Jeff Levitt wrote:
I can tell you that this is a problem that could be
hard to fix. The single-html books are not a normal
result of the DITA transform. However, some time ago
Andrew found an xsl file on the web that we were able
to use to transform the half-processed PDF transforms
from the DITA process and convert them to html.
Unfortunately, since that XSL file came from outside
of DITA, it is not supported by their project. If
someone has time they can try to modify that xsl file
to fix this problem, but I think the single-html files
were put in place as sort of a bonus when Andrew found
that xsl file, so I wouldnt expect to see as much
progress on the quality of those books as the other
two types of output, unless someone else knows what
they are doing and can fix this (I dont!) :)
Also, Andrew, do you know if the xsl file is legally
modifiable?
I had a look at this. The XSL file "fo2html.xsl" comes from
www.renderx.com. I downloaded a copy (it is free) and compared it to the
one in our repos. They are identical, so nothing has been updated.
Although I had to obtain a license to download the XSL file, the
following is stated in the XSL file itself:
<!-- =============================================================== -->
<!-- -->
<!-- Convert XSL FO (as in REC 2001-10-15) to HTML -->
<!-- -->
<!-- © RenderX 2000-2001 -->
<!-- Permission to copy and modify is granted, provided this notice -->
<!-- is included in all copies and/or derived work. -->
<!-- -->
<!-- Author: Nikolai Grigoriev, [EMAIL PROTECTED] -->
<!-- -->
<!-- =============================================================== -->
Just for completeness, my (modified) 'license.xml' looks like this:
<?xml version="1.0" encoding="UTF-8"?>
<license issuer="RenderX" issued="2006-02-28">
<licensee>
<person>My Name</person>
<organization/>
</licensee>
<product name="Fo2Html">
<limit name="full">1</limit>
</product>
<signature>[snipped]</signature></license>
Don't know if you need to have this around, or if it is only a mean to
trace/control downloads.
I'm thinking about updating the Jira issue (DERBY-753) with some
information and maybe resolve/close it, but I will wait a little longer
to see if anyone else has something to add to this topic.
--
Kristian
--- Kristian Waagan <[EMAIL PROTECTED]> wrote:
John Embretsen wrote:
Kristian Waagan wrote:
Hello,
Some time ago I created DERBY-753
(http://issues.apache.org/jira/browse/DERBY-753),
but there has not
been any activity on it since then. It says that
the links in the
table of contents of the HTML book manuals do not
work.
Can anyone confirm that the problem (still)
exists?
I can confirm that the problem still exists with
the 10.2 reference
manual. I have not tried the others.
I also tried the link (from the Jira issue) that
was supposed to work,
but it did not work either. I.e.:
http://db.apache.org/derby/docs/dev/ref/ref-single.html#crefsqlj95081
does not work;
http://db.apache.org/derby/docs/dev/ref/ref-single.html#N1223F
does not work; but
http://db.apache.org/derby/docs/dev/ref/ref-single.html#N1227F
works.
Okay, thanks for confirming the problem.
For the links, the point was to illustrate that
there are anchors in the
files, but they are not used. To me, it looks like
there has been some
kind of mapping error or something during the build
process.
I do not know how the anchors are defined and how
the documentation
build process works. Can anyone with more Derby
documentation building
experience see the reason for this problem?
Since I believe we don't exactly score "Top rated"
on our documentation,
I would say this problem should be fixed pretty
quickly! It hinders
navigation severely.
--
Kristian
