References: <FE802152-1BAD-40A0-B211-F2194DAC39E2@gmx.at>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:45.0) Gecko/20100101
            Thunderbird/45.8.0
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8; format=flowed
Message-ID:  <67607fa8-d730-2502-ce09-7dfa9b800f54@latex-project.org>
Date:         Fri, 21 Apr 2017 17:35:09 +0200
Reply-To:     Mailing list for the LaTeX3 project
              <LATEX-L@LISTSERV.UNI-HEIDELBERG.DE>
Sender:       Mailing list for the LaTeX3 project <LATEX-L@LISTSERV.UNI-HEIDELBERG.DE>
From:         Frank Mittelbach <frank.mittelbach@LATEX-PROJECT.ORG>
Subject: Re: Using d<> and babel with option spanish
To:           LATEX-L@LISTSERV.UNI-HEIDELBERG.DE
In-Reply-To:  <FE802152-1BAD-40A0-B211-F2194DAC39E2@gmx.at>
Precedence: list
Envelope-To: <rainer.schoepf@GMX.NET>
Content-Transfer-Encoding: 8bit
Status: R

Am 21.04.2017 um 16:49 schrieb Benedikt Vitecek:
> For a command I used the argument specifier d<> (which I thought was a
> good idea),
> which doesn’t work properly in combination with ‚babel‘ and the language
> option
> ‚spanish‘:
>
>   \documentclass[spanish]{scrartcl}
>   \usepackage{babel}
>   \usepackage{xparse}
>   \NewDocumentCommand \Something { d<> m }
>     {
>       Optional: #1 \\
>       Mandatory: #2
>     }
>   \begin{document}
>   \Something<Hello>{World}
>   \end{document}


the problem is that babel only enables < as an active character after 
begin document, i.e., when you make that definition in the preamble "<" 
is still a normal character (\catcode 12 presumably, not checked).

After \begin{document} it is an active char and therefore the xparse 
scanning no longer matches.

If you move the definition there your document works as expected.

In my opinion there is nothing xparse or expl could do here, it needs a 
different shorthand interface in babel --- it might be that there is 
something there to enable shortcuts already in the preamble

This is a tricky area as babel hides the fact for a reason: there are 
probably a good number of  packages that would fall over if the 
mechanism is exposed while they are setting up their internals.

Basically, this is an issue with legacy code that is difficult to 
resolve in a perfect way as long as those interfaces aren't universally 
available but are added by packages.

This is all a bit of hand-waving as I haven't looked at the babel 
internals for a while; so perhaps Javier or somebody else can give a 
more accurate answer.

frank