From 5034693e49f1dfd9e4594b2ba45044c35c5670b0 Mon Sep 17 00:00:00 2001 From: Michael Stapelberg Date: Sat, 12 Nov 2011 09:35:15 +0000 Subject: [PATCH] update docs for 4.1 --- _docs/Makefile | 5 +- _docs/multi-monitor | 14 +- _docs/testsuite | 540 ++++++++++++++++++++++++++++++++++++ _docs/userguide | 418 +++++++++++++++++++++++++--- docs/multi-monitor.html | 12 +- docs/testsuite.html | 591 ++++++++++++++++++++++++++++++++++++++++ docs/userguide.html | 501 ++++++++++++++++++++++++++++++---- 7 files changed, 1985 insertions(+), 96 deletions(-) create mode 100644 _docs/testsuite create mode 100644 docs/testsuite.html diff --git a/_docs/Makefile b/_docs/Makefile index 7def7e4..259dd22 100644 --- a/_docs/Makefile +++ b/_docs/Makefile @@ -1,9 +1,12 @@ -all: hacking-howto.html debugging.html repositories.html userguide.html ipc.html multi-monitor.html wsbar.html i3status.html i3.html i3-config-wizard.html i3-nagbar.html i3-migrate-config-to-v4.html i3-msg.html +all: hacking-howto.html debugging.html repositories.html userguide.html ipc.html multi-monitor.html wsbar.html i3status.html i3.html i3-config-wizard.html i3-nagbar.html i3-migrate-config-to-v4.html i3-msg.html testsuite.html hacking-howto.html: hacking-howto asciidoc -a linkcss -a stylesdir=/css -a scriptsdir=/js --backend=xhtml11 -f conf/i3html.conf -a toc -n $< +testsuite.html: testsuite + asciidoc -a linkcss -a stylesdir=/css -a scriptsdir=/js --backend=xhtml11 -f conf/i3html.conf -a toc -n $< + debugging.html: debugging asciidoc -a linkcss -a stylesdir=/css -a scriptsdir=/js --backend=xhtml11 -f conf/i3html.conf -n $< diff --git a/_docs/multi-monitor b/_docs/multi-monitor index ec0256c..a1fd6dc 100644 --- a/_docs/multi-monitor +++ b/_docs/multi-monitor @@ -1,7 +1,7 @@ The multi-monitor situation =========================== Michael Stapelberg -March 2010 +September 2011 …or: oh no, I have an nVidia graphics card! @@ -16,6 +16,8 @@ i3, like so: exec i3 --force-xinerama -V >>~/.i3/i3log 2>&1 ---------------------------------------------- +…or use +force_xinerama yes+ in your configuration file. + == The explanation Starting with version 3.ε, i3 uses the RandR (Rotate and Resize) API instead @@ -50,9 +52,13 @@ these are two screens). For this very reason, we decided to implement the following workaround: As long as the nVidia driver does not support RandR, an option called -+--force-xinerama+ is available in i3. This option gets the list of screens -*once* when starting, and never updates it. As the nVidia driver cannot do -dynamic configuration anyways, this is not a big deal. ++--force-xinerama+ is available in i3 (alternatively, you can use the ++force_xinerama+ configuration file directive). This option gets the list of +screens *once* when starting, and never updates it. As the nVidia driver cannot +do dynamic configuration anyways, this is not a big deal. + +Also note that your output names are not descriptive (like +HDMI1+) when using +Xinerama, instead they are counted up, starting at 0: +xinerama-0+, +xinerama-1+, … == See also diff --git a/_docs/testsuite b/_docs/testsuite new file mode 100644 index 0000000..720ff39 --- /dev/null +++ b/_docs/testsuite @@ -0,0 +1,540 @@ +i3 testsuite +============ +Michael Stapelberg +September 2011 + +This document explains how the i3 testsuite works, how to use it and extend it. +It is targeted at developers who not necessarily have been doing testing before +or have not been testing in Perl before. In general, the testsuite is not of +interest for end users. + + +== Introduction + +The i3 testsuite is a collection of files which contain testcases for various +i3 features. Some of them test if a certain workflow works correctly (moving +windows, focus behaviour, …). Others are regression tests and contain code +which previously made i3 crash or lead to unexpected behaviour. They then check +if i3 still runs (meaning it did not crash) and if it handled everything +correctly. + +The goal of having these tests is to automatically find problems and to +automatically get a feel for whether a change in the source code breaks any +existing feature. After every modification of the i3 sourcecode, the developer +should run the full testsuite. If one of the tests fails, the corresponding +problem should be fixed (or, in some cases, the testcase has to be modified). +For every bugreport, a testcase should be written to test the correct +behaviour. Initially, it will fail, but after fixing the bug, it will pass. +This ensures (or increases the chance) that bugs which have been fixed once +will never be found again. + +Also, when implementing a new feature, a testcase might be a good way to be +able to easily test if the feature is working correctly. Many developers will +test manually if everything works. Having a testcase not only helps you with +that, but it will also be useful for every future change. + +== Implementation + +For several reasons, the i3 testsuite has been implemented in Perl: + +1. Perl has a long tradition of testing. Every popular/bigger Perl module which + you can find on CPAN will not only come with documentation, but also with + tests. Therefore, the available infrastructure for tests is comprehensive. + See for example the excellent http://search.cpan.org/perldoc?Test::More + and the referenced http://search.cpan.org/perldoc?Test::Tutorial. + +2. Perl is widely available and has a well-working package infrastructure. +3. The author is familiar with Perl :). + +Please do not start programming language flamewars at this point. + +=== Mechanisms + +==== Script: complete-run + +The testcases are run by a script called +complete-run.pl+. It runs all +testcases by default, but you can be more specific and let it only run one or +more testcases. Also, it takes care of starting up a separate instance of i3 +with an appropriate configuration file and creates a folder for each run +containing the appropriate i3 logfile for each testcase. The latest folder can +always be found under the symlink +latest/+. Unless told differently, it will +run the tests on a separate X server instance (using the Xdummy script). + +.Example invocation of complete-run.pl+ +--------------------------------------- +$ cd ~/i3/testcases + +$ ./complete-run.pl +# output omitted because it is very long +All tests successful. +Files=78, Tests=734, 27 wallclock secs ( 0.38 usr 0.48 sys + 17.65 cusr 3.21 csys = 21.72 CPU) +Result: PASS + +$ ./complete-run.pl t/04-floating.t +[:3] i3 startup: took 0.07s, status = 1 +[: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 +[:3] t/04-floating.t finished +[:3] killing i3 +output for t/04-floating.t: +ok 1 - use X11::XCB::Window; +ok 2 - The object isa X11::XCB::Window +ok 3 - Window is mapped +ok 4 - i3 raised the width to 75 +ok 5 - i3 raised the height to 50 +ok 6 - i3 did not map it to (0x0) +ok 7 - The object isa X11::XCB::Window +ok 8 - i3 let the width at 80 +ok 9 - i3 let the height at 90 +ok 10 - i3 mapped it to x=1 +ok 11 - i3 mapped it to y=18 +ok 12 - The object isa X11::XCB::Window +ok 13 - i3 let the width at 80 +ok 14 - i3 let the height at 90 +1..14 + +All tests successful. +Files=1, Tests=14, 0 wallclock secs ( 0.01 usr 0.00 sys + 0.19 cusr 0.03 csys = 0.23 CPU) +Result: PASS + +$ less latest/i3-log-for-04-floating.t +---------------------------------------- + +==== IPC interface + +The testsuite makes extensive use of the IPC (Inter-Process Communication) +interface which i3 provides. It is used for the startup process of i3, for +terminating it cleanly and (most importantly) for modifying and getting the +current state (layout tree). + +See [http://i3wm.org/docs/ipc.html] for documentation on the IPC interface. + +==== X11::XCB + +In order to open new windows, change attributes, get events, etc., the +testsuite uses X11::XCB, a new (and quite specific to i3 at the moment) Perl +module which uses the XCB protocol description to generate Perl bindings to +X11. They work in a very similar way to libxcb (which i3 uses) and provide +relatively high-level interfaces (objects such as +X11::XCB::Window+) aswell as +access to the low-level interface, which is very useful when testing a window +manager. + +=== Filesystem structure + +In the git root of i3, the testcases live in the folder +testcases+. This +folder contains the +complete-run.pl+ and +Xdummy+ scripts and a base +configuration file which will be used for the tests. The different testcases +(their file extension is .t, not .pl) themselves can be found in the +conventionally named subfolder +t+: + +.Filesystem structure +-------------------------------------------- +├── testcases +│   ├── complete-run.pl +│   ├── i3-test.config +│   ├── lib +│   │   ├── i3test.pm +│   │   ├── SocketActivation.pm +│   │   └── StartXDummy.pm +│   ├── t +│   │   ├── 00-load.t +│   │   ├── 01-tile.t +│   │   ├── 02-fullscreen.t +│   │   ├── ... +│   │   ├── omitted for brevity +│   │   ├── ... +│   │   └── 74-regress-focus-toggle.t +│   └── Xdummy +-------------------------------------------- + +== Anatomy of a testcase + +Learning by example is definitely a good strategy when you are wondering how to +write a testcase. Let's take +t/11-goto.t+ as an easy example and go through it +step by step: + +.t/11-goto.t: Boilerplate +---------------------- +#!perl +# vim:ts=4:sw=4:expandtab + +use i3test; +use File::Temp; + +my $x = X11::XCB::Connection->new; +----------------------- + +This is what we call boilerplate. It exists at the top of every test file (to +some extent). The first line is the shebang, which specifies that this file is +a Perl script. The second line contains VIM specific settings on how to +edit/format this file (use spaces instead of tabs, indent using 4 spaces). +Afterwards, the +i3test+ module is used. This module contains i3 testsuite +specific functions which you are strongly encouraged to use. They make writing +testcases a lot easier and will make it easier for other people to read your +tests. + +The next line uses the +File::Temp+ module. This is specific to this testcase, +because it needs to generate a temporary name during the test. Many testcases +use only the +i3test+ module. + +The last line opens a connection to X11. You might or might not need this in +your testcase, depending on whether you are going to open windows (etc.) or +only use i3 commands. + +.t/11-goto.t: Setup +---------------------- +my $tmp = fresh_workspace; + +cmd 'split h'; +---------------------- + +The first line calls i3test's +fresh_workspace+ function which looks for a +currently unused workspace, switches to it, and returns its name. The variable ++$tmp+ will end up having a value such as +"/tmp/87kBVcHbA9"+. Note that this +is not (necessarily) a valid path, it's just a random workspace name. + +So, now that we are on a new workspace, we ensure that the workspace uses +horizontal orientation by issuing the +split h+ command (see the i3 User's +Guide for a list of commands). This is not strictly necessary, but good style. +In general, the +cmd+ function executes the specified i3 command by using the +IPC interface and returns once i3 acknowledged the command. + +.t/11-goto.t: Setup +---------------------- +##################################################################### +# Create two windows and make sure focus switching works +##################################################################### + +my $top = open_window($x); +my $mid = open_window($x); +my $bottom = open_window($x); +---------------------- + +In every major section of a testcase, you should put a comment like the one +above. This makes it immediately clear how the file is structured. + +The +open_window+ function opens a standard window, which will then be put into +tiling mode by i3. If you want a floating window, use the ++open_floating_window+ function. These functions accept the same parameters as ++X11::XCB::Window->new+, see the i3test documentation at TODO. + +.t/11-goto.t: Helper function +---------------------- +# +# Returns the input focus after sending the given command to i3 via IPC +# and syncing with i3 +# +sub focus_after { + my $msg = shift; + + cmd $msg; + sync_with_i3 $x; + return $x->input_focus; +} +---------------------- + +This section defines a helper function which will be used over and over in this +testcase. If you have code which gets executed more than once or twice +(depending on the length of your test, use your best judgement), please put it +in a function. Tests should be short, concise and clear. + +The +focus_after+ function executes a command and returns the X11 focus after +the command was executed. The +sync_with_i3+ command makes sure that i3 could +push its state to X11. See <> to learn how this works exactly. + +.t/11-goto.t: Test assumptions +---------------------- +$focus = $x->input_focus; +is($focus, $bottom->id, "Latest window focused"); + +$focus = focus_after('focus left'); +is($focus, $mid->id, "Middle window focused"); +---------------------- + +Now, we run the first two real tests. They use +Test::More+'s +is+ function, +which compares two values and prints the differences if they are not the same. +After the arguments, we supply a short comment to indicate what we are testing +here. This makes it vastly more easy for the developer to spot which testcase +is the problem in case one fails. + +The first test checks that the most recently opened window is focused. +Afterwards, the command +focus left+ is issued and it is verified that the +middle window now has focus. + +Note that this is not a comprehensive test of the +focus+ command -- we would +have to test wrapping, focus when using a more complex layout, focusing the +parent/child containers, etc. But that is not the point of this testcase. +Instead, we just want to know if +$x->input_focus+ corresponds with what we are +expecting. If not, something is completely wrong with the test environment and +this trivial test will fail. + +.t/11-goto.t: Test that the feature does not work (yet) +---------------------- +##################################################################### +# Now goto a mark which does not exist +##################################################################### + +my $random_mark = mktemp('mark.XXXXXX'); + +$focus = focus_after(qq|[con_mark="$random_mark"] focus|); +is($focus, $mid->id, "focus unchanged"); +---------------------- + +Syntax hint: The qq keyword is the interpolating quote operator. It lets you +chose a quote character (in this case the +|+ character, a pipe). This makes +having double quotes in our string easy. + +In this new major section, a random mark (mark is an identifier for a window, +see "VIM-like marks" in the i3 User’s Guide) will be generated. Afterwards, we +test that trying to focus that mark will not do anything. This is important: Do +not only test that using a feature has the expected outcome, but also test that +using it without properly initializing it does no harm. This command could for +example have changed focus anyways (a bug) or crash i3 (obviously a bug). + +.t/11-goto.t: Test that the feature does work +---------------------- +cmd "mark $random_mark"; + +$focus = focus_after('focus left'); +is($focus, $top->id, "Top window focused"); + +$focus = focus_after(qq|[con_mark="$random_mark"] focus|); +is($focus, $mid->id, "goto worked"); +---------------------- + +Remember: Focus was on the middle window (we verified that earlier in "Test +assumptions"). We now mark the middle window with our randomly generated mark. +Afterwards, we switch focus away from the middle window to be able to tell if +focusing it via its mark will work. If the test works, the goto command seems +to be working. + +.t/11-goto.t: Test corner case +---------------------- +# check that we can specify multiple criteria + +$focus = focus_after('focus left'); +is($focus, $top->id, "Top window focused"); + +$focus = focus_after(qq|[con_mark="$random_mark" con_mark="$random_mark"] focus|); +is($focus, $mid->id, "goto worked"); +---------------------- + +Now we test the same feature, but specifying the mark twice in the command. +This should have no effect, but let’s be sure: test it and see if things go +wrong. + +.t/11-goto.t: Test second code path +---------------------- +##################################################################### +# Check whether the focus command will switch to a different +# workspace if necessary +##################################################################### + +my $tmp2 = fresh_workspace; + +is(focused_ws(), $tmp2, 'tmp2 now focused'); + +cmd qq|[con_mark="$random_mark"] focus|; + +is(focused_ws(), $tmp, 'tmp now focused'); +---------------------- + +This part of the test checks that focusing windows by mark works across +workspaces. It uses i3test's +focused_ws+ function to get the current +workspace. + +.t/11-goto.t: Test second code path +---------------------- +done_testing; +---------------------- + +The end of every testcase has to contain the +done_testing+ line. This tells ++complete-run.pl+ that the test was finished successfully. If it does not +occur, the test might have crashed during execution -- some of the reasons why +that could happen are bugs in the used modules, bugs in the testcase itself or +an i3 crash resulting in the testcase being unable to communicate with i3 via +IPC anymore. + +[[i3_sync]] +== Appendix A: The i3 sync protocol + +Consider the following situation: You open two windows in your testcase, then +you use +focus left+ and want to verify that the X11 focus has been updated +properly. Sounds simple, right? Let’s assume you use this straight-forward +implementation: + +.Racey focus testcase +----------- +my $left = open_window($x); +my $right = open_window($x); +cmd 'focus left'; +is($x->input_focus, $left->id, 'left window focused'); +---------- + +However, the test fails. Sometimes. Apparantly, there is a race condition in +your test. If you think about it, this is because you are using two different +pieces of software: You tell i3 to update focus, i3 confirms that, and then you +ask X11 to give you the current focus. There is a certain time i3 needs to +update the X11 state. If the testcase gets CPU time before X11 processed i3's +requests, the test will fail. + +image::i3-sync.png["Diagram of the race condition", title="Diagram of the race condition"] + +One way to "solve" this would be to add +sleep 0.5;+ after the +cmd+ call. +After 0.5 seconds it should be safe to assume that focus has been updated, +right? + +In practice, this usually works. However, it has several problems: + +1. This is obviously not a clean solution, but a workaround. Ugly. +2. On very slow machines, this might not work. Unlikely, but in different + situations (a delay to wait for i3 to startup) the necessary time is much + harder to guess, even for fast machines. +3. This *wastes a lot of time*. Usually, your computer is much faster than 0.5s + to update the status. However, sometimes, it might take 0.4s, so we can’t + make it +sleep 0.1+. + +To illustrate how grave the problem with wasting time actually is: Before +removing all sleeps from the testsuite, a typical run using 4 separate X +servers took around 50 seconds on my machine. After removing all the sleeps, +we achieved times of about 25 seconds. This is very significant and influences +the way you think about tests -- the faster they are, the more likely you are +to check whether everything still works quite often (which you should). + +What I am trying to say is: Delays adds up quickly and make the test suite +less robust. + +The real solution for this problem is a mechanism which I call "the i3 sync +protocol". The idea is to send a request (which does not modify state) via X11 +to i3 which will then be answered. Due to the request's position in the event +queue (*after* all previous events), you can be sure that by the time you +receive the reply, all other events have been dealt with by i3 (and, more +importantly, X11). + +image::i3-sync-working.png["Diagram of the i3 sync solution", title="Diagram of the i3 sync solution"] + +=== Implementation details + +The client which wants to sync with i3 initiates the protocol by sending a +ClientMessage to the X11 root window: + +.Send ClientMessage +------------------- +# Generate a ClientMessage, see xcb_client_message_t +my $msg = pack "CCSLLLLLLL", + CLIENT_MESSAGE, # response_type + 32, # format + 0, # sequence + $root, # destination window + $x->atom(name => 'I3_SYNC')->id, + + $_sync_window->id, # data[0]: our own window id + $myrnd, # data[1]: a random value to identify the request + 0, + 0, + 0; + +# Send it to the root window -- since i3 uses the SubstructureRedirect +# event mask, it will get the ClientMessage. +$x->send_event(0, $root, EVENT_MASK_SUBSTRUCTURE_REDIRECT, $msg); +------------------- + +i3 will then reply with the same ClientMessage, sent to the window specified in ++data[0]+. In the reply, +data[0]+ and +data[1]+ are exactly the same as in the +request. You should use a random value in +data[1]+ and check that you received +the same one when getting the reply. + +== Appendix B: Socket activation + +Socket activation is a mechanism which was made popular by systemd, an init +replacement. It basically describes creating a listening socket before starting +a program. systemd will invoke the program only when an actual connection to +the socket is made, hence the term socket activation. + +The interesting part of this (in the i3 context) is that you can very precisely +detect when the program is ready (finished its initialization). + +=== Preparing the listening socket + ++complete-run.pl+ will create a listening UNIX socket which it will then pass +to i3. This socket will be used by i3 as an additional IPC socket, just like +the one it will create on its own. Passing the socket happens implicitly +because children will inherit the parent’s sockets when fork()ing and sockets +will continue to exist after an exec() call (unless CLOEXEC is set of course). + +The only explicit things +complete-run.pl+ has to do is setting the +LISTEN_FDS+ +environment variable to the number of sockets which exist (1 in our case) and +setting the +LISTEN_PID+ environment variable to the current process ID. Both +variables are necessary so that the program (i3) knows how many sockets it +should use and if the environment variable is actually intended for it. i3 will +then start looking for sockets at file descriptor 3 (since 0, 1 and 2 are used +for stdin, stdout and stderr, respectively). + +The actual Perl code which sets up the socket, fork()s, makes sure the socket +has file descriptor 3 and sets up the environment variables follows (shortened +a bit): + + +.Setup socket and environment +----------------------------- +my $socket = IO::Socket::UNIX->new( + Listen => 1, + Local => $args{unix_socket_path}, +); + +my $pid = fork; +if ($pid == 0) { + $ENV{LISTEN_PID} = $$; + $ENV{LISTEN_FDS} = 1; + + # Only pass file descriptors 0 (stdin), 1 (stdout), + # 2 (stderr) and 3 (socket) to the child. + $^F = 3; + + # If the socket does not use file descriptor 3 by chance + # already, we close fd 3 and dup2() the socket to 3. + if (fileno($socket) != 3) { + POSIX::close(3); + POSIX::dup2(fileno($socket), 3); + } + + exec "/usr/bin/i3"; +} +----------------------------- + +=== Waiting for a reply + +In the parent process, we want to know when i3 is ready to answer our IPC +requests and handle our windows. Therefore, after forking, we immediately close +the listening socket (i3 will handle this side of the socket) and connect to it +(remember, we are talking about a named UNIX socket) as a client. This connect +call will immediately succeed because the kernel buffers it. Then, we send a +request (of type GET_TREE, but that is not really relevant). Writing data to +the socket will also succeed immediately because, again, the kernel buffers it +(only up to a certain amount of data of course). + +Afterwards, we just blockingly wait until we get an answer. In the child +process, i3 will setup the listening socket in its event loop. Immediately +after actually starting the event loop, it will notice a new client connecting +(the parent process) and handle its request. Since all initialization has been +completed successfully by the time the event loop is entered, we can now assume +that i3 is ready. + +=== Timing and conclusion + +A beautiful feature of this mechanism is that it does not depend on timing. It +does not matter when the child process gets CPU time or when the parent process +gets CPU time. On heavily loaded machines (or machines with multiple CPUs, +cores or unreliable schedulers), this makes waiting for i3 much more robust. + +Before using socket activation, we typically used a +sleep(1)+ and hoped that +i3 was initialized by that time. Of course, this breaks on some (slow) +computers and wastes a lot of time on faster computers. By using socket +activation, we decreased the total amount of time necessary to run all tests +(72 files at the time of writing) from > 100 seconds to 16 seconds. This makes +it significantly more attractive to run the test suite more often (or at all) +during development. + +An alternative approach to using socket activation is polling for the existance +of the IPC socket and connecting to it. While this might be slightly easier to +implement, it wastes CPU time and is considerably uglier than this solution +:). After all, +lib/SocketActivation.pm+ contains only 54 SLOC. diff --git a/_docs/userguide b/_docs/userguide index 865ee4b..3d78d16 100644 --- a/_docs/userguide +++ b/_docs/userguide @@ -1,7 +1,7 @@ i3 User’s Guide =============== Michael Stapelberg -August 2011 +October 2011 This document contains all the information you need to configure and use the i3 window manager. If it does not, please contact us on IRC (preferred) or post your @@ -357,7 +357,8 @@ it to the position you want. When holding the floating modifier, you can resize a floating window by pressing the right mouse button on it and moving around while holding it. If -you hold the shift button as well, the resize will be proportional. +you hold the shift button as well, the resize will be proportional (the aspect +ratio will be preserved). *Syntax*: -------------------------------- @@ -431,7 +432,7 @@ change their border style, for example. *Syntax*: ----------------------------- -for_window [criteria] command +for_window command ----------------------------- *Examples*: @@ -478,37 +479,59 @@ configuration file and run it before starting i3 (for example in your [[assign_workspace]] -Specific windows can be matched by window class and/or window title. It is -recommended that you match on window classes instead of window titles whenever -possible because some applications first create their window, and then worry -about setting the correct title. Firefox with Vimperator comes to mind. The -window starts up being named Firefox, and only when Vimperator is loaded does -the title change. As i3 will get the title as soon as the application maps the +To automatically make a specific window show up on a specific workspace, you +can use an *assignment*. You can match windows by using any criteria, +see <>. It is recommended that you match on window classes +(and instances, when appropriate) instead of window titles whenever possible +because some applications first create their window, and then worry about +setting the correct title. Firefox with Vimperator comes to mind. The window +starts up being named Firefox, and only when Vimperator is loaded does the +title change. As i3 will get the title as soon as the application maps the window (mapping means actually displaying it on the screen), you’d need to have to match on 'Firefox' in this case. -You can prefix or suffix workspaces with a `~` to specify that matching clients -should be put into floating mode. If you specify only a `~`, the client will -not be put onto any workspace, but will be set floating on the current one. - *Syntax*: ------------------------------------------------------------ -assign ["]window class[/window title]["] [→] [workspace] +assign [→] workspace ------------------------------------------------------------ *Examples*: ---------------------- -assign urxvt 2 -assign urxvt → 2 -assign urxvt → work -assign "urxvt" → 2 -assign "urxvt/VIM" → 3 -assign "gecko" → 4 +# Assign URxvt terminals to workspace 2 +assign [class="URxvt"] 2 + +# Same thing, but more precise (exact match instead of substring) +assign [class="^URxvt$"] 2 + +# Same thing, but with a beautiful arrow :) +assign [class="^URxvt$"] → 2 + +# Assignment to a named workspace +assign [class="^URxvt$"] → work + +# Start urxvt -name irssi +assign [class="^URxvt$" instance="^irssi$"] → 3 ---------------------- Note that the arrow is not required, it just looks good :-). If you decide to use it, it has to be a UTF-8 encoded arrow, not `->` or something like that. +To get the class and instance, you can use +xprop+. After clicking on the +window, you will see the following output: + +*xwininfo*: +----------------------------------- +WM_CLASS(STRING) = "irssi", "URxvt" +----------------------------------- + +The first part of the WM_CLASS is the instance ("irssi" in this example), the +second part is the class ("URxvt" in this example). + +Should you have any problems with assignments, make sure to check the i3 +logfile first (see http://i3wm.org/docs/debugging.html). It includes more +details about the matching process and the window’s actual class, instance and +title when starting up. + === Automatically starting applications on i3 startup By using the +exec+ keyword outside a keybinding, you can configure @@ -519,16 +542,21 @@ keyword. These commands will be run in order. *Syntax*: ------------------- -exec command -exec_always command +exec [--no-startup-id] command +exec_always [--no-startup-id] command ------------------- *Examples*: -------------------------------- -exec i3status | i3bar -d +exec chromium exec_always ~/my_script.sh + +# Execute the terminal emulator urxvt, which is not yet startup-notification aware. +exec --no-startup-id urxvt -------------------------------- +The flag --no-startup-id is explained in <>. + [[workspace_screen]] === Automatically putting workspaces on specific screens @@ -541,17 +569,20 @@ the second screen and so on). *Syntax*: ---------------------------------- -workspace output +workspace output ---------------------------------- The 'output' is the name of the RandR output you attach your screen to. On a laptop, you might have VGA1 and LVDS1 as output names. You can see the available outputs by running +xrandr --current+. +If you use named workspaces, they must be quoted: + *Examples*: --------------------------- workspace 1 output LVDS1 workspace 5 output VGA1 +workspace "2: vim" output VGA1 --------------------------- === Changing colors @@ -690,6 +721,270 @@ force_focus_wrapping force_focus_wrapping yes ------------------------ +=== Forcing Xinerama + +As explained in-depth in , some X11 +video drivers (especially the nVidia binary driver) only provide support for +Xinerama instead of RandR. In such a situation, i3 must be told to use the +inferior Xinerama API explicitly and therefore don’t provide support for +reconfiguring your screens on the fly (they are read only once on startup and +that’s it). + +For people who do cannot modify their +~/.xsession+ to add the ++--force-xinerama+ commandline parameter, a configuration option is provided: + +*Syntax*: +----------------------- +force_xinerama +----------------------- + +*Example*: +------------------ +force_xinerama yes +------------------ + +Also note that your output names are not descriptive (like +HDMI1+) when using +Xinerama, instead they are counted up, starting at 0: +xinerama-0+, +xinerama-1+, … + +=== Automatic back-and-forth when switching to the current workspace + +This configuration directive enables automatic +workspace back_and_forth+ (see +<>) when switching to the workspace that is currently focused. + +For instance: Assume you are on workspace "1: www" and switch to "2: IM" using +mod+2 because somebody sent you a message. You don’t need to remember where you +came from now, you can just press mod+2 again to switch back to "1: www". + +*Syntax*: +-------------------------------------- +workspace_auto_back_and_forth +-------------------------------------- + +*Example*: +--------------------------------- +workspace_auto_back_and_forth yes +--------------------------------- + +== Configuring i3bar + +The bar at the bottom of your monitor is drawn by a separate process called +i3bar. Having this part of "the i3 user interface" in a separate process has +several advantages: + +1. It is a modular approach. If you don’t need a workspace bar at all, or if + you prefer a different one (dzen2, xmobar, maybe even gnome-panel?), you can + just remove the i3bar configuration and start your favorite bar instead. +2. It follows the UNIX philosophy of "Make each program do one thing well". + While i3 manages your windows well, i3bar is good at displaying a bar on + each monitor (unless you configure it otherwise). +3. It leads to two separate, clean codebases. If you want to understand i3, you + don’t need to bother with the details of i3bar and vice versa. + +That said, i3bar is configured in the same configuration file as i3. This is +because it is tightly coupled with i3 (in contrary to i3lock or i3status which +are useful for people using other window managers). Therefore, it makes no +sense to use a different configuration place when we already have a good +configuration infrastructure in place. + +Configuring your workspace bar starts with opening a +bar+ block. You can have +multiple bar blocks to use different settings for different outputs (monitors): + +*Example*: +--------------------------- +bar { + status_command i3status +} +--------------------------- + +=== Statusline command + +i3bar can run a program and display every line of its +stdout+ output on the +right hand side of the bar. This is useful to display system information like +your current IP address, battery status or date/time. + +The specified command will be passed to +sh -c+, so you can use globbing and +have to have correct quoting etc. + +*Syntax*: +---------------------- +status_command command +---------------------- + +*Example*: +------------------------------------------------- +status_command i3status --config ~/.i3status.conf +------------------------------------------------- + +=== Display mode + +You can have i3bar either be visible permanently at one edge of the screen +(+dock+ mode) or make it show up when you press your modifier key (+hide+ +mode). + +The hide mode maximizes screen space that can be used for actual windows. Also, +i3bar sends the +SIGSTOP+ and +SIGCONT+ signals to the statusline process to +save battery power. + +The default is dock mode. + +*Syntax*: +---------------- +mode +---------------- + +*Example*: +---------------- +mode hide +---------------- + +=== Position + +This option determines in which edge of the screen i3bar should show up. + +The default is bottom. + +*Syntax*: +--------------------- +position +--------------------- + +*Example*: +--------------------- +position top +--------------------- + +=== Output(s) + +You can restrict i3bar to one or more outputs (monitors). The default is to +handle all outputs. Restricting the outputs is useful for using different +options for different outputs by using multiple 'bar' blocks. + +*Syntax*: +--------------- +output +--------------- + +*Example*: +------------------------------- +# big monitor: everything +bar { + output HDMI2 + status_command i3status +} + +# laptop monitor: bright colors and i3status with less modules. +bar { + output LVDS1 + status_command i3status --config ~/.i3status-small.conf + colors { + background #000000 + statusline #ffffff + } +} +------------------------------- + +=== Tray output + +i3bar by default provides a system tray area where programs such as +NetworkManager, VLC, Pidgin, etc. can place little icons. + +You can configure on which output (monitor) the icons should be displayed or +you can turn off the functionality entirely. + +*Syntax*: +------------------------- +tray_output +------------------------- + +*Example*: +------------------------- +# disable system tray +tray_output none + +# show tray icons on the big monitor +tray_output HDMI2 +------------------------- + +=== Font + +Specifies the font (again, X core font, not Xft, just like in i3) to be used in +the bar. + +*Syntax*: +--------------------- +font +--------------------- + +*Example*: +-------------------------------------------------------------- +font -misc-fixed-medium-r-normal--13-120-75-75-C-70-iso10646-1 +-------------------------------------------------------------- + +=== Workspace buttons + +Specifies whether workspace buttons should be shown or not. This is useful if +you want to display a statusline-only bar containing additional information. + +The default is to show workspace buttons. + +*Syntax*: +-------------------------- +workspace_buttons +-------------------------- + +*Example*: +-------------------- +workspace_buttons no +-------------------- + +=== Colors + +As with i3, colors are in HTML hex format (#rrggbb). The following colors can +be configured at the moment: + +background:: + Background color of the bar. +statusline:: + Text color to be used for the statusline. +focused_workspace:: + Text color/background color for a workspace button when the workspace + has focus. +active_workspace:: + Text color/background color for a workspace button when the workspace + is active (visible) on some output, but the focus is on another one. + You can only tell this apart from the focused workspace when you are + using multiple monitors. +inactive_workspace:: + Text color/background color for a workspace button when the workspace + does not have focus and is not active (visible) on any output. This + will be the case for most workspaces. +urgent_workspace:: + Text color/background color for workspaces which contain at least one + window with the urgency hint set. + +*Syntax*: +---------------------------------------- +colors { + background + statusline + + colorclass +} +---------------------------------------- + +*Example*: +-------------------------------------- +colors { + background #000000 + statusline #ffffff + + focused_workspace #ffffff #285577 + active_workspace #ffffff #333333 + inactive_workspace #888888 #222222 + urgent_workspace #ffffff #900000 +} +-------------------------------------- + == List of commands Commands are what you bind to specific keypresses. You can also issue commands @@ -721,6 +1016,9 @@ which have the class Firefox, use: *Example*: ------------------------------------ bindsym mod+x [class="Firefox"] kill + +# same thing, but case-insensitive +bindsym mod+x [class="(?i)firefox"] kill ------------------------------------ The criteria which are currently implemented are: @@ -729,18 +1027,52 @@ class:: Compares the window class (the second part of WM_CLASS) instance:: Compares the window instance (the first part of WM_CLASS) +window_role:: + Compares the window role (WM_WINDOW_ROLE). id:: Compares the X11 window ID, which you can get via +xwininfo+ for example. title:: Compares the X11 window title (_NET_WM_NAME or WM_NAME as fallback). -mark:: +con_mark:: Compares the mark set for this container, see <>. con_id:: Compares the i3-internal container ID, which you can get via the IPC interface. Handy for scripting. -Note that currently all criteria are compared case-insensitive and do not -support regular expressions. This is planned to change in the future. +The criteria +class+, +instance+, +role+, +title+ and +mark+ are actually +regular expressions (PCRE). See +pcresyntax(3)+ or +perldoc perlre+ for +information on how to use them. + +[[exec]] + +=== Executing applications (exec) + +What good is a window manager if you can’t actually start any applications? +The exec command starts an application by passing the command you specify to a +shell. This implies that you can use globbing (wildcards) and programs will be +searched in your $PATH. + +*Syntax*: +------------------------------ +exec [--no-startup-id] command +------------------------------ + +*Example*: +------------------------------ +# Start the GIMP +bindsym mod+g exec gimp + +# Start the terminal emulator urxvt which is not yet startup-notification-aware +bindsym mod+enter exec --no-startup-id urxvt +------------------------------ + +The +--no-startup-id+ parameter disables startup-notification support for this +particular exec command. With startup-notification, i3 can make sure that a +window appears on the workspace on which you used the exec command. Also, it +will change the X11 cursor to +watch+ (a clock) while the application is +launching. So, if an application is not startup-notification aware (most GTK +and Qt using applications seem to be, though), you will end up with a watch +cursor for 60 seconds. === Splitting containers @@ -806,9 +1138,19 @@ mode_toggle:: For moving, use +move left+, +move right+, +move down+ and +move up+. +*Syntax*: +----------------------------------- +focus +focus +move [ px] +----------------------------------- + +Note that the amount of pixels you can specify for the +move+ command is only +relevant for floating containers. The default amount is 10 pixels. + *Examples*: ---------------------- -# Focus clients on the left, bottom, top, right: +# Focus container on the left, bottom, top, right: bindsym mod+j focus left bindsym mod+k focus down bindsym mod+l focus up @@ -820,11 +1162,15 @@ bindsym mod+u focus parent # Focus last floating/tiling container bindsym mod+g focus mode_toggle -# Move client to the left, bottom, top, right: +# Move container to the left, bottom, top, right: bindsym mod+j move left bindsym mod+k move down bindsym mod+l move up bindsym mod+semicolon move right + +# Move container, but make floating containers +# move more than the default +bindsym mod+j move left 20 px ---------------------- === Changing (named) workspaces/moving to workspaces @@ -836,7 +1182,12 @@ number or name of the workspace. To move containers to specific workspaces, use You can also switch to the next and previous workspace with the commands +workspace next+ and +workspace prev+, which is handy, for example, if you have workspace 1, 3, 4 and 9 and you want to cycle through them with a single key -combination. +combination. Similarily, you can use +move workspace next+ and +move workspace +prev+ to move a container to the next/previous workspace. + +[[back_and_forth]] +To switch back to the previously focused workspace, use +workspace +back_and_forth+. To move a container to another xrandr output such as +LVDS1+ or +VGA1+, you can use the +move output+ command followed by the name of the target output. You @@ -852,6 +1203,9 @@ bindsym mod+2 workspace 2 bindsym mod+Shift+1 move workspace 1 bindsym mod+Shift+2 move workspace 2 ... + +# switch between the current and the previously focused one +bindsym mod+b workspace back_and_forth ------------------------- ==== Named workspaces @@ -894,9 +1248,9 @@ resize [ px] [or ppt] Direction can be one of +up+, +down+, +left+ or +right+. The optional pixel argument specifies by how many pixels a *floating container* should be grown or -shrinked (the default is 10 pixels). The ppt argument means percentage points +shrunk (the default is 10 pixels). The ppt argument means percentage points and specifies by how many percentage points a *tiling container* should be -grown or shrinked (the default is 10 percentage points). +grown or shrunk (the default is 10 percentage points). I recommend using the resize command inside a so called +mode+: diff --git a/docs/multi-monitor.html b/docs/multi-monitor.html index ee42703..81ba8ab 100644 --- a/docs/multi-monitor.html +++ b/docs/multi-monitor.html @@ -31,7 +31,7 @@ window.onload = function(){asciidoc.footnotes(); asciidoc.toc(2);}

The multi-monitor situation

Michael Stapelberg
<michael+i3@stapelberg.de>
-March 2010 +September 2011
Table of Contents
@@ -53,6 +53,7 @@ i3, like so:

exec i3 --force-xinerama -V >>~/.i3/i3log 2>&1
+

…or use force_xinerama yes in your configuration file.

@@ -87,9 +88,12 @@ actually contains both of your physical screens (but it will not know that these are two screens).

For this very reason, we decided to implement the following workaround: As long as the nVidia driver does not support RandR, an option called ---force-xinerama is available in i3. This option gets the list of screens -once when starting, and never updates it. As the nVidia driver cannot do -dynamic configuration anyways, this is not a big deal.

+--force-xinerama is available in i3 (alternatively, you can use the +force_xinerama configuration file directive). This option gets the list of +screens once when starting, and never updates it. As the nVidia driver cannot +do dynamic configuration anyways, this is not a big deal.

+

Also note that your output names are not descriptive (like HDMI1) when using +Xinerama, instead they are counted up, starting at 0: xinerama-0, xinerama-1, …

diff --git a/docs/testsuite.html b/docs/testsuite.html new file mode 100644 index 0000000..35acb27 --- /dev/null +++ b/docs/testsuite.html @@ -0,0 +1,591 @@ + + + + + + +i3: i3 testsuite + + + + + + + +
+

i3 - improved tiling WM

 + +
+
+ +
+
+

This document explains how the i3 testsuite works, how to use it and extend it. +It is targeted at developers who not necessarily have been doing testing before +or have not been testing in Perl before. In general, the testsuite is not of +interest for end users.

+
+
+
+

1. Introduction

+
+

The i3 testsuite is a collection of files which contain testcases for various +i3 features. Some of them test if a certain workflow works correctly (moving +windows, focus behaviour, …). Others are regression tests and contain code +which previously made i3 crash or lead to unexpected behaviour. They then check +if i3 still runs (meaning it did not crash) and if it handled everything +correctly.

+

The goal of having these tests is to automatically find problems and to +automatically get a feel for whether a change in the source code breaks any +existing feature. After every modification of the i3 sourcecode, the developer +should run the full testsuite. If one of the tests fails, the corresponding +problem should be fixed (or, in some cases, the testcase has to be modified). +For every bugreport, a testcase should be written to test the correct +behaviour. Initially, it will fail, but after fixing the bug, it will pass. +This ensures (or increases the chance) that bugs which have been fixed once +will never be found again.

+

Also, when implementing a new feature, a testcase might be a good way to be +able to easily test if the feature is working correctly. Many developers will +test manually if everything works. Having a testcase not only helps you with +that, but it will also be useful for every future change.

+
+
+
+

2. Implementation

+
+

For several reasons, the i3 testsuite has been implemented in Perl:

+
    +
  1. +

    +Perl has a long tradition of testing. Every popular/bigger Perl module which + you can find on CPAN will not only come with documentation, but also with + tests. Therefore, the available infrastructure for tests is comprehensive. + See for example the excellent http://search.cpan.org/perldoc?Test::More + and the referenced http://search.cpan.org/perldoc?Test::Tutorial. +

    +
    2. +
  2. +

    +Perl is widely available and has a well-working package infrastructure. +

    +
    3. +
  3. +

    +The author is familiar with Perl :). +

    +
    4. +
+

Please do not start programming language flamewars at this point.

+
+

2.1. Mechanisms

+
+

2.1.1. Script: complete-run

+

The testcases are run by a script called complete-run.pl. It runs all +testcases by default, but you can be more specific and let it only run one or +more testcases. Also, it takes care of starting up a separate instance of i3 +with an appropriate configuration file and creates a folder for each run +containing the appropriate i3 logfile for each testcase. The latest folder can +always be found under the symlink latest/. Unless told differently, it will +run the tests on a separate X server instance (using the Xdummy script).

+
+
Example invocation of complete-run.pl+
+
+
$ cd ~/i3/testcases
+
+$ ./complete-run.pl
+# output omitted because it is very long
+All tests successful.
+Files=78, Tests=734, 27 wallclock secs ( 0.38 usr  0.48 sys + 17.65 cusr  3.21 csys = 21.72 CPU)
+Result: PASS
+
+$ ./complete-run.pl t/04-floating.t
+[:3] i3 startup: took 0.07s, status = 1
+[: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
+[:3] t/04-floating.t finished
+[:3] killing i3
+output for t/04-floating.t:
+ok 1 - use X11::XCB::Window;
+ok 2 - The object isa X11::XCB::Window
+ok 3 - Window is mapped
+ok 4 - i3 raised the width to 75
+ok 5 - i3 raised the height to 50
+ok 6 - i3 did not map it to (0x0)
+ok 7 - The object isa X11::XCB::Window
+ok 8 - i3 let the width at 80
+ok 9 - i3 let the height at 90
+ok 10 - i3 mapped it to x=1
+ok 11 - i3 mapped it to y=18
+ok 12 - The object isa X11::XCB::Window
+ok 13 - i3 let the width at 80
+ok 14 - i3 let the height at 90
+1..14
+
+All tests successful.
+Files=1, Tests=14,  0 wallclock secs ( 0.01 usr  0.00 sys +  0.19 cusr  0.03 csys =  0.23 CPU)
+Result: PASS
+
+$ less latest/i3-log-for-04-floating.t
+
+
+
+

2.1.2. IPC interface

+

The testsuite makes extensive use of the IPC (Inter-Process Communication) +interface which i3 provides. It is used for the startup process of i3, for +terminating it cleanly and (most importantly) for modifying and getting the +current state (layout tree).

+

See [http://i3wm.org/docs/ipc.html] for documentation on the IPC interface.

+
+
+

2.1.3. X11::XCB

+

In order to open new windows, change attributes, get events, etc., the +testsuite uses X11::XCB, a new (and quite specific to i3 at the moment) Perl +module which uses the XCB protocol description to generate Perl bindings to +X11. They work in a very similar way to libxcb (which i3 uses) and provide +relatively high-level interfaces (objects such as X11::XCB::Window) aswell as +access to the low-level interface, which is very useful when testing a window +manager.

+
+
+
+

2.2. Filesystem structure

+

In the git root of i3, the testcases live in the folder testcases. This +folder contains the complete-run.pl and Xdummy scripts and a base +configuration file which will be used for the tests. The different testcases +(their file extension is .t, not .pl) themselves can be found in the +conventionally named subfolder t:

+
+
Filesystem structure
+
+
├── testcases
+│   ├── complete-run.pl
+│   ├── i3-test.config
+│   ├── lib
+│   │   ├── i3test.pm
+│   │   ├── SocketActivation.pm
+│   │   └── StartXDummy.pm
+│   ├── t
+│   │   ├── 00-load.t
+│   │   ├── 01-tile.t
+│   │   ├── 02-fullscreen.t
+│   │   ├── ...
+│   │   ├── omitted for brevity
+│   │   ├── ...
+│   │   └── 74-regress-focus-toggle.t
+│   └── Xdummy
+
+
+
+
+
+

3. Anatomy of a testcase

+
+

Learning by example is definitely a good strategy when you are wondering how to +write a testcase. Let’s take t/11-goto.t as an easy example and go through it +step by step:

+
+
t/11-goto.t: Boilerplate
+
+
#!perl
+# vim:ts=4:sw=4:expandtab
+
+use i3test;
+use File::Temp;
+
+my $x = X11::XCB::Connection->new;
+
+

This is what we call boilerplate. It exists at the top of every test file (to +some extent). The first line is the shebang, which specifies that this file is +a Perl script. The second line contains VIM specific settings on how to +edit/format this file (use spaces instead of tabs, indent using 4 spaces). +Afterwards, the i3test module is used. This module contains i3 testsuite +specific functions which you are strongly encouraged to use. They make writing +testcases a lot easier and will make it easier for other people to read your +tests.

+

The next line uses the File::Temp module. This is specific to this testcase, +because it needs to generate a temporary name during the test. Many testcases +use only the i3test module.

+

The last line opens a connection to X11. You might or might not need this in +your testcase, depending on whether you are going to open windows (etc.) or +only use i3 commands.

+
+
t/11-goto.t: Setup
+
+
my $tmp = fresh_workspace;
+
+cmd 'split h';
+
+

The first line calls i3test’s fresh_workspace function which looks for a +currently unused workspace, switches to it, and returns its name. The variable +$tmp will end up having a value such as "/tmp/87kBVcHbA9". Note that this +is not (necessarily) a valid path, it’s just a random workspace name.

+

So, now that we are on a new workspace, we ensure that the workspace uses +horizontal orientation by issuing the split h command (see the i3 User’s +Guide for a list of commands). This is not strictly necessary, but good style. +In general, the cmd function executes the specified i3 command by using the +IPC interface and returns once i3 acknowledged the command.

+
+
t/11-goto.t: Setup
+
+
#####################################################################
+# Create two windows and make sure focus switching works
+#####################################################################
+
+my $top = open_window($x);
+my $mid = open_window($x);
+my $bottom = open_window($x);
+
+

In every major section of a testcase, you should put a comment like the one +above. This makes it immediately clear how the file is structured.

+

The open_window function opens a standard window, which will then be put into +tiling mode by i3. If you want a floating window, use the +open_floating_window function. These functions accept the same parameters as +X11::XCB::Window→new, see the i3test documentation at TODO.

+
+
t/11-goto.t: Helper function
+
+
#
+# Returns the input focus after sending the given command to i3 via IPC
+# and syncing with i3
+#
+sub focus_after {
+    my $msg = shift;
+
+    cmd $msg;
+    sync_with_i3 $x;
+    return $x->input_focus;
+}
+
+

This section defines a helper function which will be used over and over in this +testcase. If you have code which gets executed more than once or twice +(depending on the length of your test, use your best judgement), please put it +in a function. Tests should be short, concise and clear.

+

The focus_after function executes a command and returns the X11 focus after +the command was executed. The sync_with_i3 command makes sure that i3 could +push its state to X11. See [i3_sync] to learn how this works exactly.

+
+
t/11-goto.t: Test assumptions
+
+
$focus = $x->input_focus;
+is($focus, $bottom->id, "Latest window focused");
+
+$focus = focus_after('focus left');
+is($focus, $mid->id, "Middle window focused");
+
+

Now, we run the first two real tests. They use Test::More's is function, +which compares two values and prints the differences if they are not the same. +After the arguments, we supply a short comment to indicate what we are testing +here. This makes it vastly more easy for the developer to spot which testcase +is the problem in case one fails.

+

The first test checks that the most recently opened window is focused. +Afterwards, the command focus left is issued and it is verified that the +middle window now has focus.

+

Note that this is not a comprehensive test of the focus command — we would +have to test wrapping, focus when using a more complex layout, focusing the +parent/child containers, etc. But that is not the point of this testcase. +Instead, we just want to know if $x→input_focus corresponds with what we are +expecting. If not, something is completely wrong with the test environment and +this trivial test will fail.

+
+
t/11-goto.t: Test that the feature does not work (yet)
+
+
#####################################################################
+# Now goto a mark which does not exist
+#####################################################################
+
+my $random_mark = mktemp('mark.XXXXXX');
+
+$focus = focus_after(qq|[con_mark="$random_mark"] focus|);
+is($focus, $mid->id, "focus unchanged");
+
+

Syntax hint: The qq keyword is the interpolating quote operator. It lets you +chose a quote character (in this case the | character, a pipe). This makes +having double quotes in our string easy.

+

In this new major section, a random mark (mark is an identifier for a window, +see "VIM-like marks" in the i3 User’s Guide) will be generated. Afterwards, we +test that trying to focus that mark will not do anything. This is important: Do +not only test that using a feature has the expected outcome, but also test that +using it without properly initializing it does no harm. This command could for +example have changed focus anyways (a bug) or crash i3 (obviously a bug).

+
+
t/11-goto.t: Test that the feature does work
+
+
cmd "mark $random_mark";
+
+$focus = focus_after('focus left');
+is($focus, $top->id, "Top window focused");
+
+$focus = focus_after(qq|[con_mark="$random_mark"] focus|);
+is($focus, $mid->id, "goto worked");
+
+

Remember: Focus was on the middle window (we verified that earlier in "Test +assumptions"). We now mark the middle window with our randomly generated mark. +Afterwards, we switch focus away from the middle window to be able to tell if +focusing it via its mark will work. If the test works, the goto command seems +to be working.

+
+
t/11-goto.t: Test corner case
+
+
# check that we can specify multiple criteria
+
+$focus = focus_after('focus left');
+is($focus, $top->id, "Top window focused");
+
+$focus = focus_after(qq|[con_mark="$random_mark" con_mark="$random_mark"] focus|);
+is($focus, $mid->id, "goto worked");
+
+

Now we test the same feature, but specifying the mark twice in the command. +This should have no effect, but let’s be sure: test it and see if things go +wrong.

+
+
t/11-goto.t: Test second code path
+
+
#####################################################################
+# Check whether the focus command will switch to a different
+# workspace if necessary
+#####################################################################
+
+my $tmp2 = fresh_workspace;
+
+is(focused_ws(), $tmp2, 'tmp2 now focused');
+
+cmd qq|[con_mark="$random_mark"] focus|;
+
+is(focused_ws(), $tmp, 'tmp now focused');
+
+

This part of the test checks that focusing windows by mark works across +workspaces. It uses i3test’s focused_ws function to get the current +workspace.

+
+
t/11-goto.t: Test second code path
+
+
done_testing;
+
+

The end of every testcase has to contain the done_testing line. This tells +complete-run.pl that the test was finished successfully. If it does not +occur, the test might have crashed during execution — some of the reasons why +that could happen are bugs in the used modules, bugs in the testcase itself or +an i3 crash resulting in the testcase being unable to communicate with i3 via +IPC anymore.

+
+
+
+

4. Appendix A: The i3 sync protocol

+
+

Consider the following situation: You open two windows in your testcase, then +you use focus left and want to verify that the X11 focus has been updated +properly. Sounds simple, right? Let’s assume you use this straight-forward +implementation:

+
+
Racey focus testcase
+
+
my $left = open_window($x);
+my $right = open_window($x);
+cmd 'focus left';
+is($x->input_focus, $left->id, 'left window focused');
+
+

However, the test fails. Sometimes. Apparantly, there is a race condition in +your test. If you think about it, this is because you are using two different +pieces of software: You tell i3 to update focus, i3 confirms that, and then you +ask X11 to give you the current focus. There is a certain time i3 needs to +update the X11 state. If the testcase gets CPU time before X11 processed i3’s +requests, the test will fail.

+
+
+Diagram of the race condition +
+
Figure 1. Diagram of the race condition
+
+

One way to "solve" this would be to add sleep 0.5; after the cmd call. +After 0.5 seconds it should be safe to assume that focus has been updated, +right?

+

In practice, this usually works. However, it has several problems:

+
    +
  1. +

    +This is obviously not a clean solution, but a workaround. Ugly. +

    +
    2. +
  2. +

    +On very slow machines, this might not work. Unlikely, but in different + situations (a delay to wait for i3 to startup) the necessary time is much + harder to guess, even for fast machines. +

    +
    3. +
  3. +

    +This wastes a lot of time. Usually, your computer is much faster than 0.5s + to update the status. However, sometimes, it might take 0.4s, so we can’t + make it sleep 0.1. +

    +
    4. +
+

To illustrate how grave the problem with wasting time actually is: Before +removing all sleeps from the testsuite, a typical run using 4 separate X +servers took around 50 seconds on my machine. After removing all the sleeps, +we achieved times of about 25 seconds. This is very significant and influences +the way you think about tests — the faster they are, the more likely you are +to check whether everything still works quite often (which you should).

+

What I am trying to say is: Delays adds up quickly and make the test suite +less robust.

+

The real solution for this problem is a mechanism which I call "the i3 sync +protocol". The idea is to send a request (which does not modify state) via X11 +to i3 which will then be answered. Due to the request’s position in the event +queue (after all previous events), you can be sure that by the time you +receive the reply, all other events have been dealt with by i3 (and, more +importantly, X11).

+
+
+Diagram of the i3 sync solution +
+
Figure 2. Diagram of the i3 sync solution
+
+
+

4.1. Implementation details

+

The client which wants to sync with i3 initiates the protocol by sending a +ClientMessage to the X11 root window:

+
+
Send ClientMessage
+
+
# Generate a ClientMessage, see xcb_client_message_t
+my $msg = pack "CCSLLLLLLL",
+    CLIENT_MESSAGE, # response_type
+    32,     # format
+    0,      # sequence
+    $root,  # destination window
+    $x->atom(name => 'I3_SYNC')->id,
+
+    $_sync_window->id,    # data[0]: our own window id
+    $myrnd, # data[1]: a random value to identify the request
+    0,
+    0,
+    0;
+
+# Send it to the root window -- since i3 uses the SubstructureRedirect
+# event mask, it will get the ClientMessage.
+$x->send_event(0, $root, EVENT_MASK_SUBSTRUCTURE_REDIRECT, $msg);
+
+

i3 will then reply with the same ClientMessage, sent to the window specified in +data[0]. In the reply, data[0] and data[1] are exactly the same as in the +request. You should use a random value in data[1] and check that you received +the same one when getting the reply.

+
+
+
+
+

5. Appendix B: Socket activation

+
+

Socket activation is a mechanism which was made popular by systemd, an init +replacement. It basically describes creating a listening socket before starting +a program. systemd will invoke the program only when an actual connection to +the socket is made, hence the term socket activation.

+

The interesting part of this (in the i3 context) is that you can very precisely +detect when the program is ready (finished its initialization).

+
+

5.1. Preparing the listening socket

+

complete-run.pl will create a listening UNIX socket which it will then pass +to i3. This socket will be used by i3 as an additional IPC socket, just like +the one it will create on its own. Passing the socket happens implicitly +because children will inherit the parent’s sockets when fork()ing and sockets +will continue to exist after an exec() call (unless CLOEXEC is set of course).

+

The only explicit things complete-run.pl has to do is setting the LISTEN_FDS +environment variable to the number of sockets which exist (1 in our case) and +setting the LISTEN_PID environment variable to the current process ID. Both +variables are necessary so that the program (i3) knows how many sockets it +should use and if the environment variable is actually intended for it. i3 will +then start looking for sockets at file descriptor 3 (since 0, 1 and 2 are used +for stdin, stdout and stderr, respectively).

+

The actual Perl code which sets up the socket, fork()s, makes sure the socket +has file descriptor 3 and sets up the environment variables follows (shortened +a bit):

+
+
Setup socket and environment
+
+
my $socket = IO::Socket::UNIX->new(
+    Listen => 1,
+    Local => $args{unix_socket_path},
+);
+
+my $pid = fork;
+if ($pid == 0) {
+    $ENV{LISTEN_PID} = $$;
+    $ENV{LISTEN_FDS} = 1;
+
+    # Only pass file descriptors 0 (stdin), 1 (stdout),
+    # 2 (stderr) and 3 (socket) to the child.
+    $^F = 3;
+
+    # If the socket does not use file descriptor 3 by chance
+    # already, we close fd 3 and dup2() the socket to 3.
+    if (fileno($socket) != 3) {
+        POSIX::close(3);
+        POSIX::dup2(fileno($socket), 3);
+    }
+
+    exec "/usr/bin/i3";
+}
+
+
+
+

5.2. Waiting for a reply

+

In the parent process, we want to know when i3 is ready to answer our IPC +requests and handle our windows. Therefore, after forking, we immediately close +the listening socket (i3 will handle this side of the socket) and connect to it +(remember, we are talking about a named UNIX socket) as a client. This connect +call will immediately succeed because the kernel buffers it. Then, we send a +request (of type GET_TREE, but that is not really relevant). Writing data to +the socket will also succeed immediately because, again, the kernel buffers it +(only up to a certain amount of data of course).

+

Afterwards, we just blockingly wait until we get an answer. In the child +process, i3 will setup the listening socket in its event loop. Immediately +after actually starting the event loop, it will notice a new client connecting +(the parent process) and handle its request. Since all initialization has been +completed successfully by the time the event loop is entered, we can now assume +that i3 is ready.

+
+
+

5.3. Timing and conclusion

+

A beautiful feature of this mechanism is that it does not depend on timing. It +does not matter when the child process gets CPU time or when the parent process +gets CPU time. On heavily loaded machines (or machines with multiple CPUs, +cores or unreliable schedulers), this makes waiting for i3 much more robust.

+

Before using socket activation, we typically used a sleep(1) and hoped that +i3 was initialized by that time. Of course, this breaks on some (slow) +computers and wastes a lot of time on faster computers. By using socket +activation, we decreased the total amount of time necessary to run all tests +(72 files at the time of writing) from > 100 seconds to 16 seconds. This makes +it significantly more attractive to run the test suite more often (or at all) +during development.

+

An alternative approach to using socket activation is polling for the existance +of the IPC socket and connecting to it. While this might be slightly easier to +implement, it wastes CPU time and is considerably uglier than this solution +:). After all, lib/SocketActivation.pm contains only 54 SLOC.

+
+
+
+
+
+ + + diff --git a/docs/userguide.html b/docs/userguide.html index 4403ffa..5ab8111 100644 --- a/docs/userguide.html +++ b/docs/userguide.html @@ -31,7 +31,7 @@ window.onload = function(){asciidoc.footnotes(); asciidoc.toc(2);}

i3 User’s Guide

Michael Stapelberg
<michael+i3@stapelberg.de>
-August 2011 +October 2011
Table of Contents
@@ -436,7 +436,8 @@ you can press Mod1, click into a window using your left mouse button, and drag it to the position you want.

When holding the floating modifier, you can resize a floating window by pressing the right mouse button on it and moving around while holding it. If -you hold the shift button as well, the resize will be proportional.

+you hold the shift button as well, the resize will be proportional (the aspect +ratio will be preserved).

Syntax:

@@ -503,7 +504,7 @@ change their border style, for example.

Syntax:

-
for_window [criteria] command
+
for_window <criteria> command

Examples:

@@ -546,34 +547,54 @@ configuration file and run it before starting i3 (for example in your

4.10. Automatically putting clients on specific workspaces

-

Specific windows can be matched by window class and/or window title. It is -recommended that you match on window classes instead of window titles whenever -possible because some applications first create their window, and then worry -about setting the correct title. Firefox with Vimperator comes to mind. The -window starts up being named Firefox, and only when Vimperator is loaded does -the title change. As i3 will get the title as soon as the application maps the +

To automatically make a specific window show up on a specific workspace, you +can use an assignment. You can match windows by using any criteria, +see [command_criteria]. It is recommended that you match on window classes +(and instances, when appropriate) instead of window titles whenever possible +because some applications first create their window, and then worry about +setting the correct title. Firefox with Vimperator comes to mind. The window +starts up being named Firefox, and only when Vimperator is loaded does the +title change. As i3 will get the title as soon as the application maps the window (mapping means actually displaying it on the screen), you’d need to have to match on Firefox in this case.

-

You can prefix or suffix workspaces with a ~ to specify that matching clients -should be put into floating mode. If you specify only a ~, the client will -not be put onto any workspace, but will be set floating on the current one.

Syntax:

-
assign ["]window class[/window title]["] [→] [workspace]
+
assign <criteria> [→] workspace

Examples:

-
assign urxvt 2
-assign urxvt → 2
-assign urxvt → work
-assign "urxvt" → 2
-assign "urxvt/VIM" → 3
-assign "gecko" → 4
+
# Assign URxvt terminals to workspace 2
+assign [class="URxvt"] 2
+
+# Same thing, but more precise (exact match instead of substring)
+assign [class="^URxvt$"] 2
+
+# Same thing, but with a beautiful arrow :)
+assign [class="^URxvt$"] → 2
+
+# Assignment to a named workspace
+assign [class="^URxvt$"] → work
+
+# Start urxvt -name irssi
+assign [class="^URxvt$" instance="^irssi$"] → 3

Note that the arrow is not required, it just looks good :-). If you decide to use it, it has to be a UTF-8 encoded arrow, not -> or something like that.

+

To get the class and instance, you can use xprop. After clicking on the +window, you will see the following output:

+

xwininfo:

+
+
+
WM_CLASS(STRING) = "irssi", "URxvt"
+
+

The first part of the WM_CLASS is the instance ("irssi" in this example), the +second part is the class ("URxvt" in this example).

+

Should you have any problems with assignments, make sure to check the i3 +logfile first (see http://i3wm.org/docs/debugging.html). It includes more +details about the matching process and the window’s actual class, instance and +title when starting up.

4.11. Automatically starting applications on i3 startup

@@ -585,15 +606,19 @@ keyword. These commands will be run in order.

Syntax:

-
exec command
-exec_always command
+
exec [--no-startup-id] command
+exec_always [--no-startup-id] command

Examples:

-
exec i3status | i3bar -d
-exec_always ~/my_script.sh
+
exec chromium
+exec_always ~/my_script.sh
+
+# Execute the terminal emulator urxvt, which is not yet startup-notification aware.
+exec --no-startup-id urxvt
+

The flag --no-startup-id is explained in [exec].

4.12. Automatically putting workspaces on specific screens

@@ -605,16 +630,18 @@ the second screen and so on).

Syntax:

-
workspace <number> output <output>
+
workspace <workspace> output <output>

The output is the name of the RandR output you attach your screen to. On a laptop, you might have VGA1 and LVDS1 as output names. You can see the available outputs by running xrandr --current.

+

If you use named workspaces, they must be quoted:

Examples:

workspace 1 output LVDS1
-workspace 5 output VGA1
+workspace 5 output VGA1 +workspace "2: vim" output VGA1
@@ -778,10 +805,315 @@ will always wrap.

force_focus_wrapping yes
+
+

4.18. Forcing Xinerama

+

As explained in-depth in http://i3wm.org/docs/multi-monitor.html, some X11 +video drivers (especially the nVidia binary driver) only provide support for +Xinerama instead of RandR. In such a situation, i3 must be told to use the +inferior Xinerama API explicitly and therefore don’t provide support for +reconfiguring your screens on the fly (they are read only once on startup and +that’s it).

+

For people who do cannot modify their ~/.xsession to add the +--force-xinerama commandline parameter, a configuration option is provided:

+

Syntax:

+
+
+
force_xinerama <yes|no>
+
+

Example:

+
+
+
force_xinerama yes
+
+

Also note that your output names are not descriptive (like HDMI1) when using +Xinerama, instead they are counted up, starting at 0: xinerama-0, xinerama-1, …

+
+
+

4.19. Automatic back-and-forth when switching to the current workspace

+

This configuration directive enables automatic workspace back_and_forth (see +[back_and_forth]) when switching to the workspace that is currently focused.

+

For instance: Assume you are on workspace "1: www" and switch to "2: IM" using +mod+2 because somebody sent you a message. You don’t need to remember where you +came from now, you can just press mod+2 again to switch back to "1: www".

+

Syntax:

+
+
+
workspace_auto_back_and_forth <yes|no>
+
+

Example:

+
+
+
workspace_auto_back_and_forth yes
+
+
+ + +
+

5. Configuring i3bar

+
+

The bar at the bottom of your monitor is drawn by a separate process called +i3bar. Having this part of "the i3 user interface" in a separate process has +several advantages:

+
    +
  1. +

    +It is a modular approach. If you don’t need a workspace bar at all, or if + you prefer a different one (dzen2, xmobar, maybe even gnome-panel?), you can + just remove the i3bar configuration and start your favorite bar instead. +

    +
    2. +
  2. +

    +It follows the UNIX philosophy of "Make each program do one thing well". + While i3 manages your windows well, i3bar is good at displaying a bar on + each monitor (unless you configure it otherwise). +

    +
    3. +
  3. +

    +It leads to two separate, clean codebases. If you want to understand i3, you + don’t need to bother with the details of i3bar and vice versa. +

    +
    4. +
+

That said, i3bar is configured in the same configuration file as i3. This is +because it is tightly coupled with i3 (in contrary to i3lock or i3status which +are useful for people using other window managers). Therefore, it makes no +sense to use a different configuration place when we already have a good +configuration infrastructure in place.

+

Configuring your workspace bar starts with opening a bar block. You can have +multiple bar blocks to use different settings for different outputs (monitors):

+

Example:

+
+
+
bar {
+    status_command i3status
+}
+
+
+

5.1. Statusline command

+

i3bar can run a program and display every line of its stdout output on the +right hand side of the bar. This is useful to display system information like +your current IP address, battery status or date/time.

+

The specified command will be passed to sh -c, so you can use globbing and +have to have correct quoting etc.

+

Syntax:

+
+
+
status_command command
+
+

Example:

+
+
+
status_command i3status --config ~/.i3status.conf
+
+
+
+

5.2. Display mode

+

You can have i3bar either be visible permanently at one edge of the screen +(dock mode) or make it show up when you press your modifier key (hide +mode).

+

The hide mode maximizes screen space that can be used for actual windows. Also, +i3bar sends the SIGSTOP and SIGCONT signals to the statusline process to +save battery power.

+

The default is dock mode.

+

Syntax:

+
+
+
mode <dock|hide>
+
+

Example:

+
+
+
mode hide
+
+
+
+

5.3. Position

+

This option determines in which edge of the screen i3bar should show up.

+

The default is bottom.

+

Syntax:

+
+
+
position <top|bottom>
+
+

Example:

+
+
+
position top
+
+
+
+

5.4. Output(s)

+

You can restrict i3bar to one or more outputs (monitors). The default is to +handle all outputs. Restricting the outputs is useful for using different +options for different outputs by using multiple bar blocks.

+

Syntax:

+
+
+
output <output>
+
+

Example:

+
+
+
# big monitor: everything
+bar {
+        output HDMI2
+        status_command i3status
+}
+
+# laptop monitor: bright colors and i3status with less modules.
+bar {
+        output LVDS1
+        status_command i3status --config ~/.i3status-small.conf
+        colors {
+                background #000000
+                statusline #ffffff
+        }
+}
+
+
+
+

5.5. Tray output

+

i3bar by default provides a system tray area where programs such as +NetworkManager, VLC, Pidgin, etc. can place little icons.

+

You can configure on which output (monitor) the icons should be displayed or +you can turn off the functionality entirely.

+

Syntax:

+
+
+
tray_output <none|output>
+
+

Example:

+
+
+
# disable system tray
+tray_output none
+
+# show tray icons on the big monitor
+tray_output HDMI2
+
+
+
+

5.6. Font

+

Specifies the font (again, X core font, not Xft, just like in i3) to be used in +the bar.

+

Syntax:

+
+
+
font <font>
+
+

Example:

+
+
+
font -misc-fixed-medium-r-normal--13-120-75-75-C-70-iso10646-1
+
+
+
+

5.7. Workspace buttons

+

Specifies whether workspace buttons should be shown or not. This is useful if +you want to display a statusline-only bar containing additional information.

+

The default is to show workspace buttons.

+

Syntax:

+
+
+
workspace_buttons <yes|no>
+
+

Example:

+
+
+
workspace_buttons no
+
+
+
+

5.8. Colors

+

As with i3, colors are in HTML hex format (#rrggbb). The following colors can +be configured at the moment:

+
+
+background +
+
+

+ Background color of the bar. +

+
+
+statusline +
+
+

+ Text color to be used for the statusline. +

+
+
+focused_workspace +
+
+

+ Text color/background color for a workspace button when the workspace + has focus. +

+
+
+active_workspace +
+
+

+ Text color/background color for a workspace button when the workspace + is active (visible) on some output, but the focus is on another one. + You can only tell this apart from the focused workspace when you are + using multiple monitors. +

+
+
+inactive_workspace +
+
+

+ Text color/background color for a workspace button when the workspace + does not have focus and is not active (visible) on any output. This + will be the case for most workspaces. +

+
+
+urgent_workspace +
+
+

+ Text color/background color for workspaces which contain at least one + window with the urgency hint set. +

+
+
+

Syntax:

+
+
+
colors {
+    background <color>
+    statusline <color>
+
+    colorclass <foreground> <background>
+}
+
+

Example:

+
+
+
colors {
+    background #000000
+    statusline #ffffff
+
+    focused_workspace  #ffffff #285577
+    active_workspace   #ffffff #333333
+    inactive_workspace #888888 #222222
+    urgent_workspace   #ffffff #900000
+}
+
+
-

5. List of commands

+

6. List of commands

Commands are what you bind to specific keypresses. You can also issue commands at runtime without pressing a key by using the IPC interface. An easy way to @@ -807,7 +1139,10 @@ which have the class Firefox, use:

Example:

-
bindsym mod+x [class="Firefox"] kill
+
bindsym mod+x [class="Firefox"] kill
+
+# same thing, but case-insensitive
+bindsym mod+x [class="(?i)firefox"] kill

The criteria which are currently implemented are:

@@ -828,6 +1163,14 @@ instance

+window_role +
+
+

+ Compares the window role (WM_WINDOW_ROLE). +

+
+
id
@@ -844,7 +1187,7 @@ title

-mark +con_mark

@@ -861,10 +1204,39 @@ con_id

-

Note that currently all criteria are compared case-insensitive and do not -support regular expressions. This is planned to change in the future.

+

The criteria class, instance, role, title and mark are actually +regular expressions (PCRE). See pcresyntax(3) or perldoc perlre for +information on how to use them.

-

5.1. Splitting containers

+

6.1. Executing applications (exec)

+

What good is a window manager if you can’t actually start any applications? +The exec command starts an application by passing the command you specify to a +shell. This implies that you can use globbing (wildcards) and programs will be +searched in your $PATH.

+

Syntax:

+
+
+
exec [--no-startup-id] command
+
+

Example:

+
+
+
# Start the GIMP
+bindsym mod+g exec gimp
+
+# Start the terminal emulator urxvt which is not yet startup-notification-aware
+bindsym mod+enter exec --no-startup-id urxvt
+
+

The --no-startup-id parameter disables startup-notification support for this +particular exec command. With startup-notification, i3 can make sure that a +window appears on the workspace on which you used the exec command. Also, it +will change the X11 cursor to watch (a clock) while the application is +launching. So, if an application is not startup-notification aware (most GTK +and Qt using applications seem to be, though), you will end up with a watch +cursor for 60 seconds.

+
+
+

6.2. Splitting containers

The split command makes the current window a split container. Split containers can contain multiple windows. Every split container has an orientation, it is either split horizontally (a new window gets placed to the right of the current @@ -885,7 +1257,7 @@ bindsym mod+h split horizontal

-

5.2. Manipulating layout

+

6.3. Manipulating layout

Use layout default, layout stacking or layout tabbed to change the current container layout to default, stacking or tabbed layout, respectively.

To make the current window (!) fullscreen, use fullscreen, to make @@ -906,7 +1278,7 @@ bindsym mod+t floating toggle

-

5.3. Focusing/Moving containers

+

6.4. Focusing/Moving containers

To change the focus, use the focus command: focus left, focus right, focus down and focus up.

There are a few special parameters you can use for the focus command:

@@ -953,10 +1325,19 @@ mode_toggle

For moving, use move left, move right, move down and move up.

+

Syntax:

+
+
+
focus <left|right|down|up>
+focus <parent|child|floating|tiling|mode_toggle>
+move <left|right|down|up> [<px> px]
+
+

Note that the amount of pixels you can specify for the move command is only +relevant for floating containers. The default amount is 10 pixels.

Examples:

-
# Focus clients on the left, bottom, top, right:
+
# Focus container on the left, bottom, top, right:
 bindsym mod+j focus left
 bindsym mod+k focus down
 bindsym mod+l focus up
@@ -968,22 +1349,29 @@ bindsym mod+u focus parent
 # Focus last floating/tiling container
 bindsym mod+g focus mode_toggle
 
-# Move client to the left, bottom, top, right:
+# Move container to the left, bottom, top, right:
 bindsym mod+j move left
 bindsym mod+k move down
 bindsym mod+l move up
-bindsym mod+semicolon move right

+bindsym mod+semicolon move right
+
+# Move container, but make floating containers
+# move more than the default
+bindsym mod+j move left 20 px
-

5.4. Changing (named) workspaces/moving to workspaces

+

6.5. Changing (named) workspaces/moving to workspaces

To change to a specific workspace, use the workspace command, followed by the number or name of the workspace. To move containers to specific workspaces, use move workspace.

You can also switch to the next and previous workspace with the commands workspace next and workspace prev, which is handy, for example, if you have workspace 1, 3, 4 and 9 and you want to cycle through them with a single key -combination.

+combination. Similarily, you can use move workspace next and move workspace +prev to move a container to the next/previous workspace.

+

To switch back to the previously focused workspace, use workspace +back_and_forth.

To move a container to another xrandr output such as LVDS1 or VGA1, you can use the move output command followed by the name of the target output. You may also use left, right, up, down instead of the xrandr output name to @@ -997,10 +1385,13 @@ bindsym mod+2 workspace 2 bindsym mod+Shift+1 move workspace 1 bindsym mod+Shift+2 move workspace 2 -... +... + +# switch between the current and the previously focused one +bindsym mod+b workspace back_and_forth

-

5.4.1. Named workspaces

+

6.5.1. Named workspaces

Workspaces are identified by their name. So, instead of using numbers in the workspace command, you can use an arbitrary name:

Example:

@@ -1025,7 +1416,7 @@ will order them numerically.

-

5.5. Resizing containers/windows

+

6.6. Resizing containers/windows

If you want to resize containers/windows using your keyboard, you can use the resize command:

Syntax:

@@ -1035,9 +1426,9 @@ will order them numerically.

Direction can be one of up, down, left or right. The optional pixel argument specifies by how many pixels a floating container should be grown or -shrinked (the default is 10 pixels). The ppt argument means percentage points +shrunk (the default is 10 pixels). The ppt argument means percentage points and specifies by how many percentage points a tiling container should be -grown or shrinked (the default is 10 percentage points).

+grown or shrunk (the default is 10 percentage points).

I recommend using the resize command inside a so called mode:

Example: Configuration file, defining a mode for resizing
@@ -1071,7 +1462,7 @@ bindsym mod+r mode "resize"
-

5.6. Jumping to specific windows

+

6.7. Jumping to specific windows

Often when in a multi-monitor environment, you want to quickly jump to a specific window. For example, while working on workspace 3 you may want to jump to your mail client to email your boss that you’ve achieved some @@ -1092,7 +1483,7 @@ bindsym mod+a [class="urxvt" title="VIM"] focus

-

5.7. VIM-like marks (mark/goto)

+

6.8. VIM-like marks (mark/goto)

This feature is like the jump feature: It allows you to directly jump to a specific window (this means switching to the appropriate workspace and setting focus to the windows). However, you can directly mark a specific window with @@ -1117,7 +1508,7 @@ $ i3-msg '[con_mark="irssi"] focus'

-

5.8. Changing border style

+

6.9. Changing border style

To change the border of the current client, you can use border normal to use the normal border (including window title), border 1pixel to use a 1-pixel border (no window title) and border none to make the client borderless.

@@ -1131,7 +1522,7 @@ bindsym mod+u border none
-

5.9. Reloading/Restarting/Exiting

+

6.10. Reloading/Restarting/Exiting

You can make i3 reload its configuration file with reload. You can also restart i3 inplace with the restart command to get it out of some weird state (if that should ever happen) or to perform an upgrade without having to restart @@ -1148,7 +1539,7 @@ bindsym mod+Shift+e exit

-

6. Multiple monitors

+

7. Multiple monitors

As you can see in the goal list on the website, i3 was specifically developed with support for multiple monitors in mind. This section will explain how to @@ -1171,7 +1562,7 @@ create an unlimited number of workspaces in i3 and tie them to specific screens, you can have the "traditional" approach of having X workspaces per screen by changing your configuration (using modes, for example).

-

6.1. Configuring your monitors

+

7.1. Configuring your monitors

To help you get going if you have never used multiple monitors before, here is a short overview of the xrandr options which will probably be of interest to you. It is always useful to get an overview of the current screen configuration. @@ -1234,7 +1625,7 @@ only what you can see in xrandr.

See also [presentations] for more examples of multi-monitor setups.

-

6.2. Interesting configuration for multi-monitor environments

+

7.2. Interesting configuration for multi-monitor environments

There are several things to configure in i3 which may be interesting if you have more than one monitor:

    @@ -1265,10 +1656,10 @@ If you have many workspaces on many monitors, it might get hard to keep
-

7. i3 and the rest of your software world

+

8. i3 and the rest of your software world

-

7.1. Displaying a status line

+

8.1. Displaying a status line

A very common thing amongst users of exotic window managers is a status line at some corner of the screen. It is an often superior replacement to the widget approach you have in the task bar of a traditional desktop environment.

@@ -1286,14 +1677,14 @@ for positioning it at the bottom and -dtop to position it at the top of screen.

-

7.2. Giving presentations (multi-monitor)

+

8.2. Giving presentations (multi-monitor)

When giving a presentation, you typically want the audience to see what you see on your screen and then go through a series of slides (if the presentation is simple). For more complex presentations, you might want to have some notes which only you can see on your screen, while the audience can only see the slides.

-

7.2.1. Case 1: everybody gets the same output

+

8.2.1. Case 1: everybody gets the same output

This is the simple case. You connect your computer to the video projector, turn on both (computer and video projector) and configure your X server to clone the internal flat panel of your computer to the video output:

@@ -1306,7 +1697,7 @@ your screen will be left untouched (it will show the X background). So, in our example, this would be 1024x768 (my notebook has 1280x800).

-

7.2.2. Case 2: you can see more than your audience

+

8.2.2. Case 2: you can see more than your audience

This case is a bit harder. First of all, you should configure the VGA output somewhere near your internal flat panel, say right of it:

-- 2.39.5