On 1/2/07, Andrew McIntyre <[EMAIL PROTECTED]> wrote:
Yes, and yes. docs.properties at the top of the tree defines two
properties that allow Ant to locate the DITA toolkit: dita.dir, the
location of the expanded DITA toolkit directory, and dita.zip, the
location for the unexpanded zip file to allow for automatic expansion
of the DITA toolkit from the zip. Change these two properties to point
to the new locations of the toolkit/zip file and Ant will call into
the toolkit buildfiles in the new location.
andrew
Andrew - I changed the doc.properties file but I think that there
might be something wrong with this line:
dita.zip=${basedir}/lib/DITA-OT1.3.1_bin-ASL.zip
All I changed was 1.1.2.1 to 1.3.1
As I read this line, ANT is supposed to look for the zip file in the
basedir, under the lib directory... but that is not where the zip file
is located. And the name of the new zip file is
DITA-OT1.3.1_fullpackage_bin.zip ...
Also, I asked the DITA toolkit Help forum about the index issue. Here
is what I received back:
- - - - - - - - - - - - - - - - - - - - - - -
Good question, Laura. Indexing for the older FOP-based PDF generator
has not been improved lately. If you download and install the FO 1.2.1
transform plugin, it has the capability to produce very nice indexes.
It currently employs XEP as its output formatter, as this formatter
was way ahead of the function of FOP 0.20.5 to which the older FO
transform was written at the time (and which I think you are currently
using). This "pdf2" transform is clearly the single code base we want
to invest in for future maintenance. However, because that indexing
capability depends on XEP-specific extensions in the transforms, you
cannot currently >create equivalent indexes for FOP-based PDF with it.
IIRC, XEP provides a free license that allows its use for Open Source
and Standards-based projects.
If indexing for FOP-based output is a high priority for Derby, despite
the free
licensing on XEP for open source, then let me know. Indexing is a
tough problem, otherwise we would have solved it more generally by
now. But if you can suggest resources or XSLT algorithms that can help
solve issues like terms with multiple pointers to the same page, then
we might do a sizing based on that knowledge, and see if we can work
it into a future update.
- - - - - - - - - - - - - - - - - - - - - - -
I am not sure that I understand all of what he is saying. Is he
suggesting XEP as a replacement for ANT or something else?
--
Laura Stewart
