Hi Wolfgang,

Von: Wolfgang Schuster <wolfgang.schuster= .lists@gmail.com>
Gesendet: Samstag, 15. Januar 2022 20:28
An: mailing list for ConTeXt users <ntg-context@ntg.nl>; Denis= Maier via ntg-context <ntg-context@ntg.nl>
Cc: Maier, Denis Christian (UB) <denis.maier@unibe.ch>
Betreff: Re: [NTG-context] XML, dealing with whitespace

Denis Maier via ntg-context schrieb am 15.01.2022 um= 13:04:

Hi all,

I have sources that look like t= his:

%%%%%%%%%%%%%%%%%%%%%

<?xml version=3D"1.0&qu= ot; encoding=3D"UTF-8"?>

<article>

   <p>Bla Bla B= la</p>

   <p>

      <underline>

         <= ;italic>Bla</italic>

      </underline>, B= la Bla.</p>

</article>

%%%%%%%%%%%%%%%%%%%%%

Typesetting this with context g= ives me a spurious space after the underlined Bla in italics.

There is no spurious space, the line break is just converted to a space and= I see no reason why this shouldn't happen. To remove space before or after= certain parts of text within a paragraph you can use the \removeunwantedsp= ace and \ignorespaces commands.

Yes, it’s absolutely true. From tex’s po= int of view, the space is not spurious. It’s absolutely adaquate to t= reat the newline as a space here.

As I’ve outlined in my original post the problem occurs becau= se xslt adds these indentations here. FWIW, I finally found this solution, = which seems has been added to xslt 3.0 (after being available as a saxon extension: there's a new attribute &laqu= o;suppress-indentation» on xsl:output that can be used to control thi= s:
«This is a new property in XSLT 3.0 (it was previously available in S= axon as an extension). The value is a whitespace-separated list of element = names, and it typically identifies "inline" elements that should = not cause indentation; in XHTML, for example, these would be b, i, span, and the like.&ra= quo;

(https://www.saxonica.com/documentation9.5/xsl-e= lements/output.html)

So, the solution to my problem is this:

<xsl:output
   method=3D"xml"
   indent=3D"yes"
   suppress-indentation=3D"italic underline"
   />

Denis

%%%% begin example
\starttexdefinition RemovePreceding #1
    \removeunwantedspaces
    #1
\stoptexdefinition

\starttexdefinition RemoveFollowing #1
    #1
    \ignorespaces
\stoptexdefinition

\starttext

Bla \RemovePreceding{Bla} Bla

Bla \RemoveFollowing{Bla} Bla

\stoptext
%%%% end example

When only following spaces are a problem a better alternative to \ign= orespace is \autoinsertnextspace which checks the following token which ens= ures there is space when the next character is punctuation.

%%%% begin example
\starttexdefinition Italic #1
    \emphasized{#1}
    \autoinsertnextspace
\stoptexdefinition

\starttexdefinition Underbar #1
    \underbar{#1}
\stoptexdefinition

\starttext

Bla Bla Bla

\Underbar{\Italic{Bla} , Bla Bla.}

\stoptext
%%%% end example

Wolfgang