* First commit
* Add throttling of saves
Now we refuse to save a tiddler more often than once per second.
* Wait for a timeout before trying again after an error
* Modest optimisations of isDirty() method
* Synchronise system tiddlers and deletions from the server
Fixes two long-standing issues:
* Changes to system tiddlers are not synchronised from the server to the browser
* Deletions of tiddlers on the server are not propagated to browser clients
* Make sure we update the dirty status even if there isn't a task to perform
* Replace save-wiki button with popup sync menu
* Remove the "Server" control panel tab
We don't need it with the enhanced sync dropdown
* Add indentation to the save-wiki button
* Fix spacing in dropdown menu items
* Switch between cloud icons according to dirty status
* Add a menu item to copy syncer logs to the clipboard
* Improve animated icon
* Remove indentation from save-wiki button
@pmario the annoying thing is that using `\trim whitespace` trims significant whitespace too, so it means we have to use <$text text=" "/> when we need a space that won't be trimmed. For the moment, I've removed the indentation but will keep thinking about it.
* Further icon, UI and copy text tweaks
Move the icons and styles from the core into the TiddlyWeb plugin
* Clean up PR diff
* Tweak animation durations
* Break the actions from the syncer dropdown into separate tiddlers
@pmario I think this makes things a bit easier to follow
* Refactor syncadaptor creation and logging
The goal is for the syncadaptor to be able to log to the same logger as the syncer, so that the "copy syncer logs to clipboard" data is more useful.
* Don't transition the dirty indicator container colour, just the SVG's colour
* Only trigger a sync for changes to tiddlers we're interested in
Otherwise it is triggered by the creation of the alert tiddlers used to display errors.
* Restore deleting local tiddlers removed from the server
(I had commented it out for some testing and accidentally commited it).
* Guard against missing adaptor info
* We still need to trigger a timeout when there was no task to process
* Avoid repeatedly polling for changes
Instead we only trigger a timeout call at if there is a pending task (ie a tiddler that has changed but isn't yet old enough to save).
* Lazy loading: include skinny versions of lazily loaded tiddlers in the index.html
* Introduce _is_skinny field for indicating that a tiddler is subject to lazy loading
* Remove savetrail plugin from prerelease
It doesn't yet work with the new syncer
* Make the savetrail plugin work again
* Clear outstanding alerts when synchronisation is restored
* Logger: only remove alerts from the same component
Missed off 9f5c0de07
* Make the saving throttle interval configurable (#4385)
After switching Bob to use the core syncer the throttle interval makes saving feel very sluggish compared to the message queue setup that I had before.
The editing lock that I use to prevent conflicts with multiple users doesn't go away until the save is completed, and with the 1 second delay it means that if you edit a tiddler and save it than you have to wait one second before you can edit it again.
* Tweaks to appearance of alerts
* Exclude temp tiddlers from offline snapshots
Otherwise alerts will persist
* Tweak appearance of status line in dropdown
* Update release note
* Web server: Don't include full path in error messages
Fixes#3724
* In change event handler check for deletions
* Disable the official plugin library when the tiddlyweb plugin is loaded
* Hide error details from browser for /files/ route
See https://github.com/Jermolene/TiddlyWiki5/issues/3724#issuecomment-565702492 -- thanks @pmario
* Revert all the changes to the relationship between the syncer and the syncadaptor
Previously we had some major rearrangements to make it possible for the syncadaptor to route it's logging to the logger used by the syncer. The motivation is so that the "copy logs to clipboard" button is more useful.
On reflection, changing the interface this drastically is undesirable from a backwards compatibility perspective, so I'm going to investigate other ways to achieve the logger sharing
* Make the tiddlyweb adaptor use the syncer's logger
So that both are availavble when copying the syncer logs to the clipboard
* Update release note
* Support setting port=0 to get an OS assigned port
Quite useful
* Update code comment
* UI: Use "Get latest changes from server" instead of "Refresh"
* Add getUpdatedTiddlers() method to syncadaptor API
See https://github.com/Jermolene/TiddlyWiki5/pull/4373#issuecomment-573579495
* Refactor revision handling within the syncer
Thanks @pmario
* Fix typo in tiddlywebadaptor
* Improve presentation of errors
See https://github.com/Jermolene/TiddlyWiki5/pull/4373#issuecomment-573695267
* Add docs for getTiddlerRevision()
* Remove unused error animation
* Update comment for GET /recipes/default/tiddlers/tiddlers.json
* Optimise SVG cloud image
* Add optional list of allowed filters for get all tiddlers route
An attempt to address @Arlen22's concern here:
https://github.com/Jermolene/TiddlyWiki5/pull/4373#pullrequestreview-342146190
* Fix network error alert text translatability
* Fix error code and logging for GET /recipes/default/tiddlers/tiddlers.json
Thanks @Arlen22
* Flip GET /recipes/default/tiddlers/tiddlers.json allowed filter handling to be secure by default
* Validate updates received from getUpdatedTiddlers()
* Add syncer method to force loading of a tiddler from the server
* Remove the release note update to remove the merge conflict
* Fix crash when there's no config section in the tiddlywiki.info file
* Use config tiddler title to check filter query (merge into fix-syncer) (#4478)
* Use config tiddler title to check filter query
* Create config-tiddlers-filter.tid
* Add config switch to enable all filters on GET /recipes/default/tiddlers/tiddlers.json
And update docs
* Fix bug when deleting a tiddler with a shadow
Reported by @kookma at https://github.com/Jermolene/TiddlyWiki5/pull/4373#issuecomment-604027528
Co-authored-by: jed <inmysocks@fastmail.com>
Co-authored-by: Arlen22 <arlenbee@gmail.com>
* Add tests for backlinks
* Add backlinks indexer
* Use backlinks indexer in getTiddlerBacklinks if available
* Extract link extraction into its own method
This way we can provide an arbitrary parse tree, rather than just a
title, which will allow us to compare lists of outgoing links between
versions of a single tiddler
* Use new extractLinks method in backlinks indexer
...rather than copy-pasting the implementation
* Remove ES6-isms
TiddlyWiki needs to work with browsers that only support ES5
* First pass at dynamic loading/unloading
* Show warning for changes to plugins containing JS modules
* Use $:/config/RegisterPluginType/* for configuring whether a plugin type is automatically registered
Where "registered" means "the constituent shadows are loaded".
* Fix the info plugin
The previous mechanism re-read all plugin info during startup
* Don't prettify JSON in the plugin library
* Indicate in plugin library whether a plugin requires reloading
* Display the highlighted plugin name in the plugin chooser
And if there's no name field fall back to the part of the title after the final slash.
Fixes#4082
This version removes selective updating of the tag index, instead completely clearing the index on each update. I'm investigating restoring that optimisation.
* First pass at modular wiki indexes
An exploratory experiment
* Fix tests
* Faster checking for existence of index methods
We don't really need to check the type
* Use the index for the has operator
* Fix typo
* Move iterator index methods into indexer modules
Now boot.js doesn't know the core indexers
* Fix up the other iterator index functions
* Fix crash with missing index branch
* Limit the field indexer to values less than 128 characters
* Fallback to the old manual scan if the index method returns null
* Sadly, we can no longe re-use the field indexer to accelerate the `has` operator, because the index now omits tiddlers that have field values longer than the limit
Still need to make the index configuration exposed somehow
* Rearrange tests so that we can test with and without indexers
We also need to expose the list of enabled indexers as a config option
* Test the field indexer with different length fields
So that we test the indexed and non-indexed codepaths
* Fix search method for search tokens spread across fields
Addresses GH #3636, which reports that if you're searching for "test
body", and "test" only appears in the title field, and "body" only appears
in the text field, 5.1.18's search method won't yield that tiddler as a
result, which appears to be a regression from the 5.1.17 behavior
* Add test for searching for multiple tokens across fields
Verifies GH #3636:
> If I create a tiddler in the empty edition with the title "Test tiddler" and content "Body content", searching the wiki for "test body" yields no results under either "title matches" or "all matches". Searching for either word individually turns up "Test tiddler", and repeating this in an empty wiki I created from the 5.1.17 tag causes "Test tiddler" to show up under "all matches".
* Added better handling for sortByList manual placements
If manual placement specifications show up in an inconvenient order,
sortByList, will go to the trouble of processing them in that order.
* Added tests to confirm solution to (#3296)
...That custom tag ordering will not choke when tiddlers get sorted after their dependencies have been placed around them
* Corrected list-after bug when referencing external titles
* Using more error-proof $tw.utils.hop in sortByList
* minor indentation correction in test-tags.js
I switched this optimisation off back in
ed35d91be6, in October 2013, as part of a
big refactoring of the parsing and widget mechanism. I’ve been meaning
to switch it back on for some time.
My rough measurements suggest that this optimisation can reduce
rendering time by 5-10%.
Typically for JavaScript, initialising an object as an array doesn’t
break anything because an array is an object. Anyhow, it should be an
object in this case.
