Mistakenly, I had changed the tiddler widget to refresh itself when the
value of the target tiddler changed. This is not in fact necessary; it
only needs to refresh itself when the identity of the target tiddler
changes.
Fixes#744
Previously we were using a message `tw-auto-save-wiki` to trigger an
autosave. The message was generated by certain UI actions such as
saving a tiddler. The trouble was that the message was being processed
before the wiki change event for the accompanying change had had a
chance to percolate. The end result was that the dirty indicator was
staying lit when using autosave.
The new approach abandons the autosave message and instead triggers the
autosave in the wiki change event when a relevant change occurs.
One happy side effect of these changes is that the dirty indicator now
works as expected with the client server edition - ie, when typing in a
draft tiddler the dirty indicator will flash briefly, and then clear
when the sync mechanism has completed saving the draft.
Fixing problems caused by c4b76ceb0b:
* We still need to initialise the saver-handler even when syncing to a
server, otherwise offline snapshots can’t be saved
* We need to override the default save template a bit further up the
stack, to avoid the server side serving the offline version of the wiki
at `/`
Fixes#717
The issue was that under Windows we generate text nodes that contained
CRLF as a linebreak (rather than just LF as usual).
The subtle problem is that when these strings are placed in the DOM via
createTextNode(), the CR character is treated as a printable character,
not whitespace. When creating DOM notes with innerHTML or as part of a
static HTML document the HTML parser will strip out the CR characters.
The hacky solution is to manually remove CRs before building the text
node.
@pmario and @welford - I’m not presenting this as a fix for #717
because I’m still not in a position to reproduce it.
However, I found this during a review of newline handling code, and
would be interested if it is implicated in the problems you are
reporting.
Importing an encrypted wiki ordinarily doesn’t place the password in
the password vault on the basis that one ought to be able to import
from a file without automatically inheriting its password.
Now there’s a configuration option that can be used by the upgrade
plugin to cause the password vault to be updated with any password
entered by the user. The end result is that the user only needs to
enter their password once.
Previously any refreshing of the content of a modal would cause a
crash. The problem is the way that we steal the root widget for the
render trees used in the modal. The root widget is tied to the
container DOM node for the main content area, which isn’t actually a
parent of the modal DOM nodes, hence the confusion for the refresh
mechanism.
We were taking the commands expanded from the build targets and placing
them at the end of the queue. That caused a problem whereby the
prevailing output folder was always the last one in the command token
list.
Now we splice the new commands into the command token list at the
current position.
The new importvariables widget imports macro/variable definitions from
the specified tiddlers and makes them available to its children.
Allows us to split PageMacros up into separate tiddlers.
We still support loading macros from $:/core/ui/PageMacros to help
people upgrading.
Fixes#644 and #559
A bunch of little changes that together enable external image support.
Try:
```
tiddlywiki editions/tw5.com --verbose --build externalimages
```
Then open `externalimages.html`, look for the images in the more/types
tab of the sidebar, open them and verify that they are set with an
external SRC attribute, not a data URI.
Now a link to a single tiddler like http://tiddlywiki.com/#HelloThere
will just open that single tiddler (the old behaviour was to also open
the default tiddlers)
Previously, widgets were reading variables from themselves or their
cascaded ancestors. That means that if a widget sets a variable and
then reads the same variable, it will get the same variable back. That
sounds reasonable, until you consider a widget that wants to modify a
variable - eg the tiddler macro. For example:
```
<$tiddler tiddler={{!!report}}>
<$transclude mode="block" />
</$tiddler>
```
Here we first evaluate the `{{!!report}}` reference, which involves
reading the currentTiddler variable, looking up the tiddler, and
retrieving it’s `report` field. The next the tiddler widget is
refreshed, it will use the newly set currentTiddler as the basis for
resolving the `{{!!reference}}` reference.
The fix is to get variables from ancestors, but continue to set them on
ourselves.
We were parsing the boot tiddlers, making them into a widget and then
refreshing the widget tree. The problem is that subsequent chances to
the boot tiddlers themselves wouldn’t be picked up as part of the
refresh.
Now we indirectly parse those UI boot tiddlers through a transclusion,
which does get refreshed in the desired way.
Now we respond dynamically to changes in the location hash in the URL
bar. It means that you can do links in HTML as `<a
href=“#HelloThere”>go</a>` and in Markdown as `[example
link](#HelloThere).`
We still need to make startup.js more modular
At this point we respect any permalink at startup, but we don’t yet
dynamically update the permalink, nor do we respond to ongoing
permalink changes.
The permalink separator being `%00` seems like it might be a bit
controversial. It buys us not having to wrap tiddler titles in double
square brackets if they contain spaces.
Another thing is that this scheme doesn’t support tiddler filters; the
plan is to support them like this:
http://tiddlywiki.com/#!Target%00%00[tag[task]sort[created]]
1. Moved some methods out of boot.js because they are not needed until
after bootup
2. Added alternate message for editing an overridden shadow tiddler
3. Minor style tweaks
Replace this with a $tw.wiki.isModifiedTiddler(title) as part of the
wiki object. This allows it to be used outside of the current Wiki which
can change.
Should only display the confirmation if the shadow tiddler has not been
overridden in the first place. It checks this by looking for the
existence of a modified field for which the default system based shadow
do not have until a user changes them.
This addresses the second line item on issue #570
We will need new translations for the added string
`ConfirmEditShadowTiddler`
When saving a tiddler we check to see if the tiddler has changed
(isModified) if it hasn't then bounce the event to tw-cancel-tiddler
instead.
Addresses first line item in issue #570
This change is likely to break most existing scripts that call
TiddlyWiki.
TL;DR - output paths are now relative to the editions/output folder,
rather than to the current folder
See [[Notes for upgrading to 5.0.11-beta]] for details.
By rearranging the `[all[]]` operator we are able to ensure that shadow
tiddlers get processed before ordinary tiddlers. This makes it easier
to create custom stylesheets that override the core.
Simple optimisation whereby we defer the main refresh cycle when only
draft tiddlers have been modified.
We defer for 400ms, and keep extending the delay at each fresh draft
modification. The effect is that if the user is using the preview then
they’ll need to pause typing for 400ms before the preview is updated.
Fixes#470, #454, and maybe #206.
Refactoring page rendering so that we can fix the performance issues
with editing drafts. The plan is to defer the refresh cycle if all the
changes are to draft tiddlers.
We need to do a bit more of this to get startup.js into better shape
(and more extensible).
We need to send messages to the parent so that we don’t set up an
infinite loop by sending `tw-navigate` messages. Also we need to pass
along the target of the navigation.
load.js references the encoding set in boot.js when loading a file.
boot.js can now register file type with different deserialization from
their actual type
This is an experimental module to help us measure the performance of
the refresh cycle and the filter mechanism. Not intended to replace the
performance measurement features in browser developer tools, just to
make it easier to automate performance checks cross-browser.
The immediate purpose is to help in refactoring the filter mechanism.
The recent change to encapsulate the wiki store “tiddlers” object has
hurt the performance of filters, and it’s going to be helpful to have
decent measurements while refactoring that code.
I’m still not convinced that this stuff should be in the core, and may
well end up removing it after the present refactoring cycle.
