Continue to support redis < 7.2
Instead of updating individual substate expiration, in older versions of redis
any update to any substate will cause the entire state expiration to be
refreshed, which is better than not supporting older redis versions.
* add eradicate rules for commented out code
* remove output change
* fix pyi messed up indent
* fix pyi again
* fix layout docstring
* fix pyi_generator to remove commented out props from docs
* fix pyi_generator and regenerate some pyi
* fix double strip
* update all pyi
* try to fix stuff in pyi_gen
* whatever
* remove that maybe? i don't know
* fix that shit?
* fix more shit, idk
* better not see you ever again, extra line
* fix: handle default_factory in get_attribute_access_type, add tests for sqla dataclasses
* only test classes which have default_factory + add test for no default
* add test for color mode (initial and toggle)
* add css check
* add page reload in the tests
* update test to catch the appearance regression
* don't render the appearance prop of rx.theme
* Reuse the sqlalchemy engine once it's created
* Implement `rx.asession` for async database support
Requires setting `async_db_url` to the same database as `db_url`, except using
an async driver; this will vary by database.
* resolve the url first, so the key into _ENGINE is correct
* Ping db connections before returning them from pool
Move connect engine kwargs to a separate function
* the param is `echo`
* sanity check that config db_url and async_db_url are the same
throw a warning if the part following the `://` differs between these two
* create_async_engine: use sqlalchemy async API
update types
* redact ASYNC_DB_URL similarly to DB_URL when overridden in config
* update rx.asession docstring
* use async_sessionmaker
* Redact sensitive env vars instead of hiding them
* Add production-one-port example
A more complex version of simple-one-port that facilitates better layer caching
to shorten build times and multi-stage build to reduce final image size.
Harder to understand, but ultimately nicer to use.
* fix Caddyfile format to avoid complaints
* docker-examples: bump all base images to python:3.13
* rx.upload must include _var_data from props
str-casting the dropzone arguments removed any VarData they depended on, like
the state context.
update test_upload to include passing a prop from a state var
* Handle large payload delta from upload event handler
Fix update chunk chaining logic; try/catch wasn't catching errors from the
async inner function.
* Unit test updates
* test_client_storage: simulate backend state expiry
* [HOS-333] Send a "reload" message to the frontend after state expiry
1. a state instance expires on the backing store
2. frontend attempts to process an event against the expired token and gets a
fresh instance of the state without router_data set
3. backend sends a "reload" message on the websocket containing the event and
immediately stops processing
4. in response to the "reload" message, frontend sends
[hydrate, update client storage, on_load, <previous_event>]
This allows the frontend and backend to re-syncronize on the state of the app
before continuing to process regular events.
If the event in (2) is a special hydrate event, then it is processed normally
by the middleware and the "reload" logic is skipped since this indicates an
initial load or a browser refresh.
* unit tests working with redis
