pl@ninthfloor.org [Fri, 11 Nov 2016 16:45:52 +0000 (17:45 +0100)]
Make strdump(), csidump(), print to stderr
The two functions strdump(), csidump() are called to show errors and
their output is introduced by a message printed to stderr. Thus, it it
more consistent to have them print to stderr.
Moreover stderr is unbuffered (at least on Linux), making problems
immediately visible.
Quentin Rameau [Sat, 22 Oct 2016 14:36:10 +0000 (16:36 +0200)]
Do not use color when font attributes are supported
If fontconfig gives us a font without the attributes we asked for,
display an alternative color instead.
Quentin Rameau [Sat, 22 Oct 2016 08:42:46 +0000 (10:42 +0200)]
st.1: add an entry for ISO-14755 shortcut
Quentin Rameau [Sat, 8 Oct 2016 14:22:18 +0000 (16:22 +0200)]
Add support for iso14755
We launch dmenu for getting a codepoint, then convert it and send it to
the terminal.
Klemens Nanni [Thu, 13 Oct 2016 14:28:50 +0000 (16:28 +0200)]
Add missing device path to '-l' example
Also, it's ttyS0 not ttySO.
Roberto E. Vargas Caballero [Wed, 14 Sep 2016 06:27:32 +0000 (08:27 +0200)]
Add parsing of DCS q sequences
These sequences are used to operate with sixels, but they are still
str sequences, so they are finished with \a, ST or with a C1 control
code. This patch also disables utf8 handling for the case of sixels.
Roberto E. Vargas Caballero [Tue, 13 Sep 2016 12:01:18 +0000 (14:01 +0200)]
Add support for enabling/disabling utf
There are some ocasions where we want to disable the enconding/decoding of utf8, mainly
because it adds an important overhead. This is partial patch for ESC % G and ESC % @,
where they modified the way that st reads and write from/to the serial line, but it does
not modifies how it interacts with the X window part.
Roberto E. Vargas Caballero [Fri, 9 Sep 2016 11:11:06 +0000 (13:11 +0200)]
Delete ncv capability from terminfo
We do not need to disable the previous ncv definition, because
there is not previous definition.
Quentin Rameau [Thu, 8 Sep 2016 10:57:43 +0000 (12:57 +0200)]
st.info: do not prevent st from displaying attributes
With ncv set to 3, we prevent st from displaying A_STANDOUT and
A_UNDERLINE with colors while our virtual terminal is capable of it.
Christoph Lohmann [Thu, 11 Aug 2016 14:30:29 +0000 (16:30 +0200)]
Update the LICENSE.
This is for the next release.
Christoph Lohmann [Thu, 11 Aug 2016 14:25:58 +0000 (16:25 +0200)]
0.7 release
Christoph Lohmann [Wed, 20 Jul 2016 11:56:15 +0000 (13:56 +0200)]
Change who's expanding tabs.
Christoph Lohmann [Wed, 20 Jul 2016 11:52:31 +0000 (13:52 +0200)]
Add some hint to have the pseudo terminal in the right mode.
If you don't make sure that the terminal does not expand tabs to spaces, of
course such a setting won't work.
Christoph Lohmann [Wed, 20 Jul 2016 11:00:43 +0000 (13:00 +0200)]
Add comment about tabspaces.
st.info needs to be changed too, when tabspaces are changed.
Alive 4ever [Sun, 10 Jul 2016 06:38:04 +0000 (06:38 +0000)]
Consistent Alt+BackSpace behavior
The default config specifies BackSpace as "\177". The default behavior
should persist across modifier keys, commonly Mod1 (Alt or Meta) which
is widely used to delete a word on readline and text editors, notably
Emacs.
This will make Alt+BackSpace behaves as expected, i.e. sends "\033\177"
instead of "\033\010" as previous default behavior.
Signed-off-by: Christoph Lohmann <20h@r-36.net>
Christoph Lohmann [Fri, 3 Jun 2016 13:02:32 +0000 (15:02 +0200)]
Use XftFontMatch in place of FcFontMatch.
git am -s didn't like your patch:
From: Mark Edgar <medgar123@gmail.com>
XftFontMatch calls XftDefaultSubstitute which configures various match
properties according to the user's configured Xft defaults (xrdb) as well as
according to the current display and screen. Most importantly, the screen DPI
is computed [1]. Without this, st uses a "default" DPI of 75 [2].
[1]: https://cgit.freedesktop.org/xorg/lib/libXft/tree/src/xftdpy.c?id=libXft-2.3.2#n535
[2]: https://cgit.freedesktop.org/fontconfig/tree/src/fcdefault.c?id=2.11.1#n255
v4hn [Thu, 19 May 2016 10:16:57 +0000 (12:16 +0200)]
delete clipboard properties after pasting them
https://tronche.com/gui/x/icccm/sec-2.html#s-2.4 specifies:
> Once all the data in the selection has been retrieved,
> the requestor should delete the property in the SelectionNotify request
Most Clipboard-Owners ignore whether or not the property is already set,
so this is mostly a cosmetic change to keep the windows property list clean.
However, at least synergy decides to wait for the requestor to delete
the properties if they are already set by a previous paste (from synergy).
Signed-off-by: Christoph Lohmann <20h@r-36.net>
Roberto E. Vargas Caballero [Fri, 15 Apr 2016 05:58:26 +0000 (07:58 +0200)]
Remove stupid assignation in memcpy()
Tor Andersson [Wed, 9 Mar 2016 16:11:57 +0000 (17:11 +0100)]
st: Fix off-by-one error when calculating character width.
LEN(str) is one larger than strlen(str) because it also counts the zero
terminator. The original code would include the .notdef glyph (since it'll
try to encode character 0, which gets encoded to the .notdef glyph) when
measuring the average dimensions of printable ascii characters.
This causes problems with fonts like GNU Unifont where the .notdef glyph is
not the same width as the usual half-width characters.
Signed-off-by: Christoph Lohmann <20h@r-36.net>
Ton van den Heuvel [Mon, 7 Mar 2016 21:18:12 +0000 (22:18 +0100)]
Fix vertical character alignment in some cases
The y-position of a character found by asking fontconfig for a matching
font does not take the border pixels into account, resulting in a
slightly misaligned vertical position.
Signed-off-by: Ton van den Heuvel <tonvandenheuvel@gmail.com>
Signed-off-by: Christoph Lohmann <20h@r-36.net>
Christoph Lohmann [Tue, 8 Mar 2016 12:55:22 +0000 (13:55 +0100)]
I like empty lines.
Ryusei Yamaguchi [Tue, 8 Mar 2016 03:26:04 +0000 (12:26 +0900)]
Measure the single advance width with a heuristic method
This fix is needed to use dual-width fonts, which have double-width
glyphs (e.g. CJK unified ideographs).
Signed-off-by: Ryusei Yamaguchi <mandel59@gmail.com>
Signed-off-by: Christoph Lohmann <20h@r-36.net>
Christoph Lohmann [Mon, 7 Mar 2016 13:33:05 +0000 (14:33 +0100)]
xtermclear() is now done by xdrawglyphfontspecs()
Thanks Ton van den Heuvel for the proposal!
Lucas Gabriel Vuotto [Fri, 19 Feb 2016 18:59:49 +0000 (15:59 -0300)]
arg.h: fixed argv checks order
This prevents accessing to a potentially out-of-bounds memory section.
Signed-off-by: Lucas Gabriel Vuotto <l.vuotto92@gmail.com>
Signed-off-by: Christoph Lohmann <20h@r-36.net>
Quentin Rameau [Mon, 1 Feb 2016 11:20:33 +0000 (12:20 +0100)]
Fix forgotten bracket and duplicate option in usage()
Scratch the preceding patch, this one is more correct
(don't forget to 'git am --scissors' ;))
-- >8 --
Also reformat the strings in a saner layout
Signed-off-by: Christoph Lohmann <20h@r-36.net>
Quentin Rameau [Sat, 30 Jan 2016 16:54:41 +0000 (17:54 +0100)]
Extract ttyresize() out of cresize()
This way we can call cresize() to set the terminal size before creating
a tty or spawning a process, which will start with the correct size.
Signed-off-by: Christoph Lohmann <20h@r-36.net>
Christoph Lohmann [Sat, 30 Jan 2016 08:50:18 +0000 (09:50 +0100)]
Ok, no need for a separate version function.
This is just redundant metadata. Please add Java comment meta classes too.
Quentin Rameau [Fri, 29 Jan 2016 21:03:01 +0000 (22:03 +0100)]
Add -n option for setting WM_CLASS instance name
Signed-off-by: Christoph Lohmann <20h@r-36.net>
Quentin Rameau [Fri, 29 Jan 2016 21:02:40 +0000 (22:02 +0100)]
Reformat usage and separate version from it
Signed-off-by: Christoph Lohmann <20h@r-36.net>
Christoph Lohmann [Thu, 28 Jan 2016 17:09:11 +0000 (18:09 +0100)]
Enforce a terminal size to reduce race conditions in too efficient apps.
dvtm is too fast in starting up. It will then have a race condition in finding
the right. terminal size.
Christoph Lohmann [Thu, 14 Jan 2016 22:19:03 +0000 (23:19 +0100)]
Height is height.
Christoph Lohmann [Fri, 1 Jan 2016 13:18:43 +0000 (14:18 +0100)]
Bump year.
Christoph Lohmann [Sun, 20 Dec 2015 00:43:32 +0000 (01:43 +0100)]
Fixing the XClassHint setting in st.
Roberto E. Vargas Caballero [Sat, 21 Nov 2015 17:21:03 +0000 (18:21 +0100)]
Avoid recursive call to ttywrite from ttyread
Ttyread() calls to ttywrite, so if we check for reading before
that for writing in ttywrite we can get a circular call sequence.
Quentin Rameau [Sun, 8 Nov 2015 16:14:58 +0000 (17:14 +0100)]
arg.h: remove unused macros
ARGUM isn't used and ARGNUMF uses estrtol() that isn't defined anywhere.
Those were probably copied from sbase arg.h.
Signed-off-by: Christoph Lohmann <20h@r-36.net>
Christoph Lohmann [Fri, 6 Nov 2015 19:46:23 +0000 (20:46 +0100)]
Introduce lim in all ttywrite() checks.
Roberto E. Vargas Caballero [Fri, 6 Nov 2015 19:01:00 +0000 (20:01 +0100)]
Fix ttywrite()
ttywrite was assuming that if it could not write then it could
read, but this is not necessarily true, there are some situations
where you cannot read or write. The correct behaviour is to detect
if you can read or/and write.
Ivan Tham [Sun, 1 Nov 2015 02:53:56 +0000 (10:53 +0800)]
Now the mshortcuts are even more consistent.
Keep everyone happy
Signed-off-by: Christoph Lohmann <20h@r-36.net>
Christoph Lohmann [Sat, 31 Oct 2015 08:29:04 +0000 (09:29 +0100)]
Now the Shortcuts are more consistent.
Keep the debile happy.
Jason Woofenden [Sun, 11 Oct 2015 09:44:34 +0000 (11:44 +0200)]
fix bug where first selection snaps to lines
Quentin Rameau [Tue, 6 Oct 2015 08:29:44 +0000 (10:29 +0200)]
Fix the cursor colors selection
Signed-off-by: Christoph Lohmann <20h@r-36.net>
Christoph Lohmann [Mon, 5 Oct 2015 22:21:44 +0000 (00:21 +0200)]
Normalize the whole color selection in xdrawcursor.
Quentin Rameau [Mon, 5 Oct 2015 21:10:48 +0000 (23:10 +0200)]
Fix the cursor color when over selection.
If we want to show a custom selected cursor color, we must not set the
revert attribute to the drawn glyph.
Signed-off-by: Christoph Lohmann <20h@r-36.net>
Christoph Lohmann [Mon, 5 Oct 2015 21:05:38 +0000 (23:05 +0200)]
Make the cursor shapes more descriptive.
»IBeam« is now »Bar« because it's named like that in the source code.
Christoph Lohmann [Mon, 5 Oct 2015 20:59:04 +0000 (22:59 +0200)]
Snowman is everywhere.
Christoph Lohmann [Mon, 5 Oct 2015 20:54:44 +0000 (22:54 +0200)]
Style normalisation in the config.def.h.
This adds an awareness commit for the big key array too.
Christoph Lohmann [Mon, 5 Oct 2015 20:38:19 +0000 (22:38 +0200)]
The definition of the reverse cursor is now up to the user.
Christoph Lohmann [Mon, 5 Oct 2015 20:09:51 +0000 (22:09 +0200)]
Fixing the cursor movement in selections.
Before the fix the cursor wouldn't obey if it's in a selection. If it is
inside it will now change to the reverse. This patch also adds that the
defaultcs will be reversed for the manually drawn cursors.
Christoph Lohmann [Mon, 5 Oct 2015 19:18:34 +0000 (21:18 +0200)]
When the cursor is moved, clear the selection.
Christoph Lohmann [Mon, 5 Oct 2015 19:06:42 +0000 (21:06 +0200)]
Reverse the cursor on selection.
Christoph Lohmann [Mon, 5 Oct 2015 18:48:24 +0000 (20:48 +0200)]
More style changes. We forgot some switches.
Christoph Lohmann [Mon, 5 Oct 2015 18:08:21 +0000 (20:08 +0200)]
Small style change.
Christoph Lohmann [Mon, 28 Sep 2015 18:08:58 +0000 (20:08 +0200)]
There's no need for libXext to compile st.
dequis [Fri, 25 Sep 2015 03:56:15 +0000 (00:56 -0300)]
Fix extra bracketed paste markers when pasting >8kb
Before this patch, when pasting over BUFSIZE (8192 bytes here), st would
do the following:
\e[200~...8192 bytes...\e[201~\e[200~...remaining bytes...\e[201~
With this patch, the start marker is only sent when the offset is 0 (at
the beginning of selnotify) and the end marker is only sent when the
remaining bytes to read are 0 (at the end).
For short pastes, both conditions are true in the same iteration.
For long pastes, it removes the extra markers in the middle, keeping the
intended wrapping:
\e[200~...8192 bytes......remaining bytes...\e[201~
Signed-off-by: Christoph Lohmann <20h@r-36.net>
FRIGN [Tue, 22 Sep 2015 11:13:25 +0000 (13:13 +0200)]
dup() -> dup2()
gcc would warn about an unused result. We know it is 0 and dup()
can't fail in these circumstances, as we closed fd0 previously.
Using dup2() to do the same saves one line and shuts gcc up, bringing
us a clean build back.
Christoph Lohmann [Sat, 12 Sep 2015 14:43:32 +0000 (16:43 +0200)]
The times of bad fonts are over.
The antialiasing was false due to circumstances that do not exist anymore. We
need antialiasing on big screens with big fonts. Autohinting too.
Roberto E. Vargas Caballero [Thu, 10 Sep 2015 09:53:11 +0000 (11:53 +0200)]
Fix copy of line with len = 0
When a line has no any character linelen is 0, so last = &term.line[y][MIN(lastx, linelen-1)]
generated a pointer to the end of the previous line. The best thing we can do in this case
is to add a newline, because we don't have a glyph to print (and consult its state of
wrapping).
Roberto E. Vargas Caballero [Tue, 8 Sep 2015 15:00:20 +0000 (17:00 +0200)]
Change Pause to Break in shortcut for serial break
Jan Christoph Ebersbach [Tue, 8 Sep 2015 05:28:52 +0000 (07:28 +0200)]
Expose cursor shape in config.def.h
Signed-off-by: Christoph Lohmann <20h@r-36.net>
Roberto E. Vargas Caballero [Tue, 8 Sep 2015 10:10:11 +0000 (12:10 +0200)]
Add key to send a break to the serial line
Roberto E. Vargas Caballero [Tue, 8 Sep 2015 08:44:35 +0000 (10:44 +0200)]
Set default values of stty according to the man page
This configuration is basically 38400 8N1, without echo and
in raw mode. Kernel will not process any of the characters
sent by the user.
Christoph Lohmann [Mon, 7 Sep 2015 20:59:05 +0000 (22:59 +0200)]
Making st.1 more descriptive about -l and fix -l in st.c.
Christoph Lohmann [Mon, 7 Sep 2015 18:00:49 +0000 (20:00 +0200)]
Fix the st manpage for -l line.
Christoph Lohmann [Mon, 31 Aug 2015 13:26:21 +0000 (15:26 +0200)]
Reordering and adding control codes.
For completeness and documentation add all C1 control codes.
Christoph Lohmann [Sun, 30 Aug 2015 09:29:23 +0000 (11:29 +0200)]
Merge branch 'master' of ssh://suckless.org/gitrepos/st
Christoph Lohmann [Sun, 30 Aug 2015 09:28:35 +0000 (11:28 +0200)]
Don't read if we chunked the input data.
Roberto E. Vargas Caballero [Mon, 17 Aug 2015 09:25:38 +0000 (11:25 +0200)]
Do not mark as invalid UTF8 control codes
wcwidth() returns -1 for all the non visible characters, but it doesn't
necessarilly mean that they are incorrect. It only means that they are not
printable.
Quentin Rameau [Tue, 11 Aug 2015 15:47:12 +0000 (17:47 +0200)]
st.info: replace the acsc entry from xterm to urxvt
The current acsc entry, copied from xterm was not exposing capability to
display arrows.
Alex Kozadaev [Tue, 28 Jul 2015 09:30:14 +0000 (10:30 +0100)]
Adding mouse colour/shape settings
Christoph Lohmann [Fri, 24 Jul 2015 16:25:33 +0000 (18:25 +0200)]
Adding -T to the usage too.
Christoph Lohmann [Fri, 24 Jul 2015 16:24:10 +0000 (18:24 +0200)]
Add -T, as recommended by Dmitrij D. Czarkoff
Quentin Rameau [Fri, 24 Jul 2015 09:52:17 +0000 (11:52 +0200)]
Fix type for write(2) return variable.
A little fix in xwrite().
>From
3727d2e3344b57128ab51c7839795204f1f839ff Mon Sep 17 00:00:00 2001
From: Quentin Rameau <quinq@fifth.space>
Date: Fri, 24 Jul 2015 11:40:46 +0200
Subject: [PATCH] Fix type for write(2) return variable.
The allocated lengh of s fits into an integer so we can safely use
ssize_t here.
Mytchel Hammond [Fri, 24 Jul 2015 05:58:18 +0000 (17:58 +1200)]
added note about the low precedence of shell to config.def.h
Roberto E. Vargas Caballero [Fri, 24 Jul 2015 05:44:34 +0000 (07:44 +0200)]
Fix style in execsh
Christoph Lohmann [Fri, 10 Jul 2015 12:31:40 +0000 (14:31 +0200)]
Make the comment for the IM XFilter more understandable.
Christoph Lohmann [Fri, 10 Jul 2015 12:30:37 +0000 (14:30 +0200)]
Return style unification.
Christoph Lohmann [Fri, 10 Jul 2015 12:21:52 +0000 (14:21 +0200)]
Aligning the macros.
Christoph Lohmann [Fri, 10 Jul 2015 12:19:52 +0000 (14:19 +0200)]
Unix end of file.
Christoph Lohmann [Fri, 10 Jul 2015 12:19:31 +0000 (14:19 +0200)]
No inline declarations please.
Christoph Lohmann [Fri, 10 Jul 2015 12:17:51 +0000 (14:17 +0200)]
Typo and the missing incr atom from the patches.
Christoph Lohmann [Fri, 10 Jul 2015 12:15:39 +0000 (14:15 +0200)]
Implement chunked write to the cmdfd.
This is needed so big input like a paste of several megabyte does not clog our
I/O.
Christoph Lohmann [Fri, 10 Jul 2015 12:10:17 +0000 (14:10 +0200)]
Implement INCR transfers in the clipboard.
FRIGN [Fri, 10 Jul 2015 10:27:13 +0000 (12:27 +0200)]
Update LICENSE
Signed-off-by: Christoph Lohmann <20h@r-36.net>
FRIGN [Fri, 10 Jul 2015 08:29:53 +0000 (10:29 +0200)]
Don't treat clauses like functions
and add a space between the keyword and the parentheses.
Signed-off-by: Christoph Lohmann <20h@r-36.net>
FRIGN [Thu, 9 Jul 2015 21:59:50 +0000 (23:59 +0200)]
Use BSD-style function notation
Put the opening brace on a new line. This was already used for some
functions inside st.c.
Signed-off-by: Christoph Lohmann <20h@r-36.net>
FRIGN [Wed, 8 Jul 2015 21:56:55 +0000 (23:56 +0200)]
Unboolify st
This practice proved itself in sbase, ubase and a couple of other
projects.
Also remove the True and False defined in X11 and FcTrue and FcFalse
defined in Fontconfig.
Signed-off-by: Christoph Lohmann <20h@r-36.net>
FRIGN [Wed, 8 Jul 2015 21:49:25 +0000 (23:49 +0200)]
Remove insane *_FILENO and EXIT_* usage
Any system having different assignments than the usual 0, 1, 2 for
the standard file numbers and 0, 1 for the exit-statuses is broken
beyond repair.
Let's keep it simple and just use the numbers, no reason to fall
out of the window here and bend down for POSIX.
In one occasion, the ret-variable was not necessary. The check was
rewritten.
Signed-off-by: Christoph Lohmann <20h@r-36.net>
Christoph Lohmann [Tue, 7 Jul 2015 20:26:44 +0000 (22:26 +0200)]
Commit to push the 0.6 release.
Weng Xuetian [Thu, 2 Jul 2015 06:31:12 +0000 (08:31 +0200)]
Revert "Remove unnecessary XFilterEvent call."
This reverts commit
d2937b05aed9cee8d6651cd806d31682a853c773.
Quentin Rameau [Sun, 31 May 2015 10:26:11 +0000 (12:26 +0200)]
do not truncate font size when zooming
Roberto E. Vargas Caballero [Wed, 3 Jun 2015 06:07:55 +0000 (08:07 +0200)]
Revert "Optimize memory footprint of line buffers"
This reverts commit
7ab6c92e18d468968811256e808b02309c160a22.
We need 32 bits for real color support.
Jan Christoph Ebersbach [Fri, 22 May 2015 14:06:57 +0000 (16:06 +0200)]
Support UTF-8 characters as word delimiters
For a higher usefulness of the utf8strchr function, the index of the
UTF-8 character could be returned in addition with a Rune instead of a
char*. Since utf8strchr is currently only used by ISDELIM I didn't
bother to increase the complexity.
Roberto E. Vargas Caballero [Fri, 15 May 2015 05:51:58 +0000 (07:51 +0200)]
Merge branch 'master' of ssh://suckless.org/gitrepos/st
v4hn [Thu, 14 May 2015 13:46:07 +0000 (15:46 +0200)]
set selection to IDLE on clear
Otherwise a tangling bmotion event will consider
the selection still valid and selnormalize segfaults
because of an invalid sel.ob.y index.
suigin [Sat, 9 May 2015 22:22:40 +0000 (15:22 -0700)]
Small bugfix for makeglyphfontspecs call in drawregion
Here's a patch that fixes a bug when calling `makedrawglyphfontspecs'
in `drawregion'. Wasn't offseting the pointer into the input glyphs
array by `x1'. The bug isn't causing any problems currently, because
`drawregion' is always called with `x1' and `y1' values of 0, but if
this ever changes in the future, the bug would certainly cause some
problems.
Christoph Lohmann [Sun, 10 May 2015 13:19:48 +0000 (15:19 +0200)]
Fix the new -e handling. An empty cmd has to work for backwards compatibility.
suigin [Wed, 6 May 2015 03:24:00 +0000 (20:24 -0700)]
Clean up xdraws and optimize glyph drawing with non-unit kerning values
I have another patch here for review that optimizes the performance of
glyph drawing, primarily when using non-unit kerning values, and fixes a
few other minor issues. It's dependent on the earlier patch from me that
stores unicode codepoints in a Rune type, typedef'd to uint_least32_t.
This patch is a pretty big change to xdraws so your scrutiny is
appreciated.
First, some performance numbers. I used Yu-Jie Lin termfps.sh shell
script to benchmark before and after, and you can find it in the
attachments. On my Kaveri A10 7850k machine, I get the following
results:
Before Patch
============
1) Font: "Liberation Mono:pixelsize=12:antialias=false:autohint=false"
cwscale: 1.0, chscale: 1.0
For 273x83 100 frames.
Elapsed time : 1.553
Frames/second: 64.352
Chars /second: 1,458,159
2) Font: "Inconsolata:pixelsize=14:antialias=true:autohint=true"
cwscale: 1.001, chscale: 1.001
For 239x73 100 frames.
Elapsed time : 159.286
Frames/second: 0.627
Chars /second: 10,953
After Patch
===========
3) Font: "Liberation Mono:pixelsize=12:antialias=false:autohint=false"
cwscale: 1.0, chscale: 1.0
For 273x83 100 frames.
Elapsed time : 1.544
Frames/second: 64.728
Chars /second: 1,466,690
4) Font: "Inconsolata:pixelsize=14:antialias=true:autohint=true"
cwscale: 1.001, chscale: 1.001
For 239x73 100 frames.
Elapsed time : 1.955
Frames/second: 51.146
Chars /second: 892,361
As you can see, while the improvements for fonts with unit-kerning is
marginal, there's a huge ~81x performance increase with the patch when
using kerning values other than 1.0.
So what does the patch do?
The `xdraws' function would render each glyph one at a time if non-unit
kerning values were configured, and this was the primary cause of the
slow down. Xft provides a handful of functions which allow you to render
multiple characters or glyphs at time, each with a unique <x,y> position,
so it was simply a matter of massaging the data into a format that would
allow us to use one of these functions.
I've split `xdraws' up into two functions. In the first pass with
`xmakeglyphfontspecs' it will iterate over all of the glyphs in a given
row and it will build up an array of corresponding XftGlyphFontSpec
records. Much of the old logic for resolving fonts for glyphs using Xft
and fontconfig went into this function.
The second pass is done with `xrenderglyphfontspecs' which contains the
old logic for determining colors, clearing the background, and finally
rendering the array of XftGlyphFontSpec records.
There's a couple of other things that have been improved by this patch.
For instance, the UTF-32 codepoints in the Line's were being re-encoded
back into UTF-8 strings to be passed to `xdraws' which in turn would then
decode back to UTF-32 to verify that the Font contained a matching glyph
for the code point. Next, the UTF-8 string was being passed to
`XftDrawStringUtf8' which internally mallocs a scratch buffer and decodes
back to UTF-32 and does the lookup of the glyphs all over again.
This patch gets rid of all of this redundant round-trip encoding and
decoding of characters to be rendered and only looks up the glyph index
once (per font) during the font resolution phase. So this is probably
what's responsible for the marginal improvements seen when kerning values
are kept to 1.0.
I imagine there are other performance improvements here too, not seen in
the above benchmarks, if the user has lots of non-ASCII code plane characters
on the screen, or several different fonts are being utilized during
screen redraw.
Anyway, if you see any problems, please let me know and I can fix them.
suigin [Tue, 5 May 2015 20:13:21 +0000 (13:13 -0700)]
Changed type for UTF-32 codepoints from long to uint_least32_t
noname [Sun, 3 May 2015 19:28:10 +0000 (19:28 +0000)]
Fix empty selection highlighting bug.
When user clicks LMB, one character is selected, but will not be copied
to selection until the user moves cursor a bit. Therefore, the character
should not be highlighted as selected yet.
Before the patch, the trick was not to mark line as dirty to avoid
highlighting it. However, if user has already selected something and
clicks in line that contains selection, selclear sets the line as dirty
and one character is highlighted when it should not.
This patch replaces dirty trick with explicit check for sel.mode inside
selected().
noname [Sun, 3 May 2015 02:49:23 +0000 (02:49 +0000)]
Fix indentation.