Use Vertico + consult to do the search for an exact word.

Use Vertico + consult to do the search for an exact word.

[Hongyi Zhao]

[-- Attachment #1: Type: text/plain, Size: 702 bytes --]

Hi here,

In Emacs, I use Vertico + consult to do the in-buffer and minibuffer
search and bind
bound the command `consult-line' to M-s l.

What puzzles me is that this cannot let me search for an exact word.
For example, if I want to sear all the occurrences of the word `id',
the result obtained is shown in the attached file, which obviously is
not what I want.

How to achieve my goal based on Vertico + consult for utilizing the
more intuitive minibuffer display?

Regards,
Zhao
-- 
Assoc. Prof. Hongsheng Zhao <hongyi.zhao@gmail.com>
Theory and Simulation of Materials
Hebei Vocational University of Technology and Engineering
No. 473, Quannan West Street, Xindu District, Xingtai, Hebei province

[-- Attachment #2: image.png --]
[-- Type: image/png, Size: 171979 bytes --]

Re: Use Vertico + consult to do the search for an exact word.

[Tassilo Horn]

Hongyi Zhao <hongyi.zhao@gmail.com> writes:

> In Emacs, I use Vertico + consult to do the in-buffer and minibuffer
> search and bind bound the command `consult-line' to M-s l.
>
> What puzzles me is that this cannot let me search for an exact word.
> For example, if I want to sear all the occurrences of the word `id',
> the result obtained is shown in the attached file, which obviously is
> not what I want.
>
> How to achieve my goal based on Vertico + consult for utilizing the
> more intuitive minibuffer display?

I don't use consult myself but I guess the search string is a regular
expression.  So when you want to search for an exact word, use \bid\b to
find all occurrences of the word id.

See (info "(elisp) Regexp Backslash") for details:

--8<---------------cut here---------------start------------->8---
‘\b’
     matches the empty string, but only at the beginning or end of a
     word.  Thus, ‘\bfoo\b’ matches any occurrence of ‘foo’ as a
     separate word.  ‘\bballs?\b’ matches ‘ball’ or ‘balls’ as a
     separate word.

     ‘\b’ matches at the beginning or end of the buffer (or string)
     regardless of what text appears next to it.
--8<---------------cut here---------------end--------------->8---

HTH,
Tassilo

Re: Use Vertico + consult to do the search for an exact word.

[Hongyi Zhao]

[-- Attachment #1: Type: text/plain, Size: 1682 bytes --]

On Thu, Jun 1, 2023 at 2:54 PM Tassilo Horn <tsdh@gnu.org> wrote:
>
> Hongyi Zhao <hongyi.zhao@gmail.com> writes:
>
> > In Emacs, I use Vertico + consult to do the in-buffer and minibuffer
> > search and bind bound the command `consult-line' to M-s l.
> >
> > What puzzles me is that this cannot let me search for an exact word.
> > For example, if I want to sear all the occurrences of the word `id',
> > the result obtained is shown in the attached file, which obviously is
> > not what I want.
> >
> > How to achieve my goal based on Vertico + consult for utilizing the
> > more intuitive minibuffer display?
>
> I don't use consult myself but I guess the search string is a regular
> expression.  So when you want to search for an exact word, use \bid\b to
> find all occurrences of the word id.

I've tried this trick, but it cannot match the results like `$id$', as
shown in the attached screenshot, from where you can see that the
entry indexed by the circle numbered 1 doesn't appear in the
minibuffer's matching list.

> See (info "(elisp) Regexp Backslash") for details:
>
> --8<---------------cut here---------------start------------->8---
> ‘\b’
>      matches the empty string, but only at the beginning or end of a
>      word.  Thus, ‘\bfoo\b’ matches any occurrence of ‘foo’ as a
>      separate word.  ‘\bballs?\b’ matches ‘ball’ or ‘balls’ as a
>      separate word.
>
>      ‘\b’ matches at the beginning or end of the buffer (or string)
>      regardless of what text appears next to it.
> --8<---------------cut here---------------end--------------->8---
>
> HTH,
> Tassilo

Regards,
Zhao

[-- Attachment #2: image.png --]
[-- Type: image/png, Size: 310052 bytes --]

Re: Use Vertico + consult to do the search for an exact word.

[Tassilo Horn]

Hongyi Zhao <hongyi.zhao@gmail.com> writes:

>> I don't use consult myself but I guess the search string is a regular
>> expression.  So when you want to search for an exact word, use \bid\b
>> to find all occurrences of the word id.
>
> I've tried this trick, but it cannot match the results like `$id$', as
> shown in the attached screenshot, from where you can see that the
> entry indexed by the circle numbered 1 doesn't appear in the
> minibuffer's matching list.

That's because $ is a word-character in the syntax-table of the mode you
are using in that buffer.  You can try
\([^a-zA-Z]\|\b\)id\([^a-zA-Z]\|\b\) which should also find $id$ and
1id02.

Bye,
Tassilo

Re: Use Vertico + consult to do the search for an exact word.

[Hongyi Zhao]

[-- Attachment #1: Type: text/plain, Size: 972 bytes --]

On Thu, Jun 1, 2023 at 5:52 PM Tassilo Horn <tsdh@gnu.org> wrote:
>
> Hongyi Zhao <hongyi.zhao@gmail.com> writes:
>
> >> I don't use consult myself but I guess the search string is a regular
> >> expression.  So when you want to search for an exact word, use \bid\b
> >> to find all occurrences of the word id.
> >
> > I've tried this trick, but it cannot match the results like `$id$', as
> > shown in the attached screenshot, from where you can see that the
> > entry indexed by the circle numbered 1 doesn't appear in the
> > minibuffer's matching list.
>
> That's because $ is a word-character in the syntax-table of the mode you
> are using in that buffer.  You can try
> \([^a-zA-Z]\|\b\)id\([^a-zA-Z]\|\b\) which should also find $id$ and
> 1id02.

Yes. This works, as shown in the attached screenshot. But the input is
rather cumbersome. Are there more efficient/concise regexp patterns
for this goal?

> Bye,
> Tassilo

Regards,
Zhao

[-- Attachment #2: image.png --]
[-- Type: image/png, Size: 220060 bytes --]

Re: Use Vertico + consult to do the search for an exact word.

[Tassilo Horn]

Hongyi Zhao <hongyi.zhao@gmail.com> writes:

>> That's because $ is a word-character in the syntax-table of the mode
>> you are using in that buffer.  You can try
>> \([^a-zA-Z]\|\b\)id\([^a-zA-Z]\|\b\) which should also find $id$ and
>> 1id02.
>
> Yes. This works, as shown in the attached screenshot. But the input is
> rather cumbersome. Are there more efficient/concise regexp patterns
> for this goal?

Well, \bid\b should be enough, actually.  I don't understand why it
seems like the $s in $id$ in your latex buffer have word syntax.  I've
just tried both the stock latex-mode and AUCTeX and with both the $s
don't have word syntax.

Could you please put point on such a $ and report the output of M-x
describe-char RET?  I'm mostly interested in the line

  syntax: $ 	which means: math

which might be

  syntax: w 	which means: word

on your side for whatever reason.

Also, please try if regexp isearch works as expected, i.e., C-u C-s
\bid\b should also find the id in $id$.  If you both have $ in math
syntax and isearch works but consult doesn't, then I'd ask at the
consult community.

Bye,
Tassilo

Re: Use Vertico + consult to do the search for an exact word.

[Hongyi Zhao]

[-- Attachment #1: Type: text/plain, Size: 2364 bytes --]

On Fri, Jun 2, 2023 at 2:56 AM Tassilo Horn <tsdh@gnu.org> wrote:
>
> Hongyi Zhao <hongyi.zhao@gmail.com> writes:
>
> >> That's because $ is a word-character in the syntax-table of the mode
> >> you are using in that buffer.  You can try
> >> \([^a-zA-Z]\|\b\)id\([^a-zA-Z]\|\b\) which should also find $id$ and
> >> 1id02.
> >
> > Yes. This works, as shown in the attached screenshot. But the input is
> > rather cumbersome. Are there more efficient/concise regexp patterns
> > for this goal?
>
> Well, \bid\b should be enough, actually.  I don't understand why it
> seems like the $s in $id$ in your latex buffer have word syntax.  I've
> just tried both the stock latex-mode and AUCTeX and with both the $s
> don't have word syntax.
>
> Could you please put point on such a $ and report the output of M-x
> describe-char RET?  I'm mostly interested in the line
>
>   syntax: $     which means: math
>
> which might be
>
>   syntax: w     which means: word
>
> on your side for whatever reason.

See below:

;;; output begin here
             position: 14582 of 24975 (58%), column: 6
            character: $ (displayed as $) (codepoint 36, #o44, #x24)
              charset: ascii (ASCII (ISO646 IRV))
code point in charset: 0x24
               script: latin
               syntax: $     which means: math
             category: .:Base, a:ASCII, l:Latin, r:Roman
             to input: type "C-x 8 RET 24" or "C-x 8 RET DOLLAR SIGN"
          buffer code: #x24
            file code: #x24 (encoded by coding system utf-8-unix)
              display: by this font (glyph code):
    ftcrhb:-PfEd-DejaVuSansMono Nerd Font
Mono-regular-normal-normal-*-20-*-*-*-m-0-iso10646-1 (#x07)

Character code properties: customize what to show
  name: DOLLAR SIGN
  general-category: Sc (Symbol, Currency)
  decomposition: (36) ('$')

There are text properties here:
  face                 font-latex-math-face
  fontified            t
  wrap-prefix          " "
;;; output end here

> Also, please try if regexp isearch works as expected, i.e., C-u C-s
> \bid\b should also find the id in $id$.  If you both have $ in math
> syntax and isearch works but consult doesn't, then I'd ask at the
> consult community.

This method works, as shown in the attached screenshot.

> Bye,
> Tassilo

Regards,
Zhao

[-- Attachment #2: image.png --]
[-- Type: image/png, Size: 53660 bytes --]