Mitch already e-mailed me about it and pointed at the URL with the form and patch process. I signed the form and might as well try the official route.
There were a couple of technical issues as well. Mitch said that he does not want to use tee. Also, Fedora cloud users had concerns about using this much space for a temporary file. Finally, the code was changed in the trunk, so my patch does not apply (it was based on 1.3.1 that Garrett packages for us).
So, I am going to rework the fix so that it stores the compressed tarball as before, then decompresses it to compute SHA1 digest. This will trade CPU for I/O and disk space.
Mitch already e-mailed me about it and pointed at the URL with the form and patch process. I signed the form and might as well try the official route.
There were a couple of technical issues as well. Mitch said that he does not want to use tee. Also, Fedora cloud users had concerns about using this much space for a temporary file. Finally, the code was changed in the trunk, so my patch does not apply (it was based on 1.3.1 that Garrett packages for us).
So, I am going to rework the fix so that it stores the compressed tarball as before, then decompresses it to compute SHA1 digest. This will trade CPU for I/O and disk space.