texinfo-devel
[Top][All Lists]
Advanced

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

Re: @U attempt


From: Patrice Dumas
Subject: Re: @U attempt
Date: Mon, 9 Feb 2015 00:47:26 +0100
User-agent: Mutt/1.5.20 (2009-12-10)

On Sun, Feb 08, 2015 at 10:37:00PM +0000, Karl Berry wrote:
> So, I attempted to implement @U (didn't do the user documentation or
> tests yet).  What I did seems to get the right output, but I confess I'm
> not at all sure I did the right thing -- especially to get the text of
> the argument.  I read the pod docs that I could find, but probably
> missed some crucial stuff; the elements of the overall tree+list data
> structure remain opaque to me, I'm afraid.  Patrice, if you wrote up
> some overall documentation about that, please send a pointer.

No documentation, but I sent a reply on your commit message that you
could use what is used for errormsg in the Parser as the command text is
used for that command right in the Parser.

> Also, I just assumed the arg would be literal hex digits, that is, not
> supporting things like @address@hidden or @address@hidden  I
> know the code could easily handle such, but I felt like enough time had
> passed already without delving into that.

Unless I missed something, since you take the text from the tree
correctly, @value and @somemacro will be expanded.  And in fact it would
be harder to avoid expansion of @value or @somemacro.  I checked that
it is also expanded in @errormsg, so it should be expanded in the text
you get if you use similar code to get the teext in the parser to do an
error message there.  (The reason why @value and @somemacro are expanded
is that the 'errormsg' command is handled in the Parser.pm after the
closing brace has been found and the command has been put in the tree,
so with aall @value and @somemacro expanded).

> Below are the important diffs (not in applyable form; I committed the
> actual changes).  Patrice, if you could give it a quick look, that would
> be greatly appreciated.  If I'm totally on the wrong track, just say so,
> and I'll revert it and we won't have @U in the forthcoming release.  Not
> the end of the world.

Lokks pretty good to me except that the check of argument should be in the 
parser.

-- 
Pat



reply via email to

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