Few bugs (perhaps)

Few bugs (perhaps)

[Sebastian Urban]

Hello list,

here is a list of few bugs I found, but because I'm not sure if they
are bugs, so I'm sending them here in one e-mail.  Just reply with
"It's a bug" if it's a bug and I'll send report to bug-gnu-emacs.

Tested on GNU Emacs 26.2 (build 1, i686-w64-mingw32) of 2019-04-13 and
emacs-27.0.50-snapshot-2019-06-08-i686.

* HL line mode
==============

Shouldn't 'hl-line-mode' highlight screen line instead of logical line
when 'visual-line-mode' is active?  For example 'C-a' and 'C-e' change
behaviour.  Internet says that people are using 'hl-line-range-function'
to change default behaviour.  Maybe make it choice list:
- nil for default behaviour,
- some other option for screen line,
- "function" for user to write what he wants,
(I'm writing from perspective of "customize")?

* Secondary selection
=====================

When text is selected, deselecting it is done by clicking 'M-mouse-1',
is this how it suppose to work?  Shouldn't it deselect by simply
clicking 'mouse-1'?

* 'C-z' and resuming
====================

'C-z' runs 'suspend-frame', which on text terminal suspends Emacs and
to resume I should type '%emacs' in terminal.  The problem is this
doesn't work on MS Windows Command line (cmd.exe), I can suspend Emacs
(started with 'emacs -nw') but how do I resume it?  And if there is no
way, then maybe 'C-z' should be disabled for Emacs in MS Windows
Console?

* Version numbers of packages
=============================

Descriptions of 'ada-mode' and 'svg' (both available) show "Other
versions: builtin" without number AND there are no "builtin"
'ada-mode' and 'svg' on the package list *Packages*.

Also 'cl-generic (0.3)' and 'cl-lib (0.6.1)' (both available) show
version number lower than builtin (1.0) - shouldn't it be equal or
greater?

* 'sentence-end-base'
=====================

What are the boundaries for this variable?  Because when I was looking
at 'unicode-bmp' there was few other symbols that could be placed here.

But this time I just wanted to point out to perhaps add this:
- '>' - GREATER-THAN SIGN
(codepoint 62, #o76, #x3e).

And maybe these:
- '»' - RIGHT-POINTING DOUBLE ANGLE QUOTATION MARK
(codepoint 187, #o273, #xbb),
- '›' - SINGLE RIGHT-POINTING ANGLE QUOTATION MARK
(codepoint 8250, #o20072, #x203a).

* Indentation in latex-mode
===========================

If I use 'fill-paragraph' (M-q) on paragraph, inside things like
\footnote{} or \emph{}, if they're long, text is indented with 2
spaces - look below.  I was able to overcome this by setting
'tex-indent-basic' from 2 to 0, but I think it will affect other
indentations I may want.  Is there any other method for this?

Lacus tincidunt ultrices.  Lorem ipsum dolor sit amet, consectetuer
adipiscing elit.\footnote{Quisque ullamcorper placerat ipsum.  Cras
   nibh.  Morbi vel justo vitae lacus tincidunt ultrices.  Lorem ipsum
   dolor sit amet, consectetuer adipiscing elit.}  In hac habitasse
platea dictumst.  Integer tempus convallis augue.  Etiam facilisis.
Nunc elementum fermentum wisi.  Integer tempus convallis augue.

* Ghostscript program name
==========================

Variable 'doc-view-ghostscript-program' has value - 'gs' - which is OK
for Unix and VMS, but NOT OK for MS Windows.  It should be 'gswin32c'.
For details look at the table in "How to Use Ghostscript", chapter
"2.1 Help at the command line: gs -h".  Perhaps it should be done as
list of choice?


S. U.

C-z in MSWindows shell (was: Few bugs (perhaps))

[Óscar Fuentes]

It is more convenient to give each issue its own thread. I'll address one of
your questions below.

Sebastian Urban <mrsebastianurban@gmail.com> writes:

> * 'C-z' and resuming
> ====================
>
> 'C-z' runs 'suspend-frame', which on text terminal suspends Emacs and
> to resume I should type '%emacs' in terminal.  The problem is this
> doesn't work on MS Windows Command line (cmd.exe), I can suspend Emacs
> (started with 'emacs -nw') but how do I resume it?  And if there is no
> way, then maybe 'C-z' should be disabled for Emacs in MS Windows
> Console?

The documentation of C-z says:

  Do whatever is right to suspend the current frame.
  Calls ‘suspend-emacs’ if invoked from the controlling tty device,
  ‘suspend-tty’ from a secondary tty device, and
  ‘iconify-or-deiconify-frame’ from a graphical frame.

The documentation of suspend-emacs says:

  Stop Emacs and return to superior process.  You can resume later.
  If ‘cannot-suspend’ is non-nil, or if the system doesn’t support job
  control, run a subshell instead.

MSWindows' shell has no job control, so probably cannot-suspend is t,
which means that C-z runs a subshell. If you exit from that subshell (by
typing `exit', for instance) you should go back to Emacs.

Re: C-z in MSWindows shell

[Sebastian Urban]

> It is more convenient to give each issue its own thread.

Well, when I have several small problems, sending them separately
looks like a spam.  But if you really prefer it this way, from now on,
I will (unless they will be related).

> ... so probably cannot-suspend is t...

Actually in both versions it is set to nil.

> If you exit from that subshell (by typing `exit', for instance) you
> should go back to Emacs.

And this does the job, i.e. `exit' brings me back to emacs.  Thanks.

But also, maybe a line about this in the Emacs manual?  I mean if we
have `%emacs' in the manual, so why not a line about times when there
is no job control, which leads to subshell and for example `exit' as
a way out back to Emacs.  Maybe just putting "or `exit'" next to
`%emacs' would be sufficient (I'm just guessing here).

Re: C-z in MSWindows shell

[Óscar Fuentes]

Sebastian Urban <mrsebastianurban@gmail.com> writes:

>> ... so probably cannot-suspend is t...
>
> Actually in both versions it is set to nil.

cannot-suspend being nil in MSWindows qualifies like a bug to me. Please
use M-x report-emacs-bug to report it.

I tried C-z on MSWindows and it does not run a subshell. The console
blinks but nothing else happens. I'm using a development version,
though.

>> If you exit from that subshell (by typing `exit', for instance) you
>> should go back to Emacs.
>
> And this does the job, i.e. `exit' brings me back to emacs.  Thanks.
>
> But also, maybe a line about this in the Emacs manual?  I mean if we
> have `%emacs' in the manual, so why not a line about times when there
> is no job control, which leads to subshell and for example `exit' as
> a way out back to Emacs.  Maybe just putting "or `exit'" next to
> `%emacs' would be sufficient (I'm just guessing here).

Proposing improvements like this also warrant M-x report-emacs-bug. The
maintainers try to keep the manual thin, but sometimes an addition is
justified.

Re: Few bugs (perhaps)

[Stefan Monnier]

> Shouldn't 'hl-line-mode' highlight screen line instead of logical line
> when 'visual-line-mode' is active?

That's a question of user's opinions and expectations.
IOW, it's not clearly a bug, but it sounds like you should `M-x
report-emacs-bug` anyway and request this (whether it's called a bug or
a feature request doesn't really matter).

> * Secondary selection
> =====================
>
> When text is selected, deselecting it is done by clicking 'M-mouse-1',
> is this how it suppose to work?  Shouldn't it deselect by simply
> clicking 'mouse-1'?

mouse-1 only affects the primary selection, so wouldn't it be annoying
if mouse-1 deselected the secondary selection?
[ I never use the secondary selection, so I really have no idea.  ]

> * Version numbers of packages
> =============================
>
> Descriptions of 'ada-mode' and 'svg' (both available) show "Other
> versions: builtin" without number AND there are no "builtin"
> 'ada-mode' and 'svg' on the package list *Packages*.

That's a bug.

> Also 'cl-generic (0.3)' and 'cl-lib (0.6.1)' (both available) show
> version number lower than builtin (1.0) - shouldn't it be equal or
> greater?

No, this is normal and correct:

The packages in GNU ELPA are forward compatibility packages, making it
possible to use in older Emacsen (some of) the functionality of those
packages that are builtin in more recent Emacsen.

Hence the lower version number.

> If I use 'fill-paragraph' (M-q) on paragraph, inside things like
> \footnote{} or \emph{}, if they're long, text is indented with 2
> spaces - look below.  I was able to overcome this by setting
> 'tex-indent-basic' from 2 to 0, but I think it will affect other
> indentations I may want.  Is there any other method for this?

Yes, sometimes it's annoying.  But setting tex-indent-basic from 2 to
0 is likely too drastic.

You should probably `M-x report-emacs-bug` asking for a new feature that
uses a 0 "basic indentation" for those cases (the hard part being to
characterize what are "those" cases), i.e. at least when the \emph{ (or
similar) is in the middle of a text line, as in your example.

> * Ghostscript program name
> ==========================
>
> Variable 'doc-view-ghostscript-program' has value - 'gs' - which is OK
> for Unix and VMS, but NOT OK for MS Windows.  It should be 'gswin32c'.
> For details look at the table in "How to Use Ghostscript", chapter
> "2.1 Help at the command line: gs -h".  Perhaps it should be done as
> list of choice?

Probably best handled as a bug: the default could simply
(executable-find "gs") and (executable-find "gswin32c") and use
whichever was found, but I'll let people more in the know of the Windows
ways decide the details.


        Stefan "thanks for the VMS thought"

Re: C-z in MSWindows shell

[VanL]

Óscar Fuentes <ofv@wanadoo.es> writes:

> Proposing improvements like this also warrant
> M-x report-emacs-bug.

The CONTRIBUTE file explains the feature request
process using 'report-emacs-bug'.

-- 
© 2019 VanL
gpg using EEF2 37E9 3840 0D5D 9183  251E 9830 384E 9683 B835
          'The mouth is the dirtiest part on the human body.' - Bill Wallace

RE: Few bugs (perhaps)

[Drew Adams]

> * Secondary selection
> =====================
> 
> When text is selected, deselecting it is done by clicking 'M-mouse-1',
> is this how it suppose to work?  Shouldn't it deselect by simply
> clicking 'mouse-1'?

No, definitely not.  The secondary selection should
remain as is, until you explicitly redefine it
(including the possibility of redefining it to be
empty).

Unlike the region, the secondary selection has no
relation to point (or mark).  It can be completely
outside the window or even not currently displayed
in any buffer (the buffer still needs to exist, but
it need not be displayed).

Re: Few bugs (perhaps)

[Eli Zaretskii]

> From: Sebastian Urban <mrsebastianurban@gmail.com>
> Date: Wed, 19 Jun 2019 10:55:38 +0200
> 
> * 'C-z' and resuming
> ====================
> 
> 'C-z' runs 'suspend-frame', which on text terminal suspends Emacs and
> to resume I should type '%emacs' in terminal.  The problem is this
> doesn't work on MS Windows Command line (cmd.exe), I can suspend Emacs
> (started with 'emacs -nw') but how do I resume it?

By typing "exit RET" at the shell prompt.

Re: Few bugs (perhaps)

[Sebastian Urban]

> The packages in GNU ELPA are forward compatibility packages, making
> it possible to use in older Emacsen (some of) the functionality of
> those packages that are builtin in more recent Emacsen.

So, an older version was left (behind)/made of - for example -
'cl-generic' to be available for emacs<25, to provide partial
(compared to current version 1.0) set of features of current
'cl-generic'.

Although there is a line "Summary: Forward cl-generic compatibility
for Emacs<25", except 'cl-lib' has "Summary: Properly prefixed CL
functions and macros", which I think should be changed to something
like in case of 'cl-generic', I'm still not sure about
naming/indication...

Separating them more explicitly from the rest, could be good idea.

Re: Few bugs (perhaps)

[Stefan Monnier]

> Although there is a line "Summary: Forward cl-generic compatibility
> for Emacs<25", except 'cl-lib' has "Summary: Properly prefixed CL
> functions and macros", which I think should be changed to something
> like in case of 'cl-generic', I'm still not sure about
> naming/indication...

Indeed, the naming for cl-lib was less explicit than that for
cl-generic.  I just pushed a change to try and improve that.


        Stefan

Re: Few bugs (perhaps)

[Sebastian Urban]

Alright - Ó. Fuentes, S. Monnier, D. Adams, E. Zaretskii - thank you
for help.  In the end I posted above problems as separate bug reports
(links for convenience):

HL line mode:
http://debbugs.gnu.org/cgi/bugreport.cgi?bug=36361

Secondary selection:
https://debbugs.gnu.org/cgi/bugreport.cgi?bug=36365

'C-z' and resuming:
http://debbugs.gnu.org/cgi/bugreport.cgi?bug=36296

Version numbers of packages:
http://debbugs.gnu.org/cgi/bugreport.cgi?bug=36360

'sentence-end-base':
http://debbugs.gnu.org/cgi/bugreport.cgi?bug=36359

Indentation in latex-mode:
http://debbugs.gnu.org/cgi/bugreport.cgi?bug=36358

Ghostscript program name:
http://debbugs.gnu.org/cgi/bugreport.cgi?bug=36357


S. U.

Re: Few bugs (perhaps)

[Xavier Maillard]

Well done !

        - xma

GPG:  BA4909B7 8F04DE1B