silentbicycle / theft

property-based testing for C: generate input to find obscure bugs, then reduce to minimal failing input
ISC License
611 stars 31 forks source link

Makefile: add DESTDIR to install rules #41

Closed richardipsum closed 6 years ago

richardipsum commented 6 years ago

Hi there,

I've found this package really useful and noticed that it wasn't available in Debian, I've filed an ITP with Debian for this package https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=910296 and have packaged theft locally with one minor change required to the existing Makefile.

If this change is acceptable to you then once it's merged I should be able to continue with the process of getting theft packaged for Debian.

Thanks, Richard

silentbicycle commented 6 years ago

This looks great, thanks. Do you mind if I re-target it against the develop branch, so I can incorporate it into a versioned release?

I'm planning on doing a bugfix release fairly soon (0.4.4), which should include this, and would likely be a better version for packaging. I've been working on version 0.5.0 for quite a while (it adds multi-core support, but requires most of the internals to be rewritten), but Makefile improvements and a couple bug fixes shouldn't wait.

Also, I'd love to hear about how you're using theft.

richardipsum commented 6 years ago

Sure, sounds good to me. I'll keep an eye out for 0.4.4 and update the packaging accordingly.

My uses for theft aren't too interesting to be honest, just some internal tests for a linked list, though I did write up an article on theft a while back that might be of interest, it's pretty basic, https://yakking.branchable.com/posts/property-testing-in-c/

silentbicycle commented 6 years ago

Excellent. I'm planning on doing the 0.4.4 release this weekend, and this will get cherry-picked to it. Thanks!

Also, your article covers the core API well, that's what use should look like in general moving forward. :+1: (Except the autoshrink_config.enable part will go away in 0.5.0 -- I'm removing custom shrinkers entirely, because writing good custom shrinkers that aren't quadratic can be quite tricky, and I'd rather focus on improving autoshrinking. Also, supporting custom shrinkers massively complicates multi-core shrinking.)

richardipsum commented 6 years ago

@silentbicycle In case you were wondering I have uploaded a debian package for this library to mentors, it requires sponsorship, sadly it turns out it could be stuck in this state for sometime which I didn't realise when I started this.

If you know anyone in debian who is a maintainer and might be interested in helping out the RFS is here https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=910608

Thanks, Richard

silentbicycle commented 6 years ago

I'll ask around, thanks for getting the ball rolling and letting me know.

algernon commented 6 years ago

I'll take a look at the debian package tomorrow, will see if I can lend a hand with sponsorship (I'm a DD). Mind you, even if I can sponsor this upload, I can't commit to continued sponsorship. I'd encourage @richardipsum to apply for debian maintainership in the not too distant future, to remove the need for a sponsor :) (Mind you, having a package in the archive will certainly help with that)

richardipsum commented 6 years ago

Thanks, and I'll definitely take a look into what is required for maintainership.