Alex

> Those that are present are correct. 🎉 🙇 to #69 > 42662 missing files. It seems part of the missing files are actually not in the index, so it...

👍 I just have started another round of `pga get` with #69 and will be happy to make another full run, as soon as this issue is resolved.

🙏 @zurk Were you thinking about adding it to https://github.com/src-d/ml/tree/master/doc ?

3 solutions discussed for storing re-try state: - in memory, global Map of URLs - RDBMS, add re-try count - external system: Queue or Redis first 2 solution would be...

What do you think would be a better place to take this information from, after moving binary to `src-d/lookout-sdk`?

Using https://github.com/src-d/style-analyzer/compare/c99dcdff172f1cb5505603a45d054998cb4dd606...src-d:3a9d78 example from https://github.com/src-d/lookout/issues/144#issuecomment-413528797 it's not immediately clear why changes related to submodule in `core/server` are even detected by `DiffTreeScanner`. After they are detected in a Trees, of course...

1. and 2. is preliminary done - App and ML teams have consensus that we do not want to get submodules changes as changes in lookout. This is only about...

After a bit more investigation - indeed, a submodule seems to have a entry that is different - it's a "Commit" (!) under `core/server` ``` $ git diff-tree --raw -r...

Was able to reproduce it on https://github.com/bzz/submodule/commit/f46a1e3c7c90d63c07bf47beb13c9ffe5fab3124 🎉 Funny enough, this does not happen on [git-fixtures/submodule](https://github.com/git-fixtures/submodule/compare/c7431b5bc9d45fb64a87d4a895ce3d1073c898d2...47770b26e71b0f69c0ecd494b1066f8d1da4fc03) repo, as it does not include any commit that does both: - modifies regular...

putting back to backlog due to priority of SDK tasks