MenuBar/IconUsers.png "IconUsers.png")
MenuBar/gsoc_2010.png "gsoc_2010.png")
MenuBar/IconGlobe.png "IconGlobe.png")
Contributor - Improve PDF generation
Description
From Scilab, it is possible to generate the PDF documentation from the docbook files (see xmltopdf). This needs to be improved.
Ideas
The idea is to get rid of saxon and use only fop. Next, the PDF rendering should be improved to reach the same level of quality as the other Scilab documentations.
It must be possible to write a Java class similar to: http://gitweb.scilab.org/?p=scilab.git;a=blob;f=scilab/modules/helptools/src/java/org/scilab/modules/helptools/HTMLDocbookTagConverter.java;h=d025b53c7c992515394c08532f8c0d3e9069bcb3;hb=HEAD
which could be called FODocbookTagConverter to generate xml-fo code. The resulting PDF must contains hyperlinks to make the navigation easier and must have a navigation tree on the left.
Actually all our HTML-based documentation is generated by HTMLDocbookTagConverter or by a derivated class.
In this project, a second step could be to introduce an "on-the-fly" generation of image from Scilab code. With a specific docbook tag, we could write a Scilab code, code which would generate an image, this email being automatically integrated in the Scilab documentation. For example, the "hand made" example in the help page of plot3d could be automatically generated.