Fix a problem where sparse_seek could get left non-zero when we
did not finish writing all the data that would take us to that
sparse gap. Issue pointed out by David Taylor.
When checking a checksum that refers to a part of an --inplace file that
has been overwritten w/o getting SUMFLG_SAME_OFFSET set, we remove the
checksum from the list. This will speed up files that have a lot of
identical checksum blocks (e.g. sequences of zeros) that we can't use
due to them not getting marked as being the same. Patch provided by
Michael Chapman.
Added the client rsync-ssl script and various client/daemon support
files needed for talking to an rsync daemon over SSL on port 874 (no
tls support). This uses an elegant stunnel setup that was detailed
by dozzie (see the resources page) now that stunnel4 has improved
command-spawning support. Also incorporates some tweaks by devzero
(e.g. the nice no-tmpfile-config client-side code) and a few by me
(including logging of the actual remote IP that came in to the
stunnel process). This probably still needs a little work.
The cleanup code will try to flush the output buffer in some
circumstances, which is not valid if we're handling an async signal
(since it might have interrupted some partial I/O in the main thread).
These signals now set a flag and try to let the main I/O handler take
care of the exit strategy. Fixes a protocol error that could happen
when trying to exit after a kill signal.
- If iconv() returns EINVAL or EILSEQ and the error is being ignored, make
sure that there is room in the output buffer to store the erroneous char.
- When accepting an erroneous char, be sure to break if there are no more
input characters (without calling iconv() with a zero input length).
