http://www.perlmonks.org?node_id=560562


in reply to Re^3: ReorgXSD.pl
in thread ReorgXSD.pl

Hi! No, I don't want to tidy the file, there's plenty of that in most any generator. The problem is, and I point out it is most specific to XSD, that generators of XSD don't generally care about "reading format", their concern is standards compliance. Typically, they'll generate a file and simply alphabetize the entries. That's okay if you already know the entries and can tell the generator which entry is the root when it goes to create an example/sample/template XML. But what if you don't know? What if you're looking at some other XSD file (also generated) from another source. Without careful study and, perhaps, a few trial and error passes, you may not get a correct XML example/sample/template. What I've tried to do with ReorgXSD is do some of the "study" and reorganize the XSD into a file that more readily reveals what the structure of the XML will be, most importantly, what is the root element?

It might be profitable for you to take the code, apply it to a less than decipherable XSD and see if you don't think the output is easier to read/understand.

I've tried it with various XSD's from Altova, Stylus Studio, and Microsoft, both existing in their downloads and generated by them. I've found that most of the time, the reorg makes the XSD easier to read/understand. But that, of course, is subjective. Perhaps for many people who've been around XSD since it's inception, ReorgXSD files are no better than the original.

thx,
Dave S.