emacs-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Interactive hat. (Patch V2)


From: Alan Mackenzie
Subject: Re: Interactive hat. (Patch V2)
Date: Wed, 29 Apr 2009 11:36:02 +0000
User-agent: Mutt/1.5.9i

Hi, Eli!

On Wed, Apr 29, 2009 at 10:22:09AM +0300, Eli Zaretskii wrote:
> > Date: Tue, 28 Apr 2009 22:33:03 +0000
> > Cc: address@hidden, address@hidden, address@hidden,
> >   address@hidden
> > From: Alan Mackenzie <address@hidden>


> > > That's not what I meant.  I meant that you in effect have here
> > > @item's without the text after them.  A @table is not supposed to
> > > be like that, so who knows what will the output be?  In particular,
> > > HTML and XML outputs may assume there always be some text, and if
> > > not, fail to properly close the markup.

> > Surely it's not unusual (in any markup language) to have a blank cell
> > in a table.  I've generated HTML, and it's fine.  I've generated XML,
> > and it looks fine too, as much as XML ever looks fine (though I don't
> > know offhand if I've got a suitable viewing program for it).

> I had no doubt you tested your changes.  That is why I said that this
> might be a problem _in_some_future_version_ of Texinfo, see above.  In
> particular, the next version of Texinfo is expected to toss the C
> implementation of makeinfo and instead use texi2html, which is a Perl
> program.  That means all the undocumented (mis)features of makeinfo
> will give way to other undocumented (mis)features.

Hmm.  I can foresee there's going to be a lot of gnashing of teeth, lots
of rants like mine written, perhaps a few divorces caused by this coming
change.  I don't think the Info format is quite as simple as it appears.
A lot of projects will just stick with the old C version (i.e. recommend
it to its users), much like Emacs recommends the fairly old Texinfo 4.3
or later.  I think Stefan M.'s approach, face the problem when we come to
it, is the best here.

> > Have you never written @example code yourself, where the enforced
> > indentation has caused lines to become "too long"?

> Sure, but I always find a way to break those into several lines.

> > Anyhow, I've put @example in now (as requested by Miles and
> > yourself), and split some more lines up.  There are, however, one or
> > two lines which go over C74 which can't sensibly be split.

> If you show those lines, perhaps someone can suggest a way of
> splitting them that does make sense.

OK.  Here's a bit from page "Non-string Interactive", exactly as it
appears in the info file after @verbatim has been replaced by @example:

                                                                     70  74
                                                                      |   |
#################################################################################

    `@' - Select window of mouse event
               (interactive
                (let ((events (this-command-keys-vector))
                      e w
                      (i 0))
                  (while (and (< i (length events))
                              (not (consp (setq e (aref events i)))))
                    (setq i (1+ i)))
                  (when (consp e)
                    (setq w (car (cadr e)))
                    (if (windowp w)
                        (if (and (window-minibuffer-p w)
                                 (> (minibuffer-depth) 0))
                            (error
                             "Attempt to select inactive minibuffer window")))
                    (run-hooks mouse-leave-buffer-hook)
                    (select-window w))
                  nil))
          Note that this form only works when the KEYS parameter to
          `call-interactively' (see Interactive Call) is `nil'
          (which it almost always is).

#################################################################################
                                                                      |   |
                                                                     70  74

The "long" string, "Attempt to select inactive minibuffer window" was
taken from Fcall_interactively in callint.c.  But, on the other hand, why
the fuss?  There are menu items which are longer, e.g. this one on page
"Keymaps":

* Scanning Keymaps::            Looking through all keymaps, for printing help.

-- 
Alan Mackenzie (Nuremberg, Germany).




reply via email to

[Prev in Thread] Current Thread [Next in Thread]