Re: [lmi] help with orphan-control in xsl-fo

[Greg Chicares]

On 2009-05-17 11:46, Vaclav Slavik wrote:
> On Sun, 2009-05-17 at 03:36 +0000, Greg Chicares wrote:
>>   In collapsing border model a table does not have padding
> 
> Finally, here's a fix for this one; I now have clean FOP output with
> both nasd.xsl and reg_d_individual.xsl.
> 
> The patch below is unreadable, but the transformation I did was simple:
> for every <fo:table> that had any padding attribute, I enclosed it in
> <fo:block> and moved padding attribute to it.
> 
> For reference, here are the changes as a diff -d -w output:

I was glad to have the patch in two different formats, because this one
applied cleanly now (seven years later) whereas the other didn't. Then,
for commit cc39acff51c64801fcc662ccc591adcc31d30f07, I adjusted the
indentation manually.

> And here's the version that doesn't ignore indentation changes -- it's
> unreadable, but it has the advantage that it can be applied with
> patch(1):

I applied that locally with --fuzz=1000, which mostly worked, and then
diffed the results against my change above.

AFAICT, this (like the preceding change) fixes an actual XSL error
that isn't detected by the ancient version of fop that we use in
production, but it was supposed to have been applied years ago, and
further delay would just make it harder to apply. The real solution
is to create PDF files in a completely different, more modern way,
but it's impossible to say when we'll be able to do that.