Thanks Hussein. I'll see if we can create a small example to reproduce the issue.
Weldon -----Original Message----- From: Hussein Shafie [mailto:[email protected]] Sent: Tuesday, June 05, 2012 3:37 PM To: Sams, Weldon Cc: [email protected] Subject: Re: [ditac] Extraneous files after a convert On 06/05/2012 08:16 PM, Sams, Weldon wrote: > > After converting a ditamap that references several topicrefs to xhtml > 1.1 we're seeing several documents like this, > > __abdz-2.xhtml > > Some begin with underscores and have a -$i, where $i is an integer, > appended to the base file name. We do get a correct abdz.xhtml in the > bunch. For the roughly 250 correct xhtml documents we get more than > 1000 of these extra documents. Does anyone have any idea why this > might be happening? You don't explain to reproduce this behavior here at XMLmind. Therefore it's almost impossible to help you. * I don't remember having seen this in my own DITA documents. * May be these files correspond to what follows in the .ditamap: --- <part navtitle="Using XMLmind DITA Converter"> <chapter href="install.dita"/> ... --- or to this: --- <chapter navtitle="Getting started"> <topicref href="quickStart.dita"/> </chapter> --- That is, an anonymous topic having just a title and no body is automatically generated by ditac for a topicref (of any kind) having a navtitle and no href. If you specify chunk="by-topic" in your .ditamap, I would say that ditac will automatically generate topics having quite strange filenames for the above kind of topicrefs. If what I've described really corresponds to your use case, then this behavior is is normal and it should not disturb you. (If you don't like this behavior, then there are plenty of ways to modify your DITA source in order to better control the files generated by ditac.) > > We recently wrote the list and patched the > PreProcessor.setUniqueId(Element element, String id, Map<String, > Element> idToElement)method because a counter was hard coded to 1000. > > I also noticed ImageCopier.javaand Chunker.javahave a similar hard > coding of 1000. Maybe Chunker.java needs updated as well as best I can tell. Yes, that's right. Ditac 2.2.2, which should be released in at most a couple of weeks, cleanly fixes this problem. -- XMLmind DITA Converter Support List [email protected] http://www.xmlmind.com/mailman/listinfo/ditac-support
