Defuns in comint

Defuns in comint

[Daniel Colascione]

[-- Attachment #1: Type: text/plain, Size: 437 bytes --]

How would people feel about defining defuns in comint mode as subprocesses outputs? This way, the same navigation keys people already have become more useful in comint mode without breaking anything, because defun navigation in comint is useless today.

If not defuns, then paragraphs? Sentences? I feel like there should be *some* way of using existing sets of movement and marking commands to talk about outputs, and probably inputs.

[-- Attachment #2: Type: text/html, Size: 504 bytes --]

Re: Defuns in comint

[Jim Porter]

On 10/3/2024 4:04 PM, Daniel Colascione wrote:
> How would people feel about defining defuns in comint mode as 
> subprocesses outputs? This way, the same navigation keys people already 
> have become more useful in comint mode without breaking anything, 
> because defun navigation in comint is useless today.
> 
> If not defuns, then paragraphs? Sentences? I feel like there should be 
> *some* way of using existing sets of movement and marking commands to 
> talk about outputs, and probably inputs.

You might be interested in lisp/eshell/em-prompt.el, which remaps 
forward/backward-paragraph so that they stop at each empty line *or* prompt.

Re: Defuns in comint

[Daniel Colascione]

On October 3, 2024 4:32:14 PM PDT, Jim Porter <jporterbugs@gmail.com> wrote:
>On 10/3/2024 4:04 PM, Daniel Colascione wrote:
>> How would people feel about defining defuns in comint mode as subprocesses outputs? This way, the same navigation keys people already have become more useful in comint mode without breaking anything, because defun navigation in comint is useless today.
>> 
>> If not defuns, then paragraphs? Sentences? I feel like there should be *some* way of using existing sets of movement and marking commands to talk about outputs, and probably inputs.
>
>You might be interested in lisp/eshell/em-prompt.el, which remaps forward/backward-paragraph so that they stop at each empty line *or* prompt.

Sure, but 1) I don't want to stop at paragraph boundaries, and 2) remapping movement commands is less powerful and consistent than changing the definitions of those movements, which is why we have beginning-of-X-function variables.

Re: Defuns in comint

[Yuri Khan]

On Fri, 4 Oct 2024 at 06:05, Daniel Colascione <dancol@dancol.org> wrote:
>
> How would people feel about defining defuns in comint mode as subprocesses outputs? This way, the same navigation keys people already have become more useful in comint mode without breaking anything, because defun navigation in comint is useless today.
>
> If not defuns, then paragraphs? Sentences? I feel like there should be *some* way of using existing sets of movement and marking commands to talk about outputs, and probably inputs.

Maybe pages? (Might even work out of the box if the prompt contains ^L)

Re: Defuns in comint

[Daniel Colascione]

On October 3, 2024 9:40:34 PM PDT, Yuri Khan <yuri.v.khan@gmail.com> wrote:
>On Fri, 4 Oct 2024 at 06:05, Daniel Colascione <dancol@dancol.org> wrote:
>>
>> How would people feel about defining defuns in comint mode as subprocesses outputs? This way, the same navigation keys people already have become more useful in comint mode without breaking anything, because defun navigation in comint is useless today.
>>
>> If not defuns, then paragraphs? Sentences? I feel like there should be *some* way of using existing sets of movement and marking commands to talk about outputs, and probably inputs.
>
>Maybe pages? (Might even work out of the box if the prompt contains ^L)

We'd need to add forward and backward page function hooks. Right now, page boundaries are still regexp-only, as many things were in days of yore.