Search Elisp buffers with pcase patterns

Search Elisp buffers with pcase patterns

[Michael Heerdegen]

[-- Attachment #1: Type: text/plain, Size: 299 bytes --]

Hello,

I implemented a buffer search tool for emacs-lisp buffers based on pcase
patterns.  I would like to add it to Gnu Elpa if it's welcome when it is
completed.

Here is the prototype; for instructions see the header.  Comments on the
realization and improvement suggestions are very welcome!



[-- Attachment #2: el-search.el --]
[-- Type: application/emacs-lisp, Size: 20107 bytes --]

[-- Attachment #3: Type: text/plain, Size: 21 bytes --]




Thanks,

Michael.

Re: Search Elisp buffers with pcase patterns

[Stefan Monnier]

> I implemented a buffer search tool for emacs-lisp buffers based on pcase
> patterns.  I would like to add it to Gnu Elpa if it's welcome when it is
> completed.

Cool!  Can you put it up in elpa.git?


        Stefan

Re: Search Elisp buffers with pcase patterns

[Michael Heerdegen]

Stefan Monnier <monnier@iro.umontreal.ca> writes:

> Cool!  Can you put it up in elpa.git?

Done, thanks.

I've released this one quite early because feedback is very valuable in
this case.  The thing can be used beneficially, but expect some raw
edges (or how was that saying?).


Thanks,

Michael.

Re: Search Elisp buffers with pcase patterns

[Michael Heerdegen]

Michael Heerdegen <michael_heerdegen@web.de> writes:

> Done, thanks.

Why doesn't it appear in the package manager?  Is the reason that the
version "0.0.2" starts with a zero?  The README says only a version of
"0" would indicate that a package should not yet be released.


Thanks,

Michael.

Re: Search Elisp buffers with pcase patterns

[Stefan Monnier]

>> Done, thanks.
> Why doesn't it appear in the package manager?  Is the reason that the
> version "0.0.2" starts with a zero?  The README says only a version of
> "0" would indicate that a package should not yet be released.

No, it was the failure to follow the usual formatting conventions
(missing semi-colon on the first line, and missing final line).
M-x checkdoc-current-buffer would have helped.


        Stefan

Re: Search Elisp buffers with pcase patterns

[Michael Heerdegen]

Stefan Monnier <monnier@iro.umontreal.ca> writes:

> M-x checkdoc-current-buffer would have helped.

Indeed.  Sorry, I just forgot that.  Thanks.


Michael.

Re: Search Elisp buffers with pcase patterns

[Ted Zlatanov]

On Sat, 01 Aug 2015 17:57:55 +0200 Michael Heerdegen <michael_heerdegen@web.de> wrote: 

MH> I implemented a buffer search tool for emacs-lisp buffers based on pcase
MH> patterns.  I would like to add it to Gnu Elpa if it's welcome when it is
MH> completed.

MH> Here is the prototype; for instructions see the header.  Comments on the
MH> realization and improvement suggestions are very welcome!

I would love a lot more examples.  This is a terrific tool, thank you
for writing it.

I'd love it to be generalized for other languages. I don't know if you
think that makes sense, but I can imagine doing similar search or
search+replace work against XML, HTML, JSON, or other structured data
formats.

Ted

Re: Search Elisp buffers with pcase patterns

[Michael Heerdegen]

Ted Zlatanov <tzz@lifelogs.com> writes:

> I would love a lot more examples.

You mean examples of usage?  That should be doable.  Any suggestions?

> I'd love it to be generalized for other languages. I don't know if you
> think that makes sense, but I can imagine doing similar search or
> search+replace work against XML, HTML, JSON, or other structured data
> formats.

The package cries for that generalization, but I don't know how to do it
in a reasonable and efficient way.

For Elisp, I can just call `read' to get a suitable syntax tree.  For
other languages, I would also need something like an abstract syntax
tree.

Prior versions of semantic could generate ASTs, but they dismissed it
because it was too slow.  AFAIK I can maximally get something like a
flat tag list from semantic.  That's not enough to expand the thing to
other languages in a reasonable way.

But if someone has an enlightening idea...


Regards,

Michael.

Re: Search Elisp buffers with pcase patterns

[raman]

That would be awesome -- support for other structured data
formats. Another thing I'd love to have would be occur-like filtering
but using pcase patterns - would let one filter down large  data files
and explore efficiently -- 
--