How to change the behavior of quitting cscope display buffer

[wanghaiqiangk]

When press q, the cscope display buffer is closed and replaced with a second source code buffer. Therefore, there are two buffers for the source code and I have to manually kill one of them.

I hope, when press q, the cscope display buffer can be completely closed and return the focus to source code buffer. I think this can be customized by modifying cscope-display-buffer-args, do you have any idea about how to change and achieve this feature?

[dkogan]

Hi. First of all, `cscope-display-buffer-args` controls how the *cscope* buffer pops up, not how it disappears. So this variable doesn't do anything here. Currently 'q' effectively calls (bury-buffer), which produces the behavior you describe. I just ran a few quick experiments, and in all cases, you end up with two identical source windows visible. I tried (quit-window) and I tried to just kill the *cscope* buffer. What EXACTLY do you think 'q' should do? And better yet, can you point to a built-in emacs function that has that behavior?

[wanghaiqiangk]

I managed to figure out how and that's what you said.

I changed (bury-buffer) to (quit-window) in cscope-bury-buffer. Since quit-window also evokes bury-buffer, the *cscope* is the least candidate for switching buffer. To bring *cscope* to the first one, maybe a simple (delete-window) can work.

I prefer the behaviour of killing the window. Because I want to focus on the source code window. After finishing the use of cscope, I don't want to invoke extra typing to just close its buffer/window. But maybe some people prefer the original one. Thereby, an option to control such behaviour can be a choice.

I created a PR and please check it.

[dkogan]

So to be clear: (quit-window) does what you want and (bury-window) doesn't?

[wanghaiqiangk]

That's it.

[dkogan]

Fixed in 8e441ef