search

lehmannro / assert.sh

bash unit testing framework
GNU Lesser General Public License v3.0
487 stars 57 forks source link

CI build failed! #22

Open pahaz opened 8 years ago

pahaz commented 8 years ago

I add some doc lines, but CI tell me that build failed. (https://github.com/lehmannro/assert.sh/pull/21)

Which state of this project? Is it supported?

joseluis commented 7 years ago

For what it's worth, since maintenance seems pretty stalled and apparently no more features will be added, I've recently applied several of pending PRs and fixes in a fork, and also decided to continue developing it, as a a new different project.

lehmannro commented 7 years ago

As I've outlined elsewhere (see @joseluis' comment) the project is feature-complete from my side. I'm happy to accept pull requests for regressions, but I want to keep the base project simple (i.e., no more assert functions.)

I'm particularly unhappy about the fork(s!) because I don't see any good reasons — from a quick glance, correct me if I'm wrong — why they're full forks: An additional source file with a dependency on this project would've done as well. But that's a topic for another discussion forum.

For the CI failures, these are related more to Travis than to assert.sh. See #19 as well. I see @joseluis mentions those as fixed in his fork, and I'd be happy to accept a PR in that regard.

joseluis commented 7 years ago

Hi @lehmannro, regarding the CI fix, if I recall correctly applying #4 would solve it. Only problem is the current merge conflict, which I can't fix in that PR, but I could create a new one if you want.

And regarding the forks, in my case at least, what I really wanted was complete freedom to modify, experiment and evolve this idea in new directions, and since I'm already doing that in aserta, I don't mind removing the original fork that only has the pending PRs applied, so it doesn't create any confusion for your users. Just tell me.

As an idea, maybe having some paragraph in the README stating the preferred way for creating extensions, or maybe solving #13 and use it as a canonical example would clarify this issue for future contributions.