W3C home > Mailing lists > Public > public-qt-comments@w3.org > February 2004

ORA-XQ-153-B: rules for permissible comments etc. are not kind to humans

From: Stephen Buxton <Stephen.Buxton@oracle.com>
Date: 16 Feb 04 13:32:52
Message-Id: <200402162132.i1GLWqb04553@rgmgw4.us.oracle.com>
To: public-qt-comments@w3.org

SECTION A.2.2: lexical rules

The rules in the tables do not appear to allow comments,
pragmas and must-know extensions as ignorable whitespace 
in the middle of certain sequences of tokens.  For example,
in the DEFAULT state table, first row, last list item,
<"declare" "validation">, or more importantly, fourth row,
items <"for" "$">, <"let" "$">, <"some" "$">, <"every" "$">.
I believe the intention is to permit comments, pragmas and 
must-know extensions in the midst of these token sequences.
Note that there is no recognition of just the first token
in these sequences ("declare", "for", "let", "some", "every")
so that for example

  for (: set up loop :) $i in fn:doc("...")

is unrecognizable by these tables, and presumably a lexical
error.  Glancing at the tables, it seems that whenever
you enclose a sequence of patterns in angle brackets <...>,
that it is not permitted to have a comment.  This seems far
too limiting.  How will users ever learn where they can insert

- Steve B.
Received on Monday, 16 February 2004 16:32:54 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:45:17 UTC