1 <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN"
\r
2 "http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">
\r
3 <html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en">
\r
5 <link rel="icon" type="image/png" href="/favicon.png">
\r
6 <meta http-equiv="Content-Type" content="application/xhtml+xml; charset=UTF-8" />
\r
7 <meta name="generator" content="AsciiDoc 8.6.7" />
\r
8 <title>i3: i3 testsuite</title>
\r
9 <link rel="stylesheet" href="/css/style.css" type="text/css" />
\r
10 <link rel="stylesheet" href="/css/xhtml11.css" type="text/css" />
\r
11 <script type="text/javascript">
\r
13 window.onload = function(){asciidoc.footnotes(); asciidoc.toc(2);}
\r
16 <script type="text/javascript" src="/js/asciidoc-xhtml11.js"></script>
\r
18 <body class="article">
\r
21 <a href="/"><h1 id="title">i3 - improved tiling WM</h1></a>
\r
23 <li><a style="border-bottom: 2px solid #fff" href="/docs">Docs</a></li>
\r
24 <li><a href="/screenshots">Screens</a></li>
\r
25 <li><a href="/contact">Contact</a></li>
\r
26 <li><a href="http://bugs.i3wm.org/">Bugs</a></li>
\r
28 <br style="clear: both">
\r
31 <h1>i3 testsuite</h1>
\r
32 <span id="author">Michael Stapelberg</span><br />
\r
33 <span id="email"><tt><<a href="mailto:michael@i3wm.org">michael@i3wm.org</a>></tt></span><br />
\r
34 <span id="revdate">September 2012</span>
\r
36 <div id="toctitle">Table of Contents</div>
37 <noscript><p><b>JavaScript must be enabled in your browser to display the table of contents.</b></p></noscript>
41 <div class="sectionbody">
\r
42 <div class="paragraph"><p>This document explains how the i3 testsuite works, how to use it and extend it.
\r
43 It is targeted at developers who not necessarily have been doing testing before
\r
44 or have not been testing in Perl before. In general, the testsuite is not of
\r
45 interest for end users.</p></div>
\r
49 <h2 id="_introduction">1. Introduction</h2>
\r
50 <div class="sectionbody">
\r
51 <div class="paragraph"><p>The i3 testsuite is a collection of files which contain testcases for various
\r
52 i3 features. Some of them test if a certain workflow works correctly (moving
\r
53 windows, focus behaviour, …). Others are regression tests and contain code
\r
54 which previously made i3 crash or lead to unexpected behaviour. They then check
\r
55 if i3 still runs (meaning it did not crash) and if it handled everything
\r
56 correctly.</p></div>
\r
57 <div class="paragraph"><p>The goal of having these tests is to automatically find problems and to
\r
58 automatically get a feel for whether a change in the source code breaks any
\r
59 existing feature. After every modification of the i3 sourcecode, the developer
\r
60 should run the full testsuite. If one of the tests fails, the corresponding
\r
61 problem should be fixed (or, in some cases, the testcase has to be modified).
\r
62 For every bugreport, a testcase should be written to test the correct
\r
63 behaviour. Initially, it will fail, but after fixing the bug, it will pass.
\r
64 This ensures (or increases the chance) that bugs which have been fixed once
\r
65 will never be found again.</p></div>
\r
66 <div class="paragraph"><p>Also, when implementing a new feature, a testcase might be a good way to be
\r
67 able to easily test if the feature is working correctly. Many developers will
\r
68 test manually if everything works. Having a testcase not only helps you with
\r
69 that, but it will also be useful for every future change.</p></div>
\r
73 <h2 id="_relevant_documentation">2. Relevant documentation</h2>
\r
74 <div class="sectionbody">
\r
75 <div class="paragraph"><p>Apart from this document, you should also have a look at:</p></div>
\r
76 <div class="olist arabic"><ol class="arabic">
\r
79 The "Modern Perl" book, which can be found at
\r
80 <a href="http://onyxneon.com/books/modern_perl/modern_perl_a4.pdf">http://onyxneon.com/books/modern_perl/modern_perl_a4.pdf</a>
\r
85 The latest Perl documentation of the "i3test" (general testcase setup) and
\r
86 "i3test::Test" (additional test instructions) modules:
\r
87 <a href="http://build.i3wm.org/docs/lib-i3test.html">http://build.i3wm.org/docs/lib-i3test.html</a> respectively
\r
88 <a href="http://build.i3wm.org/docs/lib-i3test-test.html">http://build.i3wm.org/docs/lib-i3test-test.html</a>
\r
93 The latest documentation on i3’s IPC interface:
\r
94 <a href="http://build.i3wm.org/docs/ipc.html">http://build.i3wm.org/docs/ipc.html</a>
\r
100 <div class="sect1">
\r
101 <h2 id="_implementation">3. Implementation</h2>
\r
102 <div class="sectionbody">
\r
103 <div class="paragraph"><p>For several reasons, the i3 testsuite has been implemented in Perl:</p></div>
\r
104 <div class="olist arabic"><ol class="arabic">
\r
107 Perl has a long tradition of testing. Every popular/bigger Perl module which
\r
108 you can find on CPAN will not only come with documentation, but also with
\r
109 tests. Therefore, the available infrastructure for tests is comprehensive.
\r
110 See for example the excellent <a href="http://search.cpan.org/perldoc?Test::More">http://search.cpan.org/perldoc?Test::More</a>
\r
111 and the referenced <a href="http://search.cpan.org/perldoc?Test::Tutorial">http://search.cpan.org/perldoc?Test::Tutorial</a>.
\r
116 Perl is widely available and has a well-working package infrastructure.
\r
121 The author is familiar with Perl :).
\r
126 It is a good idea to use a different language for the tests than the
\r
127 implementation itself.
\r
131 <div class="paragraph"><p>Please do not start programming language flamewars at this point.</p></div>
\r
132 <div class="sect2">
\r
133 <h3 id="_installing_the_dependencies">3.1. Installing the dependencies</h3>
\r
134 <div class="paragraph"><p>As usual with Perl programs, the testsuite ships with a <tt>Makefile.PL</tt>.
\r
135 This file specifies which Perl modules the testsuite depends on and can be used
\r
136 to install all of them.</p></div>
\r
137 <div class="paragraph"><p>Perl modules are distributed via CPAN, and there is the official, standard CPAN
\r
138 client, simply called <tt>cpan</tt>. It comes with every Perl installation and can be
\r
139 used to install the testsuite. Many users prefer to use the more modern
\r
140 <tt>cpanminus</tt> instead, though (because it asks no questions and just works):</p></div>
\r
141 <div class="listingblock">
\r
142 <div class="title">Installing testsuite dependencies using cpanminus (preferred)</div>
\r
143 <div class="content">
\r
144 <pre><tt>$ cd ~/i3/testcases
\r
145 $ sudo apt-get install cpanminus
\r
146 $ sudo cpanm .</tt></pre>
\r
148 <div class="paragraph"><p>If you don’t want to use cpanminus for some reason, the same works with cpan:</p></div>
\r
149 <div class="listingblock">
\r
150 <div class="title">Installing testsuite dependencies using cpan</div>
\r
151 <div class="content">
\r
152 <pre><tt>$ cd ~/i3/testcases
\r
153 $ sudo cpan .</tt></pre>
\r
155 <div class="paragraph"><p>In case you don’t have root permissions, you can also install into your home
\r
156 directory, see <a href="http://michael.stapelberg.de/cpan/">http://michael.stapelberg.de/cpan/</a></p></div>
\r
158 <div class="sect2">
\r
159 <h3 id="_mechanisms">3.2. Mechanisms</h3>
\r
160 <div class="sect3">
\r
161 <h4 id="_script_complete_run">3.2.1. Script: complete-run</h4>
\r
162 <div class="paragraph"><p>The testcases are run by a script called <tt>complete-run.pl</tt>. It runs all
\r
163 testcases by default, but you can be more specific and let it only run one or
\r
164 more testcases. Also, it takes care of starting up a separate instance of i3
\r
165 with an appropriate configuration file and creates a folder for each run
\r
166 containing the appropriate i3 logfile for each testcase. The latest folder can
\r
167 always be found under the symlink <tt>latest/</tt>. Unless told differently, it will
\r
168 run the tests on a separate X server instance (using the Xdummy script).</p></div>
\r
169 <div class="listingblock">
\r
170 <div class="title">Example invocation of complete-run.pl+</div>
\r
171 <div class="content">
\r
172 <pre><tt>$ cd ~/i3/testcases
\r
174 $ ./complete-run.pl
\r
175 # output omitted because it is very long
\r
176 All tests successful.
\r
177 Files=78, Tests=734, 27 wallclock secs ( 0.38 usr 0.48 sys + 17.65 cusr 3.21 csys = 21.72 CPU)
\r
180 $ ./complete-run.pl t/04-floating.t
\r
181 [:3] i3 startup: took 0.07s, status = 1
\r
182 [:3] Running t/04-floating.t with logfile testsuite-2011-09-24-16-06-04-4.0.2-226-g1eb011a/i3-log-for-04-floating.t
\r
183 [:3] t/04-floating.t finished
\r
185 output for t/04-floating.t:
\r
186 ok 1 - use X11::XCB::Window;
\r
187 ok 2 - The object isa X11::XCB::Window
\r
188 ok 3 - Window is mapped
\r
189 ok 4 - i3 raised the width to 75
\r
190 ok 5 - i3 raised the height to 50
\r
191 ok 6 - i3 did not map it to (0x0)
\r
192 ok 7 - The object isa X11::XCB::Window
\r
193 ok 8 - i3 let the width at 80
\r
194 ok 9 - i3 let the height at 90
\r
195 ok 10 - i3 mapped it to x=1
\r
196 ok 11 - i3 mapped it to y=18
\r
197 ok 12 - The object isa X11::XCB::Window
\r
198 ok 13 - i3 let the width at 80
\r
199 ok 14 - i3 let the height at 90
\r
202 All tests successful.
\r
203 Files=1, Tests=14, 0 wallclock secs ( 0.01 usr 0.00 sys + 0.19 cusr 0.03 csys = 0.23 CPU)
\r
206 $ less latest/i3-log-for-04-floating.t</tt></pre>
\r
209 <div class="sect3">
\r
210 <h4 id="_ipc_interface">3.2.2. IPC interface</h4>
\r
211 <div class="paragraph"><p>The testsuite makes extensive use of the IPC (Inter-Process Communication)
\r
212 interface which i3 provides. It is used for the startup process of i3, for
\r
213 terminating it cleanly and (most importantly) for modifying and getting the
\r
214 current state (layout tree).</p></div>
\r
215 <div class="paragraph"><p>See [<a href="http://i3wm.org/docs/ipc.html">http://i3wm.org/docs/ipc.html</a>] for documentation on the IPC interface.</p></div>
\r
217 <div class="sect3">
\r
218 <h4 id="_x11_xcb">3.2.3. X11::XCB</h4>
\r
219 <div class="paragraph"><p>In order to open new windows, change attributes, get events, etc., the
\r
220 testsuite uses X11::XCB, a new (and quite specific to i3 at the moment) Perl
\r
221 module which uses the XCB protocol description to generate Perl bindings to
\r
222 X11. They work in a very similar way to libxcb (which i3 uses) and provide
\r
223 relatively high-level interfaces (objects such as <tt>X11::XCB::Window</tt>) aswell as
\r
224 access to the low-level interface, which is very useful when testing a window
\r
228 <div class="sect2">
\r
229 <h3 id="_filesystem_structure">3.3. Filesystem structure</h3>
\r
230 <div class="paragraph"><p>In the git root of i3, the testcases live in the folder <tt>testcases</tt>. This
\r
231 folder contains the <tt>complete-run.pl</tt> and <tt>Xdummy</tt> scripts and a base
\r
232 configuration file which will be used for the tests. The different testcases
\r
233 (their file extension is .t, not .pl) themselves can be found in the
\r
234 conventionally named subfolder <tt>t</tt>:</p></div>
\r
235 <div class="listingblock">
\r
236 <div class="title">Filesystem structure</div>
\r
237 <div class="content">
\r
238 <pre><tt>├── testcases
\r
239 │ ├── complete-run.pl
\r
240 │ ├── i3-test.config
\r
243 │ │ ├── SocketActivation.pm
\r
244 │ │ └── StartXDummy.pm
\r
248 │ │ ├── 02-fullscreen.t
\r
250 │ │ ├── omitted for brevity
\r
252 │ │ └── 74-regress-focus-toggle.t
\r
253 │ └── Xdummy</tt></pre>
\r
258 <div class="sect1">
\r
259 <h2 id="_anatomy_of_a_testcase">4. Anatomy of a testcase</h2>
\r
260 <div class="sectionbody">
\r
261 <div class="paragraph"><p>Learning by example is definitely a good strategy when you are wondering how to
\r
262 write a testcase. Let’s take <tt>t/11-goto.t</tt> as an easy example and go through it
\r
263 step by step:</p></div>
\r
264 <div class="listingblock">
\r
265 <div class="title">t/11-goto.t: Boilerplate</div>
\r
266 <div class="content">
\r
268 # vim:ts=4:sw=4:expandtab
\r
273 my $x = X11::XCB::Connection->new;</tt></pre>
\r
275 <div class="paragraph"><p>This is what we call boilerplate. It exists at the top of every test file (to
\r
276 some extent). The first line is the shebang, which specifies that this file is
\r
277 a Perl script. The second line contains VIM specific settings on how to
\r
278 edit/format this file (use spaces instead of tabs, indent using 4 spaces).
\r
279 Afterwards, the <tt>i3test</tt> module is used. This module contains i3 testsuite
\r
280 specific functions which you are strongly encouraged to use. They make writing
\r
281 testcases a lot easier and will make it easier for other people to read your
\r
283 <div class="paragraph"><p>The next line uses the <tt>File::Temp</tt> module. This is specific to this testcase,
\r
284 because it needs to generate a temporary name during the test. Many testcases
\r
285 use only the <tt>i3test</tt> module.</p></div>
\r
286 <div class="paragraph"><p>The last line opens a connection to X11. You might or might not need this in
\r
287 your testcase, depending on whether you are going to open windows (etc.) or
\r
288 only use i3 commands.</p></div>
\r
289 <div class="listingblock">
\r
290 <div class="title">t/11-goto.t: Setup</div>
\r
291 <div class="content">
\r
292 <pre><tt>my $tmp = fresh_workspace;
\r
294 cmd 'split h';</tt></pre>
\r
296 <div class="paragraph"><p>The first line calls i3test’s <tt>fresh_workspace</tt> function which looks for a
\r
297 currently unused workspace, switches to it, and returns its name. The variable
\r
298 <tt>$tmp</tt> will end up having a value such as <tt>"/tmp/87kBVcHbA9"</tt>. Note that this
\r
299 is not (necessarily) a valid path, it’s just a random workspace name.</p></div>
\r
300 <div class="paragraph"><p>So, now that we are on a new workspace, we ensure that the workspace uses
\r
301 horizontal orientation by issuing the <tt>split h</tt> command (see the i3 User’s
\r
302 Guide for a list of commands). This is not strictly necessary, but good style.
\r
303 In general, the <tt>cmd</tt> function executes the specified i3 command by using the
\r
304 IPC interface and returns once i3 acknowledged the command.</p></div>
\r
305 <div class="listingblock">
\r
306 <div class="title">t/11-goto.t: Setup</div>
\r
307 <div class="content">
\r
308 <pre><tt>#####################################################################
\r
309 # Create two windows and make sure focus switching works
\r
310 #####################################################################
\r
312 my $top = open_window($x);
\r
313 my $mid = open_window($x);
\r
314 my $bottom = open_window($x);</tt></pre>
\r
316 <div class="paragraph"><p>In every major section of a testcase, you should put a comment like the one
\r
317 above. This makes it immediately clear how the file is structured.</p></div>
\r
318 <div class="paragraph"><p>The <tt>open_window</tt> function opens a standard window, which will then be put into
\r
319 tiling mode by i3. If you want a floating window, use the
\r
320 <tt>open_floating_window</tt> function. These functions accept the same parameters as
\r
321 <tt>X11::XCB::Window→new</tt>, see the i3test documentation at TODO.</p></div>
\r
322 <div class="listingblock">
\r
323 <div class="title">t/11-goto.t: Helper function</div>
\r
324 <div class="content">
\r
326 # Returns the input focus after sending the given command to i3 via IPC
\r
327 # and syncing with i3
\r
334 return $x->input_focus;
\r
337 <div class="paragraph"><p>This section defines a helper function which will be used over and over in this
\r
338 testcase. If you have code which gets executed more than once or twice
\r
339 (depending on the length of your test, use your best judgement), please put it
\r
340 in a function. Tests should be short, concise and clear.</p></div>
\r
341 <div class="paragraph"><p>The <tt>focus_after</tt> function executes a command and returns the X11 focus after
\r
342 the command was executed. The <tt>sync_with_i3</tt> command makes sure that i3 could
\r
343 push its state to X11. See <a href="#i3_sync">[i3_sync]</a> to learn how this works exactly.</p></div>
\r
344 <div class="listingblock">
\r
345 <div class="title">t/11-goto.t: Test assumptions</div>
\r
346 <div class="content">
\r
347 <pre><tt>$focus = $x->input_focus;
\r
348 is($focus, $bottom->id, "Latest window focused");
\r
350 $focus = focus_after('focus left');
\r
351 is($focus, $mid->id, "Middle window focused");</tt></pre>
\r
353 <div class="paragraph"><p>Now, we run the first two real tests. They use <tt>Test::More</tt>'s <tt>is</tt> function,
\r
354 which compares two values and prints the differences if they are not the same.
\r
355 After the arguments, we supply a short comment to indicate what we are testing
\r
356 here. This makes it vastly more easy for the developer to spot which testcase
\r
357 is the problem in case one fails.</p></div>
\r
358 <div class="paragraph"><p>The first test checks that the most recently opened window is focused.
\r
359 Afterwards, the command <tt>focus left</tt> is issued and it is verified that the
\r
360 middle window now has focus.</p></div>
\r
361 <div class="paragraph"><p>Note that this is not a comprehensive test of the <tt>focus</tt> command — we would
\r
362 have to test wrapping, focus when using a more complex layout, focusing the
\r
363 parent/child containers, etc. But that is not the point of this testcase.
\r
364 Instead, we just want to know if <tt>$x→input_focus</tt> corresponds with what we are
\r
365 expecting. If not, something is completely wrong with the test environment and
\r
366 this trivial test will fail.</p></div>
\r
367 <div class="listingblock">
\r
368 <div class="title">t/11-goto.t: Test that the feature does not work (yet)</div>
\r
369 <div class="content">
\r
370 <pre><tt>#####################################################################
\r
371 # Now goto a mark which does not exist
\r
372 #####################################################################
\r
374 my $random_mark = mktemp('mark.XXXXXX');
\r
376 $focus = focus_after(qq|[con_mark="$random_mark"] focus|);
\r
377 is($focus, $mid->id, "focus unchanged");</tt></pre>
\r
379 <div class="paragraph"><p>Syntax hint: The qq keyword is the interpolating quote operator. It lets you
\r
380 chose a quote character (in this case the <tt>|</tt> character, a pipe). This makes
\r
381 having double quotes in our string easy.</p></div>
\r
382 <div class="paragraph"><p>In this new major section, a random mark (mark is an identifier for a window,
\r
383 see "VIM-like marks" in the i3 User’s Guide) will be generated. Afterwards, we
\r
384 test that trying to focus that mark will not do anything. This is important: Do
\r
385 not only test that using a feature has the expected outcome, but also test that
\r
386 using it without properly initializing it does no harm. This command could for
\r
387 example have changed focus anyways (a bug) or crash i3 (obviously a bug).</p></div>
\r
388 <div class="listingblock">
\r
389 <div class="title">t/11-goto.t: Test that the feature does work</div>
\r
390 <div class="content">
\r
391 <pre><tt>cmd "mark $random_mark";
\r
393 $focus = focus_after('focus left');
\r
394 is($focus, $top->id, "Top window focused");
\r
396 $focus = focus_after(qq|[con_mark="$random_mark"] focus|);
\r
397 is($focus, $mid->id, "goto worked");</tt></pre>
\r
399 <div class="paragraph"><p>Remember: Focus was on the middle window (we verified that earlier in "Test
\r
400 assumptions"). We now mark the middle window with our randomly generated mark.
\r
401 Afterwards, we switch focus away from the middle window to be able to tell if
\r
402 focusing it via its mark will work. If the test works, the goto command seems
\r
403 to be working.</p></div>
\r
404 <div class="listingblock">
\r
405 <div class="title">t/11-goto.t: Test corner case</div>
\r
406 <div class="content">
\r
407 <pre><tt># check that we can specify multiple criteria
\r
409 $focus = focus_after('focus left');
\r
410 is($focus, $top->id, "Top window focused");
\r
412 $focus = focus_after(qq|[con_mark="$random_mark" con_mark="$random_mark"] focus|);
\r
413 is($focus, $mid->id, "goto worked");</tt></pre>
\r
415 <div class="paragraph"><p>Now we test the same feature, but specifying the mark twice in the command.
\r
416 This should have no effect, but let’s be sure: test it and see if things go
\r
418 <div class="listingblock">
\r
419 <div class="title">t/11-goto.t: Test second code path</div>
\r
420 <div class="content">
\r
421 <pre><tt>#####################################################################
\r
422 # Check whether the focus command will switch to a different
\r
423 # workspace if necessary
\r
424 #####################################################################
\r
426 my $tmp2 = fresh_workspace;
\r
428 is(focused_ws(), $tmp2, 'tmp2 now focused');
\r
430 cmd qq|[con_mark="$random_mark"] focus|;
\r
432 is(focused_ws(), $tmp, 'tmp now focused');</tt></pre>
\r
434 <div class="paragraph"><p>This part of the test checks that focusing windows by mark works across
\r
435 workspaces. It uses i3test’s <tt>focused_ws</tt> function to get the current
\r
436 workspace.</p></div>
\r
437 <div class="listingblock">
\r
438 <div class="title">t/11-goto.t: Test second code path</div>
\r
439 <div class="content">
\r
440 <pre><tt>done_testing;</tt></pre>
\r
442 <div class="paragraph"><p>The end of every testcase has to contain the <tt>done_testing</tt> line. This tells
\r
443 <tt>complete-run.pl</tt> that the test was finished successfully. If it does not
\r
444 occur, the test might have crashed during execution — some of the reasons why
\r
445 that could happen are bugs in the used modules, bugs in the testcase itself or
\r
446 an i3 crash resulting in the testcase being unable to communicate with i3 via
\r
447 IPC anymore.</p></div>
\r
450 <div class="sect1">
\r
451 <h2 id="i3_sync">5. Appendix A: The i3 sync protocol</h2>
\r
452 <div class="sectionbody">
\r
453 <div class="paragraph"><p>Consider the following situation: You open two windows in your testcase, then
\r
454 you use <tt>focus left</tt> and want to verify that the X11 focus has been updated
\r
455 properly. Sounds simple, right? Let’s assume you use this straight-forward
\r
456 implementation:</p></div>
\r
457 <div class="listingblock">
\r
458 <div class="title">Racey focus testcase</div>
\r
459 <div class="content">
\r
460 <pre><tt>my $left = open_window($x);
\r
461 my $right = open_window($x);
\r
463 is($x->input_focus, $left->id, 'left window focused');</tt></pre>
\r
465 <div class="paragraph"><p>However, the test fails. Sometimes. Apparantly, there is a race condition in
\r
466 your test. If you think about it, this is because you are using two different
\r
467 pieces of software: You tell i3 to update focus, i3 confirms that, and then you
\r
468 ask X11 to give you the current focus. There is a certain time i3 needs to
\r
469 update the X11 state. If the testcase gets CPU time before X11 processed i3’s
\r
470 requests, the test will fail.</p></div>
\r
471 <div class="imageblock">
\r
472 <div class="content">
\r
473 <img src="i3-sync.png" alt="Diagram of the race condition" />
\r
475 <div class="title">Figure 1. Diagram of the race condition</div>
\r
477 <div class="paragraph"><p>One way to "solve" this would be to add <tt>sleep 0.5;</tt> after the <tt>cmd</tt> call.
\r
478 After 0.5 seconds it should be safe to assume that focus has been updated,
\r
480 <div class="paragraph"><p>In practice, this usually works. However, it has several problems:</p></div>
\r
481 <div class="olist arabic"><ol class="arabic">
\r
484 This is obviously not a clean solution, but a workaround. Ugly.
\r
489 On very slow machines, this might not work. Unlikely, but in different
\r
490 situations (a delay to wait for i3 to startup) the necessary time is much
\r
491 harder to guess, even for fast machines.
\r
496 This <strong>wastes a lot of time</strong>. Usually, your computer is much faster than 0.5s
\r
497 to update the status. However, sometimes, it might take 0.4s, so we can’t
\r
498 make it <tt>sleep 0.1</tt>.
\r
502 <div class="paragraph"><p>To illustrate how grave the problem with wasting time actually is: Before
\r
503 removing all sleeps from the testsuite, a typical run using 4 separate X
\r
504 servers took around 50 seconds on my machine. After removing all the sleeps,
\r
505 we achieved times of about 25 seconds. This is very significant and influences
\r
506 the way you think about tests — the faster they are, the more likely you are
\r
507 to check whether everything still works quite often (which you should).</p></div>
\r
508 <div class="paragraph"><p>What I am trying to say is: Delays adds up quickly and make the test suite
\r
509 less robust.</p></div>
\r
510 <div class="paragraph"><p>The real solution for this problem is a mechanism which I call "the i3 sync
\r
511 protocol". The idea is to send a request (which does not modify state) via X11
\r
512 to i3 which will then be answered. Due to the request’s position in the event
\r
513 queue (<strong>after</strong> all previous events), you can be sure that by the time you
\r
514 receive the reply, all other events have been dealt with by i3 (and, more
\r
515 importantly, X11).</p></div>
\r
516 <div class="imageblock">
\r
517 <div class="content">
\r
518 <img src="i3-sync-working.png" alt="Diagram of the i3 sync solution" />
\r
520 <div class="title">Figure 2. Diagram of the i3 sync solution</div>
\r
522 <div class="sect2">
\r
523 <h3 id="_implementation_details">5.1. Implementation details</h3>
\r
524 <div class="paragraph"><p>The client which wants to sync with i3 initiates the protocol by sending a
\r
525 ClientMessage to the X11 root window:</p></div>
\r
526 <div class="listingblock">
\r
527 <div class="title">Send ClientMessage</div>
\r
528 <div class="content">
\r
529 <pre><tt># Generate a ClientMessage, see xcb_client_message_t
\r
530 my $msg = pack "CCSLLLLLLL",
\r
531 CLIENT_MESSAGE, # response_type
\r
534 $root, # destination window
\r
535 $x->atom(name => 'I3_SYNC')->id,
\r
537 $_sync_window->id, # data[0]: our own window id
\r
538 $myrnd, # data[1]: a random value to identify the request
\r
543 # Send it to the root window -- since i3 uses the SubstructureRedirect
\r
544 # event mask, it will get the ClientMessage.
\r
545 $x->send_event(0, $root, EVENT_MASK_SUBSTRUCTURE_REDIRECT, $msg);</tt></pre>
\r
547 <div class="paragraph"><p>i3 will then reply with the same ClientMessage, sent to the window specified in
\r
548 <tt>data[0]</tt>. In the reply, <tt>data[0]</tt> and <tt>data[1]</tt> are exactly the same as in the
\r
549 request. You should use a random value in <tt>data[1]</tt> and check that you received
\r
550 the same one when getting the reply.</p></div>
\r
554 <div class="sect1">
\r
555 <h2 id="_appendix_b_socket_activation">6. Appendix B: Socket activation</h2>
\r
556 <div class="sectionbody">
\r
557 <div class="paragraph"><p>Socket activation is a mechanism which was made popular by systemd, an init
\r
558 replacement. It basically describes creating a listening socket before starting
\r
559 a program. systemd will invoke the program only when an actual connection to
\r
560 the socket is made, hence the term socket activation.</p></div>
\r
561 <div class="paragraph"><p>The interesting part of this (in the i3 context) is that you can very precisely
\r
562 detect when the program is ready (finished its initialization).</p></div>
\r
563 <div class="sect2">
\r
564 <h3 id="_preparing_the_listening_socket">6.1. Preparing the listening socket</h3>
\r
565 <div class="paragraph"><p><tt>complete-run.pl</tt> will create a listening UNIX socket which it will then pass
\r
566 to i3. This socket will be used by i3 as an additional IPC socket, just like
\r
567 the one it will create on its own. Passing the socket happens implicitly
\r
568 because children will inherit the parent’s sockets when fork()ing and sockets
\r
569 will continue to exist after an exec() call (unless CLOEXEC is set of course).</p></div>
\r
570 <div class="paragraph"><p>The only explicit things <tt>complete-run.pl</tt> has to do is setting the <tt>LISTEN_FDS</tt>
\r
571 environment variable to the number of sockets which exist (1 in our case) and
\r
572 setting the <tt>LISTEN_PID</tt> environment variable to the current process ID. Both
\r
573 variables are necessary so that the program (i3) knows how many sockets it
\r
574 should use and if the environment variable is actually intended for it. i3 will
\r
575 then start looking for sockets at file descriptor 3 (since 0, 1 and 2 are used
\r
576 for stdin, stdout and stderr, respectively).</p></div>
\r
577 <div class="paragraph"><p>The actual Perl code which sets up the socket, fork()s, makes sure the socket
\r
578 has file descriptor 3 and sets up the environment variables follows (shortened
\r
580 <div class="listingblock">
\r
581 <div class="title">Setup socket and environment</div>
\r
582 <div class="content">
\r
583 <pre><tt>my $socket = IO::Socket::UNIX->new(
\r
585 Local => $args{unix_socket_path},
\r
590 $ENV{LISTEN_PID} = $$;
\r
591 $ENV{LISTEN_FDS} = 1;
\r
593 # Only pass file descriptors 0 (stdin), 1 (stdout),
\r
594 # 2 (stderr) and 3 (socket) to the child.
\r
597 # If the socket does not use file descriptor 3 by chance
\r
598 # already, we close fd 3 and dup2() the socket to 3.
\r
599 if (fileno($socket) != 3) {
\r
601 POSIX::dup2(fileno($socket), 3);
\r
604 exec "/usr/bin/i3";
\r
608 <div class="sect2">
\r
609 <h3 id="_waiting_for_a_reply">6.2. Waiting for a reply</h3>
\r
610 <div class="paragraph"><p>In the parent process, we want to know when i3 is ready to answer our IPC
\r
611 requests and handle our windows. Therefore, after forking, we immediately close
\r
612 the listening socket (i3 will handle this side of the socket) and connect to it
\r
613 (remember, we are talking about a named UNIX socket) as a client. This connect
\r
614 call will immediately succeed because the kernel buffers it. Then, we send a
\r
615 request (of type GET_TREE, but that is not really relevant). Writing data to
\r
616 the socket will also succeed immediately because, again, the kernel buffers it
\r
617 (only up to a certain amount of data of course).</p></div>
\r
618 <div class="paragraph"><p>Afterwards, we just blockingly wait until we get an answer. In the child
\r
619 process, i3 will setup the listening socket in its event loop. Immediately
\r
620 after actually starting the event loop, it will notice a new client connecting
\r
621 (the parent process) and handle its request. Since all initialization has been
\r
622 completed successfully by the time the event loop is entered, we can now assume
\r
623 that i3 is ready.</p></div>
\r
625 <div class="sect2">
\r
626 <h3 id="_timing_and_conclusion">6.3. Timing and conclusion</h3>
\r
627 <div class="paragraph"><p>A beautiful feature of this mechanism is that it does not depend on timing. It
\r
628 does not matter when the child process gets CPU time or when the parent process
\r
629 gets CPU time. On heavily loaded machines (or machines with multiple CPUs,
\r
630 cores or unreliable schedulers), this makes waiting for i3 much more robust.</p></div>
\r
631 <div class="paragraph"><p>Before using socket activation, we typically used a <tt>sleep(1)</tt> and hoped that
\r
632 i3 was initialized by that time. Of course, this breaks on some (slow)
\r
633 computers and wastes a lot of time on faster computers. By using socket
\r
634 activation, we decreased the total amount of time necessary to run all tests
\r
635 (72 files at the time of writing) from > 100 seconds to 16 seconds. This makes
\r
636 it significantly more attractive to run the test suite more often (or at all)
\r
637 during development.</p></div>
\r
638 <div class="paragraph"><p>An alternative approach to using socket activation is polling for the existance
\r
639 of the IPC socket and connecting to it. While this might be slightly easier to
\r
640 implement, it wastes CPU time and is considerably uglier than this solution
\r
641 :). After all, <tt>lib/SocketActivation.pm</tt> contains only 54 SLOC.</p></div>
\r
646 <div id="footnotes"><hr /></div>
\r
647 <div id="footer" lang="de">
\r
648 © 2009-2011 Michael Stapelberg, <a href="/impress.html">Impressum</a>
\r
projects / i3 / i3.github.io / blob