- From: David Booth <david@dbooth.org>
- Date: Sat, 26 May 2012 21:58:01 -0400
- To: Andy Seaborne <andy.seaborne@epimorphics.com>
- Cc: public-rdf-dawg-comments@w3.org
Yes, Steve Harris pointed out my mistake already. Sorry for the error. I misread the BNF. David On Sat, 2012-05-26 at 14:43 +0100, Andy Seaborne wrote: > David, > > It is allowed already: > > [29] Update = Prologue ( Update1 ( ';' Update )? )? > > 'Update' can be empty hence > > ( ';' <empty> )? > > which is a trailing ';' > > http://www.sparql.org/update-validator.html > > Andy > > On 25/05/12 17:56, David Booth wrote: > > The SPARQL Update syntax currently PROHIBITS a trailing semicolon at the > > end of the last update statement: > > > > [30] Update ::= Prologue ( Update1 ( ';' Update )? )? > > > > although it REQUIRES a semicolon after all other update statements in an > > update operation. I realize that the semicolon was intended to be a > > separator rather than a terminator, but in practice this restriction > > just becomes a pointless annoyance to users. Programming language > > designers learned this lesson many years ago. > > > > Please relax this restriction to allow extra semicolons. > > > > Thanks! > > > > > > -- David Booth, Ph.D. http://dbooth.org/ Opinions expressed herein are those of the author and do not necessarily reflect those of his employer.
Received on Sunday, 27 May 2012 01:58:31 UTC