cborutil: change buffering strategy
Profiling revealed that we were spending a lot of time on the
line that was concatenating the old buffer with the incoming data
when attempting to decode long byte strings, such as manifest
revisions.
Essentially, we were feeding N chunks of size len(X) << len(Y) into
decode() and continuously allocating a new, larger buffer to hold
the undecoded input. This created substantial memory churn and
slowed down execution.
Changing the code to aggregate pending chunks in a list until we
have enough data to fully decode the next atom makes things much
more efficient.
I don't have exact data, but I recall the old code spending >1s
on manifest fulltexts from the mozilla-unified repo. The new code
doesn't significantly appear in profile output.
Differential Revision: https://phab.mercurial-scm.org/D4854
#!/bin/sh
#
# This is an example of using HGEDITOR to create of diff to review the
# changes while committing.
# If you want to pass your favourite editor some other parameters
# only for Mercurial, modify this:
case "${EDITOR}" in
"")
EDITOR="vi"
;;
emacs)
EDITOR="$EDITOR -nw"
;;
gvim|vim)
EDITOR="$EDITOR -f -o"
;;
esac
HGTMP=""
cleanup_exit() {
rm -rf "$HGTMP"
}
# Remove temporary files even if we get interrupted
trap "cleanup_exit" 0 # normal exit
trap "exit 255" HUP INT QUIT ABRT TERM
HGTMP=$(mktemp -d ${TMPDIR-/tmp}/hgeditor.XXXXXX)
[ x$HGTMP != x -a -d $HGTMP ] || {
echo "Could not create temporary directory! Exiting." 1>&2
exit 1
}
(
grep '^HG: changed' "$1" | cut -b 13- | while read changed; do
"$HG" diff "$changed" >> "$HGTMP/diff"
done
)
cat "$1" > "$HGTMP/msg"
MD5=$(which md5sum 2>/dev/null) || \
MD5=$(which md5 2>/dev/null)
[ -x "${MD5}" ] && CHECKSUM=`${MD5} "$HGTMP/msg"`
if [ -s "$HGTMP/diff" ]; then
$EDITOR "$HGTMP/msg" "$HGTMP/diff" || exit $?
else
$EDITOR "$HGTMP/msg" || exit $?
fi
[ -x "${MD5}" ] && (echo "$CHECKSUM" | ${MD5} -c >/dev/null 2>&1 && exit 13)
mv "$HGTMP/msg" "$1"
exit $?