2012/10/3 Jürgen Spitzmüller <[email protected]>: > 2012/10/3 Pavel Sanda <[email protected]>: >> does it mean that #5941 is more easy to fix now? P > > No, this seems unrelated. We could probably work around it by breaking > in PDFOptions::writeLaTeX if runparams.is_child is true, but the > actual bug is that the preamble is at all generated in the master > here. No idea why this happens.
The problem is that we validate the children's buffer params. We shouldn't do this, since the children inherit their parent's buffer params. Fix attached. Jürgen
5941.diff
Description: Binary data
