- From: Alessandro Vernet <avernet@orbeon.com>
- Date: Wed, 4 Apr 2007 20:52:50 -0700
- To: public-xml-processing-model-wg <public-xml-processing-model-wg@w3.org>
On 4/4/07, Norman Walsh <Norman.Walsh@sun.com> wrote: > I don't see any way to assure that db-update runs before db-query without > adding some sort of dependency control to the language. In our implementation we follow the following rule: step with no output are run in the order in which they appear in a block. In practice, the rule is not used very often, and when it is, it works fairly well, even if it does have some drawbacks. Henry, I understand your "nightmare!" reaction to this. I understand and like the workaround you propose. But what should we do with the code Norm wrote? Should it be considered invalid? Should the execution order in this case be implementation dependent? Alex -- Orbeon Forms - Web 2.0 Forms for the Enterprise http://www.orbeon.com/
Received on Thursday, 5 April 2007 03:52:53 UTC