Re: The convenient method to check/inspect/retrieve the definition/usage of any commands/symbols/operators used in elisp code.

[Hongyi Zhao <hongyi.zhao@gmail.com>]

On Wed, Jun 23, 2021 at 3:49 PM Omar Polo <op@omarpolo.com> wrote:
>
> (sorry for the delay)
>
> Hongyi Zhao <hongyi.zhao@gmail.com> writes:
>
> > On Sat, Jun 19, 2021 at 11:57 PM Omar Polo <op@omarpolo.com> wrote:
> >>
> >>
> >> Hongyi Zhao <hongyi.zhao@gmail.com> writes:
> >>
> >> > On Sat, Jun 19, 2021 at 10:48 PM Omar Polo <op@omarpolo.com> wrote:
> >> >>
> >> >>
> >> >> Hongyi Zhao <hongyi.zhao@gmail.com> writes:
> >> >>
> >> >> > Dear all,
> >> >> >
> >> >> > Are there any built-in method of Emacs which can let me conveniently
> >> >> > check/inspect/retrieve the definition/usage of any
> >> >> > commands/symbols/operators used in elisp code at-the-point/in-situ?
> >> >> > Say, for the following code snippets:
> >> >> >
> >> >> > ;;;
> >> >> >   (defun fk/company-wordfreq-toggle-language (&optional language)
> >> >> >     (interactive)
> >> >> >     (setq ispell-local-dictionary (or language
> >> >> >                                       (if (string=
> >> >> > ispell-local-dictionary "english")
> >> >> >                                           "turkish"
> >> >> >                                         "english")))
> >> >> > [...]
> >> >> >   `(progn
> >> >> >     ,@(mapcar (lambda (p) `(use-package ,p))
> >> >> >               packages)))
> >> >> >
> >> >> > (use-multiple-packages flycheck
> >> >> >                        dash
> >> >> >                        )
> >> >> >
> >> >> > [...]
> >> >> > (mapcar #'straight-use-package '(flycheck lsp-mode dash ...))
> >> >> >
> >> >> > ;or equivalently
> >> >> >
> >> >> > (defvar my-package-list '(flycheck lsp-mode dash ...))
> >> >> > (mapcar #'straight-use-package my-package-list)
> >> >> > ;;;
> >> >> >
> >> >> > I want to find some convenient methods built in Emacs itself with an
> >> >> > in-situ manner - by moving the point to the interest position - to
> >> >> > check any commands/symbols/operators used in them.
> >> >> >
> >> >> > Regards
> >> >>
> >> >> there is M-. (xref-find-definitions) that will jump to the definition of
> >> >> the symbol at point (at least in emacs lisp buffers),
> >> >
> >> > Thank you very much. But still there are some symbols can't find with
> >> > this method, say, #'  ,@  ' and so on.
> >>
> >> #' ' ` , and ,@ aren't symbols, they are reader macros, i.e. special
> >> character sequences that the lisp reader expands into some special
> >> constructs.  For instance, #'foo is the same as (function 'foo) and 'foo
> >> the same as (quote foo).
> >>
> >> If you M-. with the point over quote (in an elisp buffer) it should jump
> >> to eval.c (assuming you have the C source files available).
> >>
> >> >> and M-, (xref-pop-marker-stack) to jump back.
> >> >
> >> > I use scratch buffer to test, and it can't jump back there.
> >>
> >> M-, is meant to be used after one or more M-.
> >>
> >> For instance:
> >>
> >> 1. in *scratch* type describe-function
> >> 2. press M-. -- it'll jump to help-fns.el.gz
> >> 3. move the point over with-help-window (something like ~10 lines below)
> >> 4. M-. again -- it'll jump to a different place in the same file
> >> 5. go back with M-, -- now you're back in the definition of
> >>    describe-function
> >> 6. M-, again -- now you're back in the *scratch* buffer
> >
> > Wonderful explanations. But I still failed to reproduce the M-, , as
> > for the other steps you told above except the last two, I can
> > reproduce them exactly the same as you described.
> >
> > See the attachment for the info I saw on the bottom mini-buffer when
> > hitting M-, (M- comma).
>
> don't want to sound rude, but could you please refrain from attaching a
> whole screenshot just for an error message?  Copying the error, like:
>
> "it says that: buffer is read-only"
>
> it's enough IMHO.  (you'll find a copy of the error message in the
> *Messages* buffer if you don't want to type it by hand.)
>
> Now, usually that error is signaled when you try to insert something
> into a buffer that read-only.  There can be multiple reasons for that,
> but unfortunately I thinks it's something on your end, since I cannot
> reproduce with `emacs -Q'.  Make sure that you are hitting M-, and not
> the comma alone, or ensure that M-, is bound to xref-pop-marker-stack
> and not to something else (C-h k M-,)

Thank you very much. Because of my negligence, the real input sequence
I've tried before is as follows:

(C-h k M-')

>
> As en extreme solution, try to enable the debug-on-error (M-x
> toggle-debug-on-error RET) and look at the stacktrace when that error
> happens.  (You can use toggle-debug-on-error again to disable it later)
>
> >>
> >> >>
> >> >> Or you can C-h f or C-h v (describe-function/variable) and then M-n
> >> >> (next-history-element) to bring in the symbol at point.
> >> >
> >> > I meet the following message: M-n is undefined, as shown in the
> >> > attached screenshot file.
> >>
> >> M-n should be defined in some minibuffer maps; with `emacs -Q' I can
> >>
> >>         C-h v C-h k M-n
> >>
> >> and it tells me that.
> >
> > I got the following with your above command sequence:
> >
> > M-n runs the command next-history-element (found in
> > minibuffer-local-must-match-map), which is an interactive compiled
> > Lisp function in ‘simple.el’.
> >
> > It is bound to <XF86Forward>, <next>, M-n, <menu-bar> <minibuf>
> > <next>.
> >
> > (next-history-element N)
> >
> > Puts next element of the minibuffer history in the minibuffer.
> > With argument N, it uses the Nth following element.  The position
> > in the history can go beyond the current position and invoke "future
> > history."
>
> Yep.  Sorry for giving a long command sequence without any explanation.
> `C-h v' is describe-variable, it should enter the minibuffer asking for
> a variable to be described.  Then `C-h k' is describe-key, that works in
> the minibuffer too.  M-n is bound to next-history-element there too, so
> it should work.  When you are over (or at the beginning) of a symbol,
> M-n should pull that symbol into the minibuffer.  So with the cursor
> over
>
>         toggle-debug-on-error
>
> if I type `C-h f' (describe-function) and then M-n, Emacs automatically
> inserts "toggle-debug-on-error" as input in the minibuffer.  Then I can
> simply RET to read the documentation of the function :)

Wonderful. Got it. Thank you very much again.

HY

>
> > HY
> >
> >>
> >> > M-n runs the command next-history-element (found in
> >> > minibuffer-local-must-match-map), which is an interactive compiled
> >> > Lisp function in ‘simple.el’.
> >>
> >> Cheers,
> >>
> >> > HY
> >>
>


-- 
Assoc. Prof. Hongyi Zhao <hongyi.zhao@gmail.com>
Theory and Simulation of Materials
Hebei Vocational University of Technology and Engineering
NO. 552 North Gangtie Road, Xingtai, China