In general, you really can't expect to retain CDATA unless you build a DOM tree, and maybe not then. It's a fundamental principle of XML that CDATA isn't part of 'the infoset' -- the data that is represented by the file. A parser is under no obligation to faithfully report this stuff, so long as the right text ends up in the right place.
However, forget xpp3 and all that ancient lumber. Use Woodstox, or the Xerces DOM, and you can probably find enough options to ask it to do what you need. On Tue, Aug 2, 2011 at 6:00 PM, John Casey <jdca...@commonjava.org> wrote: > Hi all, > > I'm working on some tooling for $dayjob that needs to manipulate POM files > according to certain rules. > > The problem I'm running into is that some of the POMs it much manipulate > contain CDATA sections, comments, etc. Also, since the modified POM often > will be used as the basis for a patch file, I'd like to preserve as much of > the ordering and existing whitespace in the file as possible, to minimize > the patchfile size. > > I'm currently using the JDom-driven, Modello-generated writer, coupled with > the stock XPP3-driven reader (not the best, I know). It's losing the CDATA > (big problem) and injecting ^M (wrong line ending, little problem)... > > Does anyone have experience with this? Anyone maybe have an advanced POM > reader/writer stashed somewhere that can preserve CDATA and the like? > > Thanks, > > -john > > -- > John Casey > Developer, PMC Chair - Apache Maven (http://maven.apache.org) > Blog: http://www.johnofalltrades.name/ > > --------------------------------------------------------------------- > To unsubscribe, e-mail: dev-unsubscr...@maven.apache.org > For additional commands, e-mail: dev-h...@maven.apache.org > > --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@maven.apache.org For additional commands, e-mail: dev-h...@maven.apache.org