From: Fredrik L. <Fre...@im...> - 2010-05-05 16:32:28
|
Apparently the mzML 1.1 schema is fine, but the idx needs to have one space removed. Since it was recently updated to 1.1.1 it should be ok to update it once again? Maybe with version bump, even if it is not an actual functional schema change (more than for buggy software). Fredrik On 2010-05-05 18:26, Matthew Chambers wrote: > Ah, I see that the change actually was made in the XSD in the CVS repo > on 6/3/2009, but it was never propagated to the official location. I > guess perhaps we agreed to do a silent update at that time? I can't > remember the discussion in the call. > > -Matt > > > On 5/5/2010 11:21 AM, Matthew Chambers wrote: > >> Hi David, >> >> I'm sure we want to accommodate reasonable limitations on the schema due >> to validation software bugs, but a change to a released schema might be >> going too far. We should have gotten this into 1.1 (it was mentioned in >> the next WG conference call notes), but it unfortunately got lost in the >> ether. :( >> >> Is it a reasonable solution for people stuck with an affected version of >> Xerces to simply catch and ignore the buggy error? Otherwise we'd be >> looking at either changing the schema with no version bump (thus >> violating our own guidelines) or bumping to mzML 1.1.1 and mzML_idx 1.1.2. >> >> -Matt >> >> >> On 5/1/2009 6:27 AM, David Creasy wrote: >> >> >>> Hi, >>> >>> I've got a couple of issues with the example docs and schema not >>> validating. The first is that we use a rather old version of Xerces. >>> There's a known problem: >>> >>> http://marc.info/?l=xerces-c-dev&m=114345502019802&w=2 >>> >>> The solution is simple, just change, for example: >>> <xs:selector xpath=".//dx:spectrum | .//dx:chromatogram ... >>> >>> to >>> >>> <xs:selector xpath=".//dx:spectrum |.//dx:chromatogram ... >>> >>> (i.e. remove the space after the | ) >>> >>> It's not so easy for us (and possibly others) to update to a later >>> Xerces. There's quite a few occurrences of this, so if you could change >>> them, that would be appreciated. >>> >>> >>> >> ... >> >> >>> Cheers, >>> >>> David >>> > ------------------------------------------------------------------------------ > _______________________________________________ > Psidev-ms-dev mailing list > Psi...@li... > https://lists.sourceforge.net/lists/listinfo/psidev-ms-dev > |