Re: weird German error

[David Kastrup]

Patrick McCarty <address@hidden> writes:

> On 2009-09-29, Patrick McCarty wrote:
>> On 2009-09-29, Graham Percival wrote:
>> > On Tue, Sep 29, 2009 at 06:26:08PM +0200, David Kastrup wrote:
>> > > Graham Percival <address@hidden> writes:
>> > > 
>> > > > l.2850 \entry {\code \"
>> > > >                        Uber die Nicht-Schachtelung von Klammern und 
>> > > > Bindeb\"...
>> > > 
>> > > It would appear that \code takes an argument, so the usual usage would
>> > > be \code{...}.  In the absence of explicit braces, the next token is
>> > > digested.  This gets turned into
>> > > \code{\"}Uber die Nicht-...
>> > > effectively, and the \" finds nothing to apply to.
>> > 
>> > I'll be spending tomorrow on more GUB stuff, so if somebody could
>> > look into this in more detail, that would be great.
>> > (unless Patrick's recent modifications to python/auxiliar and
>> > stepmake/stepmake fixes this!)
>> 
>> I just reproduced it again with `make -j1 doc'.  I'll look through the
>> logs again to see if I can figure it out.
>
> I just pushed a commit that works for me.
>
> I don't know why it works, since those lines were changed back in
> July, and I've had many successful builds since then.

You should probably change it back.

> I *do* suspect there's a problem with the @lilyvindex macro, though.
> But I can't pinpoint a reliable case for failure.

Well, isn't it obvious from my guess above?

diff --git a/Documentation/macros.itexi b/Documentation/macros.itexi
index 33b154e..16cf580 100644
--- a/Documentation/macros.itexi
+++ b/Documentation/macros.itexi
@@ -523,7 +523,7 @@ For LilyPond version @version{}
 @c delimits arguments in the standard way (i.e., with braces).
 
 @tex
-\gdef\lilyvindex#1{\doind{vr}{\code #1}\ignorespaces}
+\gdef\lilyvindex#1{\doind{vr}{\code{#1}}\ignorespaces}
 @end tex
 
 @macro rglos{TEXT}

-- 
David Kastrup