1 |
root |
1.1 |
#! perl |
2 |
|
|
|
3 |
root |
1.16 |
#:META:X_RESOURCE:%.expr:string:background expression |
4 |
root |
1.46 |
#:META:X_RESOURCE:%.border:boolean:respect the terminal border |
5 |
|
|
#:META:X_RESOURCE:%.interval:seconds:minimum time between updates |
6 |
root |
1.33 |
|
7 |
root |
1.41 |
=head1 NAME |
8 |
root |
1.33 |
|
9 |
root |
1.79 |
background - manage terminal background |
10 |
root |
1.41 |
|
11 |
|
|
=head1 SYNOPSIS |
12 |
root |
1.33 |
|
13 |
root |
1.36 |
urxvt --background-expr 'background expression' |
14 |
|
|
--background-border |
15 |
root |
1.46 |
--background-interval seconds |
16 |
root |
1.33 |
|
17 |
root |
1.75 |
=head1 QUICK AND DIRTY CHEAT SHEET |
18 |
|
|
|
19 |
|
|
Just load a random jpeg image and tile the background with it without |
20 |
|
|
scaling or anything else: |
21 |
|
|
|
22 |
|
|
load "/path/to/img.jpg" |
23 |
|
|
|
24 |
|
|
The same, but use mirroring/reflection instead of tiling: |
25 |
|
|
|
26 |
|
|
mirror load "/path/to/img.jpg" |
27 |
|
|
|
28 |
|
|
Load an image and scale it to exactly fill the terminal window: |
29 |
|
|
|
30 |
|
|
scale keep { load "/path/to/img.jpg" } |
31 |
|
|
|
32 |
|
|
Implement pseudo-transparency by using a suitably-aligned root pixmap |
33 |
|
|
as window background: |
34 |
|
|
|
35 |
|
|
rootalign root |
36 |
|
|
|
37 |
|
|
Likewise, but keep a blurred copy: |
38 |
|
|
|
39 |
|
|
rootalign keep { blur 10, root } |
40 |
|
|
|
41 |
root |
1.41 |
=head1 DESCRIPTION |
42 |
root |
1.33 |
|
43 |
root |
1.36 |
This extension manages the terminal background by creating a picture that |
44 |
|
|
is behind the text, replacing the normal background colour. |
45 |
|
|
|
46 |
|
|
It does so by evaluating a Perl expression that I<calculates> the image on |
47 |
|
|
the fly, for example, by grabbing the root background or loading a file. |
48 |
|
|
|
49 |
|
|
While the full power of Perl is available, the operators have been design |
50 |
|
|
to be as simple as possible. |
51 |
|
|
|
52 |
|
|
For example, to load an image and scale it to the window size, you would |
53 |
|
|
use: |
54 |
|
|
|
55 |
root |
1.64 |
urxvt --background-expr 'scale keep { load "/path/to/mybg.png" }' |
56 |
root |
1.36 |
|
57 |
|
|
Or specified as a X resource: |
58 |
|
|
|
59 |
root |
1.64 |
URxvt.background-expr: scale keep { load "/path/to/mybg.png" } |
60 |
root |
1.36 |
|
61 |
root |
1.41 |
=head1 THEORY OF OPERATION |
62 |
root |
1.36 |
|
63 |
|
|
At startup, just before the window is mapped for the first time, the |
64 |
|
|
expression is evaluated and must yield an image. The image is then |
65 |
|
|
extended as necessary to cover the whole terminal window, and is set as a |
66 |
|
|
background pixmap. |
67 |
|
|
|
68 |
|
|
If the image contains an alpha channel, then it will be used as-is in |
69 |
|
|
visuals that support alpha channels (for example, for a compositing |
70 |
|
|
manager). In other visuals, the terminal background colour will be used to |
71 |
|
|
replace any transparency. |
72 |
|
|
|
73 |
|
|
When the expression relies, directly or indirectly, on the window size, |
74 |
|
|
position, the root pixmap, or a timer, then it will be remembered. If not, |
75 |
|
|
then it will be removed. |
76 |
|
|
|
77 |
|
|
If any of the parameters that the expression relies on changes (when the |
78 |
|
|
window is moved or resized, its position or size changes; when the root |
79 |
|
|
pixmap is replaced by another one the root background changes; or when the |
80 |
|
|
timer elapses), then the expression will be evaluated again. |
81 |
|
|
|
82 |
root |
1.64 |
For example, an expression such as C<scale keep { load "$HOME/mybg.png" |
83 |
|
|
}> scales the image to the window size, so it relies on the window size |
84 |
|
|
and will be reevaluated each time it is changed, but not when it moves for |
85 |
root |
1.36 |
example. That ensures that the picture always fills the terminal, even |
86 |
sf-exg |
1.51 |
after its size changes. |
87 |
root |
1.36 |
|
88 |
root |
1.41 |
=head2 EXPRESSIONS |
89 |
root |
1.36 |
|
90 |
|
|
Expressions are normal Perl expressions, in fact, they are Perl blocks - |
91 |
|
|
which means you could use multiple lines and statements: |
92 |
|
|
|
93 |
root |
1.64 |
scale keep { |
94 |
|
|
again 3600; |
95 |
|
|
if (localtime now)[6]) { |
96 |
|
|
return load "$HOME/weekday.png"; |
97 |
|
|
} else { |
98 |
|
|
return load "$HOME/sunday.png"; |
99 |
|
|
} |
100 |
root |
1.36 |
} |
101 |
|
|
|
102 |
root |
1.68 |
This inner expression is evaluated once per hour (and whenever the |
103 |
sf-exg |
1.73 |
terminal window is resized). It sets F<sunday.png> as background on |
104 |
root |
1.68 |
Sundays, and F<weekday.png> on all other days. |
105 |
root |
1.36 |
|
106 |
|
|
Fortunately, we expect that most expressions will be much simpler, with |
107 |
|
|
little Perl knowledge needed. |
108 |
|
|
|
109 |
|
|
Basically, you always start with a function that "generates" an image |
110 |
|
|
object, such as C<load>, which loads an image from disk, or C<root>, which |
111 |
|
|
returns the root window background image: |
112 |
|
|
|
113 |
|
|
load "$HOME/mypic.png" |
114 |
|
|
|
115 |
|
|
The path is usually specified as a quoted string (the exact rules can be |
116 |
|
|
found in the L<perlop> manpage). The F<$HOME> at the beginning of the |
117 |
|
|
string is expanded to the home directory. |
118 |
|
|
|
119 |
|
|
Then you prepend one or more modifiers or filtering expressions, such as |
120 |
|
|
C<scale>: |
121 |
|
|
|
122 |
|
|
scale load "$HOME/mypic.png" |
123 |
|
|
|
124 |
|
|
Just like a mathematical expression with functions, you should read these |
125 |
|
|
expressions from right to left, as the C<load> is evaluated first, and |
126 |
|
|
its result becomes the argument to the C<scale> function. |
127 |
|
|
|
128 |
|
|
Many operators also allow some parameters preceding the input image |
129 |
|
|
that modify its behaviour. For example, C<scale> without any additional |
130 |
|
|
arguments scales the image to size of the terminal window. If you specify |
131 |
root |
1.43 |
an additional argument, it uses it as a scale factor (multiply by 100 to |
132 |
|
|
get a percentage): |
133 |
root |
1.36 |
|
134 |
root |
1.43 |
scale 2, load "$HOME/mypic.png" |
135 |
root |
1.36 |
|
136 |
|
|
This enlarges the image by a factor of 2 (200%). As you can see, C<scale> |
137 |
|
|
has now two arguments, the C<200> and the C<load> expression, while |
138 |
|
|
C<load> only has one argument. Arguments are separated from each other by |
139 |
|
|
commas. |
140 |
|
|
|
141 |
|
|
Scale also accepts two arguments, which are then separate factors for both |
142 |
|
|
horizontal and vertical dimensions. For example, this halves the image |
143 |
|
|
width and doubles the image height: |
144 |
|
|
|
145 |
root |
1.43 |
scale 0.5, 2, load "$HOME/mypic.png" |
146 |
root |
1.36 |
|
147 |
root |
1.68 |
IF you try out these expressions, you might suffer from some sluggishness, |
148 |
sf-exg |
1.73 |
because each time the terminal is resized, it loads the PNG image again |
149 |
root |
1.68 |
and scales it. Scaling is usually fast (and unavoidable), but loading the |
150 |
|
|
image can be quite time consuming. This is where C<keep> comes in handy: |
151 |
root |
1.39 |
|
152 |
root |
1.64 |
scale 0.5, 2, keep { load "$HOME/mypic.png" } |
153 |
root |
1.39 |
|
154 |
root |
1.64 |
The C<keep> operator executes all the statements inside the braces only |
155 |
|
|
once, or when it thinks the outcome might change. In other cases it |
156 |
|
|
returns the last value computed by the brace block. |
157 |
root |
1.39 |
|
158 |
root |
1.64 |
This means that the C<load> is only executed once, which makes it much |
159 |
sf-exg |
1.65 |
faster, but also means that more memory is being used, because the loaded |
160 |
root |
1.64 |
image must be kept in memory at all times. In this expression, the |
161 |
|
|
trade-off is likely worth it. |
162 |
root |
1.39 |
|
163 |
root |
1.64 |
But back to effects: Other effects than scaling are also readily |
164 |
|
|
available, for example, you can tile the image to fill the whole window, |
165 |
|
|
instead of resizing it: |
166 |
root |
1.39 |
|
167 |
root |
1.64 |
tile keep { load "$HOME/mypic.png" } |
168 |
root |
1.39 |
|
169 |
root |
1.64 |
In fact, images returned by C<load> are in C<tile> mode by default, so the |
170 |
|
|
C<tile> operator is kind of superfluous. |
171 |
root |
1.39 |
|
172 |
root |
1.64 |
Another common effect is to mirror the image, so that the same edges |
173 |
|
|
touch: |
174 |
root |
1.36 |
|
175 |
root |
1.64 |
mirror keep { load "$HOME/mypic.png" } |
176 |
root |
1.36 |
|
177 |
root |
1.64 |
Another common background expression is: |
178 |
root |
1.63 |
|
179 |
root |
1.64 |
rootalign root |
180 |
root |
1.39 |
|
181 |
root |
1.64 |
This one first takes a snapshot of the screen background image, and then |
182 |
sf-exg |
1.65 |
moves it to the upper left corner of the screen (as opposed to the upper |
183 |
root |
1.64 |
left corner of the terminal window)- the result is pseudo-transparency: |
184 |
|
|
the image seems to be static while the window is moved around. |
185 |
|
|
|
186 |
root |
1.71 |
=head2 COLOUR SPECIFICATIONS |
187 |
|
|
|
188 |
sf-exg |
1.73 |
Whenever an operator expects a "colour", then this can be specified in one |
189 |
root |
1.71 |
of two ways: Either as string with an X11 colour specification, such as: |
190 |
|
|
|
191 |
|
|
"red" # named colour |
192 |
|
|
"#f00" # simple rgb |
193 |
|
|
"[50]red" # red with 50% alpha |
194 |
|
|
"TekHVC:300/50/50" # anything goes |
195 |
|
|
|
196 |
|
|
OR as an array reference with one, three or four components: |
197 |
|
|
|
198 |
|
|
[0.5] # 50% gray, 100% alpha |
199 |
|
|
[0.5, 0, 0] # dark red, no green or blur, 100% alpha |
200 |
|
|
[0.5, 0, 0, 0.7] # same with explicit 70% alpha |
201 |
|
|
|
202 |
root |
1.64 |
=head2 CACHING AND SENSITIVITY |
203 |
|
|
|
204 |
|
|
Since some operations (such as C<load> and C<blur>) can take a long time, |
205 |
|
|
caching results can be very important for a smooth operation. Caching can |
206 |
|
|
also be useful to reduce memory usage, though, for example, when an image |
207 |
|
|
is cached by C<load>, it could be shared by multiple terminal windows |
208 |
|
|
running inside urxvtd. |
209 |
|
|
|
210 |
|
|
=head3 C<keep { ... }> caching |
211 |
|
|
|
212 |
|
|
The most important way to cache expensive operations is to use C<keep { |
213 |
|
|
... }>. The C<keep> operator takes a block of multiple statements enclosed |
214 |
|
|
by C<{}> and keeps the return value in memory. |
215 |
|
|
|
216 |
|
|
An expression can be "sensitive" to various external events, such as |
217 |
sf-exg |
1.65 |
scaling or moving the window, root background changes and timers. Simply |
218 |
sf-exg |
1.67 |
using an expression (such as C<scale> without parameters) that depends on |
219 |
root |
1.64 |
certain changing values (called "variables"), or using those variables |
220 |
|
|
directly, will make an expression sensitive to these events - for example, |
221 |
|
|
using C<scale> or C<TW> will make the expression sensitive to the terminal |
222 |
|
|
size, and thus to resizing events. |
223 |
root |
1.39 |
|
224 |
root |
1.64 |
When such an event happens, C<keep> will automatically trigger a |
225 |
|
|
reevaluation of the whole expression with the new value of the expression. |
226 |
root |
1.39 |
|
227 |
root |
1.64 |
C<keep> is most useful for expensive operations, such as C<blur>: |
228 |
root |
1.39 |
|
229 |
root |
1.68 |
rootalign keep { blur 20, root } |
230 |
root |
1.39 |
|
231 |
root |
1.64 |
This makes a blurred copy of the root background once, and on subsequent |
232 |
|
|
calls, just root-aligns it. Since C<blur> is usually quite slow and |
233 |
|
|
C<rootalign> is quite fast, this trades extra memory (for the cached |
234 |
|
|
blurred pixmap) with speed (blur only needs to be redone when root |
235 |
|
|
changes). |
236 |
root |
1.39 |
|
237 |
root |
1.64 |
=head3 C<load> caching |
238 |
root |
1.36 |
|
239 |
root |
1.64 |
The C<load> operator itself does not keep images in memory, but as long as |
240 |
|
|
the image is still in memory, C<load> will use the in-memory image instead |
241 |
|
|
of loading it freshly from disk. |
242 |
root |
1.63 |
|
243 |
root |
1.64 |
That means that this expression: |
244 |
root |
1.63 |
|
245 |
root |
1.64 |
keep { load "$HOME/path..." } |
246 |
root |
1.63 |
|
247 |
root |
1.64 |
Not only caches the image in memory, other terminal instances that try to |
248 |
|
|
C<load> it can reuse that in-memory copy. |
249 |
root |
1.63 |
|
250 |
root |
1.41 |
=head1 REFERENCE |
251 |
root |
1.33 |
|
252 |
root |
1.41 |
=head2 COMMAND LINE SWITCHES |
253 |
root |
1.36 |
|
254 |
|
|
=over 4 |
255 |
|
|
|
256 |
|
|
=item --background-expr perl-expression |
257 |
|
|
|
258 |
|
|
Specifies the Perl expression to evaluate. |
259 |
|
|
|
260 |
|
|
=item --background-border |
261 |
|
|
|
262 |
|
|
By default, the expression creates an image that fills the full window, |
263 |
|
|
overwriting borders and any other areas, such as the scrollbar. |
264 |
|
|
|
265 |
|
|
Specifying this flag changes the behaviour, so that the image only |
266 |
|
|
replaces the background of the character area. |
267 |
|
|
|
268 |
root |
1.46 |
=item --background-interval seconds |
269 |
|
|
|
270 |
sf-exg |
1.51 |
Since some operations in the underlying XRender extension can effectively |
271 |
root |
1.46 |
freeze your X-server for prolonged time, this extension enforces a minimum |
272 |
|
|
time between updates, which is normally about 0.1 seconds. |
273 |
|
|
|
274 |
|
|
If you want to do updates more often, you can decrease this safety |
275 |
|
|
interval with this switch. |
276 |
|
|
|
277 |
root |
1.36 |
=back |
278 |
|
|
|
279 |
root |
1.33 |
=cut |
280 |
root |
1.12 |
|
281 |
root |
1.52 |
our %_IMG_CACHE; |
282 |
root |
1.36 |
our $HOME; |
283 |
root |
1.63 |
our ($self, $frame); |
284 |
root |
1.29 |
our ($x, $y, $w, $h); |
285 |
root |
1.3 |
|
286 |
root |
1.16 |
# enforce at least this interval between updates |
287 |
root |
1.46 |
our $MIN_INTERVAL = 6/59.951; |
288 |
root |
1.9 |
|
289 |
root |
1.1 |
{ |
290 |
|
|
package urxvt::bgdsl; # background language |
291 |
|
|
|
292 |
root |
1.63 |
sub FR_PARENT() { 0 } # parent frame, if any - must be #0 |
293 |
|
|
sub FR_CACHE () { 1 } # cached values |
294 |
|
|
sub FR_AGAIN () { 2 } # what this expr is sensitive to |
295 |
|
|
sub FR_STATE () { 3 } # watchers etc. |
296 |
|
|
|
297 |
root |
1.43 |
use List::Util qw(min max sum shuffle); |
298 |
|
|
|
299 |
root |
1.15 |
=head2 PROVIDERS/GENERATORS |
300 |
|
|
|
301 |
root |
1.31 |
These functions provide an image, by loading it from disk, grabbing it |
302 |
sf-exg |
1.32 |
from the root screen or by simply generating it. They are used as starting |
303 |
root |
1.31 |
points to get an image you can play with. |
304 |
|
|
|
305 |
root |
1.15 |
=over 4 |
306 |
|
|
|
307 |
|
|
=item load $path |
308 |
|
|
|
309 |
root |
1.29 |
Loads the image at the given C<$path>. The image is set to plane tiling |
310 |
|
|
mode. |
311 |
|
|
|
312 |
sf-exg |
1.65 |
If the image is already in memory (e.g. because another terminal instance |
313 |
root |
1.64 |
uses it), then the in-memory copy us returned instead. |
314 |
root |
1.54 |
|
315 |
root |
1.64 |
=item load_uc $path |
316 |
|
|
|
317 |
|
|
Load uncached - same as load, but does not cache the image, which means it |
318 |
root |
1.72 |
is I<always> loaded from the filesystem again, even if another copy of it |
319 |
|
|
is in memory at the time. |
320 |
root |
1.29 |
|
321 |
root |
1.15 |
=cut |
322 |
|
|
|
323 |
root |
1.72 |
sub load_uc($) { |
324 |
|
|
$self->new_img_from_file ($_[0]) |
325 |
|
|
} |
326 |
|
|
|
327 |
root |
1.63 |
sub load($) { |
328 |
root |
1.54 |
my ($path) = @_; |
329 |
|
|
|
330 |
|
|
$_IMG_CACHE{$path} || do { |
331 |
root |
1.72 |
my $img = load_uc $path; |
332 |
root |
1.54 |
Scalar::Util::weaken ($_IMG_CACHE{$path} = $img); |
333 |
|
|
$img |
334 |
|
|
} |
335 |
|
|
} |
336 |
|
|
|
337 |
root |
1.31 |
=item root |
338 |
|
|
|
339 |
|
|
Returns the root window pixmap, that is, hopefully, the background image |
340 |
root |
1.62 |
of your screen. |
341 |
root |
1.31 |
|
342 |
|
|
This function makes your expression root sensitive, that means it will be |
343 |
|
|
reevaluated when the bg image changes. |
344 |
|
|
|
345 |
|
|
=cut |
346 |
|
|
|
347 |
root |
1.2 |
sub root() { |
348 |
root |
1.63 |
$frame->[FR_AGAIN]{rootpmap} = 1; |
349 |
root |
1.52 |
$self->new_img_from_root |
350 |
root |
1.1 |
} |
351 |
|
|
|
352 |
root |
1.31 |
=item solid $colour |
353 |
|
|
|
354 |
|
|
=item solid $width, $height, $colour |
355 |
|
|
|
356 |
|
|
Creates a new image and completely fills it with the given colour. The |
357 |
|
|
image is set to tiling mode. |
358 |
|
|
|
359 |
root |
1.40 |
If C<$width> and C<$height> are omitted, it creates a 1x1 image, which is |
360 |
root |
1.31 |
useful for solid backgrounds or for use in filtering effects. |
361 |
|
|
|
362 |
|
|
=cut |
363 |
|
|
|
364 |
root |
1.42 |
sub solid($;$$) { |
365 |
root |
1.31 |
my $colour = pop; |
366 |
|
|
|
367 |
root |
1.59 |
my $img = $self->new_img (urxvt::PictStandardARGB32, 0, 0, $_[0] || 1, $_[1] || 1); |
368 |
root |
1.31 |
$img->fill ($colour); |
369 |
root |
1.15 |
$img |
370 |
|
|
} |
371 |
|
|
|
372 |
root |
1.45 |
=item clone $img |
373 |
root |
1.31 |
|
374 |
root |
1.45 |
Returns an exact copy of the image. This is useful if you want to have |
375 |
|
|
multiple copies of the same image to apply different effects to. |
376 |
root |
1.31 |
|
377 |
root |
1.20 |
=cut |
378 |
|
|
|
379 |
root |
1.45 |
sub clone($) { |
380 |
|
|
$_[0]->clone |
381 |
root |
1.20 |
} |
382 |
|
|
|
383 |
root |
1.56 |
=item merge $img ... |
384 |
|
|
|
385 |
root |
1.57 |
Takes any number of images and merges them together, creating a single |
386 |
root |
1.62 |
image containing them all. The tiling mode of the first image is used as |
387 |
sf-exg |
1.65 |
the tiling mode of the resulting image. |
388 |
root |
1.56 |
|
389 |
root |
1.61 |
This function is called automatically when an expression returns multiple |
390 |
|
|
images. |
391 |
|
|
|
392 |
root |
1.56 |
=cut |
393 |
|
|
|
394 |
|
|
sub merge(@) { |
395 |
root |
1.61 |
return $_[0] unless $#_; |
396 |
|
|
|
397 |
root |
1.58 |
# rather annoyingly clumsy, but optimisation is for another time |
398 |
|
|
|
399 |
root |
1.59 |
my $x0 = +1e9; |
400 |
|
|
my $y0 = +1e9; |
401 |
root |
1.58 |
my $x1 = -1e9; |
402 |
|
|
my $y1 = -1e9; |
403 |
|
|
|
404 |
|
|
for (@_) { |
405 |
|
|
my ($x, $y, $w, $h) = $_->geometry; |
406 |
|
|
|
407 |
|
|
$x0 = $x if $x0 > $x; |
408 |
|
|
$y0 = $y if $y0 > $y; |
409 |
|
|
|
410 |
|
|
$x += $w; |
411 |
|
|
$y += $h; |
412 |
|
|
|
413 |
root |
1.59 |
$x1 = $x if $x1 < $x; |
414 |
|
|
$y1 = $y if $y1 < $y; |
415 |
root |
1.58 |
} |
416 |
|
|
|
417 |
root |
1.59 |
my $base = $self->new_img (urxvt::PictStandardARGB32, $x0, $y0, $x1 - $x0, $y1 - $y0); |
418 |
root |
1.62 |
$base->repeat_mode ($_[0]->repeat_mode); |
419 |
root |
1.58 |
$base->fill ([0, 0, 0, 0]); |
420 |
|
|
|
421 |
root |
1.59 |
$base->draw ($_) |
422 |
root |
1.58 |
for @_; |
423 |
|
|
|
424 |
|
|
$base |
425 |
root |
1.56 |
} |
426 |
|
|
|
427 |
root |
1.76 |
=back |
428 |
|
|
|
429 |
root |
1.28 |
=head2 TILING MODES |
430 |
|
|
|
431 |
|
|
The following operators modify the tiling mode of an image, that is, the |
432 |
|
|
way that pixels outside the image area are painted when the image is used. |
433 |
root |
1.15 |
|
434 |
|
|
=over 4 |
435 |
|
|
|
436 |
root |
1.28 |
=item tile $img |
437 |
|
|
|
438 |
|
|
Tiles the whole plane with the image and returns this new image - or in |
439 |
|
|
other words, it returns a copy of the image in plane tiling mode. |
440 |
|
|
|
441 |
root |
1.34 |
Example: load an image and tile it over the background, without |
442 |
|
|
resizing. The C<tile> call is superfluous because C<load> already defaults |
443 |
|
|
to tiling mode. |
444 |
|
|
|
445 |
|
|
tile load "mybg.png" |
446 |
|
|
|
447 |
root |
1.28 |
=item mirror $img |
448 |
|
|
|
449 |
|
|
Similar to tile, but reflects the image each time it uses a new copy, so |
450 |
|
|
that top edges always touch top edges, right edges always touch right |
451 |
|
|
edges and so on (with normal tiling, left edges always touch right edges |
452 |
|
|
and top always touch bottom edges). |
453 |
|
|
|
454 |
root |
1.36 |
Example: load an image and mirror it over the background, avoiding sharp |
455 |
root |
1.34 |
edges at the image borders at the expense of mirroring the image itself |
456 |
|
|
|
457 |
|
|
mirror load "mybg.png" |
458 |
|
|
|
459 |
root |
1.28 |
=item pad $img |
460 |
|
|
|
461 |
|
|
Takes an image and modifies it so that all pixels outside the image area |
462 |
|
|
become transparent. This mode is most useful when you want to place an |
463 |
|
|
image over another image or the background colour while leaving all |
464 |
|
|
background pixels outside the image unchanged. |
465 |
|
|
|
466 |
root |
1.36 |
Example: load an image and display it in the upper left corner. The rest |
467 |
sf-exg |
1.51 |
of the space is left "empty" (transparent or whatever your compositor does |
468 |
root |
1.34 |
in alpha mode, else background colour). |
469 |
|
|
|
470 |
|
|
pad load "mybg.png" |
471 |
|
|
|
472 |
root |
1.28 |
=item extend $img |
473 |
|
|
|
474 |
|
|
Extends the image over the whole plane, using the closest pixel in the |
475 |
sf-exg |
1.51 |
area outside the image. This mode is mostly useful when you use more complex |
476 |
root |
1.28 |
filtering operations and want the pixels outside the image to have the |
477 |
|
|
same values as the pixels near the edge. |
478 |
|
|
|
479 |
root |
1.34 |
Example: just for curiosity, how does this pixel extension stuff work? |
480 |
|
|
|
481 |
|
|
extend move 50, 50, load "mybg.png" |
482 |
|
|
|
483 |
root |
1.15 |
=cut |
484 |
|
|
|
485 |
root |
1.28 |
sub pad($) { |
486 |
|
|
my $img = $_[0]->clone; |
487 |
|
|
$img->repeat_mode (urxvt::RepeatNone); |
488 |
|
|
$img |
489 |
|
|
} |
490 |
|
|
|
491 |
|
|
sub tile($) { |
492 |
|
|
my $img = $_[0]->clone; |
493 |
|
|
$img->repeat_mode (urxvt::RepeatNormal); |
494 |
|
|
$img |
495 |
|
|
} |
496 |
|
|
|
497 |
|
|
sub mirror($) { |
498 |
|
|
my $img = $_[0]->clone; |
499 |
|
|
$img->repeat_mode (urxvt::RepeatReflect); |
500 |
|
|
$img |
501 |
|
|
} |
502 |
root |
1.4 |
|
503 |
root |
1.28 |
sub extend($) { |
504 |
root |
1.24 |
my $img = $_[0]->clone; |
505 |
root |
1.28 |
$img->repeat_mode (urxvt::RepeatPad); |
506 |
root |
1.24 |
$img |
507 |
|
|
} |
508 |
|
|
|
509 |
root |
1.28 |
=back |
510 |
|
|
|
511 |
root |
1.45 |
=head2 VARIABLE VALUES |
512 |
root |
1.28 |
|
513 |
root |
1.45 |
The following functions provide variable data such as the terminal window |
514 |
|
|
dimensions. They are not (Perl-) variables, they just return stuff that |
515 |
|
|
varies. Most of them make your expression sensitive to some events, for |
516 |
|
|
example using C<TW> (terminal width) means your expression is evaluated |
517 |
|
|
again when the terminal is resized. |
518 |
root |
1.28 |
|
519 |
|
|
=over 4 |
520 |
|
|
|
521 |
root |
1.45 |
=item TX |
522 |
|
|
|
523 |
|
|
=item TY |
524 |
|
|
|
525 |
|
|
Return the X and Y coordinates of the terminal window (the terminal |
526 |
|
|
window is the full window by default, and the character area only when in |
527 |
|
|
border-respect mode). |
528 |
|
|
|
529 |
|
|
Using these functions make your expression sensitive to window moves. |
530 |
|
|
|
531 |
|
|
These functions are mainly useful to align images to the root window. |
532 |
|
|
|
533 |
|
|
Example: load an image and align it so it looks as if anchored to the |
534 |
root |
1.64 |
background (that's exactly what C<rootalign> does btw.): |
535 |
root |
1.45 |
|
536 |
root |
1.64 |
move -TX, -TY, keep { load "mybg.png" } |
537 |
root |
1.45 |
|
538 |
|
|
=item TW |
539 |
|
|
|
540 |
root |
1.77 |
=item TH |
541 |
|
|
|
542 |
root |
1.45 |
Return the width (C<TW>) and height (C<TH>) of the terminal window (the |
543 |
|
|
terminal window is the full window by default, and the character area only |
544 |
|
|
when in border-respect mode). |
545 |
|
|
|
546 |
|
|
Using these functions make your expression sensitive to window resizes. |
547 |
|
|
|
548 |
|
|
These functions are mainly useful to scale images, or to clip images to |
549 |
|
|
the window size to conserve memory. |
550 |
|
|
|
551 |
|
|
Example: take the screen background, clip it to the window size, blur it a |
552 |
|
|
bit, align it to the window position and use it as background. |
553 |
|
|
|
554 |
root |
1.64 |
clip move -TX, -TY, keep { blur 5, root } |
555 |
root |
1.45 |
|
556 |
|
|
=cut |
557 |
|
|
|
558 |
root |
1.63 |
sub TX() { $frame->[FR_AGAIN]{position} = 1; $x } |
559 |
|
|
sub TY() { $frame->[FR_AGAIN]{position} = 1; $y } |
560 |
|
|
sub TW() { $frame->[FR_AGAIN]{size} = 1; $w } |
561 |
|
|
sub TH() { $frame->[FR_AGAIN]{size} = 1; $h } |
562 |
root |
1.45 |
|
563 |
|
|
=item now |
564 |
|
|
|
565 |
|
|
Returns the current time as (fractional) seconds since the epoch. |
566 |
|
|
|
567 |
|
|
Using this expression does I<not> make your expression sensitive to time, |
568 |
|
|
but the next two functions do. |
569 |
|
|
|
570 |
|
|
=item again $seconds |
571 |
|
|
|
572 |
|
|
When this function is used the expression will be reevaluated again in |
573 |
|
|
C<$seconds> seconds. |
574 |
|
|
|
575 |
|
|
Example: load some image and rotate it according to the time of day (as if it were |
576 |
|
|
the hour pointer of a clock). Update this image every minute. |
577 |
|
|
|
578 |
root |
1.64 |
again 60; |
579 |
|
|
rotate 50, 50, (now % 86400) * -72 / 8640, scale keep { load "myclock.png" } |
580 |
root |
1.28 |
|
581 |
root |
1.45 |
=item counter $seconds |
582 |
|
|
|
583 |
|
|
Like C<again>, but also returns an increasing counter value, starting at |
584 |
|
|
0, which might be useful for some simple animation effects. |
585 |
root |
1.28 |
|
586 |
|
|
=cut |
587 |
|
|
|
588 |
root |
1.45 |
sub now() { urxvt::NOW } |
589 |
|
|
|
590 |
|
|
sub again($) { |
591 |
root |
1.63 |
$frame->[FR_AGAIN]{time} = $_[0]; |
592 |
root |
1.45 |
} |
593 |
|
|
|
594 |
|
|
sub counter($) { |
595 |
root |
1.63 |
$frame->[FR_AGAIN]{time} = $_[0]; |
596 |
|
|
$frame->[FR_STATE]{counter} + 0 |
597 |
root |
1.28 |
} |
598 |
|
|
|
599 |
root |
1.45 |
=back |
600 |
|
|
|
601 |
|
|
=head2 SHAPE CHANGING OPERATORS |
602 |
|
|
|
603 |
|
|
The following operators modify the shape, size or position of the image. |
604 |
|
|
|
605 |
|
|
=over 4 |
606 |
|
|
|
607 |
root |
1.28 |
=item clip $img |
608 |
|
|
|
609 |
|
|
=item clip $width, $height, $img |
610 |
|
|
|
611 |
|
|
=item clip $x, $y, $width, $height, $img |
612 |
|
|
|
613 |
|
|
Clips an image to the given rectangle. If the rectangle is outside the |
614 |
|
|
image area (e.g. when C<$x> or C<$y> are negative) or the rectangle is |
615 |
|
|
larger than the image, then the tiling mode defines how the extra pixels |
616 |
|
|
will be filled. |
617 |
|
|
|
618 |
root |
1.78 |
If C<$x> and C<$y> are missing, then C<0> is assumed for both. |
619 |
root |
1.28 |
|
620 |
|
|
If C<$width> and C<$height> are missing, then the window size will be |
621 |
|
|
assumed. |
622 |
|
|
|
623 |
|
|
Example: load an image, blur it, and clip it to the window size to save |
624 |
|
|
memory. |
625 |
|
|
|
626 |
root |
1.64 |
clip keep { blur 10, load "mybg.png" } |
627 |
root |
1.28 |
|
628 |
|
|
=cut |
629 |
|
|
|
630 |
root |
1.20 |
sub clip($;$$;$$) { |
631 |
root |
1.7 |
my $img = pop; |
632 |
root |
1.30 |
my $h = pop || TH; |
633 |
|
|
my $w = pop || TW; |
634 |
root |
1.21 |
$img->sub_rect ($_[0], $_[1], $w, $h) |
635 |
root |
1.4 |
} |
636 |
|
|
|
637 |
root |
1.28 |
=item scale $img |
638 |
|
|
|
639 |
root |
1.43 |
=item scale $size_factor, $img |
640 |
root |
1.28 |
|
641 |
root |
1.43 |
=item scale $width_factor, $height_factor, $img |
642 |
root |
1.28 |
|
643 |
root |
1.43 |
Scales the image by the given factors in horizontal |
644 |
|
|
(C<$width>) and vertical (C<$height>) direction. |
645 |
root |
1.28 |
|
646 |
root |
1.43 |
If only one factor is give, it is used for both directions. |
647 |
root |
1.28 |
|
648 |
root |
1.43 |
If no factors are given, scales the image to the window size without |
649 |
root |
1.28 |
keeping aspect. |
650 |
|
|
|
651 |
|
|
=item resize $width, $height, $img |
652 |
|
|
|
653 |
|
|
Resizes the image to exactly C<$width> times C<$height> pixels. |
654 |
|
|
|
655 |
root |
1.43 |
=item fit $img |
656 |
|
|
|
657 |
|
|
=item fit $width, $height, $img |
658 |
|
|
|
659 |
|
|
Fits the image into the given C<$width> and C<$height> without changing |
660 |
|
|
aspect, or the terminal size. That means it will be shrunk or grown until |
661 |
|
|
the whole image fits into the given area, possibly leaving borders. |
662 |
|
|
|
663 |
|
|
=item cover $img |
664 |
|
|
|
665 |
|
|
=item cover $width, $height, $img |
666 |
|
|
|
667 |
|
|
Similar to C<fit>, but shrinks or grows until all of the area is covered |
668 |
|
|
by the image, so instead of potentially leaving borders, it will cut off |
669 |
|
|
image data that doesn't fit. |
670 |
|
|
|
671 |
root |
1.28 |
=cut |
672 |
|
|
|
673 |
root |
1.33 |
sub scale($;$;$) { |
674 |
root |
1.28 |
my $img = pop; |
675 |
|
|
|
676 |
root |
1.43 |
@_ == 2 ? $img->scale ($_[0] * $img->w, $_[1] * $img->h) |
677 |
|
|
: @_ ? $img->scale ($_[0] * $img->w, $_[0] * $img->h) |
678 |
root |
1.30 |
: $img->scale (TW, TH) |
679 |
root |
1.28 |
} |
680 |
|
|
|
681 |
root |
1.2 |
sub resize($$$) { |
682 |
root |
1.7 |
my $img = pop; |
683 |
|
|
$img->scale ($_[0], $_[1]) |
684 |
root |
1.1 |
} |
685 |
|
|
|
686 |
root |
1.43 |
sub fit($;$$) { |
687 |
|
|
my $img = pop; |
688 |
|
|
my $w = ($_[0] || TW) / $img->w; |
689 |
|
|
my $h = ($_[1] || TH) / $img->h; |
690 |
|
|
scale +(min $w, $h), $img |
691 |
|
|
} |
692 |
|
|
|
693 |
|
|
sub cover($;$$) { |
694 |
|
|
my $img = pop; |
695 |
|
|
my $w = ($_[0] || TW) / $img->w; |
696 |
|
|
my $h = ($_[1] || TH) / $img->h; |
697 |
|
|
scale +(max $w, $h), $img |
698 |
|
|
} |
699 |
|
|
|
700 |
root |
1.36 |
=item move $dx, $dy, $img |
701 |
|
|
|
702 |
|
|
Moves the image by C<$dx> pixels in the horizontal, and C<$dy> pixels in |
703 |
|
|
the vertical. |
704 |
|
|
|
705 |
|
|
Example: move the image right by 20 pixels and down by 30. |
706 |
|
|
|
707 |
|
|
move 20, 30, ... |
708 |
|
|
|
709 |
root |
1.46 |
=item align $xalign, $yalign, $img |
710 |
|
|
|
711 |
|
|
Aligns the image according to a factor - C<0> means the image is moved to |
712 |
|
|
the left or top edge (for C<$xalign> or C<$yalign>), C<0.5> means it is |
713 |
|
|
exactly centered and C<1> means it touches the right or bottom edge. |
714 |
|
|
|
715 |
|
|
Example: remove any visible border around an image, center it vertically but move |
716 |
|
|
it to the right hand side. |
717 |
|
|
|
718 |
|
|
align 1, 0.5, pad $img |
719 |
|
|
|
720 |
root |
1.44 |
=item center $img |
721 |
|
|
|
722 |
|
|
=item center $width, $height, $img |
723 |
|
|
|
724 |
|
|
Centers the image, i.e. the center of the image is moved to the center of |
725 |
|
|
the terminal window (or the box specified by C<$width> and C<$height> if |
726 |
|
|
given). |
727 |
|
|
|
728 |
root |
1.46 |
Example: load an image and center it. |
729 |
|
|
|
730 |
root |
1.64 |
center keep { pad load "mybg.png" } |
731 |
root |
1.46 |
|
732 |
root |
1.36 |
=item rootalign $img |
733 |
|
|
|
734 |
|
|
Moves the image so that it appears glued to the screen as opposed to the |
735 |
|
|
window. This gives the illusion of a larger area behind the window. It is |
736 |
|
|
exactly equivalent to C<move -TX, -TY>, that is, it moves the image to the |
737 |
|
|
top left of the screen. |
738 |
|
|
|
739 |
|
|
Example: load a background image, put it in mirror mode and root align it. |
740 |
|
|
|
741 |
root |
1.64 |
rootalign keep { mirror load "mybg.png" } |
742 |
root |
1.36 |
|
743 |
|
|
Example: take the screen background and align it, giving the illusion of |
744 |
|
|
transparency as long as the window isn't in front of other windows. |
745 |
|
|
|
746 |
root |
1.46 |
rootalign root |
747 |
root |
1.36 |
|
748 |
|
|
=cut |
749 |
|
|
|
750 |
root |
1.7 |
sub move($$;$) { |
751 |
root |
1.20 |
my $img = pop->clone; |
752 |
|
|
$img->move ($_[0], $_[1]); |
753 |
|
|
$img |
754 |
root |
1.1 |
} |
755 |
|
|
|
756 |
root |
1.46 |
sub align($;$$) { |
757 |
|
|
my $img = pop; |
758 |
|
|
|
759 |
|
|
move $_[0] * (TW - $img->w), |
760 |
|
|
$_[1] * (TH - $img->h), |
761 |
|
|
$img |
762 |
|
|
} |
763 |
|
|
|
764 |
root |
1.44 |
sub center($;$$) { |
765 |
|
|
my $img = pop; |
766 |
|
|
my $w = $_[0] || TW; |
767 |
root |
1.46 |
my $h = $_[1] || TH; |
768 |
root |
1.44 |
|
769 |
|
|
move 0.5 * ($w - $img->w), 0.5 * ($h - $img->h), $img |
770 |
|
|
} |
771 |
|
|
|
772 |
root |
1.36 |
sub rootalign($) { |
773 |
|
|
move -TX, -TY, $_[0] |
774 |
root |
1.1 |
} |
775 |
|
|
|
776 |
root |
1.64 |
=item rotate $center_x, $center_y, $degrees, $img |
777 |
root |
1.52 |
|
778 |
root |
1.64 |
Rotates the image clockwise by C<$degrees> degrees, around the point at |
779 |
|
|
C<$center_x> and C<$center_y> (specified as factor of image width/height). |
780 |
root |
1.52 |
|
781 |
root |
1.64 |
Example: rotate the image by 90 degrees around it's center. |
782 |
root |
1.52 |
|
783 |
root |
1.64 |
rotate 0.5, 0.5, 90, keep { load "$HOME/mybg.png" } |
784 |
root |
1.52 |
|
785 |
|
|
=cut |
786 |
|
|
|
787 |
root |
1.53 |
sub rotate($$$$) { |
788 |
root |
1.52 |
my $img = pop; |
789 |
|
|
$img->rotate ( |
790 |
root |
1.60 |
$_[0] * ($img->w + $img->x), |
791 |
|
|
$_[1] * ($img->h + $img->y), |
792 |
root |
1.52 |
$_[2] * (3.14159265 / 180), |
793 |
|
|
) |
794 |
|
|
} |
795 |
|
|
|
796 |
root |
1.45 |
=back |
797 |
|
|
|
798 |
|
|
=head2 COLOUR MODIFICATIONS |
799 |
|
|
|
800 |
|
|
The following operators change the pixels of the image. |
801 |
|
|
|
802 |
|
|
=over 4 |
803 |
|
|
|
804 |
root |
1.70 |
=item tint $color, $img |
805 |
|
|
|
806 |
|
|
Tints the image in the given colour. |
807 |
|
|
|
808 |
|
|
Example: tint the image red. |
809 |
|
|
|
810 |
|
|
tint "red", load "rgb.png" |
811 |
|
|
|
812 |
|
|
Example: the same, but specify the colour by component. |
813 |
|
|
|
814 |
|
|
tint [1, 0, 0], load "rgb.png" |
815 |
|
|
|
816 |
|
|
=cut |
817 |
|
|
|
818 |
|
|
sub tint($$) { |
819 |
|
|
$_[1]->tint ($_[0]) |
820 |
|
|
} |
821 |
|
|
|
822 |
root |
1.36 |
=item contrast $factor, $img |
823 |
|
|
|
824 |
|
|
=item contrast $r, $g, $b, $img |
825 |
|
|
|
826 |
|
|
=item contrast $r, $g, $b, $a, $img |
827 |
|
|
|
828 |
|
|
Adjusts the I<contrast> of an image. |
829 |
|
|
|
830 |
root |
1.45 |
The first form applies a single C<$factor> to red, green and blue, the |
831 |
|
|
second form applies separate factors to each colour channel, and the last |
832 |
|
|
form includes the alpha channel. |
833 |
|
|
|
834 |
|
|
Values from 0 to 1 lower the contrast, values higher than 1 increase the |
835 |
|
|
contrast. |
836 |
|
|
|
837 |
|
|
Due to limitations in the underlying XRender extension, lowering contrast |
838 |
|
|
also reduces brightness, while increasing contrast currently also |
839 |
|
|
increases brightness. |
840 |
root |
1.38 |
|
841 |
root |
1.45 |
=item brightness $bias, $img |
842 |
root |
1.36 |
|
843 |
|
|
=item brightness $r, $g, $b, $img |
844 |
|
|
|
845 |
|
|
=item brightness $r, $g, $b, $a, $img |
846 |
|
|
|
847 |
root |
1.38 |
Adjusts the brightness of an image. |
848 |
|
|
|
849 |
root |
1.45 |
The first form applies a single C<$bias> to red, green and blue, the |
850 |
|
|
second form applies separate biases to each colour channel, and the last |
851 |
|
|
form includes the alpha channel. |
852 |
|
|
|
853 |
|
|
Values less than 0 reduce brightness, while values larger than 0 increase |
854 |
|
|
it. Useful range is from -1 to 1 - the former results in a black, the |
855 |
|
|
latter in a white picture. |
856 |
|
|
|
857 |
sf-exg |
1.51 |
Due to idiosyncrasies in the underlying XRender extension, biases less |
858 |
root |
1.45 |
than zero can be I<very> slow. |
859 |
|
|
|
860 |
root |
1.75 |
You can also try the experimental(!) C<muladd> operator. |
861 |
|
|
|
862 |
root |
1.36 |
=cut |
863 |
root |
1.1 |
|
864 |
root |
1.2 |
sub contrast($$;$$;$) { |
865 |
root |
1.7 |
my $img = pop; |
866 |
|
|
my ($r, $g, $b, $a) = @_; |
867 |
root |
1.4 |
|
868 |
root |
1.49 |
($g, $b) = ($r, $r) if @_ < 3; |
869 |
|
|
$a = 1 if @_ < 4; |
870 |
root |
1.4 |
|
871 |
root |
1.1 |
$img = $img->clone; |
872 |
root |
1.37 |
$img->contrast ($r, $g, $b, $a); |
873 |
root |
1.1 |
$img |
874 |
|
|
} |
875 |
|
|
|
876 |
root |
1.2 |
sub brightness($$;$$;$) { |
877 |
root |
1.7 |
my $img = pop; |
878 |
|
|
my ($r, $g, $b, $a) = @_; |
879 |
root |
1.4 |
|
880 |
root |
1.49 |
($g, $b) = ($r, $r) if @_ < 3; |
881 |
|
|
$a = 1 if @_ < 4; |
882 |
root |
1.4 |
|
883 |
root |
1.1 |
$img = $img->clone; |
884 |
|
|
$img->brightness ($r, $g, $b, $a); |
885 |
|
|
$img |
886 |
|
|
} |
887 |
|
|
|
888 |
root |
1.75 |
=item muladd $mul, $add, $img # EXPERIMENTAL |
889 |
|
|
|
890 |
|
|
First multipliesthe pixels by C<$mul>, then adds C<$add>. This cna be used |
891 |
|
|
to implement brightness and contrast at the same time, with a wider value |
892 |
|
|
range than contrast and brightness operators. |
893 |
|
|
|
894 |
|
|
Due to numerous bugs in XRender implementations, it can also introduce a |
895 |
|
|
number of visual artifacts. |
896 |
|
|
|
897 |
|
|
Example: increase contrast by a factor of C<$c> without changing image |
898 |
|
|
brightness too much. |
899 |
|
|
|
900 |
|
|
muladd $c, (1 - $c) * 0.5, $img |
901 |
|
|
|
902 |
|
|
=cut |
903 |
|
|
|
904 |
|
|
sub muladd($$$) { |
905 |
|
|
$_[2]->muladd ($_[0], $_[1]) |
906 |
|
|
} |
907 |
|
|
|
908 |
root |
1.38 |
=item blur $radius, $img |
909 |
|
|
|
910 |
|
|
=item blur $radius_horz, $radius_vert, $img |
911 |
|
|
|
912 |
|
|
Gaussian-blurs the image with (roughly) C<$radius> pixel radius. The radii |
913 |
|
|
can also be specified separately. |
914 |
|
|
|
915 |
root |
1.39 |
Blurring is often I<very> slow, at least compared or other |
916 |
|
|
operators. Larger blur radii are slower than smaller ones, too, so if you |
917 |
|
|
don't want to freeze your screen for long times, start experimenting with |
918 |
|
|
low values for radius (<5). |
919 |
|
|
|
920 |
root |
1.38 |
=cut |
921 |
|
|
|
922 |
root |
1.36 |
sub blur($$;$) { |
923 |
|
|
my $img = pop; |
924 |
|
|
$img->blur ($_[0], @_ >= 2 ? $_[1] : $_[0]) |
925 |
|
|
} |
926 |
|
|
|
927 |
root |
1.52 |
=back |
928 |
|
|
|
929 |
|
|
=head2 OTHER STUFF |
930 |
root |
1.38 |
|
931 |
root |
1.56 |
Anything that didn't fit any of the other categories, even after applying |
932 |
root |
1.52 |
force and closing our eyes. |
933 |
|
|
|
934 |
|
|
=over 4 |
935 |
|
|
|
936 |
root |
1.66 |
=item keep { ... } |
937 |
root |
1.52 |
|
938 |
root |
1.66 |
This operator takes a code block as argument, that is, one or more |
939 |
root |
1.52 |
statements enclosed by braces. |
940 |
|
|
|
941 |
root |
1.68 |
The trick is that this code block is only evaluated when the outcome |
942 |
|
|
changes - on other calls the C<keep> simply returns the image it computed |
943 |
|
|
previously (yes, it should only be used with images). Or in other words, |
944 |
|
|
C<keep> I<caches> the result of the code block so it doesn't need to be |
945 |
|
|
computed again. |
946 |
|
|
|
947 |
|
|
This can be extremely useful to avoid redoing slow operations - for |
948 |
|
|
example, if your background expression takes the root background, blurs it |
949 |
|
|
and then root-aligns it it would have to blur the root background on every |
950 |
|
|
window move or resize. |
951 |
|
|
|
952 |
|
|
Another example is C<load>, which can be quite slow. |
953 |
root |
1.52 |
|
954 |
root |
1.63 |
In fact, urxvt itself encloses the whole expression in some kind of |
955 |
root |
1.68 |
C<keep> block so it only is reevaluated as required. |
956 |
root |
1.63 |
|
957 |
root |
1.68 |
Putting the blur into a C<keep> block will make sure the blur is only done |
958 |
|
|
once, while the C<rootalign> is still done each time the window moves. |
959 |
root |
1.52 |
|
960 |
sf-exg |
1.73 |
rootalign keep { blur 10, root } |
961 |
root |
1.52 |
|
962 |
root |
1.63 |
This leaves the question of how to force reevaluation of the block, |
963 |
|
|
in case the root background changes: If expression inside the block |
964 |
|
|
is sensitive to some event (root background changes, window geometry |
965 |
|
|
changes), then it will be reevaluated automatically as needed. |
966 |
root |
1.38 |
|
967 |
|
|
=cut |
968 |
|
|
|
969 |
root |
1.68 |
sub keep(&) { |
970 |
root |
1.63 |
my $id = $_[0]+0; |
971 |
|
|
|
972 |
|
|
local $frame = $self->{frame_cache}{$id} ||= [$frame]; |
973 |
|
|
|
974 |
|
|
unless ($frame->[FR_CACHE]) { |
975 |
|
|
$frame->[FR_CACHE] = [ $_[0]() ]; |
976 |
|
|
|
977 |
|
|
my $self = $self; |
978 |
|
|
my $frame = $frame; |
979 |
|
|
Scalar::Util::weaken $frame; |
980 |
|
|
$self->compile_frame ($frame, sub { |
981 |
|
|
# clear this frame cache, also for all parents |
982 |
|
|
for (my $frame = $frame; $frame; $frame = $frame->[0]) { |
983 |
|
|
undef $frame->[FR_CACHE]; |
984 |
|
|
} |
985 |
|
|
|
986 |
|
|
$self->recalculate; |
987 |
|
|
}); |
988 |
root |
1.55 |
}; |
989 |
|
|
|
990 |
|
|
# in scalar context we always return the first original result, which |
991 |
|
|
# is not quite how perl works. |
992 |
|
|
wantarray |
993 |
root |
1.63 |
? @{ $frame->[FR_CACHE] } |
994 |
|
|
: $frame->[FR_CACHE][0] |
995 |
root |
1.52 |
} |
996 |
|
|
|
997 |
root |
1.68 |
# sub keep_clear() { |
998 |
|
|
# delete $self->{frame_cache}; |
999 |
|
|
# } |
1000 |
root |
1.36 |
|
1001 |
root |
1.15 |
=back |
1002 |
|
|
|
1003 |
|
|
=cut |
1004 |
|
|
|
1005 |
root |
1.1 |
} |
1006 |
|
|
|
1007 |
|
|
sub parse_expr { |
1008 |
root |
1.63 |
my $expr = eval |
1009 |
|
|
"sub {\n" |
1010 |
|
|
. "package urxvt::bgdsl;\n" |
1011 |
|
|
. "#line 0 'background expression'\n" |
1012 |
|
|
. "$_[0]\n" |
1013 |
|
|
. "}"; |
1014 |
root |
1.1 |
die if $@; |
1015 |
|
|
$expr |
1016 |
|
|
} |
1017 |
|
|
|
1018 |
|
|
# compiles a parsed expression |
1019 |
|
|
sub set_expr { |
1020 |
|
|
my ($self, $expr) = @_; |
1021 |
|
|
|
1022 |
root |
1.74 |
$self->{root} = []; # the outermost frame |
1023 |
root |
1.1 |
$self->{expr} = $expr; |
1024 |
|
|
$self->recalculate; |
1025 |
|
|
} |
1026 |
|
|
|
1027 |
root |
1.63 |
# takes a hash of sensitivity indicators and installs watchers |
1028 |
|
|
sub compile_frame { |
1029 |
|
|
my ($self, $frame, $cb) = @_; |
1030 |
|
|
|
1031 |
|
|
my $state = $frame->[urxvt::bgdsl::FR_STATE] ||= {}; |
1032 |
|
|
my $again = $frame->[urxvt::bgdsl::FR_AGAIN]; |
1033 |
|
|
|
1034 |
|
|
# don't keep stuff alive |
1035 |
|
|
Scalar::Util::weaken $state; |
1036 |
|
|
|
1037 |
|
|
if ($again->{nested}) { |
1038 |
|
|
$state->{nested} = 1; |
1039 |
|
|
} else { |
1040 |
|
|
delete $state->{nested}; |
1041 |
|
|
} |
1042 |
|
|
|
1043 |
|
|
if (my $interval = $again->{time}) { |
1044 |
|
|
$state->{time} = [$interval, urxvt::timer->new->after ($interval)->interval ($interval)] |
1045 |
|
|
if $state->{time}[0] != $interval; |
1046 |
|
|
|
1047 |
|
|
# callback *might* have changed, although we could just rule that out |
1048 |
|
|
$state->{time}[1]->cb (sub { |
1049 |
|
|
++$state->{counter}; |
1050 |
|
|
$cb->(); |
1051 |
|
|
}); |
1052 |
|
|
} else { |
1053 |
|
|
delete $state->{time}; |
1054 |
|
|
} |
1055 |
|
|
|
1056 |
|
|
if ($again->{position}) { |
1057 |
|
|
$state->{position} = $self->on (position_change => $cb); |
1058 |
|
|
} else { |
1059 |
|
|
delete $state->{position}; |
1060 |
|
|
} |
1061 |
|
|
|
1062 |
|
|
if ($again->{size}) { |
1063 |
|
|
$state->{size} = $self->on (size_change => $cb); |
1064 |
|
|
} else { |
1065 |
|
|
delete $state->{size}; |
1066 |
|
|
} |
1067 |
|
|
|
1068 |
|
|
if ($again->{rootpmap}) { |
1069 |
|
|
$state->{rootpmap} = $self->on (rootpmap_change => $cb); |
1070 |
|
|
} else { |
1071 |
|
|
delete $state->{rootpmap}; |
1072 |
|
|
} |
1073 |
|
|
} |
1074 |
|
|
|
1075 |
root |
1.1 |
# evaluate the current bg expression |
1076 |
|
|
sub recalculate { |
1077 |
root |
1.33 |
my ($arg_self) = @_; |
1078 |
root |
1.1 |
|
1079 |
root |
1.10 |
# rate limit evaluation |
1080 |
|
|
|
1081 |
root |
1.33 |
if ($arg_self->{next_refresh} > urxvt::NOW) { |
1082 |
|
|
$arg_self->{next_refresh_timer} = urxvt::timer->new->after ($arg_self->{next_refresh} - urxvt::NOW)->cb (sub { |
1083 |
|
|
$arg_self->recalculate; |
1084 |
root |
1.9 |
}); |
1085 |
root |
1.12 |
return; |
1086 |
root |
1.9 |
} |
1087 |
|
|
|
1088 |
root |
1.33 |
$arg_self->{next_refresh} = urxvt::NOW + $MIN_INTERVAL; |
1089 |
root |
1.9 |
|
1090 |
root |
1.10 |
# set environment to evaluate user expression |
1091 |
root |
1.6 |
|
1092 |
root |
1.63 |
local $self = $arg_self; |
1093 |
|
|
local $HOME = $ENV{HOME}; |
1094 |
root |
1.74 |
local $frame = $self->{root}; |
1095 |
root |
1.1 |
|
1096 |
root |
1.63 |
($x, $y, $w, $h) = $self->background_geometry ($self->{border}); |
1097 |
root |
1.22 |
|
1098 |
root |
1.10 |
# evaluate user expression |
1099 |
|
|
|
1100 |
root |
1.63 |
my @img = eval { $self->{expr}->() }; |
1101 |
root |
1.61 |
die $@ if $@; |
1102 |
root |
1.63 |
die "background-expr did not return anything.\n" unless @img; |
1103 |
|
|
die "background-expr: expected image(s), got something else.\n" |
1104 |
|
|
if grep { !UNIVERSAL::isa $_, "urxvt::img" } @img; |
1105 |
root |
1.1 |
|
1106 |
root |
1.63 |
my $img = urxvt::bgdsl::merge @img; |
1107 |
root |
1.10 |
|
1108 |
root |
1.63 |
$frame->[FR_AGAIN]{size} = 1 |
1109 |
root |
1.55 |
if $img->repeat_mode != urxvt::RepeatNormal; |
1110 |
|
|
|
1111 |
root |
1.63 |
# if the expression is sensitive to external events, prepare reevaluation then |
1112 |
|
|
$self->compile_frame ($frame, sub { $arg_self->recalculate }); |
1113 |
root |
1.9 |
|
1114 |
root |
1.10 |
# clear stuff we no longer need |
1115 |
|
|
|
1116 |
root |
1.63 |
# unless (%{ $frame->[FR_STATE] }) { |
1117 |
|
|
# delete $self->{state}; |
1118 |
|
|
# delete $self->{expr}; |
1119 |
|
|
# } |
1120 |
root |
1.5 |
|
1121 |
root |
1.34 |
# set background pixmap |
1122 |
root |
1.1 |
|
1123 |
root |
1.33 |
$self->set_background ($img, $self->{border}); |
1124 |
root |
1.1 |
$self->scr_recolour (0); |
1125 |
|
|
$self->want_refresh; |
1126 |
|
|
} |
1127 |
|
|
|
1128 |
|
|
sub on_start { |
1129 |
|
|
my ($self) = @_; |
1130 |
|
|
|
1131 |
root |
1.47 |
my $expr = $self->x_resource ("%.expr") |
1132 |
root |
1.33 |
or return; |
1133 |
|
|
|
1134 |
root |
1.48 |
$self->has_render |
1135 |
|
|
or die "background extension needs RENDER extension 0.10 or higher, ignoring background-expr.\n"; |
1136 |
|
|
|
1137 |
root |
1.33 |
$self->set_expr (parse_expr $expr); |
1138 |
root |
1.47 |
$self->{border} = $self->x_resource_boolean ("%.border"); |
1139 |
root |
1.1 |
|
1140 |
root |
1.47 |
$MIN_INTERVAL = $self->x_resource ("%.interval"); |
1141 |
root |
1.46 |
|
1142 |
root |
1.1 |
() |
1143 |
|
|
} |
1144 |
|
|
|