Thank you very much for your answer!
In fact I was hoping something more "practical" and automated was possible...
We currently have something in place with Docbook5 which works really fine: we have an abbreviation.xml file which contains all our glossary
entries, with unique xmlIDs, and when we write a new document we simply insert
Code: Select all
<xi:include href="abbreviations.xml" xpointer="SCR"/>
wherever we have the "SCR" abbreviation.
At the beginning of the Docbook file, we insert the xml files as references like this:
Code: Select all
<title>Definitions & Abbreviations</title>
<xi:include href="abbreviations.xml" xpointer="element(/1)"/>
And in the transformation process, the complete docbook file is scanned and only the abbreviations that are used are displayed in this section.
Same for the references (we have to deal with thousands of reference docs). We have a "library" docbook file (basically a table where each row is a document, with a unique doc number, xmlID, and attributes which define where this document is referred to in our documents). At the beginning of the transformation for one of our report (with its own unique document number), this huge table is stripped from all the documents that do not need to be displayed in the references section, and a table is built, giving each one a sequential number (, , etc...) a unique xmlID that allows to refer to it in the document...
In short: the process is as automated as possible, and we only have to maintain a few files (abbreviations, definitions, document library) that are used in ALL of our documents.
In fact my main reason for considering a switch to DITA is the availability of the DITA to responsive WebHelp and DITA to PDF based on CSS which are much more attractive and modern than what is available with DocBook file...
Perhaps you would have some suggestion to have a "modern" output available for DocBook? or to develop some equivalent solution with DITA?
Thanks a lot for your help!