From Flask#

ASGI vs WSGI#

Flask is a WSGI framework, whereas Starlite is built using the modern ASGI standard. A key difference is that ASGI is built with async in mind.

While Flask has added support for async/await, it remains synchronous at its core; The async support in Flask is limited to individual endpoints. What this means is that while you can use async def to define endpoints in Flask, they will not run concurrently - requests will still be processed one at a time. Flask handles asynchronous endpoints by creating an event loop for each request, run the endpoint function in it and then return its result.

ASGI on the other hand does the exact opposite; It runs everything in a central event loop. Starlite then adds support for synchronous functions by running them in a non-blocking way on the event loop. What this means is that synchronous and asynchronous code both run concurrently.

Routing#

from flask import Flask


app = Flask(_name_)


@app.route("/")
def index():
    return "Index Page"


@app.route("/hello")
def hello():
    return "Hello, World"
from starlite import Starlite, get


@get("/")
def index() -> str:
    return "Index Page"


@get("/hello")
def hello() -> str:
    return "Hello, World"


app = Starlite([index, hello])

Path parameters#

from flask import Flask


app = Flask(_name_)


@app.route("/user/<username>")
def show_user_profile(username):
    return f"User {username}"


@app.route("/post/<int:post_id>")
def show_post(post_id):
    return f"Post {post_id}"


@app.route("/path/<path:subpath>")
def show_subpath(subpath):
    return f"Subpath {subpath}"
from starlite import Starlite, get
from pathlib import Path


@get("/user/{username:str}")
def show_user_profile(username: str) -> str:
    return f"User {username}"


@get("/post/{post_id:int}")
def show_post(post_id: int) -> str:
    return f"Post {post_id}"


@get("/path/{subpath:path}")
def show_subpath(subpath: Path) -> str:
    return f"Subpath {subpath}"


app = Starlite([show_user_profile, show_post, show_subpath])

See also

To learn more about path parameters, check out this chapter in the documentation: Path parameters

Request object#

In Flask, the current request can be accessed through a global request variable. In Starlite, the request can be accessed through an optional parameter in the handler function.

from flask import Flask, request


app = Flask(_name_)


@app.get("/")
def index():
    print(request.method)
from starlite import get, Request


@get("/")
def index(request: Request) -> None:
    print(request.method)

Request methods#

Flask

Starlite

request.args

request.query_params

request.base_url

request.base_url

request.authorization

request.auth

request.cache_control

request.headers.get("cache-control")

request.content_encoding

request.headers.get("content-encoding")

request.content_length

request.headers.get("content-length")

request.content_md5

request.content_type

request.content_type

request.cookies

request.cookies

request.data

request.body()

request.date

request.headers.get("date")

request.endpoint

request.route_handler

request.environ

request.scope

request.files

Use `UploadFile <usage/4-request-data/#file-uploads>`__

request.form

request.form(), prefer `Body <usage/4-request-data/#specifying-a-content-type>`__

request.get_json

request.json(), prefer the `data keyword argument <usage/4-request-data/#request-body>`__

request.headers

request.headers

request.host

request.host_url

request.if_match

request.headers.get("if-match")

request.if_modified_since

request.headers.get("if_modified_since")

request.if_none_match

request.headers.get("if_none_match")

request.if_range

request.headers.get("if_range")

request.if_unmodified_since

request.headers.get("if_unmodified_since")

request.method

request.method

request.mimetype

request.mimetype_params

request.origin

request.path

request.scope["path"]

request.query_string

request.scope["query_string"]

request.range

request.headers.get("range")

request.referrer

request.headers.get("referrer")

request.remote_addr

request.remote_user

request.root_path

request.scope["root_path"]

request.server

request.scope["server"]

request.stream

request.stream

request.url

request.url

request.url_charset

request.user_agent

request.headers.get("user-agent")

request.user_agent

request.headers.get("user-agent")

See also

To learn more about requests,, check out these chapters in the documentation

Static files#

Like Flask, Starlite also has capabilities for serving static files, but while Flask will automatically serve files from a static folder, this has to be configured explicitly in Starlite.

from starlite import Starlite, StaticFilesConfig

app = Starlite(
    [], static_files_config=StaticFilesConfig(path="/static", directories=["static"])
)

See also

To learn more about static files, check out this chapter in the documentation: Static files

Templates#

Flask comes with the Jinja templating engine built-in. You can use Jinja with Starlite as well, but you’ll need to install it explicitly. You can do by installing Starlite with pip install starlite[jinja]. In addition to Jinja, Starlite supports Mako templates as well.

from flask import Flask, render_template


app = Flask(_name_)


@app.route("/hello/<name>")
def hello(name):
    return render_template("hello.html", name=name)
from starlite import Starlite, get, TemplateConfig, Template
from starlite.contrib.jinja import JinjaTemplateEngine


@get("/hello/{name:str}")
def hello(name: str) -> Template:
    return Template(name="hello.html", context={"name": name})


app = Starlite(
    [hello],
    template_config=TemplateConfig(directory="templates", engine=JinjaTemplateEngine),
)

See also

To learn more about templates, check out this chapter in the documentation: Template engines

Setting cookies and headers#

from flask import Flask, make_response

app = Flask(_name_)


@app.get("/")
def index():
    response = make_response("hello")
    response.set_cookie("my-cookie", "cookie-value")
    response.headers["my-header"] = "header-value"
    return response
from starlite import get, ResponseHeader, Cookie, Response


@get(
    "/static",
    response_headers={"my-header": ResponseHeader(value="header-value")},
    response_cookies=[Cookie("my-cookie", "cookie-value")],
)
def static() -> str:
    # you can set headers and cookies when defining handlers
    ...


@get("/dynamic")
def dynamic() -> Response[str]:
    # or dynamically, by returning an instance of Response
    return Response(
        "hello",
        headers={"my-header": "header-value"},
        cookies=[Cookie("my-cookie", "cookie-value")],
    )

See also

To learn more about response headers and cookies, check out these chapters in the documentation:

Redirects#

For redirects, instead of redirect use Redirect:

from flask import Flask, redirect, url_for

app = Flask(_name_)


@app.get("/")
def index():
    return "hello"


@app.get("/hello")
def hello():
    return redirect(url_for("index"))
from starlite import Starlite, get, Redirect


@get("/")
def index() -> str:
    return "hello"


@get("/hello")
def hello() -> Redirect:
    return Redirect(path="index")


app = Starlite([index, hello])

Raising HTTP errors#

Instead of using the abort function, raise an HTTPException:

from flask import Flask, abort

app = Flask(_name_)


@app.get("/")
def index():
    abort(400, "this did not work")
from starlite import Starlite, get, HTTPException


@get("/")
def index() -> None:
    raise HTTPException(status_code=400, detail="this did not work")


app = Starlite([index])

See also

To learn more about exceptions, check out this chapter in the documentation: Exceptions

Setting status codes#

from flask import Flask


app = Flask(_name_)


@app.get("/")
def index():
    return "not found", 404
from starlite import Starlite, get, Response


@get("/static", status_code=404)
def static_status() -> str:
    return "not found"


@get("/dynamic")
def dynamic_status() -> Response[str]:
    return Response("not found", status_code=404)


app = Starlite([static_status, dynamic_status])

Serialization#

Flask uses a mix of explicit conversion (such as jsonify) and inference (i.e. the type of the returned data) to determine how data should be serialized. Starlite instead assumes the data returned is intended to be serialized into JSON and will do so unless told otherwise.

from flask import Flask, Response

app = Flask(_name_)


@app.get("/json")
def get_json():
    return {"hello": "world"}


@app.get("/text")
def get_text():
    return "hello, world!"


@app.get("/html")
def get_html():
    return Response("<strong>hello, world</strong>", mimetype="text/html")
from starlite import Starlite, get, MediaType


@get("/json")
def get_json() -> dict[str, str]:
    return {"hello": "world"}


@get("/text", media_type=MediaType.TEXT)
def get_text() -> str:
    return "hello, world"


@get("/html", media_type=MediaType.HTML)
def get_html() -> str:
    return "<strong>hello, world</strong>"


app = Starlite([get_json, get_text, get_html])

Error handling#

from flask import Flask
from werkzeug.exceptions import HTTPException


app = Flask(_name_)


@app.errorhandler(HTTPException)
def handle_exception(e):
    ...
from starlite import Starlite, HTTPException, Request, Response


def handle_exception(request: Request, exception: Exception) -> Response:
    ...


app = Starlite([], exception_handlers={HTTPException: handle_exception})

See also

To learn more about exception handling, check out this chapter in the documentation: Exception handling