The function will unfold anything completely within the range that also touches one of the range’s end points.
This is because the layout will show adjoining folded ranges as a single image, so if the user clicks this, it is unclear which of the ranges to actually unfold (the layout return a range that encompass all the adjoining folds).
A file setting like “src/*.cc” is arguably more specific (and thus more local) than targeting the scope “source.c++”, so it makes sense that the file settings rank higher than scope settings.
Some users also find it natural to target file extensions over scopes, but as TM will store some learned settings with a scope selector, it would previously lead to confusion, when a user was unable to e.g. change tab size for “*.php” (because TM had a learned setting for “source.php”).
This changes the authentication mechanism from passing an authorization token / API key to using basic auth. Basic auth didn’t work when I initially tested the upload API, but hopefully GitHub has since added it (I did request it, but didn’t get a reply, so no idea if the omission was an oversight or by design).
This can be used to force a file to open in a new window, e.g.:
TM_PROJECT_UUID=$(uuidgen) mate README.md
Likewise, it can be used to force multiple invocations of mate to use the initially opened window (by re-using the UUID).
Closes#879
We already commit the values when focus is lost, but it’s possible to initiate a find operation while focus is still in the find dialog text fields (via the menu).
Fixes#1183
Creating a release with attachment is non-atomic in that the release is first created and then we upload to that release but our build file has it as a single action, so this action should be able to handle partial success from a previous execution.
The problem is that with a bad internet connection there is a very high probability that “codesign --deep” will fail due to having to contact the timestamp server for each embedded executable object.
Long-term we probably want to sign the sub-targets as they are produced and/or link the entire thing together so that there is only a single sign target (plus embedded plug-ins).
The services support in the file browser was mainly added to allow Quicksilver to grab the selection, I do not recall if this is why the selection was (also) made available as a regular string (in addition to the “file names pasteboard type”).
Previously when constructing a scm URL from a path, if scm was not enabled
we returned a file URL. This could result in some unexpected behavior.
For instance, when a directory was selected in the file browser,
activating "SCM Status" would just open the untracked directory.
This patch, insteads, adds a query to the scm URL that can be parsed to
determine how to handle the scm status request later on.
Accessibility visual bounds for a range of a single softline should always
be also visually restricted to the softline. This is in order to be consistent
with implementation in Apple's apps (e.g. TextEdit).
Fixes#1149.
VoiceOver expects AXLineForIndex and AXRangeForLine to be consistent with
the movement by line using arrow up/down. That means it expects "line"
to mean softline.
VoiceOver also distinguishes commands to move by:
* paragraph (hardline): VO-Shift-PgUp/PgDown
* line (softline): VO-up/down
These are available when interacting with the text area (VO-Shift-down).
This commit therefore fixes the AXLineForIndex and AXRangeForLine
to make "line" mean "softline" instead of "hardline" as was the case
until now.
Now links in text (e.g. links to RFCs after installing the Hyperlink Helper bundle)
are announced as such with VoiceOver and it is possible to search for them with
VO-Cmd-L.
TODO:
* launching links with VO-Space is not supported yet, but Enter (Fn-Return)
does the same thing and is even more reliable.
* link URL is not supported, so it is not possible to learn the URL
with VO-Shift-U.
This is a document with information I found most useful as a
new contributor to TextMate - basic stuff a new contributor is likely
to get in contact with.
It's of course incomplete, sparse, can be not well worded and even contain
mistakes. But someone had to make the first shot :-)
Previously we only scanned the folder when creating the favorites chooser singleton which meant items added or removed after the chooser had been used, were not reflected before a relaunch.
Fixes#1182
Normally a line matched by both patterns will be decreased by one unit compared to the non-ignored lines immediately above and below. If however the line above was not indented, the line would in theory be decreased to -1 and thus just go back to zero indent for the line following it.
Previously only the text view’s background would affect the window’s opaque property, which made it impossible to have a solid text view background with a transparent gutter.
This would happen if last line was not newline terminated.
The fix does not take into consideration that last line might contain non-ASCII, in which case it should be highlighted.
With responsive scrolling we need to return NO from isOpaque in order to get transparency and we cannot overload the scrollWheel: method (which we use to forward mouse events from the gutter to the text area).
I have also received a few crash reports that might be caused by responsive scrolling and one user has reported “sticky” scroll behavior.
So for now better just disable it, and see if it fixes all these issues.
Fixes#1135
We were using gnutar to avoid the extended header keywords that bsdtar add to the archive (for non-ASCII file names) as those will cause older versions of gnutar (like the one included with 10.8 and earlier) to give warnings/errors when extracting the archive.
Issue #1180
