On Sat, Apr 11, 2009 at 01:08:15PM -0500, Dan McGee wrote:
On Mon, Apr 6, 2009 at 11:15 PM, Loui Chang <louipc.ist@gmail.com> wrote:
The index in the for loop wasn't being incremented, so if the first file wasn't found, the second file would be compared to the first checksum, rather than the second.
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
This seems fine, but what the hell is going on around line 580? Is it just me that thinks a double check on "$file" is odd?
Yeah it did seem odd to me.
Originally when this came in it looked like the following (commit 142225fd): + + if [ ! -f "$file" ] ; then + if [ ! -f "$SRCDEST/$file" ] ; then + error "$(gettext "Unable to find source file %s to generate checksum.")" "$file" + plain "$(gettext "Aborting...")" + exit 1 + else + file=$SRCDEST/$file + fi + fi +
git-blame is lying to me here and I can't find when this was changed. Anyone else know what the heck happened here?
Hmm. I'm not sure what's going on there. The relevant section for 142225fd looks like this for me: @@ -541,14 +551,22 @@ check_checksums() { file="$(strip_url "$file")" echo -n " $file ... " >&2 + if [ ! -f "$file" ] ; then + if [ ! -f "$file" ] ; then + echo "$(gettext "NOT FOUND")" >&2 + errors=1 + continue + else + file=$SRCDEST/$file + fi + fi + if echo "${integrity_sums[$idx]} $file" | ${integ}sum --status -c - &>/dev/null; then echo "$(gettext "Passed")" >&2 else echo "$(gettext "FAILED")" >&2 errors=1 fi - - idx=$(($idx+1)) done if [ $errors -gt 0 ]; then