bug#21284: 25.0.50; `completion-file-name-table' should/not use `nreverse'?

bug#21284: 25.0.50; `completion-file-name-table' should/not use `nreverse'?

[Keith David Bershatsky]

The function `completion-file-name-table' uses `(setq all (nreverse comp))' instead of `(setq all comp)'.

This has the effect of reversing what was accomplished when `pcomplete-compare-entry-function' is set to `'string-lessp'.

In a nutshell, the completions are in reverse alphabetical order as a result of `(setq all (nreverse comp))'.

This was tested in an eshell buffer by typing at a prompt:  rm [space tab]; and, with `lawlist-eshell-cmpl-compare-entry-function` set to `'string-lessp'.

Thanks,

Keith

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

In GNU Emacs 25.0.50.1 (x86_64-apple-darwin10.8.0, NS appkit-1038.36 Version 10.6.8 (Build 10K549))
 of 2015-08-15 on server.local
Repository revision: 118363 mina86@mina86.com-20141117204835-paxrwolqd1cgwtdp
Windowing system distributor `Apple', version 10.3.1038
Configured using:
 `configure --with-ns --without-imagemagick'

Configured features:
ACL LIBXML2 ZLIB

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

Major mode: ELISP

Minor modes in effect:
  tb-mode: t
  sb-mode: t
  ml-mode: t
  fl-mode: t
  ds-mode: t
  sd-mode: t
  bc-mode: t
  +-mode: t
  buffer-read-only: t

Recent messages:
Wrote /Users/HOME/.0.data/.0.emacs/.lock
Emacs:  completion-file-name-table

Load-path shadows:
None found.

Features:
(shadow emacsbug sendmail lawlist-ztree lawlist-yas lawlist-ws
lawlist-wl elmo-imap4 elmo-localdir modb-standard modb-legacy
elmo-internal elmo-flag mmelmo-imap mmelmo-buffer elsp-generic mel-u
ps-print ps-def lpr parse-time enriched lawlist-w3m browse-url
doc-view image-mode ccl lawlist-vl lawlist-view lawlist-undo
lawlist-txt lawlist-tm lawlist-tex compare-w lawlist-tabbar
lawlist-speedbar lawlist-shell info esh-groups ehelp ange-ftp
lawlist-sgml lawlist-sb lawlist-saveplace lawlist-ruler mule-util
lawlist-replace lawlist-rectangle lawlist-re-builder lawlist-python
skeleton lawlist-profiler lawlist-print lawlist-php lawlist-perl
lawlist-parens lawlist-org lawlist-calendar org-agenda org org-macro
org-footnote org-pcomplete org-list org-faces org-entities org-version
ob-emacs-lisp ob ob-tangle ob-ref ob-lob ob-table ob-exp org-src
ob-keys ob-comint ob-core ob-eval org-compat org-macs org-loaddefs
holidays hol-loaddefs cal-menu calendar cal-loaddefs lawlist-neotree
lawlist-movement lawlist-mouse lawlist-ml lawlist-minibuffer
lawlist-misc lawlist-messages lawlist-mc rect lawlist-markdown
noutline outline lawlist-lorem lawlist-ln lawlist-keymap lawlist-js
lawlist-ispell lawlist-isearch lawlist-imenu lawlist-ibuffer
lawlist-hl lawlist-grep lawlist-git ido view vc-git vc vc-dispatcher
url-http tls url url-proxy url-privacy url-expand url-methods
url-history mailcap url-auth url-cookie url-domsuf url-gw thingatpt
time-stamp server log-view log-edit message mml mml-sec mm-decode
mm-bodies mm-encode gmm-utils mailheader pcvs-util add-log
lawlist-help quail help-mode json grep compile find-lisp ediff-merg
ediff-wind ediff-diff ediff-mult ediff-help ediff-init ediff-util
ediff ert find-func ewoc eieio-base diff-mode conf-mode autorevert
filenotify lawlist-frame lawlist-frame-bufs lawlist-font-lock
lawlist-fm lawlist-env lawlist-elscreen lawlist-elisp lawlist-dv
jka-compr lawlist-image lawlist-files zeroconf url-util url-parse
auth-source eieio byte-opt bytecomp byte-compile cl-extra cconv
eieio-core password-cache url-vars dbus xml lawlist-ds lawlist-dired
epa epg epg-config format-spec edmacro kmacro dired lawlist-desktop
frameset lawlist-debug lawlist-window debug lawlist-css smie
lawlist-compile rx lawlist-color lawlist-cm pcase lawlist-cc cc-langs
cc-mode cc-fonts cc-guess cc-menus cc-cmds cc-styles cc-align
cc-engine cc-vars cc-defs cc-bytecomp lawlist-calc lawlist-calccomp
lawlist-calcalg3 lawlist-calcalg2 lawlist-calc+ easymenu lawlist-bk
lawlist-bc lawlist-bbdb gnus gnus-ems nnheader gnus-util mail-utils
wid-edit mail-parse rfc2231 rfc2047 rfc2045 ietf-drums mm-util
mail-prsvr mailabbrev mail-extr rfc822 cl timezone lawlist-as
lawlist-archive cl-macs gv cl-loaddefs cl-lib lawlist-+ disp-table
easy-mmode derived advice help-fns shell pcomplete comint ansi-color
ring savehist time-date tooltip eldoc electric uniquify ediff-hook
vc-hooks lisp-float-type mwheel ns-win tool-bar dnd fontset image
regexp-opt fringe tabulated-list newcomment elisp-mode lisp-mode
prog-mode register page menu-bar rfn-eshadow timer select scroll-bar
mouse jit-lock font-lock syntax facemenu font-core frame cham georgian
utf-8-lang misc-lang vietnamese tibetan thai tai-viet lao korean
japanese hebrew greek romanian slovak czech european ethiopic indian
cyrillic chinese case-table epa-hook jka-cmpr-hook help simple abbrev
minibuffer nadvice loaddefs button faces cus-face macroexp files
text-properties overlay sha1 md5 base64 format env code-pages mule
custom widget hashtable-print-readable backquote make-network-process
cocoa ns multi-tty emacs)

Memory information:
((conses 16 2078627 153097)
 (symbols 48 83114 0)
 (miscs 40 560 467)
 (strings 32 191705 23218)
 (string-bytes 1 6865927)
 (vectors 16 40524)
 (vector-slots 8 960694 10088)
 (floats 8 2574 897)
 (intervals 56 730 61)
 (buffers 976 12))

bug#21284: 25.0.50; `completion-file-name-table' should/not use `nreverse'?

[Marcin Borkowski]

On 2015-08-17, at 10:25, Keith David Bershatsky <esq@lawlist.com> wrote:

> The function `completion-file-name-table' uses `(setq all (nreverse comp))' instead of `(setq all comp)'.
>
> This has the effect of reversing what was accomplished when `pcomplete-compare-entry-function' is set to `'string-lessp'.
>
> In a nutshell, the completions are in reverse alphabetical order as a result of `(setq all (nreverse comp))'.
>
> This was tested in an eshell buffer by typing at a prompt:  rm [space tab]; and, with `lawlist-eshell-cmpl-compare-entry-function` set to `'string-lessp'.

Hi,

I tried to reproduce it on GNU Emacs 25.1.50.9 (commit fafdfcb) and
failed.  However, I'm not sure whether I understood your report
correctly.  My Emacs doesn't seem to have
`lawlist-eshell-cmpl-compare-entry-function', only
`eshell-cmpl-compare-entry-function'.  Did you mean that variable?
I tried to set it to `'string-lessp', but nothing seemed to change after
"rm [space] [tab]" in Eshell.

> Thanks,
>
> Keith

Best,

-- 
Marcin

bug#21284: 25.0.50; `completion-file-name-table' should/not use `nreverse'?

[Keith David Bershatsky]

Thank you, Marcin, for taking a look at 21284.

I apologize for mislabeling the variable.  It turns out that the same erroneous behavior is observable with Emacs -Q.  The `*Completions*` are in reverse alphabetical order.

Step 1:  Launch the graphical version of Emacs -Q.

Step 2:  M-x eshell

Step 3:  cd to a directory containing files and folders with a variety of different names beginning with letters a to z.

Step 4:  type rm

Step 5:  hit the space bar

Step 6:  hit the tab key

Step 7:  The *Completions* are in reverse alphabetical order.

Thanks,

Keith

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

> Hi,
> 
> I tried to reproduce it on GNU Emacs 25.1.50.9 (commit fafdfcb) and
> failed.  However, I'm not sure whether I understood your report
> correctly.  My Emacs doesn't seem to have
> `lawlist-eshell-cmpl-compare-entry-function', only
> `eshell-cmpl-compare-entry-function'.  Did you mean that variable?
> I tried to set it to `'string-lessp', but nothing seemed to change after
> "rm [space] [tab]" in Eshell.
> 
> > Thanks,
> >
> > Keith
> 
> Best,
> 
> -- 
> Marcin

bug#21284: eshell does not sort completion candidates

[Lars Ingebrigtsen]

Keith David Bershatsky <esq@lawlist.com> writes:

> Step 1:  Launch the graphical version of Emacs -Q.
>
> Step 2:  M-x eshell
>
> Step 3: cd to a directory containing files and folders with a variety
> of different names beginning with letters a to z.
>
> Step 4:  type rm
>
> Step 5:  hit the space bar
>
> Step 6:  hit the tab key
>
> Step 7:  The *Completions* are in reverse alphabetical order.

(I'm going through old bug reports that unfortunately weren't resolved
at the time.)

I tried to reproduce this in Emacs 29, but was unable to.  Are you still
seeing this problem in recent Emacs versions?

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no

bug#21284: eshell does not sort completion candidates

[Lars Ingebrigtsen]

Lars Ingebrigtsen <larsi@gnus.org> writes:

> I tried to reproduce this in Emacs 29, but was unable to.  Are you still
> seeing this problem in recent Emacs versions?

More information was requested, but no response was given within a
month, so I'm closing this bug report.  If the problem still exists,
please respond to this email and we'll reopen the bug report.

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no

bug#21284: 25.0.50; `completion-file-name-table' should/not use `nreverse'?

[Noam Postavsky]

tag 21284 unreproducible
quit

I'm getting completions in what looks like a random order, here are
the first few lines for completions in the emacs src directory:

Possible completions are:
undo.o     ChangeLog.12     eval.o     insdel.o
xfaces.o     sysdep.o     category.o     marker.o
search.c     ChangeLog.10     fontset.o     dynlib.h
kqueue.c     deps/     lastfile.o     dired.o
insdel.c     w32fns.c     textprop.o     xselect.c
inotify.c     doprnt.c     cmds.c     xdisp.c
data.o     fontset.c     xsettings.h     character.o
keymap.c     xsettings.o     emacsgtkfixed.h     systime.h
syntax.h     bytecode.o     character.h     deps.mk

bug#21284: 25.0.50; `completion-file-name-table' should/not use `nreverse'?

[Eli Zaretskii]

> From: Noam Postavsky <npostavs@users.sourceforge.net>
> Date: Sun, 5 Jun 2016 22:50:10 -0400
> Cc: Keith David Bershatsky <esq@lawlist.com>, Marcin Borkowski <mbork@mbork.pl>
> I'm getting completions in what looks like a random order, here are
> the first few lines for completions in the emacs src directory:

I suspect that there's a hidden factor here: the order in which the
'readdir' libc function returns the directory entries.  On GNU/Linux,
the order is arbitrary (AFAIK, it's the order of the entries in the
directory file).  On MS-Windows, the order is alphabetic, so reversing
will produce the reverse alphabetic order.  I don't know what happens
on OS X.

bug#21284: 25.0.50; `completion-file-name-table' should/not use `nreverse'?

[Noam Postavsky]

tag 21284 - unreproducible
severity 21284 minor
quit

On Mon, Jun 6, 2016 at 10:43 AM, Eli Zaretskii <eliz@gnu.org> wrote:
> I suspect that there's a hidden factor here: the order in which the
> 'readdir' libc function returns the directory entries.  On GNU/Linux,
> the order is arbitrary (AFAIK, it's the order of the entries in the
> directory file).  On MS-Windows, the order is alphabetic, so reversing
> will produce the reverse alphabetic order.  I don't know what happens
> on OS X.

Oh yeah, I see reversed order on Windows.

On Mon, Jun 6, 2016 at 12:19 PM, Keith David Bershatsky <esq@lawlist.com> wrote:
> I am on OS X and the default sort order is alphabetic, which gets reversed by `completion-file-name-table` because it uses `nreverse'.

Actually, the result from `file-name-all-completions` is already
reversed. `completion-file-name-table` then reverses it twice: once in
the loop (dolist (tem all) (... (push tem comp))), and the `nreverse`
call to undo it.

bug#21284: 25.0.50; `completion-file-name-table' should/not use `nreverse'?

[Keith David Bershatsky]

I am on OS X and the default sort order is alphabetic, which gets reversed by `completion-file-name-table` because it uses `nreverse'.

Keith

At Mon, 06 Jun 2016 17:43:04 +0300,
Eli Zaretskii wrote:
> 
> . . .
> 
> I suspect that there's a hidden factor here: the order in which the
> 'readdir' libc function returns the directory entries.  On GNU/Linux,
> the order is arbitrary (AFAIK, it's the order of the entries in the
> directory file).  On MS-Windows, the order is alphabetic, so reversing
> will produce the reverse alphabetic order.  I don't know what happens
> on OS X.

bug#21284: 25.0.50; `completion-file-name-table' should/not use `nreverse'?

[Eli Zaretskii]

> Date:  Mon, 06 Jun 2016 09:19:07 -0700
> From:  Keith David Bershatsky <esq@lawlist.com>
> Cc:  21284@debbugs.gnu.org,Noam Postavsky <npostavs@users.sourceforge.net>,Marcin Borkowski <mbork@mbork.pl>
> 
> I am on OS X and the default sort order is alphabetic, which gets reversed by `completion-file-name-table` because it uses `nreverse'.

Then what exactly is the problem here?  Emacs doesn't know in what
order the files are returned, it considers the order random.
Reversing a randomly ordered list produces another randomly ordered
list.  Applications that need to present the list in some specific
order need to do that explicitly.  For example, "C-x C-f TAB TAB" does
display the files in alphabetic order for me, not in reverse
alphabetic order.

bug#21284: 25.0.50; `completion-file-name-table' should/not use `nreverse'?

[Noam Postavsky]

retitle 21284 eshell does not sort completion candidates
quit

On Mon, Jun 6, 2016 at 2:50 PM, Eli Zaretskii <eliz@gnu.org> wrote:
> Then what exactly is the problem here?  Emacs doesn't know in what
> order the files are returned, it considers the order random.
> Reversing a randomly ordered list produces another randomly ordered
> list.  Applications that need to present the list in some specific
> order need to do that explicitly.  For example, "C-x C-f TAB TAB" does
> display the files in alphabetic order for me, not in reverse
> alphabetic order.

So the problem is that eshell's completion doesn't sort its candidates.

bug#21284: 25.0.50; `completion-file-name-table' should/not use `nreverse'?

[Keith David Bershatsky]

If there is/are valid reason(s) to reverse the order, then perhaps consider mentioning one or more of those reasons in a comment within `completion-file-name-table' along with a mention of bug 21284.

If there is no valid reason whatsoever to reverse the order, then perhaps consider removing `nreverse` from `completion-file-name-table`.

If the issue is "de minimus", then the bug report can be closed without further ado.

Keith

At Mon, 06 Jun 2016 21:50:05 +0300,
Eli Zaretskii wrote:
> 
> . . .
> 
> Then what exactly is the problem here?  Emacs doesn't know in what
> order the files are returned, it considers the order random.
> Reversing a randomly ordered list produces another randomly ordered
> list.  Applications that need to present the list in some specific
> order need to do that explicitly.  For example, "C-x C-f TAB TAB" does
> display the files in alphabetic order for me, not in reverse
> alphabetic order.

bug#21284: 25.0.50; `completion-file-name-table' should/not use `nreverse'?

[Eli Zaretskii]

> Date:  Mon, 06 Jun 2016 13:00:40 -0700
> From:  Keith David Bershatsky <esq@lawlist.com>
> Cc:  21284@debbugs.gnu.org,Marcin Borkowski <mbork@mbork.pl>,Noam Postavsky <npostavs@users.sourceforge.net>
> 
> If there is/are valid reason(s) to reverse the order, then perhaps consider mentioning one or more of those reasons in a comment within `completion-file-name-table' along with a mention of bug 21284.
> 
> If there is no valid reason whatsoever to reverse the order, then perhaps consider removing `nreverse` from `completion-file-name-table`.
> 
> If the issue is "de minimus", then the bug report can be closed without further ado.

I believe the reason is the usual one: we create the list by pushing
each new element onto it, so the result comes in reverse order of the
files.  The call to nreverse then restores the original order.  IOW,
this is an implementation detail.