Ensure that even if FormatMessage fails we (a) don't crash, and (b) provide something useful.

Bugfix candidate.

all along.  Before instr_lb tended to be too high.

I don't think this actually makes any difference, given what the compiler
produces, but it makes me a bit happier.

Bugfix candidate.

allow "testlist1" to be snagged.

Trivial fix to the pep277 checkin: ensure that exceptions always have a filename attribute (previously did only when string filenames were passed, but not when unicode)

enough to fit in one line.

patch #617312, both on the trunk and the 22-maint branch.

Also added a test case, and ported the test_trace I wrote for HEAD
to 2.2.2 (with all those horrible extra 'line' events ;-).

CCompiler may rely on the presence of self.verbose (SciPy's distutils
appears to).

they'll end up in the right directory when --dir is used.

needs to be expanded by mkhowto, since the shell won't touch it except
at the start of the complete argument.

default when used for 3rd-party docs.  This requires the standard Python docs
to specify a location, but they are a bit of a special case in sharing one set
of icons among several documents.

Barry, please add a reasonable abstract!

from an attributes description.

\verbatiminput instead of the verbatim environment -- this does the "right
thing" regarding page breaks in long examples for the typeset formats, and
has nice benefits for the HTML version as well.

I'll do next).

Strangely, two out of three patches there seem already committed; but
the essential one (get rid of the assert in object_filenames in
ccompiler.py) was not yet applied.

This makes the build procedure for Twisted work again.

This is *not* a backport candidate despite the fact that identical
code appears to exist in 2.2.2; Twisted builds fine there, so there
must have been a change elsewhere.