Matt

> That's all fine, if you can take the performance penalty and latency from the external binary. In order to be sufficiently generic, any hook is necessarily going to have...

> One reason I've avoided hooks so far is that we sometimes want solutions that can validate the commits after the operation is done anyway I think that both pre-*...

## Upload hooks ### Pre-upload hooks Not too much to say. Pre-upload hooks should be able to mutate the commits in the stack (eg. running formatters). I imagine that they...

Note for anyone reading this thread in the future: https://github.com/martinvonz/jj/issues/1869 is the tracking bug of `jj run` > we can just rewrite the actual objects with a new transaction instead...

I've been considering this further, and also chatted to @mlcui-corp offline. My conclusions are: * Pre-upload hooks are an absolute must * While the user could achieve them with `jj...

@mlcui-corp also mentioned to me that pre-describe hooks could be made more difficult by `jj describe -m foo`, or anything where you specify the description in the command-line. I don't...

Currently, pre-upload hooks refer only to `jj git push` in the public build. However, `jj gerrit send` (#2845) is currently in progress, and would presumably also use this same mechanism....

Ah, I get you now, thanks for clarifying. > Does your view take into account solo projects, not pushed to a remote? Or contributors that are stuck offline for awhile,...

Also, @PhilipMetzger, I'd like to hear your thoughts on my updated plans, since you seemed to have issue with hooks before. I understand that you don't want hooks in the...

Here's a simple proposal for an example implementation. I think it's quite nice, but as I said before, I'm open to other ideas. # Step 1: Multi-layered config files We...