alt currently overrides the default datatype of att.pointing's @target to require 2+ teidata.pointers—presumably (and sensibly) because alternation requires that there be at least two elements to alternate:
We should probably relax this constraint somehow: simplest would be to just revert@target back to its original definition (i.e. getting rid alts attDef for it) and put the requirement for 2+ things in remarks; more complicated would be try and write a constraint such that alt/@target must have either 1+ xpath pointers OR 2+ pointers otherwise.
altcurrently overrides the default datatype ofatt.pointing's@targetto require 2+teidata.pointers—presumably (and sensibly) because alternation requires that there be at least two elements to alternate:https://github.com/TEIC/TEI/blob/dcfee0087066d4e5e8b507499280460565c30370/P5/Source/Specs/alt.xml#L30-L32
However, that means that an
#xpathpointer which returns a sequence of nodes is invalid (see https://listserv.brown.edu/cgi-bin/wa?A2=ind2309&L=TEI-L&D=0&P=64 for context).We should probably relax this constraint somehow: simplest would be to just revert
@targetback to its original definition (i.e. getting ridaltsattDeffor it) and put the requirement for 2+ things in remarks; more complicated would be try and write a constraint such thatalt/@targetmust have either 1+ xpath pointers OR 2+ pointers otherwise.