enhancements for this module

[grawk]

Hi. I've written a grunt task plugin called grunt-loop-mocha to run highly configurable parallel mocha processes.

I'm looking to achieve a similar functionality in gulp. I'm still in reading mode on your module, but it looks like you achieve parallelism on files only.

I'm wondering, have you given any thought to how you might achieve other parallel options such as:

• segregate by folder
• segregate by file groupings (glob)
• segregate by mocha grep

It looks like you've already got the ability to customize the spawned environment variables, which is a big one for my use case.

Anyway, I'm going to spend some time over the next few days exploring how I might achieve these things using this module as a starting point. I wanted to start a conversation here in case you are interested in molding the direction and possibly folding it back into your module.

Thanks!

[sebv]

You can have a look here to get some ideas.

I use gulp built in glob feature and promisePipe to spread the tests among different VMs (I guess you could use stream directly but I like promises).

Anyway if there is way we can improve this I am all for it.

[grawk]

Great. Thanks for that. I'm rather new to streams (sad to admit since I've been programming in Node for almost 4 years) so it's not immediately clear how this exactly works. But it appears that you are creating several parallel mocha runs with different mocha options and different file sets. That's essentially what I'm looking for. So I'll take some time to understand this pattern. Thanks again.

[sebv]

You may also find this interesting, this is an alternative to grep: https://github.com/sebv/mocha-bdd-with-opts.