[BUG] window mode alignment issue with i3 scratchpad windows

[infokiller]

Version

Output of rofi -v: Version: 1.6.0

Configuration

Output of rofi -help (in a gist, please paste the full output): https://gist.github.com/infokiller/c27758987dc51c929aff2cd49c1d4d49

Launch Command

The commandline used to launch **rofi**: rofi -no-config -show window

Steps to reproduce

• Install i3
• Create a window and send it to the scratchpad
• Invoke rofi -no-config -show window

What behaviour you see

The columns of lines of scratchpad windows are not aligned:

What behaviour you expect to see

• Workspace names are aligned

Workaround

Using this setting in my xresources file fixes the issue, though it also adds a lot of space after the desktop name in the entries:

rofi.window-format:               {w}\011{c} {t}

Note that \011 is interpreted as a tab character by xrdb [1].

[1] https://jlk.fjfi.cvut.cz/arch/manpages/man/XrmGetDatabase.3#FILE_SYNTAX

[DaveDavenport]

There is already a working solution for this, in window format you can set the length of each element, see manpage.

This is what I use (in my config.rasi):

window-format: "{w:20} {c:-40} {t}";

It will pad w to 20 char width.

[infokiller]

Using rofi -window-format '{w:20} {c:-40} {t}' -show window still results in the alignment issue above. Am I missing something?

[DaveDavenport]

hmm under i3 I get no alignment issue for the workspace .. (and I have scratchpad windows)

[DaveDavenport]

do you use monospace font? (I assume so, it called mono :D)

[infokiller]

Haha yup, it should be monospace. One thing that is odd in my setup is that the workspace names have unicode zero width chars in them (it's an implementation detail of i3-workspace-groups). For example, a workspace can be named '1:\u200bgroup\u200b\u200b:1'. If you execute printf '1:\u200bgroup\u200b\u200b:1' in bash/zsh you'll notice that it only shows as 1:group:1 because the unicode zero width char is not rendered. However, the scratchpad workspace name is not modified in any way. Could that be the issue?

[DaveDavenport]

Yeah, that could be the issue. It uses 'g_utf8_strlen' to count chars, and that will most likely count them and screw up alignment.

Not sure how to solve this.

[github-actions[bot]]

This issue has been automatically locked since there has not been any recent activity after it was closed. Please open a new issue for related bugs.