| Age | Commit message (Collapse) | Author |
|---|
|
Probably a good idea to allow more than 128 runs of a program.
|
|
Search results will now be sorted by history first, then by the position
of the match, such that a match at the start of a result will appear
before a match later in the result.
|
|
This was caused by some unneeded leftover logic that resulted in
surface_draw being called without a corresponding entry_update.
|
|
|
|
|
|
Tofi should now appear on whatever the compositor's default output is
(usually the focused one), unless --output has been given.
|
|
|
|
|
|
The --output option can be passed to select a specific output to appear
on. Which output is chosen otherwise is currently random, as I don't
know yet how to determine which one the user's currently on.
|
|
|
|
|
|
Also change --font-name option to just --font.
|
|
|
|
|
|
|
|
Also improve text on non-hidpi screens.
|
|
This improves start-up performance for large windows.
|
|
|
|
Implement horizontal mode for Pango, and add min-input-width option.
|
|
|
|
Currently the clipping region is still a plain rectangle bounded by the
outer rounded rectangle. A rounded rectangle clip would be nice, but
unfortunately Cairo leaves antialiasing artifacts if we do this.
|
|
|
|
Tofi now uses Harfbuzz if a file is passed to --font-name, and Pango
otherwise.
|
|
The scale factor is now only used to scale font sizes, not all Cairo
drawing operations. This makes pixel-sized options correct.
|
|
Invoking 'tofi' now acts like dmenu, expecting newline-separated options
on stdin.
Also fix history file handling and add an option to disable it.
|
|
|
|
|
|
|
|
|
|
Single letter style arguments have been removed.
|
|
There are too many really to use single-character args, so the next step
should be a config file.
|
|
A list of commands is now stored in $XDG_CACHE_HOME/.cache/tofi-compgen,
and regenerated as necessary.
|
|
- Split the compgen and history sorting parts of compgen(), for future
dmenu-like work.
- Add a separate tofi-compgen executable.
- Remove harfbuzz-glib usage, as we shouldn't be doing any complicated
unicode stuff.
|
|
|
|
|
|
|
|
|
|
Previously integers were used, effectively removing any sub-pixel
positioning.
|
|
We don't actually need a separate pango layout for each piece of text,
as we only use them one-by-one anyway, so just use one instead.
|
|
|
|
We don't actually need to initialise our second Cairo context / surface
until after the first one has been painted to the screen. This commit
therefore delays this initialisation (or at least the expensive memcpy),
granting a significant reduction in startup time.
The downside is that main() and entry_init() are now tied together
somewhat, but hopefully the comments help.
|
|
By pointing Cairo at the mmap-ed file used to create wl_shm buffers, we
can eliminate a memcpy() on every draw, providing a decent speedup
(especially for large window sizes).
This comes at the expense of having to keep track of two Cairo contexts,
one for each of our two buffers used for double buffering. Additionally,
a single memcpy() is still required for initialisation of the second
buffer, so the startup latency isn't affected much.
|
|
|
|
Rather than using cairo_paint() and re-drawing the middle of the window
multiple times, use cairo_rectangle() + cairo_stroke(). This especially
helps for large (e.g. fullscreen) windows.
|
|
Another source of slow startup is initialising Pango. If the user
supplies a ttf file, we can skip any Pango stuff and use Harfbuzz
directly with Cairo to do our font rendering, providing a large speedup.
|
|
eglInitialize() is slow (~50-100ms), and uses a fair amount of memory
(~100 MB). For such a small, simple program that just wants to launch as
quickly as possible, wl_shm performs better.
|
|
This implements a rofi-like run cache. Other smaller changes include
simplification of resize logic now that there's only one surface.
|
|
- Remove the background image and libpng dependency
- Add a prompt
- Add xmalloc with out-of-memory handling
- Add beginnings of a rofi-like run cache
|
|
The subsurface was playing havoc with layer shell stuff, and I don't
think any potential efficiency gain is worth the complicated code to
work out how big the subsurface should be. Instead, the entry code now
just draws directly onto the main surface.
Damage information should be passed to glTexSubImage2D() in future, to
avoid redrawing the entire window every keypress.
|
|
|
