This makes x(un)loadfonts internal to x.c. Needed to reorder includes
and move a typedef to keep the compiler happy.
Signed-off-by: Devin J. Pohly <djpohly@gmail.com>
This makes xsetenv internal to x.c, and allows iso14755's external
command to use $WINDOWID instead of having to snprintf it again. (The
same benefit will apply to the externalpipe patch.) The xwinid function
is no longer needed.
Signed-off-by: Devin J. Pohly <djpohly@gmail.com>
Non-printable characters, such as line breaks, in a base64 encoded
string violate the "string length must be a multiple of four" rule.
This patch pads the result buffer by one extra unit of four bytes,
and skips over non-printable characters found in the input string.
XftFontMatch does display-specific font configuration (commit 528241a).
Nice. Unfortunately, when we switched from FcFontMatch, we also stopped
storing the post-Fc{Config,Default}Substitute FcPattern for future
lookups. The result is that if a glyph isn't found in the primary font,
secondary font lookups use the original FcPattern, not the configured
one. If you have custom fontconfig rules (like me), this can be
disappointing.
I basically just copied the guts out of XftFontMatch[1] and saved
the intermediate configured FcPattern. Could be related to the bug that
inspired commit 4242027.
[1]: https://cgit.freedesktop.org/xorg/lib/libXft/tree/src/xftfont.c
Hi,
When I specify a font by point size (I'm using "Inconsolata:size=12"),
characters that are substituted from another font because they are not in the
main one appear too small. Doing a zoom reset fixes it. For example:
Before: http://i.imgur.com/G4Mfv4X.png
After: http://i.imgur.com/PMDhfQA.png
I found that adding the pixel size (acquired from the initial font load) to the
pattern then reloading the font fixes the problem. I'm not sure if this is a
proper fix, though.
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.
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.
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.
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>
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>
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>
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>
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>
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>
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.
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>
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.
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>
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.
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).
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.
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.
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>
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>
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.
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.
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.
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().
This patch also prevents sel.mode from increasing beyond 2. It is almost
impossible, but sel.mode may overflow if mouse is moved around for too
long while selecting.
st.c:1321:2: warning: ignoring return value of function declared with warn_unused_result attribute [-Wunused-result]
system(cmd);
^~~~~~ ~~~
Debatable whether an error here should case exit(EXIT_FAILURE). Just
preserving the existing behaviour for now.
Not always is desirable to create a pseudo terminal, and some times
we want to open a terminal emulator over a tty line. With this new
patch is possible to do someting like:
$ st -l /dev/ttyS0 115200
Without this option was needed to launch another terminal emulator
over st (for example minicom, picocom, cu, ...).
ICCCM mandates the use of real timestamps to interact with the
selection, to rule out race conditions if the clients are run at
different speeds. I have implemented the low hanging fruit, putting the
timestamps into text selection. Also, ICCCM mandates a check for whether
XSetSelectionOwner() worked. Not sure my version is correct, though.
tmoveto resets CURSOR_WRAPNEXT.
Simple testcase:
for i in $(seq 1 200); do
printf '\t.';
usleep 100000;
printf '\t@';
usleep 100000;
done
In st executing this script causes @ and . to overwrite each other in
the last column.
XFilterEvent usually filters KeyPress events according to input method.
At this point the window is not mapped. The only events that we process
are ConfigureNotify and MapNotify. They should not be filtered by input
method.
strsep() is not a POSIX function, and it means that every system
needs different defines to expose it. If the prototype of strsep
is not exposed then an ugly int/pointer is done and it might mean
a crash. The best solution?, to remove the strsep and make a custom
loop. If C programmers cannot do this kind of loops without calling
a library function, then maybe we should move all the suckless
software to Java.
Some programs can only deal with XA_STRING, and it makes impossible st
be able of copying to them. This patch makes st answer also to XA_STRING,
althought it sends utf8 strings. It is not a problem because moderm
applications must support utf8.
Thanks to Alex Pilon <alp@alexpilon.ca>!
Now there is a distinction between the primary and clipboard selection. With
Mod + Shift + c/v the clipboard is handled. The old Insert behavious does
reside.
