unofficial mirror of bug-gnu-emacs@gnu.org 
 help / color / mirror / code / Atom feed
* bug#25670: Feature request: ignore --help in --script mode
@ 2017-02-09 21:00 Clément Pit--Claudel
  2017-02-10  4:40 ` Glenn Morris
                   ` (2 more replies)
  0 siblings, 3 replies; 9+ messages in thread
From: Clément Pit--Claudel @ 2017-02-09 21:00 UTC (permalink / raw)
  To: 25670


[-- Attachment #1.1: Type: text/plain, Size: 619 bytes --]

Hi bug-gnu-emacs,

Event with --script, Emacs still displays its help when passed --help.  This is confusing for users (they might write emacs --script some-script.el --help expecting to get help on the script, but instead they get Emacs' help), and it force ELisp scripts to use --usage or some other harder-to-guess synonym.

Could this be changed so that emacs --script a.el --help does not show Emacs' help?

More generally, could we guarantee that options that appear *after* "--script" aren't processed by Emacs itself (and instead left to the script)?  Or would that break things?

Thanks!
Clément.


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 819 bytes --]

^ permalink raw reply	[flat|nested] 9+ messages in thread

* bug#25670: Feature request: ignore --help in --script mode
  2017-02-09 21:00 bug#25670: Feature request: ignore --help in --script mode Clément Pit--Claudel
@ 2017-02-10  4:40 ` Glenn Morris
  2017-02-10 15:05 ` Ted Zlatanov
  2017-02-18  9:59 ` Eli Zaretskii
  2 siblings, 0 replies; 9+ messages in thread
From: Glenn Morris @ 2017-02-10  4:40 UTC (permalink / raw)
  To: Clément Pit--Claudel; +Cc: 25670


AFAICS a simple shell-script wrapper removes the need for --script
altogether, so it's hard to get motivated to improve it. Eg

#!/bin/sh

... --help parsing goes here ...

exec emacs -batch -l script.el





^ permalink raw reply	[flat|nested] 9+ messages in thread

* bug#25670: Feature request: ignore --help in --script mode
  2017-02-09 21:00 bug#25670: Feature request: ignore --help in --script mode Clément Pit--Claudel
  2017-02-10  4:40 ` Glenn Morris
@ 2017-02-10 15:05 ` Ted Zlatanov
  2017-02-10 16:52   ` Clément Pit--Claudel
  2017-02-18  9:59 ` Eli Zaretskii
  2 siblings, 1 reply; 9+ messages in thread
From: Ted Zlatanov @ 2017-02-10 15:05 UTC (permalink / raw)
  To: Clément Pit--Claudel; +Cc: 25670

On Thu, 9 Feb 2017 16:00:59 -0500 Clément Pit--Claudel <clement.pitclaudel@live.com> wrote: 

CP> More generally, could we guarantee that options that appear *after* "--script" aren't processed by Emacs itself (and instead left to the script)?  Or would that break things?

Typically I've seen "--" used to indicate the end of parseable options.
Would it help?

If so, I think it makes sense to support that in Emacs. There may be
other future uses of it besides scripting.

Ted





^ permalink raw reply	[flat|nested] 9+ messages in thread

* bug#25670: Feature request: ignore --help in --script mode
  2017-02-10 15:05 ` Ted Zlatanov
@ 2017-02-10 16:52   ` Clément Pit--Claudel
  2017-02-10 17:36     ` Glenn Morris
  0 siblings, 1 reply; 9+ messages in thread
From: Clément Pit--Claudel @ 2017-02-10 16:52 UTC (permalink / raw)
  To: Ted Zlatanov; +Cc: 25670-done


[-- Attachment #1.1: Type: text/plain, Size: 691 bytes --]

On 2017-02-10 10:05, Ted Zlatanov wrote:
> On Thu, 9 Feb 2017 16:00:59 -0500 Clément Pit--Claudel <clement.pitclaudel@live.com> wrote: 
> 
> CP> More generally, could we guarantee that options that appear *after* "--script" aren't processed by Emacs itself (and instead left to the script)?  Or would that break things?
> 
> Typically I've seen "--" used to indicate the end of parseable options.
> Would it help?

Yes! And in fact it seems to work :) Neat.

$ emacs -Q --batch --eval '(print argv)' -- --help
("--" "--help")

I see now that it was mentioned in http://www.lunaryorn.com/posts/emacs-script-pitfalls.html.  Not sure how I missed it.

Thanks Ted!
Clément.


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 819 bytes --]

^ permalink raw reply	[flat|nested] 9+ messages in thread

* bug#25670: Feature request: ignore --help in --script mode
  2017-02-10 16:52   ` Clément Pit--Claudel
@ 2017-02-10 17:36     ` Glenn Morris
  2017-02-10 20:23       ` Clément Pit--Claudel
  0 siblings, 1 reply; 9+ messages in thread
From: Glenn Morris @ 2017-02-10 17:36 UTC (permalink / raw)
  To: 25670; +Cc: clement.pitclaudel

Clément Pit--Claudel wrote:

> $ emacs -Q --batch --eval '(print argv)' -- --help
> ("--" "--help")

If you can train your users to insert -- before --help, you could get
them to use any other flag for help. And if you are inserting the -- for
them, you are already using a wrapper.





^ permalink raw reply	[flat|nested] 9+ messages in thread

* bug#25670: Feature request: ignore --help in --script mode
  2017-02-10 17:36     ` Glenn Morris
@ 2017-02-10 20:23       ` Clément Pit--Claudel
  0 siblings, 0 replies; 9+ messages in thread
From: Clément Pit--Claudel @ 2017-02-10 20:23 UTC (permalink / raw)
  To: Glenn Morris, 25670


[-- Attachment #1.1: Type: text/plain, Size: 636 bytes --]

On 2017-02-10 12:36, Glenn Morris wrote:
> Clément Pit--Claudel wrote:
> 
>> $ emacs -Q --batch --eval '(print argv)' -- --help
>> ("--" "--help")
> 
> If you can train your users to insert -- before --help, you could get
> them to use any other flag for help. And if you are inserting the -- for
> them, you are already using a wrapper.

I'm using this as my header:

#!/bin/sh
":"; exec emacs -Q --script "$0" -- "$@" # -*- mode: emacs-lisp; lexical-binding: t; -*-

I don't understand the wrapper point: I want to do the argument parsing in ELisp, not in bash.  Did I misunderstand your suggestion?

Clément.



[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 819 bytes --]

^ permalink raw reply	[flat|nested] 9+ messages in thread

* bug#25670: Feature request: ignore --help in --script mode
  2017-02-09 21:00 bug#25670: Feature request: ignore --help in --script mode Clément Pit--Claudel
  2017-02-10  4:40 ` Glenn Morris
  2017-02-10 15:05 ` Ted Zlatanov
@ 2017-02-18  9:59 ` Eli Zaretskii
  2017-02-18 13:46   ` Clément Pit--Claudel
  2 siblings, 1 reply; 9+ messages in thread
From: Eli Zaretskii @ 2017-02-18  9:59 UTC (permalink / raw)
  To: Clément Pit--Claudel; +Cc: 25670

> From: Clément Pit--Claudel <clement.pitclaudel@live.com>
> Date: Thu, 9 Feb 2017 16:00:59 -0500
> 
> Event with --script, Emacs still displays its help when passed --help.  This is confusing for users (they might write emacs --script some-script.el --help expecting to get help on the script, but instead they get Emacs' help), and it force ELisp scripts to use --usage or some other harder-to-guess synonym.
> 
> Could this be changed so that emacs --script a.el --help does not show Emacs' help?

What should be done with --help in that case?

(In general, I'd expect such script not to use --help or --version at
all.)

> More generally, could we guarantee that options that appear *after* "--script" aren't processed by Emacs itself (and instead left to the script)?  Or would that break things?

Which other options should be processed when --script is given?  Are
there any requirements for what command-line-args-left should have
when the script is loaded?

IOW, it is not clear to me what are the expectations in this use case.
Please take a look at how the command-line arguments are processed at
startup, in emacs.c and in startup.el, and make a more detailed
proposal.  Then we could see if it's easy to implement without
potentially breaking anything.





^ permalink raw reply	[flat|nested] 9+ messages in thread

* bug#25670: Feature request: ignore --help in --script mode
  2017-02-18  9:59 ` Eli Zaretskii
@ 2017-02-18 13:46   ` Clément Pit--Claudel
  2017-02-18 14:18     ` Eli Zaretskii
  0 siblings, 1 reply; 9+ messages in thread
From: Clément Pit--Claudel @ 2017-02-18 13:46 UTC (permalink / raw)
  To: Eli Zaretskii; +Cc: 25670


[-- Attachment #1.1: Type: text/plain, Size: 559 bytes --]

On 2017-02-18 04:59, Eli Zaretskii wrote:
> IOW, it is not clear to me what are the expectations in this use case.
> Please take a look at how the command-line arguments are processed at
> startup, in emacs.c and in startup.el, and make a more detailed
> proposal.  Then we could see if it's easy to implement without
> potentially breaking anything.

Ted's pointer to Emacs' support for "--" solved this problem for me: when I pass "--" before "--help", Emacs doesn't process "--".  So I don't --script to be special-cased any more :)

Clément.


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 819 bytes --]

^ permalink raw reply	[flat|nested] 9+ messages in thread

* bug#25670: Feature request: ignore --help in --script mode
  2017-02-18 13:46   ` Clément Pit--Claudel
@ 2017-02-18 14:18     ` Eli Zaretskii
  0 siblings, 0 replies; 9+ messages in thread
From: Eli Zaretskii @ 2017-02-18 14:18 UTC (permalink / raw)
  To: Clément Pit--Claudel; +Cc: 25670

> Cc: 25670@debbugs.gnu.org
> From: Clément Pit--Claudel <clement.pitclaudel@live.com>
> Date: Sat, 18 Feb 2017 08:46:07 -0500
> 
> Ted's pointer to Emacs' support for "--" solved this problem for me: when I pass "--" before "--help", Emacs doesn't process "--".  So I don't --script to be special-cased any more :)

OK, thanks.





^ permalink raw reply	[flat|nested] 9+ messages in thread

end of thread, other threads:[~2017-02-18 14:18 UTC | newest]

Thread overview: 9+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2017-02-09 21:00 bug#25670: Feature request: ignore --help in --script mode Clément Pit--Claudel
2017-02-10  4:40 ` Glenn Morris
2017-02-10 15:05 ` Ted Zlatanov
2017-02-10 16:52   ` Clément Pit--Claudel
2017-02-10 17:36     ` Glenn Morris
2017-02-10 20:23       ` Clément Pit--Claudel
2017-02-18  9:59 ` Eli Zaretskii
2017-02-18 13:46   ` Clément Pit--Claudel
2017-02-18 14:18     ` Eli Zaretskii

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).