bug#61014: 29.0.60; flymake-mode stderr warning is confusing when edit the init.el or eary-init.el

bug#61014: 29.0.60; flymake-mode stderr warning is confusing when edit the init.el or eary-init.el

[Eason Huang]

I found that when editing the init.el or eary-ini.el under
`user-emacs-directory`, flymake will report the warning as below:

```
Warning (initialization): Your `load-path' seems to contain
your `.emacs.d' directory: /Users/eason/.emacs.d/
This is likely to cause problems...
Consider using a subdirectory instead, e.g.: /Users/eason/.emacs.d/lisp
```

And this is because the variable `elisp-flymake-byte-compile-load-path`
has a defaut value `("./")`

so flymake adding the `.emacs.d` to the load-path and then also
complaining about it, it is confusing to users. May be we can improve
it.

Steps to reproduce:

1. create a new .emacs.d folder in the HOME directory
2. create an empty init.e and eary-init.el file in the .emacs.d folder
3. Start Emacs and C-x, C-f open the init.el or early.el
4. M-x flymake-mode, add some blank lines in the init.el or early.el
5. C-x b, Switch to buffer:  *stderr of elisp-flymake-byte-compile*
Now you will see the warnings.


By the way, the message "Your `load-path' seems to contain
your `.emacs.d' directory" should use `user-emacs-directory` instead,
because some user will use `~/.config/emacs/` folder instead of
`.emacs.d`

--
Eason Huang





In GNU Emacs 29.0.60 (build 1, x86_64-apple-darwin22.2.0, NS
 appkit-2299.30 Version 13.1 (Build 22C65), git sha1 846838dbab) of
 2023-01-22 built on macbook
Windowing system distributor 'Apple', version 10.3.2299
System Description:  macOS 13.1

Configured using:
 'configure --without-native-compilation --without-dbus
 'CPPFLAGS=-I/opt/local/include
 -isysroot/Library/Developer/CommandLineTools/SDKs/MacOSX.sdk'
 'LDFLAGS=-L/opt/local/lib -Wl,-headerpad_max_install_names -Wl,-rpath
 /opt/local/lib/gcc12
 -Wl,-syslibroot,/Library/Developer/CommandLineTools/SDKs/MacOSX.sdk
 -arch x86_64''

Configured features:
ACL GIF GLIB GMP GNUTLS JPEG JSON LCMS2 LIBXML2 MODULES NOTIFY KQUEUE NS
PDUMPER PNG RSVG SQLITE3 THREADS TIFF TOOLKIT_SCROLL_BARS TREE_SITTER
WEBP XIM ZLIB

Important settings:
  locale-coding-system: utf-8-unix

Major mode: ELisp/l

Minor modes in effect:
  tooltip-mode: t
  global-eldoc-mode: t
  eldoc-mode: t
  show-paren-mode: t
  electric-indent-mode: t
  mouse-wheel-mode: t
  menu-bar-mode: t
  file-name-shadow-mode: t
  global-font-lock-mode: t
  font-lock-mode: t
  blink-cursor-mode: t
  line-number-mode: t
  transient-mark-mode: t
  auto-composition-mode: t
  auto-encryption-mode: t
  auto-compression-mode: t

Load-path shadows:
None found.

Features:
(shadow sort mail-extr vc-git diff-mode easy-mmode vc-dispatcher
emacsbug message mailcap yank-media puny rfc822 mml mml-sec
password-cache epa derived epg rfc6068 epg-config gnus-util mm-decode
mm-bodies mm-encode mail-parse rfc2231 mailabbrev gmm-utils mailheader
sendmail rfc2047 rfc2045 ietf-drums mm-util mail-prsvr mail-utils
cl-extra noutline outline dired-aux dired dired-loaddefs time-date
subr-x checkdoc lisp-mnt help-mode cl-macs cl-seq flymake-proc flymake
project byte-opt gv bytecomp byte-compile compile text-property-search
comint ansi-osc ansi-color ring warnings icons thingatpt
emacs-git-version cl-loaddefs cl-lib rmc iso-transl tooltip cconv eldoc
paren electric uniquify ediff-hook vc-hooks lisp-float-type elisp-mode
mwheel term/ns-win ns-win ucs-normalize mule-util term/common-win
tool-bar dnd fontset image regexp-opt fringe tabulated-list replace
newcomment text-mode lisp-mode prog-mode register page tab-bar menu-bar
rfn-eshadow isearch easymenu timer select scroll-bar mouse jit-lock
font-lock syntax font-core term/tty-colors frame minibuffer nadvice seq
simple cl-generic indonesian philippine cham georgian utf-8-lang
misc-lang vietnamese tibetan thai tai-viet lao korean japanese eucjp-ms
cp51932 hebrew greek romanian slovak czech european ethiopic indian
cyrillic chinese composite emoji-zwj charscript charprop case-table
epa-hook jka-cmpr-hook help abbrev obarray oclosure cl-preloaded button
loaddefs theme-loaddefs faces cus-face macroexp files window
text-properties overlay sha1 md5 base64 format env code-pages mule
custom widget keymap hashtable-print-readable backquote threads kqueue
cocoa ns lcms2 multi-tty make-network-process emacs)

Memory information:
((conses 16 70613 13315)
 (symbols 48 8086 0)
 (strings 32 24199 1872)
 (string-bytes 1 729791)
 (vectors 16 16886)
 (vector-slots 8 227852 13364)
 (floats 8 38 127)
 (intervals 56 627 0)
 (buffers 984 18))

bug#61014: 29.0.60; flymake-mode stderr warning is confusing when edit the init.el or eary-init.el

[Eli Zaretskii]

> From: Eason Huang <aqua0210@foxmail.com>
> Date: Mon, 23 Jan 2023 10:15:49 +0800
> 
> 
> I found that when editing the init.el or eary-ini.el under
> `user-emacs-directory`, flymake will report the warning as below:
> 
> ```
> Warning (initialization): Your `load-path' seems to contain
> your `.emacs.d' directory: /Users/eason/.emacs.d/
> This is likely to cause problems...
> Consider using a subdirectory instead, e.g.: /Users/eason/.emacs.d/lisp
> ```
> 
> And this is because the variable `elisp-flymake-byte-compile-load-path`
> has a defaut value `("./")`
> 
> so flymake adding the `.emacs.d` to the load-path and then also
> complaining about it, it is confusing to users. May be we can improve
> it.
> 
> Steps to reproduce:
> 
> 1. create a new .emacs.d folder in the HOME directory
> 2. create an empty init.e and eary-init.el file in the .emacs.d folder
> 3. Start Emacs and C-x, C-f open the init.el or early.el
> 4. M-x flymake-mode, add some blank lines in the init.el or early.el
> 5. C-x b, Switch to buffer:  *stderr of elisp-flymake-byte-compile*
> Now you will see the warnings.

João, any comments?  The warning comes from startup.el, but it is
triggered by what Flymake does, AFAICT.

> By the way, the message "Your `load-path' seems to contain
> your `.emacs.d' directory" should use `user-emacs-directory` instead,
> because some user will use `~/.config/emacs/` folder instead of
> `.emacs.d`

I fixed this part.

bug#61014: 29.0.60; flymake-mode stderr warning is confusing when edit the init.el or eary-init.el

[João Távora]

Eli Zaretskii <eliz@gnu.org> writes:

>> 1. create a new .emacs.d folder in the HOME directory
>> 2. create an empty init.e and eary-init.el file in the .emacs.d folder
>> 3. Start Emacs and C-x, C-f open the init.el or early.el
>> 4. M-x flymake-mode, add some blank lines in the init.el or early.el
>> 5. C-x b, Switch to buffer:  *stderr of elisp-flymake-byte-compile*
>> Now you will see the warnings.
>
> João, any comments?  The warning comes from startup.el, but it is
> triggered by what Flymake does, AFAICT.

Sorry for taking so long.

Yes, I have comments.

First of all, let's make it clear that this warning happens in a hidden
Flymake buffer, an implementation detail which stores the stderr output
of the subprocess(es) invoked.  It does not appear in the interactive
seeion's *Warnings* buffer.  Unless I've missed something, the user is
never confronted with it or hindered by it in any way.  It seems to be
largely benign.

The warning only become visible when switching to a specific hidden
buffer, like Eason did, and that operation is more accurately described
as an introspection feature intended for developers, not for users.

So, to summarize, the warning is coming from the subprocess
non-interactive Emacs spawned by the interactive Emacs where
flymake-mode is turned on.  In the interactive Emacs session, nothing
special is going on.

The reason elisp-flymake-byte-compile-load-path has a default value of
"./" is to help with the primary use case for the
elisp-flymake-byte-compile Flymake backend: developing libraries that
`require` other .el libraries.  Such .el files often live in the same
directory of the file/buffer you are editing with flymake-mode turned
on.  So when you're editing x/y/z/foo.el and you type

  (require 'foo-utils)

elisp-flymake-byte-compile will understand that to syntax-check foo.el
with the byte-compiler it needs to load "x/u/z/foo-utils.el".

In Eason's case, the user happens to be editing a .el file in a special
Emacs directory where such a use case is unlikely (though not
impossible).  So every few changes the user does to the buffer,
flymake-mdoe will invoke a Emacs subprocess and give it this particular
load-path initialization, that, in the specific acse of ~/.emacs.d, is
slightly non-orthodox.  And thus the subprocess emacs complains with
this warning which is recorded in the special hidden buffer.

If we're really worries about this, we could special-case this directory
elisp-mode.el.  But I also don't see what the harm could be.  So my
advice is to only do this when someone describes real harm.

João

bug#61014: 29.0.60; flymake-mode stderr warning is confusing when edit the init.el or eary-init.el

[Eli Zaretskii]

> From: João Távora <joaotavora@gmail.com>
> Cc: Eason Huang <aqua0210@foxmail.com>,  61014@debbugs.gnu.org
> Date: Thu, 26 Jan 2023 11:16:39 +0000
> 
> First of all, let's make it clear that this warning happens in a hidden
> Flymake buffer, an implementation detail which stores the stderr output
> of the subprocess(es) invoked.  It does not appear in the interactive
> seeion's *Warnings* buffer.  Unless I've missed something, the user is
> never confronted with it or hindered by it in any way.  It seems to be
> largely benign.
> 
> The warning only become visible when switching to a specific hidden
> buffer, like Eason did, and that operation is more accurately described
> as an introspection feature intended for developers, not for users.
> 
> So, to summarize, the warning is coming from the subprocess
> non-interactive Emacs spawned by the interactive Emacs where
> flymake-mode is turned on.  In the interactive Emacs session, nothing
> special is going on.
> 
> The reason elisp-flymake-byte-compile-load-path has a default value of
> "./" is to help with the primary use case for the
> elisp-flymake-byte-compile Flymake backend: developing libraries that
> `require` other .el libraries.  Such .el files often live in the same
> directory of the file/buffer you are editing with flymake-mode turned
> on.  So when you're editing x/y/z/foo.el and you type
> 
>   (require 'foo-utils)
> 
> elisp-flymake-byte-compile will understand that to syntax-check foo.el
> with the byte-compiler it needs to load "x/u/z/foo-utils.el".
> 
> In Eason's case, the user happens to be editing a .el file in a special
> Emacs directory where such a use case is unlikely (though not
> impossible).  So every few changes the user does to the buffer,
> flymake-mdoe will invoke a Emacs subprocess and give it this particular
> load-path initialization, that, in the specific acse of ~/.emacs.d, is
> slightly non-orthodox.  And thus the subprocess emacs complains with
> this warning which is recorded in the special hidden buffer.
> 
> If we're really worries about this, we could special-case this directory
> elisp-mode.el.  But I also don't see what the harm could be.  So my
> advice is to only do this when someone describes real harm.

How about disabling this warning in non-interactive sessions?  That
should be easy, and I don't see why we would want to emit such a
warning in non-interactive sessions anyway.

Stefan, am I missing something?

bug#61014: 29.0.60; flymake-mode stderr warning is confusing when edit the init.el or eary-init.el

[Eli Zaretskii]

Ping!

> Cc: aqua0210@foxmail.com, 61014@debbugs.gnu.org
> Date: Sat, 28 Jan 2023 13:21:14 +0200
> From: Eli Zaretskii <eliz@gnu.org>
> 
> > From: João Távora <joaotavora@gmail.com>
> > Cc: Eason Huang <aqua0210@foxmail.com>,  61014@debbugs.gnu.org
> > Date: Thu, 26 Jan 2023 11:16:39 +0000
> > 
> > First of all, let's make it clear that this warning happens in a hidden
> > Flymake buffer, an implementation detail which stores the stderr output
> > of the subprocess(es) invoked.  It does not appear in the interactive
> > seeion's *Warnings* buffer.  Unless I've missed something, the user is
> > never confronted with it or hindered by it in any way.  It seems to be
> > largely benign.
> > 
> > The warning only become visible when switching to a specific hidden
> > buffer, like Eason did, and that operation is more accurately described
> > as an introspection feature intended for developers, not for users.
> > 
> > So, to summarize, the warning is coming from the subprocess
> > non-interactive Emacs spawned by the interactive Emacs where
> > flymake-mode is turned on.  In the interactive Emacs session, nothing
> > special is going on.
> > 
> > The reason elisp-flymake-byte-compile-load-path has a default value of
> > "./" is to help with the primary use case for the
> > elisp-flymake-byte-compile Flymake backend: developing libraries that
> > `require` other .el libraries.  Such .el files often live in the same
> > directory of the file/buffer you are editing with flymake-mode turned
> > on.  So when you're editing x/y/z/foo.el and you type
> > 
> >   (require 'foo-utils)
> > 
> > elisp-flymake-byte-compile will understand that to syntax-check foo.el
> > with the byte-compiler it needs to load "x/u/z/foo-utils.el".
> > 
> > In Eason's case, the user happens to be editing a .el file in a special
> > Emacs directory where such a use case is unlikely (though not
> > impossible).  So every few changes the user does to the buffer,
> > flymake-mdoe will invoke a Emacs subprocess and give it this particular
> > load-path initialization, that, in the specific acse of ~/.emacs.d, is
> > slightly non-orthodox.  And thus the subprocess emacs complains with
> > this warning which is recorded in the special hidden buffer.
> > 
> > If we're really worries about this, we could special-case this directory
> > elisp-mode.el.  But I also don't see what the harm could be.  So my
> > advice is to only do this when someone describes real harm.
> 
> How about disabling this warning in non-interactive sessions?  That
> should be easy, and I don't see why we would want to emit such a
> warning in non-interactive sessions anyway.
> 
> Stefan, am I missing something?
> 
> 
> 
> 

bug#61014: 29.0.60; flymake-mode stderr warning is confusing when edit the init.el or eary-init.el

[João Távora]

[-- Attachment #1: Type: text/plain, Size: 311 bytes --]

On Thu, Feb 2, 2023 at 10:25 AM Eli Zaretskii <eliz@gnu.org> wrote:
>
> Ping!

This ping is mostly for Stefan, right?  I'm OK with:

* Disabling the warning in non-interactive sessions
* Doing nothing
* Adding special logic to elisp-mode.el if someone explains what
  harm this is producing.

João

[-- Attachment #2: Type: text/html, Size: 480 bytes --]

bug#61014: 29.0.60; flymake-mode stderr warning is confusing when edit the init.el or eary-init.el

[Stefan Monnier via Bug reports for GNU Emacs, the Swiss army knife of text editors]

> How about disabling this warning in non-interactive sessions?  That
> should be easy, and I don't see why we would want to emit such a
> warning in non-interactive sessions anyway.

Before making any change, I'd wait to see some explanation about why
this message in a hidden " *stderr of elisp-flymake-byte-compile*" buffer
is a problem.

I think the choice of `load-path` in the subprocess is a concern (with
no reliably right answer, sadly), and maybe it would make sense to
avoid adding the CWD when we're in ~/.emacs.d (presumably using the
same code used to decide whether to emit the warning under
discussion :-), but there's no hurry for that.


        Stefan

bug#61014: 29.0.60; flymake-mode stderr warning is confusing when edit the init.el or eary-init.el

[Eli Zaretskii]

> From: Stefan Monnier <monnier@iro.umontreal.ca>
> Cc: João Távora <joaotavora@gmail.com>,
>   aqua0210@foxmail.com,
>   61014@debbugs.gnu.org
> Date: Thu, 02 Feb 2023 11:01:20 -0500
> 
> > How about disabling this warning in non-interactive sessions?  That
> > should be easy, and I don't see why we would want to emit such a
> > warning in non-interactive sessions anyway.
> 
> Before making any change, I'd wait to see some explanation about why
> this message in a hidden " *stderr of elisp-flymake-byte-compile*" buffer
> is a problem.

I don't thin it should be a problem, but I also don't see any harm in
suppressing the warning in non-interactive sessions.  That warning is
clearly meant to be presented to users when they start an interactive
session and their customizations did dubious things.

I hoped that suppressing this in non-interactive session would strike
a balance between having the warning when it matters and removing the
annoyance in the OP's case, even if the justification of that
annoyance is not necessarily clear to us.

IOW, a compromise that would leave everybody happy enough.

WDYT?

bug#61014: 29.0.60; flymake-mode stderr warning is confusing when edit the init.el or eary-init.el

[Eli Zaretskii]

> From: Stefan Monnier <monnier@iro.umontreal.ca>
> Cc: João Távora <joaotavora@gmail.com>,
>   aqua0210@foxmail.com,
>   61014@debbugs.gnu.org
> Date: Thu, 02 Feb 2023 11:01:20 -0500
> 
> > How about disabling this warning in non-interactive sessions?  That
> > should be easy, and I don't see why we would want to emit such a
> > warning in non-interactive sessions anyway.
> 
> Before making any change, I'd wait to see some explanation about why
> this message in a hidden " *stderr of elisp-flymake-byte-compile*" buffer
> is a problem.
> 
> I think the choice of `load-path` in the subprocess is a concern (with
> no reliably right answer, sadly), and maybe it would make sense to
> avoid adding the CWD when we're in ~/.emacs.d (presumably using the
> same code used to decide whether to emit the warning under
> discussion :-), but there's no hurry for that.

No further comments, so I've now made that warning be emitted only in
interactive sessions, and I'm boldly marking this bug done.

Thanks.

bug#61014: 29.0.60; flymake-mode stderr warning is confusing when edit the init.el or eary-init.el

[Eason Huang]

Eli Zaretskii <eliz@gnu.org> writes:

>> From: Stefan Monnier <monnier@iro.umontreal.ca>
>> Cc: João Távora <joaotavora@gmail.com>,
>>   aqua0210@foxmail.com,
>>   61014@debbugs.gnu.org
>> Date: Thu, 02 Feb 2023 11:01:20 -0500
>>
>> > How about disabling this warning in non-interactive sessions?  That
>> > should be easy, and I don't see why we would want to emit such a
>> > warning in non-interactive sessions anyway.
>>
>> Before making any change, I'd wait to see some explanation about why
>> this message in a hidden " *stderr of elisp-flymake-byte-compile*" buffer
>> is a problem.
>>
>> I think the choice of `load-path` in the subprocess is a concern (with
>> no reliably right answer, sadly), and maybe it would make sense to
>> avoid adding the CWD when we're in ~/.emacs.d (presumably using the
>> same code used to decide whether to emit the warning under
>> discussion :-), but there's no hurry for that.
>
> No further comments, so I've now made that warning be emitted only in
> interactive sessions, and I'm boldly marking this bug done.
>
> Thanks.

Great!

I build the latest commit(d4fc7012978) of emacs-29 branch, It works
perfect.

Thanks for working on this issue.

-- 
Eason Huang

bug#61014: 29.0.60; flymake-mode stderr warning is confusing when edit the init.el or eary-init.el

[Eli Zaretskii]

> From: Eason Huang <aqua0210@foxmail.com>
> Cc: Stefan Monnier <monnier@iro.umontreal.ca>,  joaotavora@gmail.com,
>   61014-done@debbugs.gnu.org
> Date: Sun, 12 Feb 2023 22:45:47 +0800
> 
> Eli Zaretskii <eliz@gnu.org> writes:
> 
> > No further comments, so I've now made that warning be emitted only in
> > interactive sessions, and I'm boldly marking this bug done.
> >
> > Thanks.
> 
> Great!
> 
> I build the latest commit(d4fc7012978) of emacs-29 branch, It works
> perfect.
> 
> Thanks for working on this issue.

Thanks for testing the fix (and for reporting the problem to begin
with).