/fastapi/pull/14834) by [@tiangolo](https://github.com/tiangolo).
### Internal
* 👷 Run tests with Starlette from git. PR [#14849](https://github.com/fastapi/fastapi/pull/14849) by [@tiangolo](https://github.com/tiangolo).
* 👷 Run tests with lower bound uv sync, upgrade `fastapi[all]` minimum dependencies: `ujson >=5.8.0`, `orjson >=3.9.3`. PR [#14846](https://github.com/fastapi/fastapi/pull/14846) by [@tiangolo](https://github.com/tiangolo).
## 0.128.2
### Features
* ✨ Add support for PEP695 `TypeAliasType`. PR [#13920](https://github.com/fastapi/fastapi/pull/13920) by [@cstruct](https://github.com/cstruct).
* ✨ Allow `Response` type hint as dependency annotation. PR [#14794](https://github.com/fastapi/fastapi/pull/14794) by [@jonathan-fulton](https://github.com/jonathan-fulton).
### Fixes
* 🐛 Fix using `Json[list[str]]` type (issue #10997). PR [#14616](https://github.com/fastapi/fastapi/pull/14616) by [@mkanetsuna](https://github.com/mkanetsuna).
### Docs
* 📝 Update docs for translations. PR [#14830](https://github.com/fastapi/fastapi/pull/14830) by [@tiangolo](https://github.com/tiangolo).
* 📝 Fix duplicate word in `advanced-dependencies.md`. PR [#14815](https://github.com/fastapi/fastapi/pull/14815) by [@Rayyan-Oumlil](https://github.com/Rayyan-Oumlil).
### Translations
* 🌐 Enable Traditional Chinese translations. PR [#14842](https://github.com/fastapi/fastapi/pull/14842) by [@tiangolo](https://github.com/tiangolo).
* 🌐 Enable French docs translations. PR [#14841](https://github.com/fastapi/fastapi/pull/14841) by [@tiangolo](https://github.com/tiangolo).
* 🌐 Update translations for fr (translate-page). PR [#14837](https://github.com/fastapi/fastapi/pull/14837) by [@tiangolo](https://github.com/tiangolo).
* 🌐 Update translations for de (update-outdated). PR [#14836](https://github.com/fastapi/fastapi/pull/14836) by [@tiangolo](https://github.com/tiangolo).
* 🌐 Update translations for pt (update-outdated). PR [#14833](https://github.com/fastapi/fastapi/pull/14833) by [@tiangolo](https://github.com/tiangolo).
* 🌐 Update translations for ko (update-outdated). PR [#14835](https://github.com/fastapi/fastapi/pull/14835) by [@tiangolo](https://github.com/tiangolo).
* 🌐 Update translations for es (update-outdated). PR [#14832](https://github.com/fastapi/fastapi/pull/14832) by [@tiangolo](https://github.com/tiangolo).
* 🌐 Update translations for tr (update-outdated). PR [#14831](https://github.com/fastapi/fastapi/pull/14831) by [@tiangolo](https://github.com/tiangolo).
* 🌐 Update translations for tr (add-missing). PR [#14790](https://github.com/fastapi/fastapi/pull/14790) by [@tiangolo](https://github.com/tiangolo).
* 🌐 Update translations for fr (update-outdated). PR [#14826](https://github.com/fastapi/fastapi/pull/14826) by [@tiangolo](https://github.com/tiangolo).
* 🌐 Update translations for zh-hant (update-outdated). PR [#14825](https://github.com/fastapi/fastapi/pull/14825) by [@tiangolo](https://github.com/tiangolo).
* 🌐 Update translations for uk (update-outdated). PR [#14822](https://github.com/fastapi/fastapi/pull/14822) by [@tiangolo](https://github.com/tiangolo).
* 🔨 Update docs and translations scripts, enable Turkish. PR [#14824](https://github.com/fastapi/fastapi/pull/14824) by [@tiangolo](https://github.com/tiangolo).
### Internal
* 🔨 Add max pages to translate to configs. PR [#14840](https://github.com/fastapi/fastapi/pull/14840) by [@tiangolo](https://github.com/tiangolo).
## 0.128.1
### Features
* ✨ Add `viewport` meta tag to improve Swagger UI on mobile devices. PR [#14777](https://github.com/fastapi/fastapi/pull/14777) by [@Joab0](https://github.com/Joab0).
* 🚸 Improve error message for invalid query parameter type annotations. PR [#14479](https://github.com/fastapi/fastapi/pull/14479) by [@retwish](https://github.com/retwish).
### Fixes
* 🐛 Update `ValidationError` schema to include `input` and `ctx`. PR [#14791](https://github.com/fastapi/fastapi/pull/14791) by [@jonathan-fulton](https://github.com/jonathan-fulton).
* 🐛 Fix TYPE_CHECKING annotations for Python 3.14 (PEP 649). PR [#14789](https://github.com/fastapi/fastapi/pull/14789) by [@mgu](https://github.com/mgu).
* 🐛 Strip whitespaces from `Authorization` header credentials. PR [#14786](https://github.com/fastapi/fastapi/pull/14786) by [@WaveTheory1](https://github.com/WaveTheory1).
* 🐛 Fix OpenAPI duplication of `anyOf` refs for app-level responses with specified `content` and `model` as `Union`. PR [#14463](https://github.com/fastapi/fastapi/pull/14463) by [@DJMcoder](https://github.com/DJMcoder).
### Refactors
* 🎨 Tweak types for mypy. PR [#14816](https://github.com/fastapi/fastapi/pull/14816) by [@tiangolo](https://github.com/tiangolo).
* 🏷️ Re-export `IncEx` type from Pydantic instead of duplicating it. PR [#14641](https://github.com/fastapi/fastapi/pull/14641) by [@mvanderlee](https://github.com/mvanderlee).
* 💡 Update comment for Pydantic internals. PR [#14814](https://github.com/fastapi/fastapi/pull/14814) by [@tiangolo](https://github.com/tiangolo).
### Docs
* 📝 Update docs for contributing translations, simplify title. PR [#14817](https://github.com/fastapi/fastapi/pull/14817) by [@tiangolo](https://github.com/tiangolo).
* 📝 Fix typing issue in `docs_src/app_testing/app_b` code example. PR [#14573](https://github.com/fastapi/fastapi/pull/14573) by [@timakaa](https://github.com/timakaa).
* 📝 Fix example of license identifier in documentation. PR [#14492](https://github.com/fastapi/fastapi/pull/14492) by [@johnson-earls](https://github.com/johnson-earls).
* 📝 Add banner to translated pages. PR [#14809](https://github.com/fastapi/fastapi/pull/14809) by [@YuriiMotov](https://github.com/YuriiMotov).
* 📝 Add links to related sections of docs to docstrings. PR [#14776](https://github.com/fastapi/fastapi/pull/14776) by [@YuriiMotov](https://github.com/YuriiMotov).
* 📝 Update embedded code examples to Python 3.10 syntax. PR [#14758](https://github.com/fastapi/fastapi/pull/14758) by [@YuriiMotov](https://github.com/YuriiMotov).
* 📝 Fix dependency installation command in `docs/en/docs/contributing.md`. PR [#14757](https://github.com/fastapi/fastapi/pull/14757) by [@YuriiMotov](https://github.com/YuriiMotov).
* 📝 Use return type annotation instead of `response_model` when possible. PR [#14753](https://github.com/fastapi/fastapi/pull/14753) by [@YuriiMotov](https://github.com/YuriiMotov).
* 📝 Use `WSGIMiddleware` from `a2wsgi` instead of deprecated `fastapi.middleware.wsgi.WSGIMiddleware`. PR [#14756](https://github.com/fastapi/fastapi/pull/14756) by [@YuriiMotov](https://github.com/YuriiMotov).
* 📝 Fix minor typos in release notes. PR [#14780](https://github.com/fastapi/fastapi/pull/14780) by [@whyvineet](https://github.com/whyvineet).
* 🐛 Fix copy button in custom.js. PR [#14722](https://github.com/fastapi/fastapi/pull/14722) by [@fcharrier](https://github.com/fcharrier).
* 📝 Add contribution instructions about LLM generated code and comments and automated tools for PRs. PR [#14706](https://github.com/fastapi/fastapi/pull/14706) by [@tiangolo](https://github.com/tiangolo).
* 📝 Update docs for management tasks. PR [#14705](https://github.com/fastapi/fastapi/pull/14705) by [@tiangolo](https://github.com/tiangolo).
* 📝 Update docs about managing translations. PR [#14704](https://github.com/fastapi/fastapi/pull/14704) by [@tiangolo](https://github.com/tiangolo).
* 📝 Update docs for contributing with translations. PR [#14701](https://github.com/fastapi/fastapi/pull/14701) by [@tiangolo](https://github.com/tiangolo).
* 📝 Specify language code for code block. PR [#14656](https://github.com/fastapi/fastapi/pull/14656) by [@YuriiMotov](https://github.com/YuriiMotov).
### Translations
* 🌐 Improve LLM prompt of `uk` documentation. PR [#14795](https://github.com/fastapi/fastapi/pull/14795) by [@roli2py](https://github.com/roli2py).
* 🌐 Update translations for ja (update-outdated). PR [#14588](https://github.com/fastapi/fastapi/pull/14588) by [@tiangolo](https://github.com/tiangolo).
* 🌐 Update translations for uk (update outdated, found by fixer tool). PR [#14739](https://github.com/fastapi/fastapi/pull/14739) by [@YuriiMotov](https://github.com/YuriiMotov).
* 🌐 Update translations for tr (update-outdated). PR [#14745](https://github.com/fastapi/fastapi/pull/14745) by [@tiangolo](https://github.com/tiangolo).
* 🌐 Update `llm-prompt.md` for Korean language. PR [#14763](https://github.com/fastapi/fastapi/pull/14763) by [@seuthootDev](https://github.com/seuthootDev).
* 🌐 Update translations for ko (update outdated, found by fixer tool). PR [#14738](https://github.com/fastapi/fastapi/pull/14738) by [@YuriiMotov](https://github.com/YuriiMotov).
* 🌐 Update translations for de (update-outdated). PR [#14690](https://github.com/fastapi/fastapi/pull/14690) by [@tiangolo](https://github.com/tiangolo).
* 🌐 Update LLM prompt for Russian translations. PR [#14733](https://github.com/fastapi/fastapi/pull/14733) by [@YuriiMotov](https://github.com/YuriiMotov).
* 🌐 Update translations for ru (update-outdated). PR [#14693](https://github.com/fastapi/fastapi/pull/14693) by [@tiangolo](https://github.com/tiangolo).
* 🌐 Update translations for pt (update-outdated). PR [#14724](https://github.com/fastapi/fastapi/pull/14724) by [@tiangolo](https://github.com/tiangolo).
* 🌐 Update Korean LLM prompt. PR [#14740](https://github.com/fastapi/fastapi/pull/14740) by [@hard-coders](https://github.com/hard-coders).
* 🌐 Improve LLM prompt for Turkish translations. PR [#14728](https://github.com/fastapi/fastapi/pull/14728) by [@Kadermiyanyedi](https://github.com/Kadermiyanyedi).
* 🌐 Update portuguese llm-prompt.md. PR [#14702](https://github.com/fastapi/fastapi/pull/14702) by [@ceb10n](https://github.com/ceb10n).
* 🌐 Update LLM prompt instructions file for French. PR [#14618](https://github.com/fastapi/fastapi/pull/14618) by [@tiangolo](https://github.com/tiangolo).
* 🌐 Update translations for ko (add-missing). PR [#14699](https://github.com/fastapi/fastapi/pull/14699) by [@tiangolo](https://github.com/tiangolo).
* 🌐 Update translations for ko (update-outdated). PR [#14589](https://github.com/fastapi/fastapi/pull/14589) by [@tiangolo](https://github.com/tiangolo).
* 🌐 Update translations for uk (update-outdated). PR [#14587](https://github.com/fastapi/fastapi/pull/14587) by [@tiangolo](https://github.com/tiangolo).
* 🌐 Update translations for es (update-outdated). PR [#14686](https://github.com/fastapi/fastapi/pull/14686) by [@tiangolo](https://github.com/tiangolo).
* 🔧 Add LLM prompt file for Turkish, generated from the existing translations. PR [#14547](https://github.com/fastapi/fastapi/pull/14547) by [@tiangolo](https://github.com/tiangolo).
* 🔧 Add LLM prompt file for Traditional Chinese, generated from the existing translations. PR [#14550](https://github.com/fastapi/fastapi/pull/14550) by [@tiangolo](https://github.com/tiangolo).
* 🔧 Add LLM prompt file for Simplified Chinese, generated from the existing translations. PR [#14549](https://github.com/fastapi/fastapi/pull/14549) by [@tiangolo](https://github.com/tiangolo).
### Internal
* ⬇️ Downgrade LLM translations model to GPT-5 to reduce mistakes. PR [#14823](https://github.com/fastapi/fastapi/pull/14823) by [@tiangolo](https://github.com/tiangolo).
* 🐛 Fix translation script commit in place. PR [#14818](https://github.com/fastapi/fastapi/pull/14818) by [@tiangolo](https://github.com/tiangolo).
* 🔨 Update translation script to retry if LLM-response doesn't pass validation with Translation Fixer tool. PR [#14749](https://github.com/fastapi/fastapi/pull/14749) by [@YuriiMotov](https://github.com/YuriiMotov).
* 👷 Run tests only on relevant code changes (not on docs). PR [#14813](https://github.com/fastapi/fastapi/pull/14813) by [@tiangolo](https://github.com/tiangolo).
* 👷 Run mypy by pre-commit. PR [#14806](https://github.com/fastapi/fastapi/pull/14806) by [@YuriiMotov](https://github.com/YuriiMotov).
* ⬆ Bump ruff from 0.14.3 to 0.14.14. PR [#14798](https://github.com/fastapi/fastapi/pull/14798) by [@dependabot[bot]](https://github.com/apps/dependabot).
* ⬆ Bump pyasn1 from 0.6.1 to 0.6.2. PR [#14804](https://github.com/fastapi/fastapi/pull/14804) by [@dependabot[bot]](https://github.com/apps/dependabot).
* ⬆ Bump sqlmodel from 0.0.27 to 0.0.31. PR [#14802](https://github.com/fastapi/fastapi/pull/14802) by [@dependabot[bot]](https://github.com/apps/dependabot).
* ⬆ Bump mkdocs-macros-plugin from 1.4.1 to 1.5.0. PR [#14801](https://github.com/fastapi/fastapi/pull/14801) by [@dependabot[bot]](https://github.com/apps/dependabot).
* ⬆ Bump gitpython from 3.1.45 to 3.1.46. PR [#14800](https://github.com/fastapi/fastapi/pull/14800) by [@dependabot[bot]](https://github.com/apps/dependabot).
* ⬆ Bump typer from 0.16.0 to 0.21.1. PR [#14799](https://github.com/fastapi/fastapi/pull/14799) by [@dependabot[bot]](https://github.com/apps/dependabot).
* 👥 Update FastAPI People - Contributors and Translators. PR [#14796](https://github.com/fastapi/fastapi/pull/14796) by [@tiangolo](https://github.com/tiangolo).
* 🔧 Ensure that an edit to `uv.lock` gets the `internal` label. PR [#14759](https://github.com/fastapi/fastapi/pull/14759) by [@svlandeg](https://github.com/svlandeg).
* 🔧 Update sponsors: remove Requestly. PR [#14735](https://github.com/fastapi/fastapi/pull/14735) by [@tiangolo](https://github.com/tiangolo).
* 🔧 Update sponsors, LambdaTest changes to TestMu AI. PR [#14734](https://github.com/fastapi/fastapi/pull/14734) by [@tiangolo](https://github.com/tiangolo).
* ⬆ Bump actions/cache from 4 to 5. PR [#14511](https://github.com/fastapi/fastapi/pull/14511) by [@dependabot[bot]](https://github.com/apps/dependabot).
* ⬆ Bump actions/upload-artifact from 5 to 6. PR [#14525](https://github.com/fastapi/fastapi/pull/14525) by [@dependabot[bot]](https://github.com/apps/dependabot).
* ⬆ Bump actions/download-artifact from 6 to 7. PR [#14526](https://github.com/fastapi/fastapi/pull/14526) by [@dependabot[bot]](https://github.com/apps/dependabot).
* 👷 Tweak CI input names. PR [#14688](https://github.com/fastapi/fastapi/pull/14688) by [@tiangolo](https://github.com/tiangolo).
* 🔨 Refactor translation script to allow committing in place. PR [#14687](https://github.com/fastapi/fastapi/pull/14687) by [@tiangolo](https://github.com/tiangolo).
* 🐛 Fix translation script path. PR [#14685](https://github.com/fastapi/fastapi/pull/14685) by [@tiangolo](https://github.com/tiangolo).
* ✅ Enable tests in CI for scripts. PR [#14684](https://github.com/fastapi/fastapi/pull/14684) by [@tiangolo](https://github.com/tiangolo).
* 🔧 Add pre-commit local script to fix language translations. PR [#14683](https://github.com/fastapi/fastapi/pull/14683) by [@tiangolo](https://github.com/tiangolo).
* ⬆️ Migrate to uv. PR [#14676](https://github.com/fastapi/fastapi/pull/14676) by [@DoctorJohn](https://github.com/DoctorJohn).
* 👥 Update FastAPI People - Contributors and Translators. PR [#14625](https://github.com/fastapi/fastapi/pull/14625) by [@tiangolo](https://github.com/tiangolo).
@ -247,7 +478,7 @@ hide:
### Refactors
* 🔥 Remove dangling extra condiitonal no longer needed. PR [#14435](https://github.com/fastapi/fastapi/pull/14435) by [@tiangolo](https://github.com/tiangolo).
* 🔥 Remove dangling extra conditional no longer needed. PR [#14435](https://github.com/fastapi/fastapi/pull/14435) by [@tiangolo](https://github.com/tiangolo).
* ♻️ Refactor internals, update `is_coroutine` check to reuse internal supported variants (unwrap, check class). PR [#14434](https://github.com/fastapi/fastapi/pull/14434) by [@tiangolo](https://github.com/tiangolo).
### Translations
@ -382,7 +613,7 @@ hide:
### Docs
* 📝 Upate docs for advanced dependencies with `yield`, noting the changes in 0.121.0, adding `scope`. PR [#14287](https://github.com/fastapi/fastapi/pull/14287) by [@tiangolo](https://github.com/tiangolo).
* 📝 Update docs for advanced dependencies with `yield`, noting the changes in 0.121.0, adding `scope`. PR [#14287](https://github.com/fastapi/fastapi/pull/14287) by [@tiangolo](https://github.com/tiangolo).
### Internal
@ -2610,7 +2841,7 @@ Read more in the [advisory: Content-Type Header ReDoS](https://github.com/tiango
* 🌐 Add Japanese translation for `docs/ja/docs/tutorial/handling-errors.md`. PR [#1953](https://github.com/tiangolo/fastapi/pull/1953) by [@SwftAlpc](https://github.com/SwftAlpc).
* 🌐 Add Japanese translation for `docs/ja/docs/tutorial/response-status-code.md`. PR [#1942](https://github.com/tiangolo/fastapi/pull/1942) by [@SwftAlpc](https://github.com/SwftAlpc).
* 🌐 Add Japanese translation for `docs/ja/docs/tutorial/extra-models.md`. PR [#1941](https://github.com/tiangolo/fastapi/pull/1941) by [@SwftAlpc](https://github.com/SwftAlpc).
* 🌐 Add Japanese tranlsation for `docs/ja/docs/tutorial/schema-extra-example.md`. PR [#1931](https://github.com/tiangolo/fastapi/pull/1931) by [@SwftAlpc](https://github.com/SwftAlpc).
* 🌐 Add Japanese translation for `docs/ja/docs/tutorial/schema-extra-example.md`. PR [#1931](https://github.com/tiangolo/fastapi/pull/1931) by [@SwftAlpc](https://github.com/SwftAlpc).
* 🌐 Add Japanese translation for `docs/ja/docs/tutorial/body-nested-models.md`. PR [#1930](https://github.com/tiangolo/fastapi/pull/1930) by [@SwftAlpc](https://github.com/SwftAlpc).
* 🌐 Add Japanese translation for `docs/ja/docs/tutorial/body-fields.md`. PR [#1923](https://github.com/tiangolo/fastapi/pull/1923) by [@SwftAlpc](https://github.com/SwftAlpc).
* 🌐 Add German translation for `docs/de/docs/tutorial/index.md`. PR [#9502](https://github.com/tiangolo/fastapi/pull/9502) by [@fhabers21](https://github.com/fhabers21).
@ -3986,7 +4217,7 @@ You hopefully updated to a supported version of Python a while ago. If you haven
### Fixes
* 🐛 Fix `RuntimeError` raised when `HTTPException` has a status code with no content. PR [#5365](https://github.com/tiangolo/fastapi/pull/5365) by [@iudeen](https://github.com/iudeen).
* 🐛 Fix empty reponse body when default `status_code` is empty but the a `Response` parameter with `response.status_code` is set. PR [#5360](https://github.com/tiangolo/fastapi/pull/5360) by [@tmeckel](https://github.com/tmeckel).
* 🐛 Fix empty response body when default `status_code` is empty but the a `Response` parameter with `response.status_code` is set. PR [#5360](https://github.com/tiangolo/fastapi/pull/5360) by [@tmeckel](https://github.com/tmeckel).
#### How relative imports work { #how-relative-imports-work }
@ -279,7 +279,7 @@ We are not adding the prefix `/items` nor the `tags=["items"]` to each *path ope
But we can still add _more_`tags` that will be applied to a specific *path operation*, and also some extra `responses` specific to that *path operation*:
@ -305,13 +305,13 @@ You import and create a `FastAPI` class as normally.
And we can even declare [global dependencies](dependencies/global-dependencies.md){.internal-link target=_blank} that will be combined with the dependencies for each `APIRouter`:
As the files `app/routers/users.py` and `app/routers/items.py` are submodules that are part of the same Python package `app`, we can use a single dot `.` to import them using "relative imports".
@ -374,13 +374,13 @@ the `router` from `users` would overwrite the one from `items` and we wouldn't b
So, to be able to use both of them in the same file, we import the submodules directly:
@ -420,13 +420,13 @@ It contains an `APIRouter` with some admin *path operations* that your organizat
For this example it will be super simple. But let's say that because it is shared with other projects in the organization, we cannot modify it and add a `prefix`, `dependencies`, `tags`, etc. directly to the `APIRouter`:
But we still want to set a custom `prefix` when including the `APIRouter` so that all its *path operations* start with `/admin`, we want to secure it with the `dependencies` we already have for this project, and we want to include `tags` and `responses`.
We can declare all that without having to modify the original `APIRouter` by passing those parameters to `app.include_router()`:
That way, the original `APIRouter` will stay unmodified, so we can still share that same `app/internal/admin.py` file with other projects in the organization.
@ -447,7 +447,7 @@ We can also add *path operations* directly to the `FastAPI` app.
@ -155,7 +155,7 @@ The function parameters will be recognized as follows:
FastAPI will know that the value of `q` is not required because of the default value `= None`.
The `str | None`(Python 3.10+) or `Union` in `Union[str, None]` (Python 3.9+) is not used by FastAPI to determine that the value is not required, it will know it's not required because it has a default value of `= None`.
The `str | None` is not used by FastAPI to determine that the value is not required, it will know it's not required because it has a default value of `= None`.
But adding the type annotations will allow your editor to give you better support and detect errors.
@ -46,7 +46,7 @@ But even if you **fill the data** and click "Execute", because the docs UI works
In some special use cases (probably not very common), you might want to **restrict** the cookies that you want to receive.
Your API now has the power to control its own <abbrtitle="This is a joke, just in case. It has nothing to do with cookie consents, but it's funny that even the API can now reject the poor cookies. Have a cookie. 🍪">cookie consent</abbr>. 🤪🍪
Your API now has the power to control its own <dfntitle="This is a joke, just in case. It has nothing to do with cookie consents, but it's funny that even the API can now reject the poor cookies. Have a cookie. 🍪">cookie consent</dfn>. 🤪🍪
You can use Pydantic's model configuration to `forbid` any `extra` fields:
@ -54,9 +54,9 @@ You can use Pydantic's model configuration to `forbid` any `extra` fields:
If a client tries to send some **extra cookies**, they will receive an **error** response.
Poor cookie banners with all their effort to get your consent for the <abbrtitle="This is another joke. Don't pay attention to me. Have some coffee for your cookie. ☕">API to reject it</abbr>. 🍪
Poor cookie banners with all their effort to get your consent for the <dfntitle="This is another joke. Don't pay attention to me. Have some coffee for your cookie. ☕">API to reject it</dfn>. 🍪
For example, if the client tries to send a `santa_tracker` cookie with a value of `good-list-please`, the client will receive an **error** response telling them that the `santa_tracker`<abbrtitle="Santa disapproves the lack of cookies. 🎅 Okay, no more cookie jokes.">cookie is not allowed</abbr>:
For example, if the client tries to send a `santa_tracker` cookie with a value of `good-list-please`, the client will receive an **error** response telling them that the `santa_tracker`<dfntitle="Santa disapproves the lack of cookies. 🎅 Okay, no more cookie jokes.">cookie is not allowed</dfn>:
```json
{
@ -73,4 +73,4 @@ For example, if the client tries to send a `santa_tracker` cookie with a value o
## Summary { #summary }
You can use **Pydantic models** to declare <abbrtitle="Have a last cookie before you go. 🍪">**cookies**</abbr> in **FastAPI**. 😎
You can use **Pydantic models** to declare <dfntitle="Have a last cookie before you go. 🍪">**cookies**</dfn> in **FastAPI**. 😎
The default parameters used by the `CORSMiddleware` implementation are restrictive by default, so you'll need to explicitly enable particular origins, methods, or headers, in order for browsers to be permitted to use them in a Cross-Domain context.
These dependencies will be executed/solved the same way as normal dependencies. But their value (if they return any) won't be passed to your *path operation function*.
@ -44,13 +44,13 @@ You can use the same dependency *functions* you use normally.
They can declare request requirements (like headers) or other sub-dependencies:
@ -58,7 +58,7 @@ And they can return values or not, the values won't be used.
So, you can reuse a normal dependency (that returns a value) you already use somewhere else, and even though the value won't be used, the dependency will be executed:
# Dependencies with yield { #dependencies-with-yield }
FastAPI supports dependencies that do some <abbrtitle='sometimes also called "exit code", "cleanup code", "teardown code", "closing code", "context manager exit code", etc.'>extra steps after finishing</abbr>.
FastAPI supports dependencies that do some <dfntitle='sometimes also called "exit code", "cleanup code", "teardown code", "closing code", "context manager exit code", etc.'>extra steps after finishing</dfn>.
To do this, use `yield` instead of `return`, and write the extra steps (code) after.
@ -29,15 +29,15 @@ For example, you could use this to create a database session and close it after
Only the code prior to and including the `yield` statement is executed before creating a response:
The same way, you could have some dependencies with `yield` and some other dependencies with `return`, and have some of those depend on some of the others.
@ -109,7 +109,7 @@ But it's there for you if you need it. 🤓
If you want to catch exceptions and create a custom response based on that, create a [Custom Exception Handler](../handling-errors.md#install-custom-exception-handlers){.internal-link target=_blank}.
@ -117,7 +117,7 @@ If you want to catch exceptions and create a custom response based on that, crea
If you catch an exception using `except` in a dependency with `yield` and you don't raise it again (or raise a new exception), FastAPI won't be able to notice there was an exception, the same way that would happen with regular Python:
In this case, the client will see an *HTTP 500 Internal Server Error* response as it should, given that we are not raising an `HTTPException` or similar, but the server will **not have any logs** or any other indication of what was the error. 😱
@ -127,7 +127,7 @@ If you catch an exception in a dependency with `yield`, unless you are raising a
You can re-raise the same exception using `raise`:
Now the client will get the same *HTTP 500 Internal Server Error* response, but the server will have our custom `InternalError` in the logs. 😎
@ -190,7 +190,7 @@ Normally the exit code of dependencies with `yield` is executed **after the resp
But if you know that you won't need to use the dependency after returning from the *path operation function*, you can use `Depends(scope="function")` to tell FastAPI that it should close the dependency after the *path operation function* returns, but **before** the **response is sent**.
And all the ideas in the section about [adding `dependencies` to the *path operation decorators*](dependencies-in-path-operation-decorators.md){.internal-link target=_blank} still apply, but in this case, to all of the *path operations* in the app.
**FastAPI** has a very powerful but intuitive **<abbrtitle="also known as components, resources, providers, services, injectables">Dependency Injection</abbr>** system.
**FastAPI** has a very powerful but intuitive **<dfntitle="also known as components, resources, providers, services, injectables">Dependency Injection</dfn>** system.
It is designed to be very simple to use, and to make it very easy for any developer to integrate other components with **FastAPI**.
If one of your dependencies is declared multiple times for the same *path operation*, for example, multiple dependencies have a common sub-dependency, **FastAPI** will know to call that sub-dependency only once per request.
And it will save the returned value in a <abbrtitle="A utility/system to store computed/generated values, to reuse them instead of computing them again.">"cache"</abbr> and pass it to all the "dependants" that need it in that specific request, instead of calling the dependency multiple times for the same request.
And it will save the returned value in a <dfntitle="A utility/system to store computed/generated values, to reuse them instead of computing them again.">"cache"</dfn> and pass it to all the "dependants" that need it in that specific request, instead of calling the dependency multiple times for the same request.
In an advanced scenario where you know you need the dependency to be called at every step (possibly multiple times) in the same request instead of using the "cached" value, you can set the parameter `use_cache=False` when using `Depends`: