Instead of generating a fluent.proto file with a giant enum, create
a .json file representing the translations that downstream consumers
can use for code generation.
This enables the generation of a separate method for each translation,
with a docstring that shows the actual text, and any required arguments
listed in the function signature.
The codebase is still using the old enum for now; updating it will need
to come in future commits, and the old enum will need to be kept
around, as add-ons are referencing it.
Other changes:
- move translation code into a separate crate
- store the translations on a per-file/module basis, which will allow
us to avoid sending 1000+ strings on each JS page load in the future
- drop the undocumented support for external .ftl files, that we weren't
using
- duplicate strings in translation files are now checked for at build
time
- fix i18n test failing when run outside Bazel
- drop slog dependency in i18n module
- The previous commits moved the majority of the remaining global css
into components; move the remaining @emotion/css references into
ticks.scss and the styling of the Graph.svelte. This is not as elegant
as the emotion solution, but builds a whole lot faster, and most of
our styling can be scoped to a component anyway.
- Leave the .html files in ts/ for now. AnkiMobile uses them, and
AnkiDroid likely will in the future too. In the long run we'll likely
move to loading the JS into an existing page instead of loading a
separate page, but at that point we can just exclude the .html file from
copy_files_into_group() without affecting other clients.
Closes#1074
- svelte compilation outputs a separate .css file for each component
- compilation also adds an "import foo.css" to the top of each generated
.mjs file
- when the .mjs files are bundled into app.js, esbuild creates an app.css
as well
- graphs.scss was renamed to graphs_shared.scss and imported in the
top level GraphsPage. Henrik's style refactoring would be a better path
forward, but I needed to make this change for now, as the filenames were
conflicting.
- enable resolver patch on worker binary to ensure js imports work
on subsequent worker requests
- cache ts library content, and use unified interface for cache
- prepare for separate css outputs
Just a quick hack for now to store it in memory, as the temp file
conflicts on Windows due to the lack of a sandbox, and we don't really
have a need to write it to the filesystem anyway.
based on changes from upstream rules_svelte
Their code was using run_node() instead of ctx.actions.run(), which
seems to create a new worker for every CPU core, instead of respecting
the standard limit of 4.
Svelte 3.25.0 and onwards bundle compiler.mjs, which seems to be
preferentially used over the .js file. Presumably this is only breaking
on Windows due to the lack of a sandbox. Resolve by explicitly requesting
the .js file.
Issues that need fixing:
- when the editor saves the note with perform_op(), if it isn't modified,
no new undo entry is created, and perform_op then returns the changes
made by the previous operation instead
- the approach of fetching the last action in a subsequent backend
method is unsound, as another queued operation may sneak in first before
we have a chance to query the result - it would be better if it were
returned in a single atomic action
- redrawing the current card while editing is likely to make sound
autoplay annoyingly, and it has an unpleasant redraw. We may be better off
fading it out instead
Side note: the editor cursor moves to the start of the field when the
note is updated in another window - it might be nicer to have it move
the cursor to the end instead.
Bootstrap sets font-size to inherit, so it's not inherited from top.
Before, it defaulted to 13, so this should prevent add-ons, which
feature text in their buttons (like <select> buttons).
* Ported from #1046:
* disabling buttons will clear button highlight
* enabling button will set button highlight
* move caret to end executed before enabling buttons (so button highlight will be for actual position of caret)
* move caret to end will also be executed if previousActiveElement is null, which will only be the case before the first onBlur was executed:
* so that caret will be moved to end on opening editor
- using :host-context(.nightMode) allows for applying the nightmode
scroll bar inside the component
- apply max-width: 100% to all element within editable, not just images
- draw a border between sidebar and main area
- tweak padding
Testing is a pain, because you need to check day mode on the
three platforms, and night mode as well. If you can do it better, PRs
are welcome :-)
1. Clicking away from the editor window, and back on it should not focus old field
2. Clicking on a field, which is not fully visible, should scroll it into view
This is to provide minimal closures for the mutable file-local variables:
- changeTimer
- previousActiveElement
- currentNoteId
This makes it clear, that they should not be used, but rather the
functions which wrap them in an API
To support images on that screen, we'll first need to adjust the base url
for each platform, or rewrite the local image URLs, as otherwise they
are resolved to _anki/pages/...
Committing for reference; will roll back afterwards.
This adds approximately 150k to the bundled .js file in release mode.
html-sanitizer might be useful to replace our custom paste filtering
code in the future, but for now I'm not sure it's worth the extra
page load time over doing the filtering in Rust.
This reverts commit ffcf0aa3ca and
points to a new rules_svelte commit.
It looks like we were getting away with not listing the dep on the
rules_svelte end - the failing build turned out to be because we need
to pass sass in to our local svelte_check invocation.
- Closes#976
- Added helper to apply arbitrary colour to an icon.
- Fix#979 - low res icons in night mode.
- The icons and colours are not perfect - please feel free to send
through a PR if you can improve them.
- Convert colors dictionary into module consts, so we can
use code completion.
- Added "Edited Today" and "Due Tomorrow"
- Rename camelCase attribute to snake_case and tweak the wording
of some enum constants. We've already broken compatibility with the
major sidebar add-ons, so we may as well make these changes while we
can.
- Removed Filter button. Currently there is no exposed way to toggle
the Sidebar off - wonder if we still need it?
The original reason for the catch-all message was users with bad
data such as decimal intervals, but those get automatically coerced
these days. The common case should now be invalid search strings, which
we can show verbatim.
Uses the logic from the sqltools VSCode add-on, with a workaround
for the use of 'type' in some table columns.
By detecting the presence of 'BUILD_WORKSPACE_DIRECTORY' we can tell
if the rule is running in test mode or was run directly, avoiding the
need for separate check and fix rules. It might be nice to extend this
to other formatting rules in the future as well.
examples:
* x0 = -66 and x1 = -64 should yield 65-66 days ago instead of 64-65 days ago
* x0 = -2 and x1 = 0 should yield 1-2 days ago instead of 0-1 days ago
* cards will still be mostly counted by ctype rather than queue
* if the user wants to include inactive cards buried and suspended will
be first filtered out, before the rest is counted by ctype
Rather than creating a separate rule for each package, we can just
create a generic one and reuse it. Also switch to keyword arguments
in the resulting macros, as it's easier to read.
This reverts commit 62600051ae, reversing
changes made to 88553acb0d.
- Standard graphs render incorrectly on latest version - the wrong number
of days are shown, and the grid lines look wrong. Any version after 0.8.3
seems to suffer from this problem.
- Pie graphs and stack graphs don't render - they are provided in separate
files, and plot.js in previous Anki versions has them included in the one
file. To maintain compatibility with add-ons, we'd need to create a single
file as before, instead of importing multiple files.
If the above issues are fixed I'd be happy to merge this in again, but
as the old graphs are on the way out, it's probably not worth the effort.
While 'SvelteComponent | null' seems to make it into the .tsx file
created by svelte2tsx, the subsequent tsc call seems to discard the
'| null' part when creating the .d.ts file. Hack around it with a cast
for now; this may be fixed if we move to ts_project in the future.
Allows some type errors to surface that were only being picked up
on Windows.
The root cause seems to be TypeScript picking up other .d.ts/.tsx
files in the same folder, which it can only do on Windows due to the
lack of sandboxing. On other platforms the other files can't be found,
and tsc changes the types into 'any'.
I experimented with modifying rules_svelte to build all .tsx files up
front and convert them to .d.ts in bulk, but ran into further issues
with conflicting types, as the typings in svelte2tsx seem to conflict
with Svelte's built-in types, and passing the dependencies in explicitly
causes them to be checked even though --skipLibCheck is passed in to
TypeScript.
Forcing sandboxing off is an ugly hack, and our best approach moving
forward may be to switch to ts_project for the Svelte generation -
it does appear that rules_nodejs favours it over ts_library anyway.
