OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.


Help: OASIS Mailing Lists Help | MarkMail Help

docbook message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]

Subject: Re: DOCBOOK: Re: DocBook filename extension

On Tue, Feb 12, 2002 at 05:48:07AM +0000, Matt G. wrote:
> >What is the official DocBook filename extension (assuming there is
> >one)?  I've seen .docb, .dbx, and .xml.
> I use '*.xdbk'.  I don't understand why people use '*.xml', since it's not 
> just XML - you can be far more specific than that.  I regard that as 
> somewhat like naming files containing C code as '*.txt', since they are 
> technically text files (well yeah... but OF COURSE they're text files!).  
> Maybe one reason I'm so keen to distinguish XML DocBook from other types of 
> XML files is that I have pattern rules, in my makefiles, for processing 
> them, as well as files of other XML-based formats.

An xml-docbook document is not just xml but it definitely (should) pass
the test for an xml document according to the xml recommendation. So being
an xml document is IMHO a fact that should be noted in the file's suffix.
That's why I like the idea of multiple suffixes like .docbook.xml if there
is need to express the certain kind of markup language that is used inside
an xml document in the document's filename (as Jirka already pointed out
it's all inside the document's prolog anyways). Specifying increasing
detail from right to left through multiple suffixes allows you to process
any xml document with generic xml processors, no matter what dtd (or not)
it conforms to (even if this is not always meaningful).

I think that's an important point when thinking about mime-types. Trying
to follow RFC3023, I use the following mime-type declarations and
file-suffixes (e.g. for apache):

text/xml			xml xsl xhtml
text/xml-external-parsed-entity	ent
application/xml-dtd		dtd mod
# application/xslt+xml		xsl

Now that xml browsers will be more common, it seems to be more important
to serve documents or related xml stuff with a corresponding mime-type so
that the browser knows how to proceed with what it receives. If you take
constructed suffixes like .dbx or the like you would have to enumerate
each of those in your mime-type declaration file.

> What I'd like to know is what people use for external parsed entity filename 
> conventions.  I use '*.xdbk.ent', since they are external parsed entities 
> that tend to be fairly specific to XML DocBook.


> For external parameter entities, I generally use '*_xdbk.dtd', since a DTD 
> fragment is theoretically usable as a stand-alone DTD (unlike external 
> parsed entities, which don't have to meet criteria as stringent as 
> well-formed XML files), and are more reusable from DTDs for another 
> vocabulary, but tend to specific to XML DocBook, in some way.  I regard 
> external parameter entities in much the same way as I view C header files - 
> the extension declares the format and usage model, but not the usage 
> semantics.

Personally I use .mod for (external) parameter entities, because I've got
.mod files that might only contain entity declarations so they don't meet
the criteria of a self contained dtd IMO.

My EUR 0.02,


[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]

Powered by eList eXpress LLC