From: Drew Adams <drew.adams@oracle.com>
To: Eli Zaretskii <eliz@gnu.org>
Cc: 20105@debbugs.gnu.org
Subject: bug#20105: 25.0.50; Emacs manual, `i HOME RET' sends you to `Moving Point', which is wrong
Date: Sat, 14 Mar 2015 09:44:09 -0700 (PDT) [thread overview]
Message-ID: <1cf950ce-96bf-4a06-b104-581af13b6b74@default> (raw)
In-Reply-To: <<83zj7f1rox.fsf@gnu.org>>
> Yes, "emacs -Q" here. And it's not a surprise you have those missing
> candidates in the index, that's how I can see them ;-)
>
> Are you sure you did 'i home TAB' in the latest manual, though?
Yes, exactly that, in this build:
In GNU Emacs 25.0.50.1 (i686-pc-mingw32)
of 2015-02-27 on LEG570
Repository revision: b2a590d4e3dc692a97c1b53e015b945d84b4b4c7
Windowing system distributor `Microsoft Corp.', version 6.1.7601
Configured using:
`configure --host=i686-pc-mingw32 --enable-checking=yes,glyphs'
> > > > But why is `HOME' capitalized as a candidate if it points to info
> > > > about the key?
> > >
> > > Don't know. Looks like some feature of completion.
> >
> > Really? My guess is instead that it comes from this explicit index
> > entry (which I see in Emacs 24 but not 23): (I removed some whitespace.)
> >
> > * HOME: Moving Point. (line 57)
>
> By "feature" I meant that it replaces "home" which I typed by "HOME",
> for whatever reasons. The _only_ reason that I could accept as
> legitimate is if _all_ completion candidates started with an
> upper-case "HOME". But that's not what we have here. So it looks
> like some attempt at being smart is misfiring.
Yes, I guess so.
But I'm guessing that what happens might have to do with searching the
indexes in order, where Key Index comes first. And in that index,
`HOME' is the only entry matching input `home'. It is also the only
matching entry in the Variable Index.
> > That's from the Key Index. However, note that there are also these
> > two entries in the Variable Index, which seem not to be used when
> > I do `i home TAB':
> >
> > * HOME: General Variables. (line 59)
> > * HOSTNAME: General Variables. (line 70)
>
> HOME _is_ used, except that completion removes duplicates (I guess).
I guess so too. That is a mistake. The calling function should
decide whether completion should remove duplicates (IMHO). And in
this case, it should not (IMHO).
> As for HOSTNAME -- why should it be used? I don't see it used here.
My bad for mentioning HOSTNAME. Might have been a copy&paste error.
Or it might have reflected a senior moment.
> > From what I see, the only explicit index entries that include `HOME'
> > (capitalized) refer to the home directory, not to the <home> key.
>
> No, the entry that revers to "Moving Point" is also capitalized.
Right. Do you agree that that is wrong? I doubt that that is the
only problem here. But it would be a start, to fix that entry.
It should be changed to something like `home keyboard key' or `home
key (keyboard)', I think.
(We might also want to add an entry for `<home> keyboard key'.)
Also, I question whether the quote marks should be included in these
two index entries:
* 'HOME' directory on MS-Windows: Windows HOME. (line 6)
* 'HOME' directory under MS-DOS: MS-DOS File Names. (line 35)
Et voila: That's probably why typing `host' does not match either of
these entries. And it's perhaps why it does work for you: the curly
quotes are perhaps not present for you, in the indexes.
---
Interestingly, copying and pasting those into this mail shows them
as curly quotes, but in Info they appear to be straight. They are
LEFT SINGLE QUOTATION MARK etc. I guess it is the default Info font
that makes them appear straight.
But I thought that we had arranged to use the older makeinfo or
whatever, and that Emacs was arranging to keep simple backtick and
apostrophe in the produced Info buffers. No? I see now that Isearch
for ` or ' fails miserably in Info, wrt finding such quoted names.
This makes it hard for users who want to use keyboard keys ` and '
for searching. That's a clear regression in usefulness, IMHO.
next prev parent reply other threads:[~2015-03-14 16:44 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <<265aed11-7056-45f2-afbf-1b5f8b3b0a05@default>
[not found] ` <<83egos2df6.fsf@gnu.org>
2015-03-14 14:46 ` bug#20105: 25.0.50; Emacs manual, `i HOME RET' sends you to `Moving Point', which is wrong Drew Adams
2015-03-14 15:26 ` Eli Zaretskii
[not found] ` <<b15373cd-5f12-4dde-b999-d0b54c82aa52@default>
[not found] ` <<834mpn38iu.fsf@gnu.org>
2015-03-14 15:59 ` Drew Adams
2015-03-14 16:15 ` Eli Zaretskii
[not found] ` <<9fe72841-fa75-4bbe-bb92-c14e68e0cd7a@default>
[not found] ` <<83zj7f1rox.fsf@gnu.org>
2015-03-14 16:44 ` Drew Adams [this message]
2015-03-14 17:39 ` Eli Zaretskii
2015-03-14 17:41 ` Drew Adams
[not found] <<1cf950ce-96bf-4a06-b104-581af13b6b74@default>
[not found] ` <<83twxn1nsx.fsf@gnu.org>
2015-03-14 17:44 ` Drew Adams
2015-03-14 2:23 Drew Adams
2015-03-14 8:25 ` Eli Zaretskii
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
List information: https://www.gnu.org/software/emacs/
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1cf950ce-96bf-4a06-b104-581af13b6b74@default \
--to=drew.adams@oracle.com \
--cc=20105@debbugs.gnu.org \
--cc=eliz@gnu.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
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).