Skip to content

rtags: 2.8-p1 -> 2.10 #26280

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 3 commits into from
Jun 4, 2017
Merged

rtags: 2.8-p1 -> 2.10 #26280

merged 3 commits into from
Jun 4, 2017

Conversation

acowley
Copy link
Contributor

@acowley acowley commented May 31, 2017

Motivation for this change
Things done
  • Tested using sandboxing
    (nix.useSandbox on NixOS,
    or option build-use-sandbox in nix.conf
    on non-NixOS)
  • Built on platform(s)
    • NixOS
    • macOS
    • Linux
  • Tested compilation of all pkgs that depend on this change using nix-shell -p nox --run "nox-review wip"
  • Tested execution of all binary files (usually in ./result/bin/)
  • Fits CONTRIBUTING.md.

Sorry, something went wrong.

@acowley
Copy link
Contributor Author

acowley commented May 31, 2017

I don't understand the Travis CI failure. It built successfully in two configurations, but a third ends up with an unexpected hash.

@Mic92
Copy link
Member

Mic92 commented May 31, 2017

the first build only check for evaluation errors. The last two check builds on darwin and linux.

@Mic92
Copy link
Member

Mic92 commented May 31, 2017

I get the same error as travis:

output path ‘/nix/store/5p4zqpdy61bh86l2hmjxsi07caj8zzbf-rtags’ has r:sha256 hash ‘1swcz4p5v6zqzqrxdbxbcvy009k509s4ni8qqg1wyp537qcibwdb’ when ‘0z32yncn3x1pm2l6r349kfwi6mi83vll820kka87agcla64pl1f2’ was expected  

I am also on Linux.

@acowley
Copy link
Contributor Author

acowley commented Jun 1, 2017

Thank you @Mic92! Can you think of a reason why the sha256 would be different between darwin and Linux? Doing a stdenv.isDarwin check on the hash of the source seems worrying, but I can do so if you think that's the way to go.

@Mic92
Copy link
Member

Mic92 commented Jun 1, 2017

Maybe fetchgit (with submodules ?) is not so reproducible as it should be. It could be something fs related, which leads to a deviation in the checksum. However this is the first time I encounter that.

@Mic92
Copy link
Member

Mic92 commented Jun 1, 2017

nix-prefetch-git is also consistent on my system:

nix-prefetch-git --fetch-submodules https://github.com/andersbakken/rtags.git refs/tags/v2.10
{                                 
  "url": "https://github.com/andersbakken/rtags.git",               
  "rev": "3b3ace901f53827daef81d4c13658ec4feb318b4",                
  "date": "2017-05-26T14:04:03-07:00",                              
  "sha256": "1swcz4p5v6zqzqrxdbxbcvy009k509s4ni8qqg1wyp537qcibwdb", 
  "fetchSubmodules": true         
}

@acowley
Copy link
Contributor Author

acowley commented Jun 1, 2017

I investigated a bit earlier today with some suggestions from IRC, and the most suspicious diff I could find between Linux and Darwin is a file with UTF8 characters in its name and contents. When I dumped file listings to a file on each system and opened in emacs, ediff highlighted that file name as a difference. Possibly a red herring, but it's all I came up with.

@pSub pSub added the 8.has: package (update) This PR updates a package to a newer version label Jun 2, 2017
@Mic92
Copy link
Member

Mic92 commented Jun 3, 2017

Probably HFS plus does unicode normalisation, which is not the case for other filesystems commonly found on Linux.

@Mic92
Copy link
Member

Mic92 commented Jun 3, 2017

The quick fix would be to just have checksums per platform at the moment, if this is really the root cause.

@LnL7
Copy link
Member

LnL7 commented Jun 3, 2017

Yep, it's the filesystem. I have a machine that doesn't use HFS+ for the store and I get the same hash as linux there.

@LnL7
Copy link
Member

LnL7 commented Jun 3, 2017

With #26358 and this I get the same hash everywhere.

-    sha256 = "0z32yncn3x1pm2l6r349kfwi6mi83vll820kka87agcla64pl1f2";
+    sha256 = "0rv5hz4cfc1adpxvp4j4227nfc0p0yrjdc6l9i32jj11p69a5401";
+    postFetch = ''
+      rm $out/src/rct/tests/testfile_*.txt
+    '';

@Mic92
Copy link
Member

Mic92 commented Jun 4, 2017

HFS+ will hopefully be deprecated soon.

@Mic92 Mic92 merged commit 8490c83 into NixOS:master Jun 4, 2017
@acowley acowley deleted the rtags-2.10 branch October 9, 2017 18:16
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
8.has: package (update) This PR updates a package to a newer version
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

4 participants