* Fix bug with checkbox widget and `list` field
The `list` field is stored as a list and frozen against modifications,
and getFieldList() returns it directly without creating a copy. So
before we modify it, we need to make a copy so we're not modifying a
frozen list. This bug doesn't manifest with custom fields, which are
stored as strings, only with the built-in `list` field.
* Fix checkboxes referencing non-existent tiddlers
This fixes the "tiddler is undefined" error when a checkbox's listField
property references a tiddler that doesn't (yet) exist.
* Better logic for checkbox listField handling
If the field contains an array, then it's almost certainly referenced
elsewhere and needs a defensive copy made. If it contained a string,
then it's safe to modify without making a defensive copy.
This first implementation concatenates the results of the filter (with no separator) and then wikifies the result.
The test in this commit is quite interesting...
* Implement insertafter operator (like insertbefore)
Currently, the behavior of insertafter if the target is not found is to
append the inserted tiddler to the end of the list, like insertbefore
does. In the next commit, we'll add a suffix to customize what both
insertafter and insertbefore do when the target is not found.
* Add failing tests for insertafter suffixes
Also includes tests for insertbefore suffixes (start/end), since we'll
be implementing both of those at the same time.
* Add start/end suffixes for insertafter/before
The tests that exercise the start/end suffixes now pass.
* Add `main` and `article` roles to stories and story.
* Support `role` and \aria-checked` in buttons, and integrate with sidebar tabs.
* Add `region` role to sidebar, and set caption as `aria-label`.
* Add accessibility roles and labels to static templates.
* Update test fixtures with new ARIA tab attributes.
Global variables access within attributes will automatically trigger a refresh if the attribute text changes, but that wasn't happening for transclusions.
The objective is to add a $depth attribute so that it is possible to reach up to retrieve the parameters of ancestor transclusions. However, doing so requires changing the encoding of parameter names so that it is not possible for a user parameter to clash with an attribute like $depth. So now we have to double up dollars on any attribute names seen by the parameters widget, just like with the transclude widget itself.
Can invoke any functions, not just those start with a period. And can pass zero parameters (in contrast when invoked as a custom filter operator there's no way to omit the first parameter).
So far it appears to be totally backwards compatible... In practice, I think maybe this and the conversion of the other macros should go into a separate subsequent PR.
The basic idea is that if we don't find a variable `foo` then we fallback to retrieving the value from the tiddler `$:/global/foo`, if it exists.
This allows us to replace the usual importvariables-based mechanism for global definitions, avoiding cluttering up the variable namespace with every macro.
In order to permit subprocedures to be overridden, we also introduce a mechanism for conditional definitions: preceding the word definition|procedure|function|widget with a + causes the definition only to occur if the specified variable doesn't already exist. In the next commit we'll apply this mechanism to the tabs macro
There are two big changes here:
Replace the previous "ts-wrapper" mechanism, which we had been using to redefine custom widgets inside their definitions to prevent recursive calls. Now we've got the genesis widget we can instead control recursion through a new "$remappable" attribute that allows the custom widget mechanism to be skipped.
We also extend the slot widget to allow a depth to be specified; it then reaches up by the indicated number of transclusion widgets to find the one from which it should retrieve the slot value.
* Documentation for indeterminate checkboxes
* Unit tests for indeterminate checkboxes
* Implement indeterminate checkboxes
* Simplify indeterminate checkbox example
* Slightly simplify refresh logic for indeterminate
That five-line if statement can be turned into a simple assignment.
* Use "yes" and "no" for checkbox indeterminate attr
This makes the "indeterminate" attribute of the checkbox widget work the
same way as other boolean attributes of other widgets.
* Fix bug with invertTag attribute
One place in the checkbox widget code was checking invertTag for
Javascript truthiness rather than the value "yes", which could have
produced incorrect results if anyone wrote invertTag="no". Fixed.
* fix: formatDateString with [UTC]xxx didn't use passed date
* test: for formatDateString UTC
* fix: not possible to test internal date without hijack
Expected '20220410073037515' to be '20220410073037516'.
* fix: hour
* Docs for CheckboxWidget list and filter modes
This documents the `listField` and `filter` attributes.
* Tests for checkbox widget list mode
* Implement checkbox list mode
* WIP on implementing filter attr for checkboxes
* Improve CheckboxWidget documentation
* Refactor checkbox tests: move function to top
The big findNodeOfType function belongs at the top of the describe
block, so that the checkbox tests are more compact and easy to read.
* Move checkbox widget tests to end of file
The checkbox widget tests are long and involved, so we'll move them to
the end of the file so they aren't a huge block of code you need to read
past to find the next test.
* Improve formatting of CheckboxWidget docs
The \define() calls that are short enough to fit on one line should be
put on one line, for readability. The ones that are quite long have been
kept on multiple lines, for readability.
* Added more passing tests for checkbox widget
* Add some failing tests for checkbox widget
The filter mode where neither checked nor unchecked is specified (in
which case an empty filter result means false and a non-empty result
means true) is not working yet.
* Make failing tests pass
* Uncomment (and improve) test for field mode
We're now ready to start working on making this test pass. (There was
also one small mistake in the test, which this commit corrects).
* All tests now passing
* No indeterminate checkboxes in simple modes
The simple checkbox modes (field and index) should not produce
indeterminate checkboxes. That should be reserved for the advanced modes
(list and filter).
* Minor improvement to unit tests
* Allow indeterminate checkboxes in list and filter modes
This change may require some tweaks to the unit tests to be able to test
it properly.
* Slightly easier to read tests
* Two more tests for list mode
* Greatly simplify unit test code
Turns out there's no need to jump through Object.getPrototypeOf hoops.
* Minor simplification of unit test
* Add tests for indeterminate in list & filter modes
With this, the set of tests is complete.
* More tests to specify list mode behavior
* Unfocus tests so all tests run
* Update docs to say "new in 5.2.3" insetad of 5.2.2
* Move checkbox widget tests into their own file
The test-widget.js file was getting too long with all the checkbox
tests added, so we'll move the checkbox tests into their own file.
* Add checkbox widget tests for index mode
This commit also adds tests for index list mode (with a listIndex
attribute that will parallel the listField attribute) but leaves them
commented out because they don't pass yet: the code that implements the
listIndex attribute hasn't been written yet).
* Add listIndex attribute to checkbox widget
* Remove code that lets checkboxes be indeterminate
This reverts commit 6afcb151be. We will
add this code back in a later PR.
* Remove indeterminate tests for checkbox widget
We're currently not allowing indeterminate checkboxes, so there's no
need for the tests that check for them.
* Document listIndex attribute of CheckboxWidget
* adds class tc-checkbox-checked when checked
* equivalent to #2182 (RadioWidget)
* also applies `tc-checkbox` to checkboxes by default, always
* Move macro definitions inside example text
Since the wikitext-example-without-html macro creates a new parsing
context, it's safe to have macro definitions inside it. That makes these
examples a lot easier to write, and to read.
* Remove all mention of indeterminate checkboxes
Also improve the documentation a little bit: mention what happens in
list mode if neither checked nor unchecked is specified.
* Move filter mode to bottom of checkbox docs
The `filter` attribute should be under both `listField` and `listIndex`
rather than being between them. The documentation for filter mode should
similarly be after the `listIndex` documentation.
* Improve docs for `class` attr of checkbox widget
This brings the wording of the `class` attribute more in line with how
it's worded in the RadioWidget docs.
* Fix bug with list tiddlers
If neither checked nor unchecked was specified, then the behavior should
be "empty = false, non-empty = true". But if *both* are specified yet
neither is found, then the checkbox should be unchecked (false). It had
been falling through to the "non-empty = true" behavior, which was wrong.
* Improve listIndex example of checkbox widgets
* Remove unused function from test-widget.js
Co-authored-by: Tobias Beer <beertobias@gmail.com>
* tabs activate v5.2.2 tests add whitespace trim
* tabs-macro -- add indentation and code preview
* tabs-macro -- replace substitutions with variables
* split tabs-macro macro into different elements
- tabs-button
- tabs-tab
- tabs-tab-list
- tabs-tab-body
- tabs ... main macro
* tabs: add cascade to button and reaveal widgets
This will allow users to create "default tab" configurations similar to the tiddler info tab handling.
* tabs-macro -- add code_body: yes
* adding trim: Last of the macros I think
* adding trim: EditTemplate and ItemSidebarIcon
* adding trim: control panel basics
* Another hidden space to guide the uglifier
* More consistent nested quoting
* Reconciling tests for \whitespace trim
* Support specifying insertbefore position title as parameter
Closes: #6133
* Update insertbefore calls with new variable parameter syntax
See-also: 96b52606a (Support specifying insertbefore position title as parameter, 2022-02-21)
* corrected [!is[draft]] to be a proper complement
* Ensuring [is[draft]] and [!is[draft]] are complements
* Made [is[draft]] more analogous to .isDraft()
* Support case insensitive matching in prefix/suffix operators
Support `caseinsensitive`/`caseinsensitive` suffixes in the following
filter operators:
* prefix
* suffix
* removeprefix
* removesuffix
The suffixes have the same behaviour as in the match operator.
Closes: #6407
* Do not filter titles if suffix/removesuffix operand is empty
Issue: #6407
* Failing test
* Fix underlying problem
* Less naive fix
Now we make sure we maintain the sort order of the titles array when adding a new tiddler
* Fix failing tests
* Refactor filter tests to repeat them with different store orderings
* Revert "Fix failing tests"
This reverts commit ee03ee57f5.
* Refine fix to retain stylesheet ordering
The order of tiddlers in the HTML file uses localeCompare(), and that determines the insertion order. So if we want to be compatible with older versions we have to use localeCompare() to order tiddlers, not a plain sort()
* Don't sort shadow tiddlers
Instead rely on the existing ordering
* Fix failing tests, take 2
I think that all of these changes are explained by the store no longer retaining insertion order, but now using localecompare ordering
* Fix tests from #6327
