bug#24295: 24.5; RIGHT SINGLE QUOTATION MARK has punctuation syntax in Elisp, but acts like a symbol constituent

bug#24295: 24.5; RIGHT SINGLE QUOTATION MARK has punctuation syntax in Elisp, but acts like a symbol constituent

[Drew Adams]

In an Emacs-Lisp buffer or in *scratch*, use `C-u C-x =3D' on this
character: =A1=AF (RIGHT SINGLE QUOTATION MARK).  It tells you that its
syntax is punctuation.

But (setq foo (intern "ab=A1=AFcd=A1=AFef")) shows that it acts like a
symbol-constituent character.

And the Elisp manual, node Syntax Class Table says this, about syntax
class punctuation:

  Some programming language modes, such as Emacs Lisp mode, have no
  characters in this class since the few characters that are not symbol
  or word constituents all have other uses.

What gives?  Is there a doc bug?  Should the syntax class be
symbol-constituent?  Are things as they should be (in which case, can
the doc be clarified, as this seems confusing)?


In GNU Emacs 24.5.1 (i686-pc-mingw32)
 of 2015-04-11 on LEG570
Windowing system distributor `Microsoft Corp.', version 6.1.7601
Configured using:
 `configure --prefix=3D/c/usr --host=3Di686-pc-mingw32'

bug#24295: 24.5; RIGHT SINGLE QUOTATION MARK has punctuation syntax in Elisp, but acts like a symbol constituent

[Drew Adams]

> In an Emacs-Lisp buffer or in *scratch*, use `C-u C-x =3D' on this
> character: =A1=AF (RIGHT SINGLE QUOTATION MARK).  It tells you that its
> syntax is punctuation.
> 
> But (setq foo (intern "ab=A1=AFcd=A1=AFef")) shows that it acts like a
> symbol-constituent character.

That certainly didn't paste well into my mail client, I guess.

The character in question is RIGHT SINGLE QUOTATION MARK.  It showed
up in my message as =A1=AF, for some reason.  Trying again (it _looks_
OK in my mail message, before I send it, at least):

’’’’’’’’’’’

bug#24295: 24.5; RIGHT SINGLE QUOTATION MARK has punctuation syntax in Elisp, but acts like a symbol constituent

[Drew Adams]

> > In an Emacs-Lisp buffer or in *scratch*, use `C-u C-x =3D' on this
> > character: =A1=AF (RIGHT SINGLE QUOTATION MARK).  It tells you that its
> > syntax is punctuation.
> >
> > But (setq foo (intern "ab=A1=AFcd=A1=AFef")) shows that it acts like a
> > symbol-constituent character.
> 
> That certainly didn't paste well into my mail client, I guess.
> 
> The character in question is RIGHT SINGLE QUOTATION MARK.  It showed
> up in my message as =A1=AF, for some reason.  Trying again (it _looks_
> OK in my mail message, before I send it, at least):
> 
> ’’’’’’’’’’’

OK, that worked.  What was the difference?  La voici :

For the initial bug report, I used `C-c C-c' from `report-emacs-bug',
and then I used `C-v' in the popped up MS Outlook mail message window.

For the followup, I selected the char with the region in Emacs and
then used `C-v' in the MS Outlook mail message window.  So it seems
like `report-emacs-bug' doesn't quite DTRT always.

bug#24295: 24.5; RIGHT SINGLE QUOTATION MARK has punctuation syntax in Elisp, but acts like a symbol constituent

[npostavs]

Drew Adams <drew.adams@oracle.com> writes:

[I replaced =A1=AF with ’ in the quoted text]

> In an Emacs-Lisp buffer or in *scratch*, use `C-u C-x =3D' on this
> character: ’ (RIGHT SINGLE QUOTATION MARK).  It tells you that its
> syntax is punctuation.
>
> But (setq foo (intern "ab’cd’ef")) shows that it acts like a
> symbol-constituent character.

In what way does it show that??  A symbol's name can be any arbitrary
string, there's no significance to the syntax class of the characters.

bug#24295: 24.5; RIGHT SINGLE QUOTATION MARK has punctuation syntax in Elisp, but acts like a symbol constituent

[Drew Adams]

> In what way does it show that??  A symbol's name can be any arbitrary
> string, there's no significance to the syntax class of the characters.

Silly me.  Of course that's right.

Nevertheless, there seems to be a problem, but it is not what
I thought.

M-: (global-set-key (kbd "C-o") ’forward-char)

Debugger entered--Lisp error: (void-variable ’forward-char)
  (global-set-key (kbd "C-o") ’forward-char)
  eval((global-set-key (kbd "C-o") ’forward-char) nil)
  elisp--eval-last-sexp(nil)
  eval-last-sexp(nil)
  funcall-interactively(eval-last-sexp nil)
  call-interactively(eval-last-sexp nil nil)
  command-execute(eval-last-sexp)

bug#24295: 24.5; RIGHT SINGLE QUOTATION MARK has punctuation syntax in Elisp, but acts like a symbol constituent

[Drew Adams]

> M-: (global-set-key (kbd "C-o") ’forward-char)
> 
> Debugger entered--Lisp error: (void-variable ’forward-char)

Well, I guess that's not a problem either.   Pretty much anything
that is not a paren, quote, backquote, comma, or isolated period
will be taken as part of a symbol name here.

I'll close the bug, I guess.

bug#24295: 24.5; RIGHT SINGLE QUOTATION MARK has punctuation syntax in Elisp, but acts like a symbol constituent

[Eli Zaretskii]

> Date: Tue, 23 Aug 2016 15:56:39 -0700 (PDT)
> From: Drew Adams <drew.adams@oracle.com>
> 
> For the initial bug report, I used `C-c C-c' from `report-emacs-bug',
> and then I used `C-v' in the popped up MS Outlook mail message window.
> 
> For the followup, I selected the char with the region in Emacs and
> then used `C-v' in the MS Outlook mail message window.  So it seems
> like `report-emacs-bug' doesn't quite DTRT always.

It does DTRT: this is "quoted-printable" encoding of non-ASCII
characters.  You already asked this several times in the past, and
received the same answer.

bug#24295: 24.5; RIGHT SINGLE QUOTATION MARK has punctuation syntax in Elisp, but acts like a symbol constituent

[Andreas Schwab]

On Aug 24 2016, Eli Zaretskii <eliz@gnu.org> wrote:

> It does DTRT: this is "quoted-printable" encoding of non-ASCII
> characters.  You already asked this several times in the past, and
> received the same answer.

The message was quoted-unreadable encoded *twice*.

Andreas.

-- 
Andreas Schwab, schwab@linux-m68k.org
GPG Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

bug#24295: 24.5; RIGHT SINGLE QUOTATION MARK has punctuation syntax in Elisp, but acts like a symbol constituent

[Eli Zaretskii]

> From: Andreas Schwab <schwab@linux-m68k.org>
> Cc: Drew Adams <drew.adams@oracle.com>,  24295@debbugs.gnu.org
> Date: Wed, 24 Aug 2016 08:58:17 +0200
> 
> On Aug 24 2016, Eli Zaretskii <eliz@gnu.org> wrote:
> 
> > It does DTRT: this is "quoted-printable" encoding of non-ASCII
> > characters.  You already asked this several times in the past, and
> > received the same answer.
> 
> The message was quoted-unreadable encoded *twice*.

I don't think we have any better choice, when the end-user uses
mailclient-send-it: we know nothing about how the MUA handles
non-ASCII characters.

bug#24295: 24.5; RIGHT SINGLE QUOTATION MARK has punctuation syntax in Elisp, but acts like a symbol constituent

[Drew Adams]

> > > It does DTRT: this is "quoted-printable" encoding of non-ASCII
> > > characters.  You already asked this several times in the past, and
> > > received the same answer.
> >
> > The message was quoted-unreadable encoded *twice*.
> 
> I don't think we have any better choice, when the end-user uses
> mailclient-send-it: we know nothing about how the MUA handles
> non-ASCII characters.

Perhaps have a user option (and let users set it during the same
initial how-do-you-want-to-report-this? dialog).

Many (most?) mail clients now support such characters.
Perhaps the default behavior should assume that many can,
and thus let users simply copy+paste such chars normally,
once they've said that their client supports this.

But yes, let them choose.