EINTR errors are recoverable, just need to read() again.
Randomly while looking at files in the file browser the contents view will be empty or files will not be shown when clicking a discloser triangle. Seems to happen more when running in XCode than when running by itself.
When the Subject column is not the first column in the git history table the contextual menu for refs and dragging and dropping refs don't work.
Offset the mouse click's x value by the x origin of the Subject column.
In PBRefController look up the column index of the subject column and check that it's the clicked column (don't assume it's the first one).
Apparently the setFeedURL: method is persistent, so if someone ever used
a nightly, it would keep the debug URL. That's not what I intended (I wanted
to use a different URL for debugging purposes), so let's just remove it.
Since 0b81911d we have a nicer Git version we can use, so we no longer
require this hack. It also interferes with Sparkle's updating system,
so just remove it and use the same versioning system as we use for
the short string.
I've seen this hang or crash a few times, so I hope this works better.
Instead of running a task in a separate thread, we just let it go through
the run loop and catch it when the task is done.
This ruins the second subview in the history view, but I don't think
anybody ever used that, so I'm going to remove it.
This class would use the location information provided by
regex(3) as range for for a substring. However, the information
regex(3) returns is a byte-based, while NSString works on characters.
This can cause a problem when there are UTF-8 characters in the string,
as the wrong subsstring will be returned.
This is fixed by taking the UTF bytesequence, and extracting a substring
from that, rather than using NSString's own substring method
Some people have had Serif-fonts in their interface where no specific
fonts for classes/ids were specified. This sets the default-font to
'Lucida Grande' in order to have a nice non-serif on any system.
Signed-off-by: Johannes Gilger <heipei@hackvalue.de>
This uses the same mechanisms as git to determine whether a file is
binary: By simply searching the first 8000 bytes for a 0-byte. This
gets rid of the call to "file" and is a much cleaner and shorter
implementation.
Signed-off-by: Johannes Gilger <heipei@hackvalue.de>
Git 1.6.0 was released over a year ago now, so it shouldn't be too hard
to get people to upgrade to that. We need 1.6.0 for the --children revlist
option.
Requiring Git 1.6.0 or higher means we can also revert
a94981f35 (RevList: Don't rely on the presence of %x00) sometime.
This patch prevents the plaintext display of files with binary content
in tree-view by connecting the content to the textContents attribute.
PBGitTree is extended with the method textContents, which returns the
textual representation of a PBGitTree-object. The methods first checks
the output of "git check-attr binary <file>" to see if the user
set/unset the binary attribute manually. Then it checks for common
binary file-extensions. If this method can't determine whether the file
is binary, the file-content is loaded and Unix "file" is run on the
first 100 bytes of the file to make a decision.
It also adds the -[PBGitTree fileSize] method to check the size
of the file before actually loading its contents.
Signed-off-by: Johannes Gilger <heipei@hackvalue.de>
Edited-by: Pieter de Bie <pdebie@ai.rug.nl>
* pu/jg/docs:
Update releasenotes, as autorefresh won't make it into 0.7
ReleaseNotes for v0.7 with author information
Site: W3C-Valid and CSS/HTML corrections
UserManual: Rewrite most parts for GitX 0.7
GitX 0.7 Release-Notes
UserManual: Add disclaimer as to what GitX is
Conflicts:
Site/UserManual/text/01 - Introduction.markdown
Include Dave's reload feature and author-information for the most
important parts. Also a list of contributors since v0.6.3 from the
shortlog.
Signed-off-by: Johannes Gilger <heipei@hackvalue.de>
* Makes all the pages (except "See it") W3C Valid
* Adjusts a few texts here and there a little bit
* Makes the menubar span the whole div and display correct on FF/Linux
* Moves "GitX" in the header a few px down to center it
Signed-off-by: Johannes Gilger <heipei@hackvalue.de>
This patch adds a CFBundleGitRevision key to Info.plist which is set to
the output of "git describe" when building.
The menu-entry for "About GitX" was reconnected to a custom method in
the AppController, which reads the CF-string from the .plist and also
indicates if the build is a DEBUG-build.
Signed-off-by: Johannes Gilger <heipei@hackvalue.de>
This rewrites most of the UserManual for GitX 0.7 and also updates the
screenshots to reflect the latest GitX-version.
Signed-off-by: Johannes Gilger <heipei@hackvalue.de>
Add a disclaimer about the intended use of GitX. A lot of people seem to
think GitX is a way to make using a complex system easy. Other think
that they can learn git by clicking their way around GitX.
Signed-off-by: Johannes Gilger <heipei@hackvalue.de>
This prepares the GitX 0.7 release-notes and also fixes two small bugs
in the previous notes.
It also fixes a CSS-bug which makes the gitx-website display the same
font with Firefox and Safari instead of using a serif with FF.
Signed-off-by: Johannes Gilger <heipei@hackvalue.de>
Add a disclaimer about the intended use of GitX. A lot of people seem to
think GitX is a way to make using a complex system easy. Other think
that they can learn git by clicking their way around GitX.
Signed-off-by: Johannes Gilger <heipei@hackvalue.de>
When rewriting big parts of commit.js, JD forgot to use the new function
for the "Discard hunk" functionality as well. This fixes that.
Signed-off-by: Johannes Gilger <heipei@hackvalue.de>
Nathan Kinsinger