- From: Dave Pawson <dave.pawson@gmail.com>
- Date: Tue, 11 Jan 2022 19:13:33 +0000
- To: "C. M. Sperberg-McQueen" <cmsmcq@blackmesatech.com>
- Cc: ixml <public-ixml@w3.org>
On Tue, 11 Jan 2022 at 18:46, C. M. Sperberg-McQueen <cmsmcq@blackmesatech.com> wrote: > > Dave, > > In the proposal Tomos and I brought forward, pragmas will appear > within the ixml grammar, not as a secondary input file (or stream). Clarifying (for me). An XML grammar may contain pragmas nested within the grammar. Is that correct? I'll amend. > > Any implementor may choose to accept a secondary input file, of > course, and in that case their data flow may look a lot like what > you show (assuming that the big circle labeled “iXML grammar” > pointing to “iXML text grammar” and “XML grammar” signals > that an iXML grammar takes one of those two forms). that was my intent. Clearly not clear (what!!!) OK, I would prefer to stick to one option An implementation may do .... anything (what was that game you mentioned!) I shall remove the separate stream. > > But if a processor took all its non-standard instructions from a secondary > file, there would not be much need to try to agree on a syntax for > pragmas: other processors would just not read the secondary file > in question, and they would then not need to understand its syntax. <<<<<<<<<<< Is it just me, or does this sound enticing>>>>>>>>>>>>>>>> Would it remove Stevens objections? Or at least some of them? Issue: How to link pragma 35 with line 56 of the grammar? > The reason for standardizing pragma syntax is to enable pragmas > to coexist with standard grammar notation in the grammar input to > a processor. Rev C tomorrow morning. regards -- Dave Pawson XSLT XSL-FO FAQ. Docbook FAQ.
Received on Tuesday, 11 January 2022 19:13:57 UTC