This can be toggled using the `MASONRY_TRACE_PASSES` environment
variable.
This is as-discussed in [#xilem > Scrolling is insanely
laggy](https://xi.zulipchat.com/#narrow/channel/354396-xilem/topic/Scrolling.20is.20insanely.20laggy).
There are a few overlapping issues, but this *significantly* improves
performance in debug mode.
The main pass which was problematic was the compose pass, however this
also had a significant impact on the time taken by the
accessibility/paint passes.
This only applies to the passes which traverse the entire tree, so not
the targeted passes. I also chose to exclude update_disabled and
update_stashed, as those will not necessarily happen to all widgets.
This also significantly reduces the size of the created log files - see
[#masonry > Heavy amounts of logs with large
app](https://xi.zulipchat.com/#narrow/channel/317477-masonry/topic/Heavy.20amounts.20of.20logs.20with.20large.20app).
In most cases, if you're using the log file, you will be in development,
which means that you can hopefully recreate the issue with the logging
for the passes you need enabled.
Phew, fixing (future) clippy-lints is grunt-work (I enabled
`unnameable_types` for this), this is by far not complete, mostly
xilem-side, but at least a small step towards enabling more of our lint
set.
There's a few "drive-by fixes" like hiding `View::ViewState` types in
Xilem (I don't think we generally want to expose these), or exposing
event types in xilem_web.
I would suggest disabling the `allow_attributes_without_reason` lint for
now, so that we can update the rust version. I feel like adding comments
next to these is just unnecessary extra-work. I think adding more
reasons should probably be a separate pass, after a rust version update,
so that `#[allow(..., reason="")]` is supported.
cc for awareness @mwcampbell.
That field was dead code left behind when we removed previous links
support. On that struct is probably not the right field for links to
live, since `Label` also uses `TextLayout`, but can't meaningfully
support links as part of a different interactive widget.
The whole reason the Label widget exists is as a non-interactive text
holder for higher-level widgets. Therefore, it isn't meaningful for it
to have interactivity.
This marks animation as something that's somewhere between an event
(it's triggered externally, it can't be requested synchronously as part
of rewrite passes) and an update (it takes the same context type, and
isn't tied to a user interaction).
It's possible to avoid returning the `ViewElement::Mut` in
`View::rebuild`, while having access to the element *after* a child view
was rebuilt.
This PR removes that again making the signature a little bit cleaner,
and also cleans up a lot of other stuff:
* `Mut<'_, Self::Element>` can be shortened to `Mut<Self::Element>`
* `use xilem::core::*` instead of `use xilem_core::*`, this is
especially useful in the examples, to avoid newcomers giving the
impression, that an additional dependency xilem_core is needed.
* Various cosmetic stuff, like one typo fix, empty line between
doc-comment (as (nightly) clippy currently bleats for me)
For more context see
https://xi.zulipchat.com/#narrow/stream/354396-xilem/topic/.E2.9C.94.20.60.26mut.20ViewElement.3A.3AMut.3C'_.3E.60.20in.20.60View.3A.3Arebuild.60.3F
Reorder update passes so their code is in the same order as they're run.
Split out anim pass to a separate file (since it's not run with the
other passes).
This is strictly cut-and-paste, no code has been changed.
I'm pretty sure this method never caught a single bug.
I think the idea of a "safety net" that checks invariants are upheld by
passes (without trusting pass code) has merit, but it should be based on
a survey of previous bugs in pass code.
This adds the necessary AccessKit properties and child nodes to expose
detailed information about text widgets. It also supports the
`SetTextSelection` action.
---------
Co-authored-by: Olivier FAURE <couteaubleu@gmail.com>
Remove type parameter from TextLayout.
Tweak error message in text widgets.
This makes the TextLayout code slightly simpler (though with more
invariants) in anticipation of splitting it into its own crate.
Replace with `Widget::accepts_xxx()` methods
Tweak accessibility code
Remove `BuildFocusChain` event
Remove `WidgetState::is_portal`, `register_as_portal()`, and overall
remove the idea of a first-class "portal" concept.
Create run_rewrite_passes function.
Remove redundant uses of synthetic WidgetState.
Small bugfixes.
Move RenderRoot::root_layout code into the layout pass. Rename
update_new_widgets pass to update_widget_tree.
These methods will be used with #639 to implement a fixpoint loop for
rewrite passes.
This also lets us remove `mark_changed`, since we can use
`content_changed` for the same purpose.
I missed comment artifacts in #621, added doc-comment/test for
`SvgElement` and put the `WithStyle` on the `Element` interface bounds,
as it plays better with rust-analyzer and is more consistent with the
other API.
This sketches/implements (parts of) the transform modifier API described
in [Add `Affine` transform to `Widget`
trait?](https://xi.zulipchat.com/#narrow/stream/317477-masonry/topic/Add.20.60Affine.60.20transform.20to.20.60Widget.60.20trait.3F/near/472076600)
for xilem_web.
This currently includes `rotate` and `scale`, because there are less
cases to handle for these [CSS transform
functions](https://developer.mozilla.org/en-US/docs/Web/CSS/transform-function).
In `rotate` I think we can reduce this to just radians (there are more
units for `<angle>`: `turn`, `grad`, `deg` and `rad` (used here) but
they can all be derived from radians), and percent in scale is kinda
redundant (as `200%` == `2.0`)), for example `translate` has more cases
as it includes all kinds of units (like `px`, `em`, `%` etc.), so this
may require more thought (and is thus probably better for a future PR).
This can be combined with untyped `transform: ...` as style, these
modifier views extend the `transform` style (while the untyped
`style(..)` overwrites previous set values).
The `svgtoy` example is updated to include these views.
Add a method which brute-force resets the paint and accessibility flags
for all widgets.
This is useful for rescale, and for testing purposes where you want a
quick and dirty way to repaint everything.
Each widget implementation was doing its own bespoke tracing in layout.
This removes the bespoke tracing and adds a trace inside the layout pass
logging the same information in a more centralized way.
Olivier FAURE