In TreeView:on_mouse_pressed() we need to find the directory a
relative filename belongs to from its absolute filename.
The code was using string.find to locate the relative filename
within the absolute path but in some very specific cases we can
find a pattern which is not the right-most one leading to a
wrong directory name.
Fix the error by adding a loop to make sure we find the right-most
match. The standard Lua library has not a string.rfind to make a
reverse search.
Add a check to avoid trying to updating a topdir project directory.
Groups together consecutive mouse move events like done in core.step()
lua function but on the C side.
It does not introduce any meaningful speedup but it theory is more efficient and
simplifies the Lua code.
The simplification of the Lua code alone is enough to justify this change?
On macos 11.2.3 with sdl 2.0.14 the keyup handler for cmd+w was not
enough. Maybe the quit event started to be triggered from the keydown
handler? In any case, flushing the quit event there too helped.
Use resolved executablePath instead of resourcePath to allow lanching
the lite-xl binary via a symlink, like typically done by Homebrew:
/usr/local/bin/lite-xl -> /Applications/lite-xl.app/Contents/MacOS/lite-xl
The resourcePath returns /usr/local in this case instead of
/Applications/lite-xl.app/Contents/Resources, which makes later
access to the resource files fail. Resolving the symlink to the
executable and then the relative path to the expected directory
Resources is a workaround for starting the application from both
the launcher directly and the command line via the symlink.
In order to stay simple and closer to the lite's design principles we
deprecate the core.add_save_hook function and the related mechanism.
Instead we now directly override the Doc:save() method.
The method is already overrided from core.init to add the automatic
reloading of style when user's module is saved.
The cleanup is related to the discussion in issue #229.
The bug was actually due to a presence of a ghost tab scrolling button
in all the views.
We need to check if the node has multiple views, if not there are no tabs
and therefore no scrolling button areas so we return nothing from the
method Node:get_scroll_button_index().
Close#216
When the number of files in a project directory is above the max
limit switch back to a mechanism to read directory content only
when the corresponding folder is expanded in the treeview.
When the command core:find-file is invoked the command core:open-file
is executed instead because the complete list of the project's
files is not available.
When a project search is done we search through all the files within
the project dir without indexing them.
Address issues #217#203#183.
* Indent enhancements.
* Fixed to match style guidelines.
* Added in useful explanatory comment.
* Changed which selection we're using, as we don't want this kind of wrapping to happen.
* Fixed bug involving lines full of whitespace.
* Removed unecessary commit.
* Actually reverted function, so that we don't screw up commenting.
* Fixed hard tab issue.
In some cases rencache was using a FontDesc pointer that was actually freed by
Lua giving segfaults errors.
In addition, some FontDesc object were not freed in some cases if the
rencache_end_frame was not called when performing the "restart" command.
The invalid access problem can happen because rencache keep some pointers to
FontDesc object but these are Lua userdata that Lua can dispose of. This
situation is prone to hard errors and we should avoid to keep pointers to
objects managed by Lua.
To this purpose we use luaL_ref/unref to bind the FontDesc into the Lua's
registry while rencache need them. We still keeps pointer to FontDesc object but
using luaL_ref we are assured they will not be disposed by Lua.
Since we are using luaL_ref/unref to inform the GC about when the objects are in
use we can now finalize the objects directly when Lua collects them. Previously
the GC metamethods was issuing a FREE command to rencache and the font was
actually freed only from the rencache_end_frame function.
* Cleaned up tokenizer to make subsyntax operations more clear.
* Explanatory comments.
* Made it so push_subsyntax could be safely called elsewhere.
* Unified terminology.
* Minor bug fix.
* State is an incredibly vaguely named variable. Changed convention to represent what it actually is.
* Also changed function name.
* Fixed bug.
George Sokianos
Francesco Abbate
Adam
Björn Buckwalter
Ferdinand Prantl
redtide
Janis-Leuenberger
liquidev
Alexandr "Nil" Shchelov