1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
|
Here are the problems with building a PDF version of the manual:
1. The currently "used" jade / jadetex / dvipdfm run takes much time, as
jadetex needs to be run three times[!] to find all crossreferences
correctly. The process works this way:
- jade is run on the XML source with the DSSSL print stylesheet to
create a tex version
- jadetex is run 3 times on the tex source to create a dvi
- dvipdfm is run to convert the dvi to pdf
2. The limits set while setting up jadetex on the build machine are
not sufficient, the current manual size exceeds them. [Hartmut's
observation]
What would be the goal?
We have a plan to convert all jade / DSSSL based conversions to XSLT.
The current extended CHM is already created with XSLT and xsltproc.
Xsltproc seems to be the fastest conversion tool. It is writen in C,
so there is no need to setup a JVM to use it, and is also part of
cygwin [for windows users]. Xsltproc is already slightly integrated
to the build system (it is checked for, and html_xsl and chm_xsl
make targets use it).
How would that work?
- run xsltproc (or some other xslt processor) to convert XML -> FO with XSLT
- run FOP to convert FO -> PDF
Of course for one "make pdf" command.
Cons
Last time Hartmut checked it, FOP was unable to do all the needed
conversions to generate the correct PDF file out of the FO source.
Pros
- it would be significantly faster
- we would have a PDF version finally
- XSLT is easier to modify then the more cryptic DSSSL, there are more
tools and more manuals for it
|
