Michael Haggerty

A note, in case anybody wants to attack this: Probably the first step would be to change `Revision.generate_email_body()` to call a method from the `Environment`, like maybe `Environment.summarize_commit(self)`. That would...

@sudoman: What do you think should happen in this scenario? Is the issue that the whole email was rejected; i.e., that it wasn't sent to other (non-malformed) recipients?

If you are satisfied to use `Bcc:` instead of `To:` (in which case recipients will not see other recipients _or themselves_, this is pretty easy, though it depends on which...

Storing information like this in the repo would have the disadvantage that the `.git-multimail` file would have different contents on different branches, whereas we really want a single config that...

If you give the file(s) the git attribute `-diff`, then instead of getting a diff you will just get a message "binary files differ". If you set this attribute via...

I guess the problem is that `git diff` is not respecting the `.gitattributes` file because it is run in a bare repository. Would you mind putting the same content in...

Thanks. I'll look into this more when I get the chance, especially whether it is realistic for the in-tree `.gitattributes` files to be taken into account when generating diffs.

> Was `git imerge remove --name THE_NAME` enough? It should be enough, yes. If you want to double-check manually, [this section of the README](https://github.com/mhagger/git-imerge#technical-notes) documents the references that `git-imerge` uses...

@hlovdal: sorry for my lack of response. It would be preferable for the environment variables to be set only in the subcommand that is being run rather than in the...

@tucked: thanks for all the help! :sparkles: That kindof took me forever, but I have finally merged your PR along with a couple of minor tweaks and created and uploaded...