On Mon, 31 Jan 2011, Daniel Kobras wrote:
Hi!
Am 31.01.2011 um 12:28 schrieb Reuben Thomas:
Absolutely awesome, Bob, that fixes it.
/me too! ;-) Bob, given these results, do you consider your patch
ready for inclusion in the Debian package?
I am not 100% sure but it seems ok, particularly since the old code
results in a definite crash. The only sort of bug I can think of is
when writing to a blob in memory. If not all the image objects in the
list refer to the same blob and the final blob length is obtained from
the first image in the list rather than the last to be updated then
the blob could be effectively truncated. I have not yet observed an
image format writer which works that way.
Take note that this bogus code has been in all releases of
GraphicsMagick since it was inherited from ImageMagick.
The test suite is still passing. I am enhancing the test suite to
verify multi-frame reading/writing with blobs and files.
There is another bug I am aware of which impacts Linux but it is in a
different area (pixel cache) and deserves a new bug report. That bug
results in a deadlock.
Bob
--
Bob Friesenhahn
bfrie...@simple.dallas.tx.us, http://www.simplesystems.org/users/bfriesen/
GraphicsMagick Maintainer, http://www.GraphicsMagick.org/
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
