[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]
Subject: DOCBOOK: automatic generation of revisionflag attribute,for generation of change bars
Here's the problem: I want to automatically generate change bars, between two arbitrary versions of an XML DocBook document, but I have no interest in manual use of the 'revisionflag' attribute. Not only do I regard manual use of 'revisionflag' as clumsy, I have good reason to rule it out, as an option. Requiring manual use of 'revisionflag' would be unacceptable, since my documents are composed of certain common components that get released in independent documents. Furthermore, some of these components are automatically generated, making manual use of the 'revisionflag' attribute virtually impossible. To give a more specific example, we release multiple software applications that depend on a core library to provide a significant portion of their functionality. We need such documents as the release notes (and other internal documents) to provide a means of easily discerning any functional (or implementation, in other docs) changes since the previous release of the application, including those changes to the core library, which have accumulated over the period since the previous release of a specific application. My solution: To determine the types of structural differences I'm interested in capturing, writing an XML-based diff tool (if one doesn't already exist), writing XSLT to consume the diff tool output + the document and produce a temporary (i.e. diff'd) document with the revisionflag attributes properly set, and use whatever stylesheets are most appropriate for rendering the differences. Does anyone see any problems with the above proposal (other than the challenges of writing a tree-based diff tool that produces output suitable for an XSLT stylesheet to properly set the 'revisionflag' attributes)? Does anyone have a simpler suggestion that'd still meet my requirements? Is anyone aware of some subset of this solution that's already been implemented, besides the output stylesheets? BTW, if a suitable one doesn't already exist, I plan to limit the cases my xmldiff tool handles, in order to simplify the implementation and improve the computational characteristics of the problem. Also, making it DTD-aware would increase the implementation complexity, but significantly reduce the search-space. Also, I regard it as a testament to the power of XML that my proposed solution is not only possible, but as generally applicable to similar sorts of problems as it is. Common syntax for specifying the structure of data truly is a wonderful idea, and one of the biggest step forward, in computing, since standardized encoding of character data. Thanks, in advance, for any help. Matthew Gruenke _________________________________________________________________ MSN Photos is the easiest way to share and print your photos: http://photos.msn.com/support/worldwide.aspx
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]
Powered by eList eXpress LLC