profile
viewpoint

google/nftables 546

This package manipulates Linux nftables (the iptables successor)

google/renameio 433

Package renameio provides a way to atomically create or replace a file or symbolic link.

google/gousb 407

gousb provides low-level interface for accessing USB devices

distr1/distri 363

a Linux distribution to research fast package management

google/go-safeweb 330

Secure-by-default HTTP servers in Go.

jacobsa/fuse 257

A Go package for implementing a FUSE file system.

stapelberg/airscan 112

Go package to scan paper documents 📄 from a scanner 🖨️ via the network 🕸️ using the Apple AirScan (eSCL) protocol.

google/dnsmasq_exporter 79

dnsmasq exporter for Prometheus

paultag/go-debian 52

debian golang support library

pion/obs-wormhole 37

Supercharge OBS with WebRTC. Remote WebRTC Sources and Serverless Broadcasting

issue commenti3/i3

semicolon in i3-msg commands add trailing space to new marks

I propose adding a i3-msg section.

i3-msg is just a tool to use the IPC interface (separately documented in https://i3wm.org/docs/ipc.html) to send commands, which are documented in https://i3wm.org/docs/userguide.html#list_of_commands. That section that last link points to should probably be expanded.

budRich

comment created time in 3 hours

issue commenti3/i3

building deb package fails w/ debuild

are these flags the way to go for disabling sanitizers in meson for release build?: --buildtype=release -Db_sanitize=none

looks right to me, yeah

laur89

comment created time in 3 hours

push eventi3/i3

Michael Stapelberg

commit sha 5c565d0ad46b66f0941424c06acfb67808b5164b

debian/rules: explicitly enable building manpages This fixes a build failure in dh_installwms. fixes #4136

view details

Michael Stapelberg

commit sha 8a7ac068c8359e5aae513854f0b21bce4460e8aa

meson.build: fix typo in summary: docs → mans

view details

push time in 5 hours

delete branch stapelberg/i3

delete branch : debuild

delete time in 5 hours

issue closedi3/i3

building deb package fails w/ debuild

<!-- PLEASE HELP US PROCESS GITHUB ISSUES FASTER BY PROVIDING THE FOLLOWING INFORMATION. -->

I'm submitting a…

<!-- Please check one of the following options with "x" --> <pre> [x] Bug [ ] Feature Request [ ] Documentation Request [ ] Other (Please describe in detail) </pre>

Current Behavior

<!-- Describe the current behavior, e.g., »When pressing Alt+j (focus left), the window above the current window is focused.« --> Building deb package using debuild after meson (actually unlikely it's the fault of a build tool change, only speculating here) was introduced fails on debian, see this gist for full log. Tail end of the error:

make[1]: Entering directory '/tmp/i3'
dh_installchangelogs RELEASE-NOTES-*
make[1]: Leaving directory '/tmp/i3'
   dh_systemd_enable -O--buildsystem=meson
   dh_installinit -O--buildsystem=meson
   dh_systemd_start -O--buildsystem=meson
   dh_perl -O--buildsystem=meson
   dh_link -O--buildsystem=meson
   dh_installwm -O--buildsystem=meson
dh_installwm: error: no manpage found (creating an x-window-manager alternative requires a slave symlink for the manpage)
make: *** [debian/rules:30: binary] Error 25
dpkg-buildpackage: error: fakeroot debian/rules binary subprocess returned exit status 2
debuild: fatal error at line 1182:
dpkg-buildpackage -us -uc -ui -b failed

Expected Behavior

<!-- Describe the desired behavior you expect after mitigation of the issue, e.g., »The window left next to the current window should be focused.« -->

Expecting running debuild -us -uc -b to produce a valid deb package, as used to be the case some months ago.

Reproduction Instructions

<!-- Please provide detailed instructions on how the bug can be reproduced. E.g., »Open three windows in a V[A H[B C]] layout on a new workspace« -->

  • Clone the i3 repo
  • navigate to the cloned dir
  • make sure all the build deps are installed
  • run debuild -us -uc -b

Environment

i3 ver: 51b05835

<!-- Please also answer the questions below to help us process your issue faster. If you have any other information to share, please add it here as well. --> <pre>

  • Linux Distribution & Version: Linux WH5005264 5.6.0-2-amd64 #1 SMP Debian 5.6.14-1 (2020-05-23) x86_64 GNU/Linux Debian testing (bullseye) </pre>

Bonus question:

How to modify the debian/rules file to configure for proper release build? Prior to meson, I applied this patch, ie adding --disable-sanitizers flag. How to properly do this with meson? Would this be correct (ie adding --buildtype=release -Db_sanitize=none flags)?

closed time in 5 hours

laur89

PR merged i3/i3

Fix Debian package build from git
+2 -2

0 comment

2 changed files

stapelberg

pr closed time in 5 hours

issue commenti3/i3

strndup build problem

Nothing we can do about that now.

Next time, feel free to cc me in explicitly early on so that I’m aware.

eleksir

comment created time in 5 hours

issue commenti3/i3

Support for `split tabbed|stacked` command in addition to `split h|v|t`

Obviously because it's counter-intuitive and imposes requirements on initial layout, and inability to split directly into desired layout is inconsistent artificial restriction in general.

It’s not obvious to me at all. Can you elaborate on the requirements?

I don’t split windows often; I use workspace_layout and two or three panes of windows mostly.

AMDmi3

comment created time in 7 hours

issue commenti3/i3

semicolon in i3-msg commands add trailing space to new marks

But I think you should consider adding a fix when you can (v. 5).

We don’t plan to do another major version break.

Till then i guess clarifying this behavior in the user-guide wouldn't hurt.

Agreed. Leaving this issue open to track the documentation change.

budRich

comment created time in 9 hours

issue commenti3/i3

building deb package fails w/ debuild

See https://github.com/i3/i3/pull/4235 for the fix.

laur89

comment created time in 9 hours

PR opened i3/i3

Fix Debian package build from git
+2 -2

0 comment

2 changed files

pr created time in 9 hours

create barnchstapelberg/i3

branch : debuild

created branch time in 9 hours

issue closedi3/i3

Using Pango Markup escape sequences causes the window title bar to be set to empty

<!-- PLEASE HELP US PROCESS GITHUB ISSUES FASTER BY PROVIDING THE FOLLOWING INFORMATION. -->

I'm submitting a…

<!-- Please check one of the following options with "x" --> <pre> [x] Bug [ ] Feature Request [ ] Documentation Request [ ] Other (Please describe in detail) </pre>

Current Behavior

<!-- Describe the current behavior, e.g., »When pressing Alt+j (focus left), the window above the current window is focused.« --> i3-msg title_format '&amp;' causes window title to be set to empty. The command returns [{"success":true}].

Expected Behavior

<!-- Describe the desired behavior you expect after mitigation of the issue, e.g., »The window left next to the current window should be focused.« --> Since &amp; is an escape sequence for the ampersand sign (&) in the Pango Markup, I would expect the window title to be set to &.

Reproduction Instructions

<!-- Please provide detailed instructions on how the bug can be reproduced. E.g., »Open three windows in a V[A H[B C]] layout on a new workspace« --> Just run i3-msg title_format '&amp;' in the terminal and observe the text in the window title bar.

Environment

<!-- Please include your exact i3 version. Note that we only support the latest major release and the current development version. If you are using an older version of i3, please first update to the current release version and reproduce the issue there. --> Output of i3 --moreversion 2>&-: <pre> Binary i3 version: 4.18.2 (2020-07-26) © 2009 Michael Stapelberg and contributors Running i3 version: 4.18.2 (2020-07-26) (pid 668) abort…) Loaded i3 config: /home/jacek/.config/i3/config (Last modified: Fri 24 Jul 2020 06:43:23 PM CEST, 1737182 seconds ago)

The i3 binary you just called: /usr/bin/i3 The i3 binary you are running: i3 </pre>

<!-- Please include your (complete) i3 config with which the issue occurs. You can either paste the file directly or provide a link to a service such as pastebin.

If you would like to help debugging the issue, please try to reduce the config such that it is as close to the default config as possible while still reproducing the issue. This can help us bisect the root cause. --> <details><summary>Config file</summary><pre> </pre> </details>

<!-- Providing a logfile can help us trace the root cause of an issue much quicker. You can learn how to generate the logfile here: https://i3wm.org/docs/debugging.html

Providing the logfile is optional. --> <pre> Logfile URL: </pre>

<!-- Please also answer the questions below to help us process your issue faster. If you have any other information to share, please add it here as well. --> <pre>

  • Linux Distribution & Version: Arch Linux
  • Are you using a compositor (e.g., xcompmgr or compton): compton </pre>

closed time in 9 hours

jw0k

issue commenti3/i3

Using Pango Markup escape sequences causes the window title bar to be set to empty

I think you’re running into shell quoting issues:

  1. i3-msg title_format '&amp;' is turned into ["i3-msg", "title_format", "&amp;"] by your shell
  2. i3-msg concatenates all arguments for convenience: https://github.com/i3/i3/blob/e54e88b9e58579b406c5de94b68cb9ad3211eb4e/i3-msg/main.c#L228-L241
  3. i3 receives title_format &amp;
  4. i3 commands are separated by ;, so i3 parses the command as title_format &amp, which results in invalid markup, which results in pango not displaying anything.

So how can you make things work? You need to make i3 receive the double quotes around the title_format parameter, e.g.:

i3-msg title_format '"&amp;"'

Or, a little easier to understand maybe:

i3-msg 'title_format "&amp;"'

You can verify what i3 sees by running i3-dump-log -f | grep COMMAND

jw0k

comment created time in 9 hours

issue commenti3/i3

Windows start lagging/stuttering after a few minutes unless I regularly switch workspaces.

Oh, I just remembered: your symptoms match overly-aggressive powersaving measures of modern GPUs.

Which GPU do you have?

Have you tried setting it to “maximum performance” and see if that solves the issue?

starcraft66

comment created time in 10 hours

issue commenti3/i3

Windows start lagging/stuttering after a few minutes unless I regularly switch workspaces.

Can you install awesome or some other window manager? We really need to narrow down this problem more first.

starcraft66

comment created time in 10 hours

issue commenti3/i3

Allow for including configuration partials

Every unique real path can be included at most once.

So you’re suggesting evaluating symlinks fully, like https://golang.org/pkg/path/filepath/#EvalSymlinks, yes? Do you happen to know how to do that properly in C?

One alternative (?) might be to check if the files are the same (i.e. same inode on the same file system).

Likewise, included pf the same file twice and includes of nonexistent or unreadable (no permission) files are silently ignored.

No, this is not a good failure mode. If i3 can’t read your config, it must throw an error loudly. I think I’m okay with adding an explicit flag (e.g. include --ignore_errors) to cover the use-case you mentioned, but by default it must error.

moritzheiber

comment created time in 14 hours

issue commenti3/i3

Next release (4.19)

Went through the open issues and added a few to https://github.com/i3/i3/issues/4220#issue-722082797 that need to be looked at before the release.

If any others are relevant that anyone knows of, please mention them.

orestisfl

comment created time in 16 hours

issue commenti3/i3

semicolon in i3-msg commands add trailing space to new marks

So you would have expected that i3 strips whitespace, I take it.

I’m not sure if we can make that change safely without breaking any existing configs.

budRich

comment created time in 16 hours

issue commenti3/i3

Minor modes

As an Emacs user myself, I’m sympathetic to the idea of major/minor modes, but at the same time, I’m worried about making this area of i3 more complex than it absolutely has to be.

Is there any chance that this could be implemented external to i3 using the IPC interface?

emdemir

comment created time in 16 hours

issue commenti3/i3

Enable code coverage using latest version of lcov (1.14)

Does it work if you’re using meson? We’re switching away from autotools, see #4086

colletted

comment created time in 16 hours

issue commenti3/i3

get_tree provides malformed JSON

i3 uses the yajl library to produce JSON. Either:

  1. The bug is actually in yajl, or:
  2. i3 is using yajl subtly wrong.

It would be good to figure out more direct steps to reproduce. Or maybe attach a debugger to i3 when the problem starts happening, if you’re familiar with C.

MatrixManAtYrService

comment created time in 16 hours

issue commenti3/i3

document all commands on the man pages

I don't think we should accept a manual, incremental re-write of the userguide in the manpage.

Agreed.

So we should develop it for people like us that use it, people that don't will still have the userguide

No question about it, I just don’t have time to do it :)

bart9h

comment created time in 17 hours

issue commenti3/i3

Test failures

Actually, the issue pointed out here persists:

Can you provide steps to reproduce please? I thought I fixed that issue over there (and I did verify that at the time).

Also, why does only the i3-gaps ebuild need the patch? The i3 ebuild itself seems to get along without it: https://github.com/gentoo/gentoo/blob/master/x11-wm/i3/i3-4.18.3.ebuild

ajakk

comment created time in 17 hours

issue commenti3/i3

Coloring of Active Container's Parent Pane

Can you re-take these screenshots with the i3 default config please?

In particular, I’m wondering where the focused_inactive color is involved here, and I can’t tell with your config.

H8to

comment created time in 17 hours

issue commenti3/i3

Changing workspace resets floating window size

Floating windows, specifically browser's "Save As" dialog menu, extend outside my screen.

I have noticed this starting to happen as well, and I think this needs to be fixed in the toolkits.

Seems independent of the window manager, too.

SandWoodJones

comment created time in 17 hours

issue commenti3/i3

document all commands on the man pages

Well that's a chicken and egg problem, nobody is going to regularly look at the manpage if it is empty.

It isn’t empty, it’s just not kept up-to-date as much as the userguide is.

I’m not sure we can say that it’s a chicken-and-egg problem. Given that we already have a manpage, I would expect to see more incremental contributions to improve said manpage if there was actual interest.

I think automatically appending the userguide so that it is accessible from man i3 will be useful and more conforming to the usual behavior of nix tools. E.g. man i3 and then /for_window is much more convenient than needing to open a browser tab with the userguide.

Really depends on your workflow/environment. You and I use manpages and know how to navigate them, but I personally know a number of (advanced) computer users who never use manpages.

bart9h

comment created time in 17 hours

issue closedi3/i3

Allow for_window rules to be removed

<!-- PLEASE HELP US PROCESS GITHUB ISSUES FASTER BY PROVIDING THE FOLLOWING INFORMATION. -->

I'm submitting a…

<!-- Please check one of the following options with "x" --> <pre> [ ] Bug [x] Feature Request [ ] Documentation Request [ ] Other (Please describe in detail) </pre>

Current Behavior

for_window and similar rules stick around until a reload

Desired Behavior

Rules can be removed after they are no longer needed (for example, having a launcher add and remove pid-based rules for a process before launching and after it exits - right now, you just have to hope PID reuse never happens).

closed time in 17 hours

danieldg

issue commenti3/i3

Allow for_window rules to be removed

Rules can be removed after they are no longer needed

But i3 can’t know when a rule is no longer needed.

It seems clearer to me to instruct i3 to reload its config (which you modify programmatically beforehand) to pick up changes to these rules.

danieldg

comment created time in 17 hours

issue commenti3/i3

per-workspace-layout.pl not working

per-workspace-layout.pl lives in the contrib directory, i.e. its support level is best-effort.

If it doesn’t work for you, I think you’ll have to take a closer look and send a fix.

Sorry I don’t have a better answer.

sudo-panda

comment created time in 17 hours

issue closedi3/i3

Always show titlebar in tabbed/stacked layout

I'm submitting a…

<!-- Please check one of the following options with "x" --> <pre> [ ] Bug [x] Feature Request [ ] Documentation Request [ ] Other (Please describe in detail) </pre>

Current Behavior

<!-- Describe the current behavior, e.g., »When pressing Alt+j (focus left), the window above the current window is focused.« --> Currently, when titlebars are disabled, a tabbed or stacked container will show a titlebar only when it has multiple children.

Desired Behavior

<!-- Describe the desired behavior you expect after mitigation of the issue, e.g., »The window left next to the current window should be focused.« --> For the sake of consistency, and to avoid the window size changing whenever the second tab is added or removed, I would like to have an option to always show the titlebar in tabbed and stacked containers, even when it is otherwise disabled.

If there is interest, I can try writing a pull request.

closed time in 17 hours

miseran

issue commenti3/i3

Always show titlebar in tabbed/stacked layout

For the sake of consistency, and to avoid the window size changing whenever the second tab is added or removed, I would like to have an option to always show the titlebar in tabbed and stacked containers, even when it is otherwise disabled.

When you disable title bars, i3 only shows title bars when necessary to disambiguite what’s on the screen.

I don’t think we should change this behavior—existing users are used to it, and I like the simple-to-understand model of when to expect title bars. Let’s not make it more complicated.

miseran

comment created time in 17 hours

issue closedi3/i3

Focus specific numbered window in each workspace

<!-- PLEASE HELP US PROCESS GITHUB ISSUES FASTER BY PROVIDING THE FOLLOWING INFORMATION. -->

I'm submitting a…

<!-- Please check one of the following options with "x" --> <pre> [ ] Bug [x] Feature Request [ ] Documentation Request [ ] Other (Please describe in detail) </pre>

Current Behavior

Only workspaces are reachable with $mod+n key binding.

Desired Behavior

I'd like to be able set a key bind to focus each specific window, exactly like workspaces currently work (e.g. go to workspace 2 with $mod+2, focus to window 2 with $mod+ctrl+2).

For this purpose, all the windows in each workspace should be incrementally numbered from 1 to 10, to be uniquely identifiable, and each identifier should be visible on each window to identify it, for example right-aligned on the title bar, or it could be visible at the center of each window while keeping pressed $mod .

The user should able to activate/deactivate this feature from the config file

Environment

<!-- Please include your exact i3 version. Note that we only support the latest major release and the current development version. If you are using an older version of i3, please first update to the current release version and reproduce the issue there. --> Output of i3 --moreversion 2>&-: <pre> i3 version: Binary i3 version: 4.18.2 (2020-07-26) © 2009 Michael Stapelberg and contributors Running i3 version: 4.18.2 (2020-07-26) (pid 41968)bort…) Loaded i3 config: /home/malta/.config/i3/config (Last modified: mer 2 set 2020, 22:30:40, 2885 seconds ago)

The i3 binary you just called: /usr/bin/i3 The i3 binary you are running: i3 </pre>

<!-- Please also answer the questions below to help us process your issue faster. If you have any other information to share, please add it here as well. --> <pre>

  • Linux Distribution & Version: Arch Linux 2020-09-01
  • Are you using a compositor (e.g., xcompmgr or compton): No </pre>

closed time in 17 hours

malta895

issue commenti3/i3

Focus specific numbered window in each workspace

Have a look at this script for tagging the windows. You then only need to bind keys for jumping to the corresponding marks (example).

+1 for doing this externally to i3

malta895

comment created time in 17 hours

issue commenti3/i3

Allow for including configuration partials

This feature request was last discussed and declined in https://github.com/i3/i3/issues/1197#issuecomment-195646042

If you think we should revisit this, can you please go through the points raised in the previous discussion and write down clear answers for how to handle those nuances?

moritzheiber

comment created time in 17 hours

issue closedi3/i3

Keyboard focus on system tray

I'm submitting a…

<pre> [ ] Bug [x] Feature Request [ ] Documentation Request [ ] Other (Please describe in detail) </pre>

Current Behavior

Keyboard focus cannot be given to the system tray or i3status elements on the i3bar

This appears to be incorrect:

i3-msg '[class="i3bar"] focus'  # doesn't work

Desired Behavior

I would like to move keyboard focus to i3bar or the system tray and tab through items for accessibility. I'm primarily interested in accessing system tray content without a mouse.

Environment

i3 version: 4.18.2 (2020-07-26)
i3bar version:  4.18.2 (2020-07-26)
i3status: 2.13
- distro: Arch
- compositor: yshui/picom

In most desktop environments, such as GNOME, KDE and Ubuntu, it appears to be possible to focus the status bar. However, perhaps there are limitations with a WM? From my unsuccessful search on this topic, I am assuming this feature has not been implemented.


Thank you for providing such wonderful software. I benefit tremendously from this project.

closed time in 17 hours

brettinternet

issue commenti3/i3

Keyboard focus on system tray

Yep, seems really tricky on a technical level, unfortunately.

I’ll close this, but feel free to re-open if you think we missed any evidence to the contrary.

brettinternet

comment created time in 17 hours

issue commenti3/i3

Support for `split tabbed|stacked` command in addition to `split h|v|t`

Currently this can be emulated by issuing e.g. split v (which creates a subframe) followed by layout tabbed (which changes its type to tabbed), but that's not entirely convenient.

Can you elaborate on why that isn’t convenient for you? Are you aware that you can have a single keybinding that runs multiple commands? See “multiple commands” in https://i3wm.org/docs/userguide.html#list_of_commands

AMDmi3

comment created time in 17 hours

issue commenti3/i3

document all commands on the man pages

Yeah, the manpage is lacking because nobody has time to maintain it.

I’m not sure how many people even look at the manpage. The HTML userguide is certainly where our effort is spent.

bart9h

comment created time in 17 hours

issue commenti3/i3

Wrong values when resizing floating window

If I listen to events from the Xserver through the xcb lib, I receive two ConfigureNotify events.

I haven’t had time to read through this whole issue or do any investigation, but note that i3 sends out ConfigureNotify events on its own. Search for fake_configure_notify.

aacebedo

comment created time in 17 hours

issue closedi3/i3

Show indicator when hide_edge_borders is set

Hi! I don't know if it is bug or should be feature request, but anyway...

I'm submitting a…

<!-- Please check one of the following options with "x" --> <pre> [ ] Bug [x] Feature Request [ ] Documentation Request [ ] Other (Please describe in detail) </pre>

Current Behavior

When hide_edge_borders is set (to smart in my case) and there is only one window in workspace, there is no indicator showing where the next window will appear

Expected Behavior

An indicator is drawn at the edge of where next window will appear.

closed time in 17 hours

geshido

issue commenti3/i3

Show indicator when hide_edge_borders is set

Next-window indicators are drawn over the window border, and if you don’t have any border, there is no place where we can draw.

I’m not sure how this feature request would even work on a technical level, but independently, I don’t think adding a new knob is worth it for changing this behavior.

geshido

comment created time in 17 hours

issue closedi3/i3

strndup build problem

<!-- PLEASE HELP US PROCESS GITHUB ISSUES FASTER BY PROVIDING THE FOLLOWING INFORMATION. -->

I'm submitting a…

<!-- Please check one of the following options with "x" --> <pre> [x] Bug [ ] Feature Request [ ] Documentation Request [ ] Other (Please describe in detail) </pre>

Current Behavior

<!-- Describe the current behavior, e.g., »When pressing Alt+j (focus left), the window above the current window is focused.« --> Unable to build i3-wm 4.18.2

Expected Behavior

<!-- Describe the desired behavior you expect after mitigation of the issue, e.g., »The window left next to the current window should be focused.« --> Able to build build i3-wm 4.18.2

Reproduction Instructions

<!-- Please provide detailed instructions on how the bug can be reproduced. E.g., »Open three windows in a V[A H[B C]] layout on a new workspace« -->

  1. Get slackware64 linux 14.2,
  2. pick up requirements from slackbuilds.org, build and install them
  3. get slackbuild for i3wm, change version number
  4. run slackbuild

Environment

<!-- Please include your exact i3 version. Note that we only support the latest major release and the current development version. If you are using an older version of i3, please first update to the current release version and reproduce the issue there. --> Output of i3 --moreversion 2>&-: <pre> Binary i3 version: 4.18.2 (2020-07-26) © 2009 Michael Stapelberg and contributors Running i3 version: 4.18.2 (2020-07-26) (pid 4058)abort…) Loaded i3 config: /home/eleksir/.i3/config (Last modified: Чт 18 июн 2020 01:10:30, 4392040 seconds ago)

The i3 binary you just called: /usr/bin/i3 The i3 binary you are running: i3 </pre> <!-- Please include your (complete) i3 config with which the issue occurs. You can either paste the file directly or provide a link to a service such as pastebin.

If you would like to help debugging the issue, please try to reduce the config such that it is as close to the default config as possible while still reproducing the issue. This can help us bisect the root cause. --> <details><summary>Config file</summary><pre> </pre> </details>

<!-- Providing a logfile can help us trace the root cause of an issue much quicker. You can learn how to generate the logfile here: https://i3wm.org/docs/debugging.html

Providing the logfile is optional. --> <pre> Logfile URL: http://ix.io/2tw2 </pre>

<!-- Please also answer the questions below to help us process your issue faster. If you have any other information to share, please add it here as well. --> <pre>

  • Linux Distribution & Version: Slackware64 Linux 14.2
  • Are you using a compositor (e.g., xcompmgr or compton): yup, but it is irrelevant </pre>

I last version that builds okay is 4.18

closed time in 17 hours

eleksir

issue commenti3/i3

strndup build problem

I’ll mark this as closed since the fix is already in.

Please cherry-pick the fix until the next i3 release, and sorry for the trouble.

eleksir

comment created time in 17 hours

issue commenti3/i3

strndup build problem

This is already fixed in 5ee7690af3416edf1a62acb2cc2a04dc98e1e7fb (so it will be in 4.19), but not in a stable version.

eleksir

comment created time in 17 hours

issue commenti3/i3

Windows start lagging/stuttering after a few minutes unless I regularly switch workspaces.

Does it happen without picom?

Yes, as the report states:

I originally thought this was an issue with my compositor, Picom, but it turns out I can reproduce the issue even without a compositor running.


Can you check with a different window manager that this problem is unrelated to your GPU etc.?

starcraft66

comment created time in 17 hours

pull request commentjacobsa/fuse

Add pid in all fuse ops

If the PR is squash-merged, then this is not relevant. right?

Cleanly separated commits also make the review easier, no matter if they end up in the git history or not :)

xinmeigui-db

comment created time in 17 hours

push eventjacobsa/fuse

Mei Gui

commit sha 36e01f1b3aaafad29231bd7286dac6c384110502

Add pid in all fuse ops (#92) Expose PID as metadata in CreateFile, OpenFile and FlushFile operations This will help us with kahing/goofys#273 Co-authored-by: Sai Teja Suram <pratap130492@gmail.com> Co-authored-by: Sai Teja Suram <pts@avah.dev>

view details

push time in 17 hours

PR merged jacobsa/fuse

Add pid in all fuse ops

Add pid in all fuse ops

+250 -79

4 comments

6 changed files

xinmeigui-db

pr closed time in 17 hours

created tagi3/i3lock

tag2.13

improved screen locker

created time in 2 days

push eventi3/i3lock

Michael Stapelberg

commit sha 4f76d51a3f79082e01b319e9031aad868d89449d

release v2.13

view details

push time in 2 days

push eventi3/i3.github.io

Michael Stapelberg

commit sha df346b1a2364ef5546beaccf5d1ff3d9bf166bfb

add i3lock 2.13

view details

push time in 2 days

CommitCommentEvent
CommitCommentEvent

issue commenti3/i3lock

Crash

Did you already check i3lock with valgrind? There seval potential problems, e.g.:

We do check i3lock from time to time with valgrind, but there are often false positive reports, so it’s tedious.

At a quick look at the source: is there a specific reason why pam_end is called in the callback? Is it guaranteed that the callback gets only called once?

That has already been fixed in https://github.com/i3/i3lock/issues/254, but is not yet released.

I’ll try and get out a new release over the next few days.

markus2330

comment created time in 3 days

issue commentgolang/go

cmd/go: override semantics break e.g. GOFLAGS=-ldflags

That’s fair. I agree that a command-line flag is more direct than an environment variable, and it makes sense that the environment variable supplies defaults which the command-line overrides.


More as a note to myself: the take-away is that GOFLAGS can and should still be set in my build system, but will only help with packages that are built with the go tool directly. For packages that are built indirectly, e.g. via a Makefile, package-specific care must be taken to pass through build-system-wide policy flags (such as the ELF interpreter).


I see that https://golang.org/cmd/go/ currently already documents the override semantics:

GOFLAGS A space-separated list of -flag=value settings to apply to go commands by default, when the given flag is known by the current command. Each entry must be a standalone flag. Because the entries are space-separated, flag values must not contain spaces. Flags listed on the command line are applied after this list and therefore override it.

Still left to do: the “rightmost setting wins” semantics should also be documented here.

stapelberg

comment created time in 3 days

issue commenti3/i3

Bintray (Debian/Ubuntu autobuild repository) is empty

Thanks for flagging this. I have sent https://github.com/i3/i3/pull/4229 which I believe should fix this issue.

smarek

comment created time in 3 days

PR opened i3/i3

travis: fix bintray deploy step

Commit be1065f62d156fb825f566b7e1e6ab4937dc9db4 moved the Debian/Ubuntu package builds into the distbuild/ directory.

fixes #4179

+4 -4

0 comment

2 changed files

pr created time in 3 days

push eventstapelberg/i3

Michael Stapelberg

commit sha 45feaac54c558459ab6396eed3fec0cfcb84a9c3

Introduce GET_BINDING_STATE IPC command fixes #3892

view details

Ingo Bürk

commit sha f4964faef08d6add93afbd24cc00fef0c3f6c72a

Merge pull request #4122 from stapelberg/bindingstate Introduce GET_BINDING_STATE IPC command

view details

Orestis Floros

commit sha 4d9c3131ade445c21d37dc6338a353a6923fc2f6

_con_move_to_con: Traverse below con to check for fullscreen Fixes #4124

view details

Ingo Bürk

commit sha cf09cc790ce39fc7c100407b0c8ee2695209ac51

Merge pull request #4125 from orestisfl/move-fullscreen-to-ws _con_move_to_con: Traverse below con to check for fullscreen

view details

Jason

commit sha 209bef7ea3a7cda14f9d300090a9f698a0b50a45

Update WORKSPACES reply docs

view details

Michael Stapelberg

commit sha 4085c4be3b45ddcb797f7c9026d0af0561d36ef1

the stable version can now be found in the stable branch (#4129) fixes #4127

view details

Jason

commit sha 4e676ef73eeae16dd6c1f09bc2a9d98e8813aae2

i3-msg: add missing option (#4128)

view details

Michael Stapelberg

commit sha b402ce2cc925dfd507ed9484a7d451c37c81159a

binding state post-submit review fixes

view details

Orestis Floros

commit sha 5bc5fc188b8825c1cc2b2a1afa2d1e9f7f4a33c9

Merge pull request #4131 from stapelberg/bindingstate binding state post-submit review fixes

view details

Vasily Fomin

commit sha 24e30a81c36e5ecd8d54a841b154eb94ae01b128

Fixed a typo in the docs

view details

Ingo Bürk

commit sha 51b0583578ecb8ccdb264dca9a0c7f64afca60be

Merge pull request #4133 from VasilyFomin/patch-1 Fixed a typo in the docs

view details

Orestis Floros

commit sha 025743eaf9c993e57c7fdd20127078b835bcd2c0

Fix i3bar Xorg memory leak (#4140) The `*_free_*` calls in this branch: https://github.com/i3/i3/blob/51b0583578ecb8ccdb264dca9a0c7f64afca60be/i3bar/src/xcb.c#L1854 are never called when we remove the output from the tailq during a refresh. Fixes #4123

view details

Orestis Floros

commit sha 874d862507c2f9b5912bd1d76af5e27e8fa3f897

route_click: Fix regression: release mouse Fixes #4143

view details

Ingo Bürk

commit sha a45eb8156c29b7bdb2dfda6451840558cc5e3fa7

Merge pull request #4144 from i3/orestisfl-patch-2 route_click: Fix regression: release mouse

view details

Orestis Floros

commit sha 65e468edd7cac68c569edbc1d4f62c293297c9dc

route_click: Correctly eat the event Restores original intention of 93e96f4 after #4144

view details

Lukas Kern

commit sha 2c1b18b1ac4b48acc81047650a80d668c5a943d6

Fixed a typo in the user's guide

view details

Orestis Floros

commit sha 1449d4890fc36de9b6f04dab3c6cad19d6c16acd

Merge pull request #4146 from lukaskern/next Fixed a typo in the user's guide

view details

Mark Guptill

commit sha 838b600fead202416013db5c1b57f7031f06bed6

kick tray clients before destroying the bar

view details

Orestis Floros

commit sha e67407302738fdc79d9ed1970a60cc580ccf34fe

Merge pull request #4147 from compguy284/kick_tray_clients kick tray clients before destroying the bar

view details

Orestis Floros

commit sha 96639c04106de630efa0eee1662f5f049e4a324d

floating_maybe_reassign_ws: only re-focus if previously focused Fixes #3979

view details

push time in 3 days

pull request commentjacobsa/fuse

Add pid in all fuse ops

Adding the caller process id to all ops sounds okay to me in general, but:

  1. Why do you need to disable writeback caching for the pid to become available?
  2. Your pull request seems to contain a lot of unrelated commits? Can you clean it up please? (rebase)

Thanks,

xinmeigui-db

comment created time in 3 days

push eventstapelberg/configfiles

Michael Stapelberg

commit sha 4cf9c47595c4583f3b682a72430a1c004a8b6a08

emacs: store auto save files alongside backups

view details

push time in 4 days

startedraviqqe/muffet

started time in 5 days

startedmdlayher/keylight

started time in 6 days

push eventdistr1/distri

Michael Stapelberg

commit sha 1cb2853c809c2a451d6c9a7dea4c8ea62ea47d5b

google-chrome: pick up new upstream version done using “distri scaffold -pull google-chrome”

view details

push time in 6 days

startedanatol/luks.go

started time in 8 days

issue commenti3/i3

Create a meson.build file for evaluating meson

That implies that we got this wrong for 4.18.2, too, and nobody complained?

Maybe not big enough a deal to justify another release, given that we want to get 4.19 out next?

stapelberg

comment created time in 9 days

push eventkinx-project/kint

Michael Stapelberg

commit sha 6cff1da9f07e2c1da288c88a283dd889a952d5d3

README: collect debugging/troubleshooting info

view details

push time in 9 days

issue commenti3/i3

Next release (4.18.3 or 4.19)

4.18.3 now released. Re-titling this for 4.19, but that’ll happen over the next few weeks I think

orestisfl

comment created time in 9 days

push eventi3/i3.github.io

Michael Stapelberg

commit sha 1da696092316349201fb18810a42f4b27a0bd630

save docs for 4.18.2

view details

Michael Stapelberg

commit sha aa72f6830d26e65e3ac576d220708f342ac0ec3a

add 4.18.3 release

view details

Michael Stapelberg

commit sha a13661e50f19e00e93b10c980eb7c465d10a6328

update docs for 4.18.3

view details

push time in 9 days

push eventi3/i3

Mark Guptill

commit sha 654e2cefa6effa10170977e6f5b1f065b75e9d3f

kick tray clients before destroying the bar

view details

Michael Stapelberg

commit sha b2ac041ffa6af806678baa434b57e24a8dde38f1

release i3 4.18.3

view details

Michael Stapelberg

commit sha 4da7e318e55dc8c50734fd86684c36647b5185e9

Set non-git version to 4.18.3-non-git.

view details

Michael Stapelberg

commit sha 036903e8b2b9c38dd6e4566638b39e6ca1b6e5b9

Merge branch 'release-4.18.3'

view details

Michael Stapelberg

commit sha e6b4e1314a4516e5794d280ea23b62fb8dd57396

Merge branch 'stable' into next

view details

Michael Stapelberg

commit sha ce01babc7c5df2309f14ff920ec35161d253e735

Update debian/changelog

view details

Michael Stapelberg

commit sha 83078a1e16cac3a9c7dc359d19e7191135b5140c

debian: update changelog

view details

push time in 9 days

push eventi3/i3

Michael Stapelberg

commit sha b2ac041ffa6af806678baa434b57e24a8dde38f1

release i3 4.18.3

view details

Michael Stapelberg

commit sha 4da7e318e55dc8c50734fd86684c36647b5185e9

Set non-git version to 4.18.3-non-git.

view details

Michael Stapelberg

commit sha 036903e8b2b9c38dd6e4566638b39e6ca1b6e5b9

Merge branch 'release-4.18.3'

view details

push time in 9 days

created tagi3/i3

tag4.18.3

A tiling window manager

created time in 9 days

delete branch i3/i3

delete branch : orestisfl-patch-1

delete time in 9 days

push eventi3/i3

Orestis Floros

commit sha eaac9125a9dc9586752cccfc3cc551b0b9e3c2dd

meson: Mark complete-run.pl as executable (#4224) Fixes problem described in https://github.com/i3/i3/issues/4086#issuecomment-711774506

view details

push time in 9 days

PR merged i3/i3

meson: Mark complete-run.pl as executable

This is a bit hacky but I didn't find any other way

Fixes problem described in https://github.com/i3/i3/issues/4086#issuecomment-711774506

+10 -7

0 comment

1 changed file

orestisfl

pr closed time in 9 days

PullRequestReviewEvent

push eventgoogle/renameio

Tom Payne

commit sha ad9e5e50f5274088511afd58a920bf584f0a6b3b

add maybe package (#27)

view details

push time in 9 days

PR merged google/renameio

add maybe package

This is an initial proposal for a helper package as requested so that #26 can be unblocked. The naming of the package is of course contentious, I'm happy to change it to whatever the maintainer prefers.

+30 -0

1 comment

1 changed file

twpayne

pr closed time in 9 days

PullRequestReviewEvent

issue commenti3/i3

Next release (4.18.3 or 4.19)

Great, thanks for confirming. I’ll try to get this out later today.

orestisfl

comment created time in 10 days

issue commenti3/i3

Next release (4.18.3 or 4.19)

So, just to be extra double triple clear: I have cherry-picked commit 838b600fead202416013db5c1b57f7031f06bed6 into the stable branch as the only change on top of our last stable release 4.18.2.

This is all we need to make a useful 4.18.3 release, yes? @orestisfl

orestisfl

comment created time in 10 days

push eventi3/i3

Mark Guptill

commit sha 654e2cefa6effa10170977e6f5b1f065b75e9d3f

kick tray clients before destroying the bar

view details

push time in 10 days

issue commenti3/i3

Next release (4.18.3 or 4.19)

Let’s do a minor bugfix release first, then do a 4.19 in a few weeks with the meson changes.

That way, meson won’t be in the critical path for people who are suffering from the current bugs :)

orestisfl

comment created time in 10 days

issue commenti3/i3

Create a meson.build file for evaluating meson

Ah, good catch! Can you send a PR to fix this please?

stapelberg

comment created time in 10 days

pull request commentstapelberg/airscan

Control duplex setting in airscan1

Thank you!

nenad

comment created time in 10 days

push eventstapelberg/airscan

Nenad Stojanovikj

commit sha f1eaf62bf1edf267294c3c25b6c9c161c3898b43

cmd/airscan1: Support duplex

view details

Nenad Stojanovikj

commit sha d1a9195f8a61ee8761fe9018c684d57086348a93

Document Brother MFC-L2710DN in README

view details

push time in 10 days

PR merged stapelberg/airscan

Control duplex setting in airscan1

This pull request brings support for toggling the duplex setting. My scanner would not use the ADF source when scan:Duplex was set to true and would always revert to using the flatbed. For simplex scanners, this setting should be set to false.

+10 -0

0 comment

2 changed files

nenad

pr closed time in 10 days

PullRequestReviewEvent

delete branch i3/i3

delete branch : orestisfl-patch-1

delete time in 10 days

push eventi3/i3

Orestis Floros

commit sha 71c059d03358cd040a4187df516aa410f5ba3c2a

i3-dmenu-desktop: Manually search for topdir (#4033) Since 3a672bc, using follow or follow_fast, does not set $File::Find::topdir, breaking our deduplication. Fixes #4031

view details

push time in 10 days

PR merged i3/i3

i3-dmenu-desktop: Manually search for topdir in .desktop deduplication

Fixes #4031

+12 -2

2 comments

1 changed file

orestisfl

pr closed time in 10 days

issue closedi3/i3

i3-dmenu-desktop: Nondeterministic .desktop entries ordering (unable to reliably override system/default version)

I'm submitting a…

<pre> [x] Bug [ ] Feature Request [ ] Documentation Request [ ] Other (Please describe in detail) </pre>

Current Behavior

I have two .desktop entries for thunderbird. One is located in /usr/share/applications/thunderbird.desktop (system one), the other is in /home/user/.local/share/applications/thunderbird.desktop (user generated one). I'm using i3-dmenu-desktop to display and run desktop entries.

When displaying an entry for thunderbird or running it, one of those two .desktop entries is being selected in a nondeterministic fashion. Sometimes the user generated one is being used, sometimes the system one.

Expected Behavior

User configuration should take precedence over the global system one, so the /home/user/.local/share/applications/thunderbird.desktop should be used every time.

While it is remotely possible that I've been very lucky for the last few years, I'd say that this change was introduced recently in 3a672bc930237e1b10c81a4c3a8d51eb3c4ae3f5. After rolling it back, everything seems to be working fine again (of course I might be very lucky... again ;)).

Additionally both the commit message and Perl documentation (follow_fast) mention duplicates and both .desktop entries have the same file name.

Reproduction Instructions

Copy any .desktop entry from /usr/share/applications/ to /home/user/.local/share/. Change the value of Name in the .local/share version and invoke i3-dmenu-desktop several times. You will notice that the name will change randomly.

Environment

Output of i3 --moreversion 2>&-: <pre> Binary i3 version: 4.18-85-g0bce0d86 (2020-04-15, branch "next") © 2009 Michael Stapelberg and contributors Running i3 version: 4.18-85-g0bce0d86 (2020-04-15, branch "next") (pid 42362) Loaded i3 config: /home/user/.config/i3/config (Last modified: Tue 21 Apr 2020 13:27:27 CEST, 14286 seconds ago)

The i3 binary you just called: /usr/local/bin/i3 The i3 binary you are running: i3 </pre>

<pre>

  • Linux Distribution & Version: Debian (testing)
  • Are you using a compositor (e.g., xcompmgr or compton): Yes, compton </pre>

edit: Readability improved.

closed time in 10 days

tnnn
PullRequestReviewEvent
PullRequestReviewEvent

PR opened scottprahl/RigolWFM

wfmconvert: add sigrok, add vcsv (full volts)

These two options make it easier to get Rigol Waveform data into the free sigrok logic analyzer, or other tools (vcsv).

fixes #12

+51 -2

0 comment

2 changed files

pr created time in 13 days

create barnchstapelberg/RigolWFM

branch : sigrok

created branch time in 13 days

fork stapelberg/RigolWFM

Parsers .wfm binary files created by a wide range of Rigol oscilloscopes

fork in 13 days

push eventstapelberg/hugo

Michael Stapelberg

commit sha b3d93d0123fd1926f02ae242d279b7e560f74d08

add Nuki Opener bug report

view details

push time in 13 days

push eventstapelberg/stapelberg.github.io

Michael Stapelberg

commit sha 849266a3abadba541f7d5c2425bf82498699668b

update site

view details

push time in 13 days

issue commenti3/i3lock

i3lock slow when verifying my password

@Yekutiel That is entirely unrelated to this issue. Please reach out to your Linux distribution’s support channels to learn how to configure PAM to your wishes. i3lock is merely a user of PAM.

adriencarbonaro

comment created time in 13 days

more