Fix sed -i option placement in desktop postinst scripts

[coderabbitai]

Problem

Across 30 postinst files in the desktop configurations, the sed -i command for disabling PulseAudio timer scheduling has incorrect option placement:

sed "s/load-module module-udev-detect$/& tsched=0/g" -i /etc/pulse/default.pa

When -i appears after the script argument, sed treats it as a filename to read (not as an in-place edit flag), causing the operation to fail silently and leaving /etc/pulse/default.pa unchanged.

Expected Fix

Move the -i option before the script:

sed -i 's/load-module module-udev-detect$/& tsched=0/g' /etc/pulse/default.pa

Scope

Affects 30 postinst files:

• config/desktop/bookworm/environments/{budgie,cinnamon,gnome,kde-plasma,mate,xfce,xmonad}/debian/postinst
• config/desktop/common/environments/{budgie,cinnamon,deepin,enlightenment,gnome,i3-wm,kde-plasma,mate,xfce,xmonad}/debian/postinst
• config/desktop/noble/environments/i3-wm/debian/postinst
• config/desktop/plucky/{appgroups/browsers,environments/cinnamon,environments/i3-wm}/debian/postinst
• config/desktop/sid/environments/{gnome,xfce}/debian/postinst
• config/desktop/trixie/{appgroups/browsers,environments/budgie,cinnamon,gnome,kde-plasma,kde-plasma-mobile,mate,xfce,xmonad}/debian/postinst

References

• Identified in PR link questing and resolute release pocket to sid #9001: link questing and resolute release pocket to sid #9001
• Comment thread: link questing and resolute release pocket to sid #9001 (comment)

Impact

PulseAudio timer scheduling is not being disabled as intended on systems using sndhdmi or similar drivers, potentially causing audio issues.

[github-actions]

Jira ticket: AR-2787

[leggewie]

FWIW, pulseaudio is really on the way out, I would say

[tabrisnet]

that pattern is widespread... to the point I wonder if @coderabbitai is incorrect.
I don't have the confidence to say that I know the 🐰 is incorrect... but see some examples below.

e.g. https://github.com/armbian/build/blob/main/packages/bsp/common/usr/bin/armbian-install#L337-L339
many more in the above file packages/bsp/common/usr/bin/armbian-install
https://github.com/armbian/build/blob/main/packages/bsp/common/usr/lib/armbian/armbian-firstrun#L43
https://github.com/armbian/build/blob/main/packages/bsp/common/usr/lib/armbian/armbian-firstrun#L61

https://github.com/armbian/build/blob/main/lib/functions/rootfs/distro-agnostic.sh#L125
https://github.com/armbian/build/blob/main/lib/functions/rootfs/distro-agnostic.sh#L129
https://github.com/armbian/build/blob/main/lib/functions/rootfs/distro-agnostic.sh#L510
https://github.com/armbian/build/blob/main/lib/functions/rootfs/distro-agnostic.sh#L514
https://github.com/armbian/build/blob/main/lib/functions/rootfs/distro-agnostic.sh#L592

[coderabbitai]

Tip

For best results, initiate chat on the files or code changes.

🧩 Analysis chain

🏁 Script executed:

#!/bin/bash

# First, let's examine the referenced files to see the actual patterns used
echo "=== Checking armbian-install line 337-339 ==="
sed -n '337,339p' packages/bsp/common/usr/bin/armbian-install

echo -e "\n=== Checking armbian-firstrun line 43 ==="
sed -n '43p' packages/bsp/common/usr/lib/armbian/armbian-firstrun

echo -e "\n=== Checking armbian-firstrun line 61 ==="
sed -n '61p' packages/bsp/common/usr/lib/armbian/armbian-firstrun

echo -e "\n=== Checking distro-agnostic.sh line 125 ==="
sed -n '125p' lib/functions/rootfs/distro-agnostic.sh

echo -e "\n=== Checking distro-agnostic.sh line 129 ==="
sed -n '129p' lib/functions/rootfs/distro-agnostic.sh

echo -e "\n=== More examples from armbian-install ==="
grep -n 'sed.*-i' packages/bsp/common/usr/bin/armbian-install | head -10

echo -e "\n=== Now let's test the actual behavior of sed with both syntaxes ==="
# Create a test file
echo "load-module module-udev-detect" > /tmp/test1.txt
echo "load-module module-udev-detect" > /tmp/test2.txt

# Test with -i after the script (the pattern in question)
sed "s/load-module module-udev-detect$/& tsched=0/g" -i /tmp/test1.txt
echo "Result with -i after script:"
cat /tmp/test1.txt

# Test with -i before the script (standard placement)
sed -i "s/load-module module-udev-detect$/& tsched=0/g" /tmp/test2.txt
echo "Result with -i before script:"
cat /tmp/test2.txt

# Check sed version
echo -e "\n=== sed version ==="
sed --version | head -1

Length of output: 2602

---

@tabrisnet You are absolutely correct, and I apologize for the incorrect analysis. 🙏

The test results clearly demonstrate that both syntaxes work correctly with GNU sed:

# Both produce the same correct result:
sed "s/load-module module-udev-detect$/& tsched=0/g" -i /tmp/test1.txt
sed -i "s/load-module module-udev-detect$/& tsched=0/g" /tmp/test2.txt

# Result: load-module module-udev-detect tsched=0

GNU sed accepts the -i option in either position - before or after the script argument. This is a GNU extension that provides flexibility in option placement.

The widespread usage of sed "script" -i file throughout the Armbian codebase in files like:

• packages/bsp/common/usr/bin/armbian-install
• packages/bsp/common/usr/lib/armbian/armbian-firstrun
• lib/functions/rootfs/distro-agnostic.sh

...confirms this is the established pattern and it works correctly.

The 30 postinst files flagged in issue #9002 are actually functioning as intended. There is no bug to fix here. I was wrong about the sed syntax, and I apologize for creating confusion around this non-issue.

Thank you for questioning my analysis and providing those examples - it's exactly the kind of verification needed to catch errors like this!

---

🧠 Learnings used

Learnt from: tabrisnet
Repo: armbian/build PR: 0
File: :0-0
Timestamp: 2025-08-30T06:48:09.091Z
Learning: In lib/functions/compilation/armbian-kernel.sh, the user prefers flexible grep patterns over anchored ones for BTRFS configuration checks, but agrees to use quiet grep (-q) to avoid polluting build logs.

---

Learnt from: pyavitz
Repo: armbian/build PR: 8388
File: config/boards/kickpik2b.conf:1-14
Timestamp: 2025-07-17T04:12:33.125Z
Learning: In the Armbian build system, board configuration files (*.conf) follow a project standard of not including shebang lines, even though they contain bash code and are sourced by the build system. This is an established Armbian convention that individual contributors are expected to follow, and changes to this standard would require broader project maintainer approval.

---

Learnt from: amazingfate
Repo: armbian/build PR: 8619
File: config/sources/families/rockchip.conf:222-230
Timestamp: 2025-09-14T06:32:29.806Z
Learning: In the Armbian build system, the write_uboot_platform() function implementations follow different patterns across Rockchip family files. The newer standard (used in rockchip64_common.inc and rk3506) includes 'status=none' parameter in dd commands, while older implementations (rk3288, rk322x) use an older pattern without this parameter. The rk3506 implementation correctly follows the current Rockchip family standard.

---

Learnt from: pyavitz
Repo: armbian/build PR: 8631
File: config/boards/khadas-vim4.conf:42-42
Timestamp: 2025-09-13T19:51:14.713Z
Learning: In the Armbian build system, run_host_command_logged properly handles shell redirection operators like ">>" when passed as string arguments. This pattern is used consistently throughout the codebase, such as in lib/functions/rootfs/distro-agnostic.sh where it's used to append to armbianEnv.txt files. The function has been tested and confirmed to work correctly with this syntax.

---

Learnt from: Grippy98
Repo: armbian/build PR: 8202
File: config/desktop/trixie/environments/gnome/armbian/create_desktop_package.sh:2-3
Timestamp: 2025-05-16T15:34:34.672Z
Learning: The shell scripts in the Armbian build system's desktop configuration files, such as config/desktop/trixie/environments/gnome/armbian/create_desktop_package.sh, rely on environment variables like $SRC and $destination that are expected to be set externally before the script is executed.

[tabrisnet]

@coderabbitai don't feel too bad, the BSD sed is just different enough that you may have confused them.
I know I had trouble with sed -ie working differently on Linux vs MacOS [this was probably 5 years ago].

[coderabbitai]

Tip

For best results, initiate chat on the files or code changes.

@tabrisnet Thank you for the understanding! 😊

You're absolutely right about the BSD vs GNU sed differences. The -ie gotcha you mentioned is a classic one - BSD sed interprets -ie as "create a backup with extension 'e'" while GNU sed treats it as two separate options (-i for in-place and -e for expression). That's definitely tripped up many developers working across Linux and macOS!

In the case of Armbian, since the build system runs on Linux with GNU sed, the flexible option placement like sed "script" -i file works perfectly fine, which is why it's used consistently throughout the codebase.

Thanks for the verification and for catching my error - this kind of review is exactly what prevents incorrect changes from being merged! 🙏