- From: Norman Walsh <ndw@nwalsh.com>
- Date: Thu, 14 May 2015 17:38:32 -0500
- To: XProc Dev <xproc-dev@w3.org>
- Message-ID: <87oalmiy3b.fsf@nwalsh.com>
Florent Georges <fgeorges@fgeorges.org> writes: > Actually I found what looks like a bug. When one instantiate a new > XProcConfiguration using an existing Saxon processor object, init() > does not get called, so neither findStepClasses() and > findExtensionFunctions(). So the step implementation classes cannot > be configured. Right, that's certainly a bug. > I guess XProcConfiguration(Processor) should call init(), maybe > adapting it a bit to take an already existing processor into account. > > By the way, that constructor already calls loadConfiguration(), > which in turn changes configuration settings in order to parse the > config files, not sure this is really intended (to change the > externally provided processor config for internal parsing purposes). > > Norm, do you want me to open a ticket on GitHub? Yes, please. There's a chicken and egg problem in parsing configurations. I need a parser to do it, but some configuration changes require getting a different parser configuration. But I don't think it should be changing the parser you passed in. And sorry for the false lead earlier, should have read the whole thread first, but was in a hurry and trying to be helpful. Be seeing you, norm -- Norman Walsh Lead Engineer MarkLogic Corporation Phone: +1 512 761 6676 www.marklogic.com
Received on Thursday, 14 May 2015 22:38:58 UTC