versorger at gmx.li wrote: > > I'm having a hard time with xinclude: I have two files FOO and BAR, each > representing a docbook chapter. In chapter FOO I have a table that I would > like to refer to in chapter BAR. With XMLMind this is possible by adding an > id attribute to the table element. Now I can "copy as reference" the table > from FOO and paste the reference into BAR. So far so good and my chapter BAR > is a valid instance of docbook. > > The problem comes when I process the whole book where both chapters FOO and > BAR are referenced via xinclude: After I have resolved the includes with > xmllint I end up with an invalid instance of docbook due to the fact that the > table, along with its id attribute, is now present twice.
Simply ignore this kind of duplicate ID errors. This kind of problem is *intrinsic* to the use of XInclude with the element() scheme and there is *nothing* that can be done. > By the way, I also get a (correct) warning about duplicate IDs when when I > open the book in XMLMind. As of v3.3, Options|Options, Tools|Validate section, checkbox "Filter duplicate ID errors found in modular documents" allows to discard this kind of duplicate ID errors (but not other duplicate ID errors). > > After some reading on xinclude I came up with the following solution to use > in BAR: > > <table id="newID"> > <xi:include href="FOO" xmlns:xi="http://www.w3.org/2001/XInclude"; > xpointer="xpointer(id('someID')/*)" /> > </table> > > I manually enter the table element and simply xinclude all child nodes, > leaving out the id attribute. xmllint will happily process this while XMLMind > chokes on the unsupported scheme for XPointer. * XInclude does not mandate supporting the xpointer() scheme, only the element() scheme. * Unlike the element() scheme, the xpointer() scheme is not yet a W3C recommendation but only a working draft. See http://www.w3.org/TR/xptr-xpointer/ > Now I am pretty sure that it's a common request to reuse content like I plan > to do. Sure, we do that all the time here at XMLmind. The main difference between you and us is that we do not feel the need to double-check our documents using xmllint. > I am sure there is a nice and clean way to achieve this with both XMLMind and > any other tool and I was just not able to find it ... Once again, simply *ignore* this kind of duplicate ID errors.
