Checksum fails for Cockatrice

[skwerlman@…]

When attempting to install Cockatrice under OSX 10.10 Yosemite, port fails to verify the checksum of the sourceball.

Alekss-MacBook-Pro:~ AleksBirnhak$ sudo port install cockatrice
--->  Computing dependencies for Cockatrice
--->  Fetching archive for Cockatrice
--->  Attempting to fetch Cockatrice-0.0.1_0.darwin_14.x86_64.tbz2 from http://packages.macports.org/Cockatrice
--->  Attempting to fetch Cockatrice-0.0.1_0.darwin_14.x86_64.tbz2 from http://lil.fr.packages.macports.org/Cockatrice
--->  Attempting to fetch Cockatrice-0.0.1_0.darwin_14.x86_64.tbz2 from http://mse.uk.packages.macports.org/sites/packages.macports.org/Cockatrice
--->  Fetching distfiles for Cockatrice
--->  Verifying checksums for Cockatrice
Error: Checksum (rmd160) mismatch for Cockatrice-c0bd7db658fb54d6f5834f1e99838c150b94379a.tar.gz
Error: Checksum (sha256) mismatch for Cockatrice-c0bd7db658fb54d6f5834f1e99838c150b94379a.tar.gz
Error: org.macports.checksum for port Cockatrice returned: Unable to verify file checksums
Please see the log file for port Cockatrice for details:
    /opt/local/var/macports/logs/_opt_local_var_macports_sources_rsync.macports.org_release_tarballs_ports_games_Cockatrice/Cockatrice/main.log
To report a bug, follow the instructions in the guide:
    http://guide.macports.org/#project.tickets
Error: Processing of port cockatrice failed

[ryandesign (Ryan Carsten Schmidt)]

Looks like the github organization name has changed from "Daenyth" to "Cockatrice". The organization name is included in the top-level directory name in the tarball, so this changed the checksum of the tarball. Otherwise, the contents are identical.

Let me look into why we're fetching some random git commit, and not the latest tagged release...

[ryandesign (Ryan Carsten Schmidt)]

Marko, do you recall why you used a commit hash instead of the "2014-06-25-Release" tag?

[mkae (Marko Käning)]

Replying to ryandesign@…:

Marko, do you recall why you used a commit hash instead of the "2014-06-25-Release" tag?

Hmm, I think I simply wasn't aware of this tag! :-/

[mkae (Marko Käning)]

Fixed in r129571

[skwerlman@…]

Thanks for the quick replies and fix!

[mkae (Marko Käning)]

You're welcome. :)

Only problem is now that it fails on SL, L & ML buildbots. :(

See #46248

[ryandesign (Ryan Carsten Schmidt)]

Replying to mk@…:

Fixed in r129571

This is ok, it's just a bit odd that our made-up version "0.0.2" is three days older than our prior made-up version "0.0.1".

Replying to mk@…:

Only problem is now that it fails on SL, L & ML buildbots. :(

See #46248

Is this a new problem with "0.0.2", or was it also happening with "0.0.1"?

[mkae (Marko Käning)]

Replying to ryandesign@…:

This is ok, it's just a bit odd that our made-up version "0.0.2" is three days older than our prior made-up version "0.0.1".

Yep, I should have perhaps used 0.0.0 and only increased the revision anyway, as long as the project has no version.

Is this a new problem with "0.0.2", or was it also happening with "0.0.1"?

I don't know, sorry. Must have slipped my attention, but perhaps it is the reason, why I did not use the release version!!! ;-)

[ryandesign (Ryan Carsten Schmidt)]

Replying to ryandesign@…:

Marko, do you recall why you used a commit hash instead of the "2014-06-25-Release" tag?

I only just found ticket #44173, in which Cockatrice was initially added (and an earlier ticket #39747 with a slightly different submission). I had assumed there was not a ticket, since no ticket number was mentioned in the commit message for r121529. I have now added the ticket number to the commit message so that it can be found in the future. If a ticket prompted a commit, please remember to add the ticket number to the commit message.

Both of the submission tickets used fetch.type git and did not specify a commit or version to fetch, so it was of course necessary to fix that before committing, unfortunately the tickets don't say why that particular commit was chosen.

[mkae (Marko Käning)]

Thanks for caring, Ryan.