Re: 3.19: problem compiling user manual

[Mike Dowling]

On Tue, Mar 21, 2000 at 02:38:04PM +0000, Jeffrey Howard Kingston wrote:
> 
> 
> On Tue, 21 Mar 2000, Mike Dowling wrote:
> 
> > #lout all >op
> > lout file "gra_summ" (from "gra" line 44, from "all" line 44):
> >      8,1: 23.6c object too high for 23.6c space; @Scale inserted
> > lout file "cpp_tabs" (from "cpp" line 80, from "all" line 45):
> >    56,23: prg2lout 2,1: program text ended within comment
> >    58,35: prg2lout 2,1: program text ended within comment
> > lout:
> >          internal error: assert failed in FlushGalley: dest_side != FWD
> >          || !is_indefinite(type(y))!
> > Aborted
> >
> 
> That's nasty.  Did you change anything before you did the run, such as
> the paper size or anything?  I always compile the user's guide myself
> and I obviously didn't get this, which means it's either the result of
> some change you made that I need in order to reproduce it, or else it
> is caused by an uninitialized variable somewhere.  I mean really
> uninitialized, please don't tell me that gcc reports lots of them,
> I know about those ones.

All I did was compile and install lout-3.19.  I then went to the doc/user
subdirectory, and entered 

lout all > op

as described in the README file.  No changes to anything, straight out
of the box.

> It seems to have occurred quite late in the run, and not on the first
> pass either.  Could be in the index, in which case it could have to
> do with the new index running headers.

Yes, it must have been the sixth run, I'm not quite sure.  The last page
to be printed was page 242, which was the second page of Appendix A, the
Quick Reference Guide.

Perhaps I should add that this is Linux-2.3.50.  That should not have a
baring on the matter, but on rare occasions using one of these
development kernels does have funny side effects.

Cheers,
Mike Dowling