[Slackbuilds-users] md5sum mismatch on netcdf

[Nick Warne]

David Spencer wrote, On 12/01/14 22:19:
> On 12 January 2014 18:25, Nathan Becker <nathanbecker at gmail.com> wrote:
>> Hi,
>>
>> I observe a discrepancy between the md5sum of the downloaded file
>> netcdf-4.3.0.tar.gz and the checksum listed on slackbuilds.
>>
>> Slackbuild reports: 39fc507bf2966980fa92defecd9d2167
>> downloaded file:     40c0e53433fc5dc59296ee257ff4a813
>
> Hi Nathan, hi everybody --
>
> Yes, upstream pushed a new 4.3.0 tarball at the end of November.
> 40c0e53433fc5dc59296ee257ff4a813 is now the correct md5sum.  The old
> tarball of 4.3.0 had an enormous amount of extraneous rubbish in it,
> in hidden directories, so (when they finally noticed) they pushed a
> corrected tarball, with the same version number :-/
>
> Rather than hassling the SBo admins, I was waiting for 4.3.1 to come
> out so I could submit that instead, but it's now been stuck at
> 4.3.1-rc4 since early November, so if one of our friendly admins wants
> to correct the md5sum for 4.3.0, that would be lovely ;-)

But surely the correct md5sum to a download is a remote test, therefore 
the slackbuild needs to be updated (no matter when the download changes, 
so does the md5sum) - not trying to match the md5sum to what is the 
download is wrong:  if it changes, it's an update and slackbuild needs 
changes too.

Nick
-- 
"A bug in the code is worth two in the documentation."
FSF Associate Member 5508
http://linicks.net/