[RFC 1/8] Language independent bits

[Tom Tromey]

>>>>> "Sergio" == Sergio Durigan Junior <sergiodj@redhat.com> writes:

Sergio> I already replaced `int' by `size_t'.  Do you want me to revert
Sergio> the change?

No, there's no need to do additional work.

My general rule is that we should review patches, not the context of
patches.  If the context of a patch has problems, and it matters to the
reviewer, then either (1) the review can ask for a separate patch to fix
it (and in this case accept "no" as an answer), or (2) the reviewer can
fix it himself (since it is so darn important :-).

Sometimes it can be hard to recognize that a particular gross bit in a
patch was actually just moved from elsewhere.  In this case, IMO, the
patch submitter can and probably should just push back against the
review -- whatever bad effect there was is already there.

Of course, it is nice if we improve the code as we go along.  I think it
is usually easier to understand, and thus also review and approve,
separate patches for separate issues.

Tom