Re: Derby documentation in XML DITA

6 Dec 2004 19:20:23 -0000 

Jeff Levitt wrote:

--- scott hutinger <[EMAIL PROTECTED]> wrote:



[EMAIL PROTECTED] wrote:



scott hutinger <[EMAIL PROTECTED]> wrote on


06/12/2004 05:21:24 PM:






[EMAIL PROTECTED] wrote:





scott hutinger <[EMAIL PROTECTED]> wrote on


06/12/2004 11:38:59 AM:







I haven't followed that entire thread, nor can I


find that thread

currently. But, a log plugin; which I think?


followes which files





have




changed, has been developed using


java.util.logging. I think it





allows




forrest to know which files have changed but I


think it runs on a

dynamic site, and I am uncertain if the loging


plugin is hooked into 

forrest for this yet.

I will keep my eyes on that and research it a


bit.


Also, I am uncertain about the release date of


0.7. I think Jean has

the docs up to spec for 0.6, which I think is


needed for 0.7. Cocoon

had a problem creating some graphic files which


would create errors in









site logs, that was fixed in cocoon, but I don't


know if forrest





updated












cocoon with 6.0 after the release. I can look


into that on monday.


scott







Refer to the links in the comments of the


following JIRA issue:


http://nagoya.apache.org/jira/browse/DERBY-79

It would be nice if the JIRA issue is used to


record progress on the 

documentation.







Thanks for pointing this out. What did you have


in mind for the search

on a pc? I was wondering if you were thinking of

something like the

adobe extensible metadata platform, or something


like that? I really

haven't looked at pdf search techniques. I guess


someone could attach 

derby to pdf files somehow :-)

scott





Thanks,

John







Hi Scott,

I haven't spent any time with cocoon, FOP etc, but


I have seen other

commercial software products provide a PDF document


for each of their

manuals and a searchable index that covers all the


manuals.

I'm not sure if any of the open source PDF file


generation solutions

provide something like:


http://www.adobe.co.uk/epaper/tips/acr5search/main.html


I wasn't trying to suggest that Derby be somehow


used to search PDF

documentation (not sure if that is what you thought


I said)..

I would like the ability to search within acrobat


reader across the

different derby manuals and also have the ability


to have links from one

manual to the other (as I found I had to refer to


more than one manual to 

perform some Derby tasks).





OK, I understand.  I guess I didn't read the issue
correctly all the way.



The following page describes the search


functionality and it also

mentioned PDF metadata (that you mentioned), so it


sounds like you might 

have given yourself away as the PDF expert :-)





Actually, I don't know anything about PDF. I did
notice that FOP supports PDF version 1.3. So I don't think embedded
XML within the PDF file would work. It does have Document Level
Navigation, Destinations, Bookmarks etc. I did notice in FOP if memory
problems start to exist, turn off forward references. But the point I guess
is, do any other new document formats exist that might be better than
PDF? I know we will still output to PDF, but thought some other cool
document display architecture might exist. I don't know, as I am
clueless about such things. I can look a bit though.

Anyone know of some other display architecture?
I'll look more into FOP to see what all it supports.

thanks,
scott


http://www.planetpdf.com/enterprise/article.asp?ContentID=6521


Cheers,

John





I have an idea that may just solve most if not all of
these issues.  We can use Eclipse.

Check out the Cloudscape version of the help:
http://publib.boulder.ibm.com/infocenter/cldscp10/index.jsp

That entire site was created using Eclipse's help
system package.  To learn more about creating an
Eclipse help system, read about it here:

http://dev.eclipse.org/viewcvs/index.cgi/~checkout~/org.eclipse.platform.doc.isv/reference/misc/help_standalone.html?rev=1.26.2.2

And to deploy a help system on the web:

http://dev.eclipse.org/viewcvs/index.cgi/~checkout~/org.eclipse.platform.doc.isv/reference/misc/help_infocenter.html?rev=1.27.2.2

Long story short, we could host it on the Derby site
using an Apache http server.  I'm not good with the
technical aspects of this, so hopefully someone can
read up on whether the requirements for hosting this
would be allowed by Apache.

But if we can, then all we have to do is use
saxon/cocoon to output xhtml from the DITA (I've
already tested that with success), as well as create
an XML nav tree from the ditamap, also using
saxon/cocoon (again, tested successfully), and then
throw those into an eclipse help system plugin.  The
result would be almost exactly what you see on the
Cloudscape infocenter.  We'd have full advanced
pre-indexed search on all the docs at once, a much
more advanced nav tree, and once we set it up and take
a look at how related links work within the system, we
could modify the dita files to add links between the
docs where needed.  We'd still have PDF docs for those
who want them, of course.

I took the liberty of throwing my tested xhtml and and
xml nav tree into a plugin and testing it locally, and
it looked alright.

Does this sound like it could work?

Jeff



I think it sounds like a good idea myself. I would be willing as a user, to use jetty or something else, and host the help on my own machine during development. I don't see why others wouldn't if that is what they wanted. I guess a doc on how to use it might be needed :-)

I don't know about apache.org, one would think that would be allowed since they are the builder of such things. But I think it could have dual use, a local machine and a website if it all worked out.

scott

Reply via email to