unofficial mirror of emacs-devel@gnu.org 
 help / color / mirror / code / Atom feed
From: "T.V Raman" <raman@google.com>
To: Naoya Yamashita <conao3@gmail.com>
Cc: monnier@iro.umontreal.ca,  caiohcs0@gmail.com,
	 stefankangas@gmail.com, emacs-devel@gnu.org
Subject: Re: Include leaf in Emacs distribution
Date: Fri, 23 Oct 2020 11:11:23 -0700	[thread overview]
Message-ID: <p913624vmlw.fsf@google.com> (raw)
In-Reply-To: <20201024.005302.2225963876712457330.conao3@gmail.com> (Naoya Yamashita's message of "Sat, 24 Oct 2020 00:53:02 +0900 (JST)")

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset=gb18030, Size: 5743 bytes --]

Naoya Yamashita <conao3@gmail.com> writes:
Another nice to have would be to rationalize lef/use-package vs the
custom file.

Custom-file: One file that holds all settings.

Leaf/use-package: refactors package-specific settings into
package-specific declarations

It would be nice to be able to look in one place (either
use-package/leaf or custom) rather than having to hunt down some rogue
setting in multiple places.

If leaf/use-package is built in to emacs, one possibility might be to
recommend  custom exclusively for   setting emacs builtin settings
e.g.find-file-hook -- but that "builtin definition" can become tenuous
fast.

>> I also hope we can use it to make init files "compatible" with flymake.
>> IOW make it so byte-compiling your init file doesn't result in lots and
>> lots of "spurious" warnings.
>
> I believe john say same thing, leaf and use-package have the
> feature to generate `defvar` and `declare-function` to prevent
> warnings in byte compiling.  So my init.el also uses flycheck,
> but there are no warnings.  References to undefined
> functions/variables are shown as usual, and the warnings are very
> useful.
>
> ```
> (macroexpand-1
>  '(leaf server
>     :doc "Lisp code for GNU Emacs running as server process"
>     :defvar server-temp-file-regexp
>     :defun server-running-p
>     :config
>     (setq server-temp-file-regexp
>           (rx-to-string `(or (regexp ,server-temp-file-regexp) ".DS_Store") t))
>     (unless (server-running-p)
>       (server-start))))
> ;;=> (prog1 'server
> ;;     (declare-function server-running-p "server")
> ;;     (defvar server-temp-file-regexp)
> ;;     (setq server-temp-file-regexp
> ;;           (rx-to-string `(or (regexp ,server-temp-file-regexp) ".DS_Store") t))
> ;;     (unless (server-running-p)
> ;;       (server-start)))
>
>
> (let ((byte-compile-current-file "bar"))
>   (macroexpand-1
>    '(use-package server
>       :defines server-temp-file-regexp
>       :functions server-running-p
>       :config
>       (setq server-temp-file-regexp
>             (rx-to-string `(or (regexp ,server-temp-file-regexp) ".DS_Store") t))
>       (unless (server-running-p)
>         (server-start)))))
> ;;=> (progn
> ;;     (eval-and-compile
> ;;       (defvar server-temp-file-regexp)
> ;;       (declare-function server-running-p "server")
> ;;       (eval-when-compile
> ;;         (with-demoted-errors "Cannot load server: %S"
> ;;           nil
> ;;           (unless (featurep 'server)
> ;;             (load "server" nil t)))))
> ;;     (require 'server nil nil)
> ;;     (setq server-temp-file-regexp
> ;;           (rx-to-string `(or (regexp ,server-temp-file-regexp) ".DS_Store") t))
> ;;     (unless (server-running-p)
> ;;       (server-start))
> ;;     t)
> ```
>
> I've also compared leaf and use-package.
>
> - The leaf has document keywords such as :doc.  The advantage of
>   this is that the text is easier to recognize than the comments
>   due to the color scheme.  In addition, other packages can use
>   the strings specified by this keyword.
>
> - The :defines of use-package becomes a :defvar in the leaf.
>   It's very analogous to defvar, considering that defvar prevents
>   warnings on variable references.
>
> - A :function of use-package is a :defun in the leaf.  It is a
>   function version of :defvar to suppress function warnings.
>   There is a possibility of misunderstanding because there're no
>   actual `defun`, but it's an acceptable confusion compared to
>   the disadvantage of not knowing whether `:defines` are
>   variables or functions.
>
> - use-package has an expression that is secretly output only
>   during byte-compilation.  In this case, if
>   `byte-compile-current-file` is nil, then `:defines` and
>   `:functions` will not output any expressions.  This makes
>   debugging difficult; leaf does not have this feature. It's
>   consistent.
>
> - use-package will add a `t` to the use-package's own return
>   value, as shown in this example, to make the overall return
>   value to `t`, while sometimes returning a set right-hand side
>   value.  This is a confusion, and the leaf solves it; the return
>   value returned by the leaf is always a symbol of the first
>   argument.
>
> ###
>
> Sample init.el I tested.  We (I and john) want to remove this
> bootstrap code.
>
> ```
> ;;; init.el --- Sample clean init.el  -*- lexical-binding: t; -*-
>
> ;; ~/.debug.emacs.d/leaf-byte-compile/init.el
>
> ;; you can run like 'emacs -q -l ~/.debug.emacs.d/leaf-byte-compile/init.el'
> (when load-file-name
>   (setq user-emacs-directory
>         (expand-file-name (file-name-directory load-file-name))))
>
> (eval-when-compile
>   (setq user-emacs-directory
>         (expand-file-name (file-name-directory default-directory))))
>
> (eval-and-compile
>   (prog1 "leaf"
>     (custom-set-variables
>      '(package-archives '(("org"   . "https://orgmode.org/elpa/")
>                           ("melpa" . "https://melpa.org/packages/")
>                           ("gnu"   . "https://elpa.gnu.org/packages/"))))
>     (package-initialize)
>     (unless (package-installed-p 'leaf)
>       (package-refresh-contents)
>       (package-install 'leaf))))
>
> ;; ---
>
> (leaf server
>   :doc "Lisp code for GNU Emacs running as server process"
>   :defvar server-temp-file-regexp
>   :defun server-running-p
>   :config
>   (setq server-temp-file-regexp
>         (rx-to-string `(or (regexp ,server-temp-file-regexp) ".DS_Store") t))
>   (unless (server-running-p)
>     (server-start)))
> ```
>

-- 

Thanks,

--Raman
7©4 Id: kg:/m/0285kf1  •0Ü8



  parent reply	other threads:[~2020-10-23 18:11 UTC|newest]

Thread overview: 38+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-10-08  1:37 Include leaf in Emacs distribution Naoya Yamashita
2020-10-08  9:00 ` Ergus
2020-10-08  9:22   ` Naoya Yamashita
2020-10-10 10:11     ` Eli Zaretskii
2020-10-11  5:24       ` Richard Stallman
2020-10-11  8:39         ` Naoya Yamashita
2020-10-11  9:52           ` Thibaut Verron
2020-10-11 16:50             ` Naoya Yamashita
2020-10-11 17:12               ` Thibaut Verron
2020-10-12  2:10                 ` Naoya Yamashita
2020-10-12 20:23                   ` Ergus via Emacs development discussions.
2020-10-11 17:02           ` Stefan Kangas
2020-10-11 16:51   ` Stefan Kangas
2020-10-12 20:53     ` Mingde (Matthew) Zeng
2020-10-11 17:22 ` Stefan Kangas
2020-10-12  1:35   ` Naoya Yamashita
2020-10-12 22:13     ` Stefan Kangas
2020-10-12 22:19       ` Qiantan Hong
2020-10-12 22:39       ` Caio Henrique
2020-10-13 13:23         ` Stefan Monnier
2020-10-13 14:14           ` Thibaut Verron
2020-10-13 14:29             ` Stefan Monnier
2020-10-13 15:29               ` Thibaut Verron
2020-10-18  9:32                 ` Phil Sainty
2020-10-13 15:25           ` Caio Henrique
2020-10-23  2:37             ` Naoya Yamashita
2020-10-23  3:41               ` John Wiegley
2020-10-23 14:33                 ` Stefan Monnier
2020-10-23 15:53                   ` Naoya Yamashita
2020-10-23 16:46                     ` Warnings in init files (was: Include leaf in Emacs distribution) Stefan Monnier
2020-10-23 18:11                     ` T.V Raman [this message]
2020-10-23 18:41               ` Include leaf in Emacs distribution Stefan Kangas
2020-10-23 20:04                 ` John Wiegley
2020-11-16  5:29                   ` Naoya Yamashita
2020-11-17  0:39                     ` John Wiegley
2020-11-20 11:04                       ` Naoya Yamashita
2020-11-20 11:29                         ` Pankaj Jangid
2020-11-20 15:44                         ` T.V Raman

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

  List information: https://www.gnu.org/software/emacs/

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=p913624vmlw.fsf@google.com \
    --to=raman@google.com \
    --cc=caiohcs0@gmail.com \
    --cc=conao3@gmail.com \
    --cc=emacs-devel@gnu.org \
    --cc=monnier@iro.umontreal.ca \
    --cc=stefankangas@gmail.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
Code repositories for project(s) associated with this public inbox

	https://git.savannah.gnu.org/cgit/emacs.git

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).