Re: deferred parsing issue

Yes, it's a bug! I thought I fixed the problem already but guess not. I 
will look into it. I remember I change the generating behavior so it 
always generate the .xmlc to the same name as the "class" not the 
interface.

You may want to try the setDefaultMetaDataPath pointing that to a 
default meta file. but this may not work well with I18N if you specify 
the encoding in the option file.

David

On Saturday, Apr 5, 2003, at 13:35 Asia/Shanghai, Jacob Kjome wrote:

>
> I've got deferred parsing working for a Barracuda app, but I have one 
> problem.   When I use "-generate both" in my options.xmlc file and I 
> have a file called "Mypage.html", then only .xmlc file generated by 
> the Xmlc taskdef (actually the Localize taskdef which extends the Xmlc 
> taskdef) is "MypageHTML.xmlc".  However deferred parsing throws an 
> error saying that it can't find the file "MypageHTMLImpl.xmlc".  Now, 
> if I simply copy "MypageHTML.xmlc" to "MypageHTMLImpl.xmlc" and then 
> run the page, everything works fine.
>
> So, I guess I'm wondering if there is some special flag I have to 
> provide to make the Xmlc taskdef generate the appropriate .xmlc file 
> name when using the "-generate-both" flag?  Or is it just a bug that 
> it is not doing that already without me having to specify anything?  
> This will also be a concern when I start utilizing both "-generate 
> both" and the Localize taskdef to create localized versions of the 
> DOM.  In that case, I'd need a .xmlc file named something such as 
> "MypageHTMLImpl_es.xmlc" (at least, I would think?).