Re: DOCBOOK: Re: imagedata: entityref vs fileref

[Jens Stavnstrup <js@ddre.dk>]

Patrick,

I would only recommend the use of fileref. When you use entityref (the 
reference is resolved to the physical path (when usibg XSLT with the HTML 
stylesheets), due to the XSL processors use of unparsen-entity-url.

Your HTML file are note very portable, when they refere to the image

  /home/myhome/mymages/someinage.jpg

I must confess, that this was tested at least 6 months ago, but if the 
XSLT processors do it different today. I am sure I will be corrected.

Regards

Jens

On Tue, 21 May 2002, Norman Walsh wrote:

> / Patrick Eisenacher <eisenacher@fillmore-labs.com> was heard to say:
> | I'm new to the list and I have a question regarding the use of the
> | entityref and fileref attributes when including an image via the
> | imagedata element. When do I use which attribute? Is it merely personal
> | preference? What are the benefits? What are the drawbacks? What is the
> | preferred way?
> 
> It is, to a large extent, personal preference. The use of entities
> provides a degree of indirection that is sometimes useful. It was more
> common, perhaps, in the pre-XML (and pre-web) days before URIs were a
> widespread concept.
> 
> In order to use entityref, you must setup a more complex document structure:
> 
> <!DOCTYPE article PUBLIC "..." "..." [
> <!ENTITY graphic PUBLIC "..." "..." NDATA PNG>
> ]>
> <article>
> ...
> <imagedata entityref="graphic"/>
> 
> When the document is processed, the PUBLIC and SYSTEM identifiers are
> used to resolve the actual location of the 'graphic'. This means that
> a catalog or other resolver mechanism can make dynamic selections.
> 
> It's also possible to put the entity declarations in an external file
> and switch between them wholesale. For DocBook: TDG, there are several
> hundred 'unicode character' graphics. For the HTML version of the
> book, these are all PNGs; for the print version, they're all EPSes. I
> switch between the two by selecting different sets of entities.
> 
> | The processing expectations say:
> |> There are two ways to provide content for ImageData: EntityRef or FileRef. It is best to use only one of these methods, however, if multiple sources are provided, EntityRef will be used in favor of FileRef.
> |
> | How do I have to understand 'if multiple sources are provided' - if I
> | specify a fileref and an entityref attribute for the same imagedata
> | element?
> 
> Yes.
> 
>                                         Be seeing you,
>                                           norm
> 
> -- 
> Norman Walsh <ndw@nwalsh.com>      | Extinction, n. The raw material
> http://www.oasis-open.org/docbook/ | out of which theology created the
> Chair, DocBook Technical Committee | future state.--Ambrose Bierce
>