Prettier by default tries to preserve whitespace around inline tags,
which can prevent problems such as a space before the period in
'<a>text</a>.':
https://prettier.io/blog/2018/11/07/1.15.0.html#whitespace-sensitive-formatting
Unfortunately only standard HTML block elements are excluded from this
behaviour, so all of our Svelte components are treated the same way, even
if they are block-based, or used in a way where the extra whitespace
doesn't matter. This makes the code somewhat harder to read.
Changing this option does carry the risk that rogue spaces will creep
into our UI in the future as code is formatted, but as there don't appear
to be any such issues with this initial reformat, I think the improved
readability may justify the relatively small risk.
(for upgrading users, please see the notes at the bottom)
Bazel brought a lot of nice things to the table, such as rebuilds based on
content changes instead of modification times, caching of build products,
detection of incorrect build rules via a sandbox, and so on. Rewriting the build
in Bazel was also an opportunity to improve on the Makefile-based build we had
prior, which was pretty poor: most dependencies were external or not pinned, and
the build graph was poorly defined and mostly serialized. It was not uncommon
for fresh checkouts to fail due to floating dependencies, or for things to break
when trying to switch to an older commit.
For day-to-day development, I think Bazel served us reasonably well - we could
generally switch between branches while being confident that builds would be
correct and reasonably fast, and not require full rebuilds (except on Windows,
where the lack of a sandbox and the TS rules would cause build breakages when TS
files were renamed/removed).
Bazel achieves that reliability by defining rules for each programming language
that define how source files should be turned into outputs. For the rules to
work with Bazel's sandboxing approach, they often have to reimplement or
partially bypass the standard tools that each programming language provides. The
Rust rules call Rust's compiler directly for example, instead of using Cargo,
and the Python rules extract each PyPi package into a separate folder that gets
added to sys.path.
These separate language rules allow proper declaration of inputs and outputs,
and offer some advantages such as caching of build products and fine-grained
dependency installation. But they also bring some downsides:
- The rules don't always support use-cases/platforms that the standard language
tools do, meaning they need to be patched to be used. I've had to contribute a
number of patches to the Rust, Python and JS rules to unblock various issues.
- The dependencies we use with each language sometimes make assumptions that do
not hold in Bazel, meaning they either need to be pinned or patched, or the
language rules need to be adjusted to accommodate them.
I was hopeful that after the initial setup work, things would be relatively
smooth-sailing. Unfortunately, that has not proved to be the case. Things
frequently broke when dependencies or the language rules were updated, and I
began to get frustrated at the amount of Anki development time I was instead
spending on build system upkeep. It's now about 2 years since switching to
Bazel, and I think it's time to cut losses, and switch to something else that's
a better fit.
The new build system is based on a small build tool called Ninja, and some
custom Rust code in build/. This means that to build Anki, Bazel is no longer
required, but Ninja and Rust need to be installed on your system. Python and
Node toolchains are automatically downloaded like in Bazel.
This new build system should result in faster builds in some cases:
- Because we're using cargo to build now, Rust builds are able to take advantage
of pipelining and incremental debug builds, which we didn't have with Bazel.
It's also easier to override the default linker on Linux/macOS, which can
further improve speeds.
- External Rust crates are now built with opt=1, which improves performance
of debug builds.
- Esbuild is now used to transpile TypeScript, instead of invoking the TypeScript
compiler. This results in faster builds, by deferring typechecking to test/check
time, and by allowing more work to happen in parallel.
As an example of the differences, when testing with the mold linker on Linux,
adding a new message to tags.proto (which triggers a recompile of the bulk of
the Rust and TypeScript code) results in a compile that goes from about 22s on
Bazel to about 7s in the new system. With the standard linker, it's about 9s.
Some other changes of note:
- Our Rust workspace now uses cargo-hakari to ensure all packages agree on
available features, preventing unnecessary rebuilds.
- pylib/anki is now a PEP420 implicit namespace, avoiding the need to merge
source files and generated files into a single folder for running. By telling
VSCode about the extra search path, code completion now works with generated
files without needing to symlink them into the source folder.
- qt/aqt can't use PEP420 as it's difficult to get rid of aqt/__init__.py.
Instead, the generated files are now placed in a separate _aqt package that's
added to the path.
- ts/lib is now exposed as @tslib, so the source code and generated code can be
provided under the same namespace without a merging step.
- MyPy and PyLint are now invoked once for the entire codebase.
- dprint will be used to format TypeScript/json files in the future instead of
the slower prettier (currently turned off to avoid causing conflicts). It can
automatically defer to prettier when formatting Svelte files.
- svelte-check is now used for typechecking our Svelte code, which revealed a
few typing issues that went undetected with the old system.
- The Jest unit tests now work on Windows as well.
If you're upgrading from Bazel, updated usage instructions are in docs/development.md and docs/build.md. A summary of the changes:
- please remove node_modules and .bazel
- install rustup (https://rustup.rs/)
- install rsync if not already installed (on windows, use pacman - see docs/windows.md)
- install Ninja (unzip from https://github.com/ninja-build/ninja/releases/tag/v1.11.1 and
place on your path, or from your distro/homebrew if it's 1.10+)
- update .vscode/settings.json from .vscode.dist
* Reduce font size of answer button indicators
* Increase padding of browser rows with ResizeToContents on vertical header
* Remove 0.8 scale factor for dropdown item font-size
* Remove font-size prop entirely from DropdownItem
* Revert "Remove font-size prop entirely from DropdownItem"
This reverts commit bb0a158f96183cca74e198867070c2f99af04dc4.
* Remove hard-coded Python font sizes
* Move font size and scrollbar into _root-vars.scss
* Revert editor size variable to 1.6
* Fix icon alignment
* Fix checkbox alignment for dropdown items
* Remove unused classes from Tag.svelte
* Revert "Increase padding of browser rows with ResizeToContents on vertical header"
This reverts commit 77bfc854ba140dd99aae98efcdd4af7052615fa6.
* Remove option to set font size of browser entries
* Add setting for browser row padding to preferences
* Revert "Add setting for browser row padding to preferences"
This reverts commit 75c59da65a1028e2caa3c48b247f99825c1b0b6c.
* Revert "Remove option to set font size of browser entries"
This reverts commit a543783d8ea079f39b7ae445152573c96be29841.
* Allow passing in reference into WithFloating as prop
* Fix WithAutocomplete
* Fix WithFloating for MathjaxOverlay
* Add resize-store
* Allow passing debug=True to jest_test for debugger support (#2013)
* Disable auto-closing of HTML tags
https://forums.ankiweb.net/t/set-html-editor-as-a-default-editor-instead-of-visual-editor/20988/3Closes#1963
* Add slight margin to MathjaxEditor
* Enable passing offset and shift to WithFloating
* Hide overflow of mathjax editor
* Add automatic hide functionality to sveltelib/position
* Last polishes for Surrounder class (#2017)
* Make private properties in Surrounder truly private
* Fix remove logic of Surrounder
* No reason for toggleTriggerRemove to be async
* Allow using alt-shift to set all remove formats but this one
* modifyFormat => updateFormat
* Fix formatting
* Fix field descriptions blocking cursor from being set (#2018)
- happens when focus is in HTML editor
* Remove hiding functionality again until it's really useful
* Add support for autoPlacement
* Implement new WithFloating that supports manually calling position()
* Implement hide mechanisms
* Add option in math dropdown to toggle MathJax rendering (#2014)
* Add option in math dropdown to toggle MathJax rendering
Closes#1942
* Hackily redraw the page when toggling MathJax
* Add Fluent string
* Default input setting in fields dialog (#1987) (kleinerpirat)
* Introduce field setting to use plain text editor by default (kleinerpirat)
* Remove leftover function from #1476
* Use boolean instead of string
* Simplify clear_other_field_duplicates
* Convert plain text key to camelCase
* Move HTML item below the existing checkbox, instead of to the right (dae)
Showing it on the right is more space efficient, but feels a bit
cluttered IMHO.
* Fix not being able to scroll when mouse hovers PlainTextInput (#2019)
* Remove overscroll-behavior: none for * (all elements)
* Revert "Remove overscroll-behavior: none for * (all elements)"
This reverts commit 189358908cecd03027e19d8fe47822735319ec17.
* Use body instead of *, but keep CSS rule
* Unify two CSS rules
* Remove console.logs
* Reposition mathjax menu on switching between inline/block
* Implement WithOverlay
* Implement FloatingArrow
* Display overlay with padding and brighter background
* Rename to MathjaxOverlay
* Simplify MathjaxOverlay component overall
* Rename ImageHandle to image overlay
* Generally fix ImageOverlay again
* Increase z-index of StickyContainer
* Fix setting block or inline on mathjax
* Add reasons in closing-{click,keyup}
* Have both WithFloating and WithOverlay use a simple show flag instead of a store
* Remove subscribe-trigger
* Fix clicking from one mathjax element to another
* Check before executing cleanup
* Do not wait for elements to mount before slotting in With{Floating,Overlay}
* Allow using reference slot for WithFloating and WithOveray
* Add inline argument to options
* Add support for inline slot in WithOvelay
* Use WithFloating for RemoveFormatButton
* Remove last uses of DropdownMenu and WithDropdown
* Remove all of the bootstrap dropdown components
* Fix closing behavior of several buttons and ImageOverlay
* Increase popover padding to 6px
* Find a different way to create some padding at the bottom of the fields
...before the tag editor
@kleinerpirat I think is what this css what trying to achieve?
* Satisfy tests
* Use removeStyleProperties in ImageOverlay
* Use notify function in WithOverlay and WithFloating
* Do not use portal for WithFloating and WithOverlay
Allows for scrolling
* Set hidden to default false in Rich/Plain TextInput
* Reset handle when changing mathjax elements via click
* Restrict size of empty mathjax image
* Prevent sticky labels from obscuring menus
* Remove several overflow-hidden
* Fix empty string being falsy bug when editing mathjax
* Do not import portal anymore
* Use { reason, originalEvent } instead of symbol as update to modified event store
* Fix closing behavior of image overlay (do not close after resize)
* Simplify Collapsible
* Use removeStyleProperties in Collapsible
* Satisfy eslint
* Fix latex shortcuts being mounted
* Fix mathjax overlay not focusable in first field
* Neither hide image overlay on escaped
* Fix Block ButtonDropdown wrapping
* Bring back portal to fix tag editor
* Make eslint sort our imports
* fix missing deps in eslint rule (dae)
Caught on Linux due to the stricter sandboxing
* Remove exports-last eslint rule (for now?)
* Adjust browserslist settings
- We use ResizeObserver which is not supported in browsers like KaiOS,
Baidu or Android UC
* Raise minimum iOS version 13.4
- It's the first version that supports ResizeObserver
* Apply new eslint rules to sort imports
* Add componentHook functionality
* Register package NoteEditor
* Rename OldEditorAdapter to NoteEditor
* Expose instances in component-hook as well
* Rename NoteTypeButtons to NotetypeButtons
* Move PreviewButton initialization to BrowserEditor.svelte
* Remove focusInRichText
- Same thing can be done by inspecting activeInput
* Satisfy formatter
* Fix remaining rebase issues
* Add .bazel to .prettierignore
* Rename currentField and activeInput to focused{Field,Input}
* Move identifier to lib and registration to sveltelib
* Fix Dynamic component insertion
* Simplify editingInputIsRichText
* Give extra warning in svelte/svelte.ts
- This was caused by doing a rename of a files, that only differed in
case: NoteTypeButtons.svelte to NotetypeButtons.svelte
- It was quite tough to figure out, and this console.log might make it
easier if it ever happens again
* Change signature of contextProperty
* Add ts/typings for add-on definition files
* Add Anki types in typings/common/index.d.ts
* Export without .svelte suffix
It conflicts with how Svelte types its packages
* Fix left over .svelte import from editor.py
* Rename NoteTypeButtons to unrelated to ensure case-only rename
* Rename back to NotetypeButtons.svelte
* Remove unused component-hook.ts, Fix typing in lifecycle-hooks
* Merge runtime-require and register-package into one file
+ Give some preliminary types to require
* Rename uiDidLoad to loaded
* Fix eslint / svelte-check
* Rename context imports to noteEditorContext
* Fix import name mismatch
- I wonder why these issues are not caught by svelte-check?
* Rename two missed usages of uiDidLoad
* Fix ButtonDropdown from having wrong border-radius
* Uniformly rename libraries to packages
- I don't have a strong opinion on whether to name them libraries or
packages, I just think we should have a uniform name.
- JS/TS only uses the terms "module" and "namespace", however `package`
is a reserved keyword for future use, whereas `library` is not.
* Refactor registration.ts into dynamic-slotting
- This is part of an effort to refactor the dynamic slotting (extending
buttons) functionality out of components like ButtonGroup.
* Remove dynamically-slottable logic from ButtonToolbar
* Use DynamicallySlottable in editor-toolbar
* Fix no border radius on indentation button dropdown
* Fix AddonButtons
* Remove Item/ButtonGroupItem in deck-options, where it's not necessary
* Remove unnecessary uses of Item and ButtonGroupItem
* Fix remaining tests
* Fix relative imports
* Revert change return value of remapBinToSrcDir to ./bazel/out...
* Remove typings directory
* Adjust comments for dynamic-slottings
* Move move-nodes logic into domlib
Refactor input-manager
Refactor out FocusTrap from EditingArea
Remove unnecessary selecting of node from surround
Add onInput interface to input-manager
Create MathjaxElement.svelte
- This should contain all the setup necessary for displaying
<anki-mathjax> elements in the rich text input
- Does not contain setup necessary for Mathjax Overlay
Deal with backwards deletion, when caret inside anki-mathjax
Set mathjax elements contenteditable=false
Do not undecorate mathjaxx element on disconnect
- Fixes issues, where Mathjax might undecorate when it is moved into a
different div
Add framed element custom element
Introduce iterateActions to allow global hooks for RichTextInput
Remove some old code
Deal with deletion of frame handles
Make Anki frame and frame handles restore each other
Make FrameElement restore its structure upon modification
Frame and strip off framing from MathjaxElement automatically
Move FrameHandle to separate FrameStart/FrameEnd
Refactor FrameHandle
Set data-frames on FrameElement
Fix logic error connected to FrameElement
Communicate frameHandle move{in,out} to anki-mathjax
Clear selection when blurring ContentEditable
Make sure frame is destroyed when undecorating Mathjax
Use Hairline space instead of zeroWidth
- it has better behavior in the contenteditable when placing the caret
via clicking
Allow detection of block elements with `block` attribute
- This way, anki-mathjax block="true" will make field a
field be recognized to have block content
Make block frame element operater without handles
- Clicking on the left/right side of a block mathjax will insert a br
element to that side
When deleting, remove mathajax-element not just image
Update MathjaxButtons to correctly show block state
SelectAll when moving into inline anki mathjax
Remove CodeMirror autofocus functionality
Move it to Mathjaxeditor directly
Fix getRangeAt throwing error
Update older code to use cross-browser
Fix issue with FrameHandles not being properyly removed
Satisfy formatting
Use === instead of node.isSameNode()
Fix issue of focusTrap not initialized
* Fix after rebasing
* Fix focus not being moved to first field
* Add documentation for input-manager and iterate-actions
* Export logic of ContentEditable to content-editable
* Fix issue with inserting newline right next to inline Mathjax
* Fix reframing issue of Mathjax Svelte component
* Allow for deletion of Inline Mathjax again
* Rename iterate-actions to action-list
* Add copyright header
* Split off frame-handle from frame-element
* Add some comments for framing process
* Add mising return types
* Remove unnecessary stopPropagation of mathjax-overlay events
* Use CodeMirror component for MathjaxHandle
* Refactor ResizeObserver code in MathjaxHandle
* Wrap setRange in CodeMirror in try/catch
* Add Mathjax Editor bottom margin
* Add custom Enter and Shift+Enter shortcuts for the MathjaxHandle
* Format
* Move placeCaretAfter to domlib
* Move focus back to field after editing Mathjax
* Put Cursor after Mathjax after accepting
* Add delete button for Mathjax
* Change border color of mathjax menu
* Refactor into MathjaxMenu
* Put caretKeyword in variable
* Use one ResizeObserver for all Mathjax images
* Add minmimum width for Mathjax editor
* is still smaller than minimal window width
* Add bazel directories to .prettierignore and format from root
* exclude ftl/usage (dae)
the json files that live there are output from our tooling, and
formatting them means an extra step each time we want to update them
also exclude .mypy_cache, which is output by scripts/mypy*
* minor ftl tweak: newline -> new line (dae)
