Background color gets broken on large command output #14248

Closed
opened 2026-01-31 04:05:00 +00:00 by claunia · 6 comments
Owner

Originally created by @sophec on GitHub (Jun 17, 2021).

Windows Terminal version (or Windows build number)

1.8.1521.0

Other Software

WSL2 + Void Linux
zsh shell (5.8)
Seems most frequent when using Neovim (not during, but maybe after?), GCC, clang, etc.

My configs for zsh, nvim, vim, etc. can be found here. My WT settings are attached (zipped because GitHub; note that I am almost always using my Void config, I don't use the others frequently enough to know if this happens there).

settings.zip

Steps to reproduce

Simply run a command which produces a lot of output. It's not consistent, but I can reproduce it frequently when a program's output is very long and all comes out at once (or extremely quickly, anyway). This often includes things like C++ compilers giving lots of notes, my own programs dumping a lot of debug logs (at the moment, a C++ program parsing a large JSON file and printing the parsed data), etc.

Expected Behavior

The program's output should just be there as you'd normally expect:
image
I have added the output line count for reference, though it's likely not super useful here.

Actual Behavior

Ironically I'm struggling to reproduce it at the moment, but what happens is that the background color after a newline character (i.e. at the end of each line, all the way to the right of the window, but not during the line's content itself) changes to a lighter one. I will see if I can figure out what actual steps seem to make this happen the most. Additionally, resizing the window seems to fix it (although it causes other issues with my shell's output, but that seems likely to be more related to the shell, or at the very least out of the scope of this issue). I normally use one window maximized (not fullscreen) with multiple tabs.

I will update with a screenshot the next time it happens (which should be about 10 seconds after I click "Submit", knowing how software works).

Originally created by @sophec on GitHub (Jun 17, 2021). ### Windows Terminal version (or Windows build number) 1.8.1521.0 ### Other Software WSL2 + Void Linux zsh shell (5.8) Seems most frequent when using Neovim (not during, but maybe after?), GCC, clang, etc. My configs for zsh, nvim, vim, etc. can be found [here](https://github.com/willeccles/dotfiles). My WT settings are attached (zipped because GitHub; note that I am almost always using my Void config, I don't use the others frequently enough to know if this happens there). [settings.zip](https://github.com/microsoft/terminal/files/6670632/settings.zip) ### Steps to reproduce Simply run a command which produces a lot of output. It's not consistent, but I can reproduce it frequently when a program's output is very long and all comes out at once (or extremely quickly, anyway). This often includes things like C++ compilers giving lots of notes, my own programs dumping a lot of debug logs (at the moment, a C++ program parsing a large JSON file and printing the parsed data), etc. ### Expected Behavior The program's output should just be there as you'd normally expect: ![image](https://user-images.githubusercontent.com/3360049/122406159-a0605a80-cf4e-11eb-9223-c170d3603a13.png) I have added the output line count for reference, though it's likely not super useful here. ### Actual Behavior Ironically I'm struggling to reproduce it at the moment, but what happens is that the background color after a newline character (i.e. at the end of each line, all the way to the right of the window, but not during the line's content itself) changes to a lighter one. I will see if I can figure out what actual steps seem to make this happen the most. Additionally, resizing the window seems to fix it (although it causes other issues with my shell's output, but that seems likely to be more related to the shell, or at the very least out of the scope of this issue). I normally use one window maximized (not fullscreen) with multiple tabs. I will update with a screenshot the next time it happens (which should be about 10 seconds after I click "Submit", knowing how software works).
claunia added the Needs-TriageNeeds-AttentionResolution-Duplicate labels 2026-01-31 04:05:01 +00:00
Author
Owner

@DHowett commented on GitHub (Jun 17, 2021):

Let us know if you can reproduce it and what led to that state! Some of this is likely by design -- the terminal is supposed to take on the background color that was present at the end of the line at the bottom of the viewport, but it may not do so consistently.

@DHowett commented on GitHub (Jun 17, 2021): Let us know if you can reproduce it and what led to that state! Some of this is likely by design -- the terminal is supposed to take on the background color that was present at the end of the line at the bottom of the viewport, but it may not do so consistently.
Author
Owner

@sophec commented on GitHub (Jun 18, 2021):

Absolutely will! I'll see if I can't get it to happen tomorrow--had to work on other stuff today, so I didn't get to do as much testing as I wanted.

@sophec commented on GitHub (Jun 18, 2021): Absolutely will! I'll see if I can't get it to happen tomorrow--had to work on other stuff today, so I didn't get to do as much testing as I wanted.
Author
Owner

@j4james commented on GitHub (Jun 20, 2021):

This might be related to #8823. That issue also had the background color changing on end of the line (and it doesn't happen all the time).

@j4james commented on GitHub (Jun 20, 2021): This might be related to #8823. That issue also had the background color changing on end of the line (and it doesn't happen all the time).
Author
Owner

@sophec commented on GitHub (Jun 21, 2021):

Yes, I think that this is most definitely related. That seems to be about the behavior I'm seeing.

@sophec commented on GitHub (Jun 21, 2021): Yes, I think that this is most definitely related. That seems to be about the behavior I'm seeing.
Author
Owner

@sophec commented on GitHub (Jun 22, 2021):

I have been unable to reproduce the issue (unsurprisingly, because nothing ever goes wrong when you need it to), and I am almost certain that #8823 is the same issue. I will close this issue for now, and if I can find some evidence to suggest they are different issues, I will reopen. For the time being, it's probably best to avoid the duplicate issue.

@sophec commented on GitHub (Jun 22, 2021): I have been unable to reproduce the issue (unsurprisingly, because nothing ever goes wrong when you need it to), and I am almost certain that #8823 is the same issue. I will close this issue for now, and if I can find some evidence to suggest they are different issues, I will reopen. For the time being, it's probably best to avoid the duplicate issue.
Author
Owner

@DHowett commented on GitHub (Jun 22, 2021):

Thanks for following up 😄 If it ends up being different, feel free to reopen or open anew!

@DHowett commented on GitHub (Jun 22, 2021): Thanks for following up :smile: If it ends up being different, feel free to reopen or open anew!
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: starred/terminal#14248