Lacks documentation on how it differs from MRI #2148

Closed
agrimm opened this Issue Feb 10, 2013 · 5 comments

Projects

None yet

3 participants

@agrimm
Contributor
agrimm commented Feb 10, 2013

JRuby has documentation describing how it is different from MRI, located at https://github.com/jruby/jruby/wiki/DifferencesBetweenMriAndJruby . Rubinius does not appear to have such documentation.

As it seems Rubinius has intentional differences from MRI (for example #2054 ), it'd be helpful to list how it, too, differs from MRI.

Like JRuby's documentation, it doesn't have to cover all differences, just those cases that are not bugs.

It'll be helpful for those wanting to know whether or not to use Rubinius, and also avoid bug reports being filed when something isn't actually a bug.

I've previously filed a similar issue at https://github.com/rubinius/website/issues/1 , but that's suggesting that the differences be documented in the Rubinius website, whereas this issue is requesting that the difference be documented somewhere, whether it be within the website, within the Rubinius git repo, or somewhere else.

Just so I'm not misinterpreted, I'm fine with there being intentional differences between Rubinius and MRI - I'm not complaining about such differences existing.

@jc00ke
Member
jc00ke commented Feb 10, 2013

PDI 😉 There are a lot of places you could start looking, like RubySpec. I think this kind of content would best be suited for the website, but that's in the web/ dir, not in website repo.

@brixen
Member
brixen commented Feb 13, 2013

Rubinius is a drop-in replacement for MRI. There is almost no place in Ruby code that we differ, or should differ. There are some places that we cannot support C-extensions that make use of MRI ad hoc C functions that expose MRI implementation details.

We have been as explicit as possible about Rubinius being a drop-in replacement for MRI. If you want to look at not_compliant_on :rubinius guards in RubySpec and send a documentation PR, that would be cool. Otherwise, I'm probably going to close this as too vague to be actionable.

@agrimm
Contributor
agrimm commented Feb 13, 2013

I'm willing to do this, though I'll need what I write checked by someone more technically savvy than I am.

@brixen
Member
brixen commented Feb 13, 2013

@agrimm sounds great, happy to review. Just FYI, I'm about to split out the docs into a separate gem and bundle with Rubinius using our pre-installed gem facility. Feel free to send PR here if I haven't done that yet, but if you look for web/ and it's missing, you'll know why. :)

@agrimm
Contributor
agrimm commented Sep 29, 2013

Closing, as currently I do not anticipate me writing the documentation for this. If anyone else wants the differences documented, feel free to re-open.

@agrimm agrimm closed this Sep 29, 2013
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment