Add JMH-based benchmark project with example benchmarks

[sritchie]

Closes #31. This PR also

• Updates binary compatibility information
• Removes the MIMA check from the build because we've already broken binary compatibility with 0.6.0!!
• fixes some busted info in publishSettings

[codecov-io]

Current coverage is 73.99% (diff: 100%)

Merging #194 into master will not change coverage

@@             master       #194   diff @@
==========================================
  Files            36         36          
  Lines           769        769          
  Methods         708        708          
  Messages          0          0          
  Branches         61         61          
==========================================
  Hits            569        569          
  Misses          200        200          
  Partials          0          0          

Powered by Codecov. Last update 06f311f...d1e0307

[johnynek]

How can we have broken mima compat with 0.6.0 if mima is turned on? If we broke compatibility we need to fix it. Or make a new release with mima on.

We are going to get burned if we have this at the bottom of algebird and Spire but break compatibility all the time.

[sritchie]

None of the projects were specifying their youngest compatible versions. I added those and boom, failure. On Sat, Dec 3, 2016 at 10:33 AM P. Oscar Boykin notifications@github.com wrote:

How can we have broken mima compat with 0.6.0 if mima is turned on? If we broke compatibility we need to fix it. Or make a new release with mima on.

We are going to get burned if we have this at the bottom of algebird and Spire but break compatibility all the time.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/typelevel/algebra/pull/194#issuecomment-264653201, or mute the thread https://github.com/notifications/unsubscribe-auth/AAEQA8kUT_cy8uRClvy_Mt-W82YHSXDTks5rEafZgaJpZM4LDWxA .

[non]

Oh jeez. Can you add a comment with the binary incompatibilities?

[sritchie]

yeah, one second...

[sritchie]

[info] core: found 1 potential binary incompatibilities while checking against org.typelevel:algebra_2.12:0.6.0
[error]  * method this()Unit in class algebra.instances.BigDecimalAlgebra does not have a correspondent in current version
[error]    filter with: ProblemFilters.exclude[DirectMissingMethodProblem]("algebra.instances.BigDecimalAlgebra.this")
[info] Done updating.
[info] benchmark: previous-artifact not set, not analyzing binary compatibility
[info] laws: found 0 potential binary incompatibilities while checking against org.typelevel:algebra-laws_2.12:0.6.0
[info] laws: found 0 potential binary incompatibilities while checking against org.typelevel:algebra-laws_2.12:0.6.0
[info] core: found 1 potential binary incompatibilities while checking against org.typelevel:algebra_2.12:0.6.0
[error]  * method this()Unit in class algebra.instances.BigDecimalAlgebra does not have a correspondent in current version
[error]    filter with: ProblemFilters.exclude[DirectMissingMethodProblem]("algebra.instances.BigDecimalAlgebra.this")

[johnynek]

Ahh, this was Thomas's change to accept a parameter to the constructor. We can fix that by adding a default constructor with no params. Problem solved.

def this() = this(BigDecimal... On Sat, Dec 3, 2016 at 10:28 Sam Ritchie notifications@github.com wrote:

[info] core: found 1 potential binary incompatibilities while checking against org.typelevel:algebra_2.12:0.6.0 [error] method this()Unit in class algebra.instances.BigDecimalAlgebra does not have a correspondent in current version [error] filter with: ProblemFilters.excludeDirectMissingMethodProblem [info] Done updating. [info] benchmark: previous-artifact not set, not analyzing binary compatibility [info] laws: found 0 potential binary incompatibilities while checking against org.typelevel:algebra-laws_2.12:0.6.0 [info] laws: found 0 potential binary incompatibilities while checking against org.typelevel:algebra-laws_2.12:0.6.0 [info] core: found 1 potential binary incompatibilities while checking against org.typelevel:algebra_2.12:0.6.0 [error] method this()Unit in class algebra.instances.BigDecimalAlgebra does not have a correspondent in current version [error] filter with: ProblemFilters.excludeDirectMissingMethodProblem

— You are receiving this because you commented.

Reply to this email directly, view it on GitHub https://github.com/typelevel/algebra/pull/194#issuecomment-264663436, or mute the thread https://github.com/notifications/unsubscribe-auth/AAEJdmZ07IbUnRhT5J9RVh3wjP5WQWP8ks5rEdDjgaJpZM4LDWxA .

[non]

👍 Thanks for catching this @sritchie

[sritchie]

@johnynek @non that addition worked great! I changed it and re-enabled mima. let's see how the build goes.

[non]

If that works I can try to do a rush 0.6.1 release to preserve 0.5.x compatibility.

[sritchie]

Ah, is this 0.5 compatible too? If so feel free to change the minimum version in this PR! I've gotta pull myself away for the weekend. On Sat, Dec 3, 2016 at 1:53 PM Erik Osheim notifications@github.com wrote:

If that works I can try to do a rush 0.6.1 release to preserve 0.5.x compatibility.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/typelevel/algebra/pull/194#issuecomment-264664803, or mute the thread https://github.com/notifications/unsubscribe-auth/AAEQA01qAZiDpOCQ62W_chg-a41rPsi3ks5rEda8gaJpZM4LDWxA .

[non]

No, sorry, I got confused on which version we were on. Ignore me!

[sritchie]

Hmmm, looks like we have another legit binary compatibility issue. This commit is the culprit: https://github.com/typelevel/algebra/commit/24fe85a71d58056d07c954f9a1a900159a053ea8

it renames Ring.fromDouble to defaultFromDouble.

[error]  * method defaultFromDouble(Double,algebra.ring.Ring,algebra.ring.MultiplicativeGroup)java.lang.Object in trait algebra.ring.RingFunctions is present only in current version
[error]    filter with: ProblemFilters.exclude[ReversedMissingMethodProblem]("algebra.ring.RingFunctions.defaultFromDouble")
[info] core: found 1 potential binary incompatibilities while checking against org.typelevel:algebra_2.11:0.6.0
[error]  * method defaultFromDouble(Double,algebra.ring.Ring,algebra.ring.MultiplicativeGroup)java.lang.Object in trait algebra.ring.RingFunctions is present only in current version
[error]    filter with: ProblemFilters.exclude[ReversedMissingMethodProblem]("algebra.ring.RingFunctions.defaultFromDouble")

@non @johnynek, do one of you want to fix this in a separate PR, or commit to this PR? Might be better to use this one to leverage the mima check.

[johnynek]

Maybe it is best to make a PR to turn Mima back on and fix the issues (or quickly make a 0.7.0 of algebra).

One idea to fix issues would be to add methods to RingFunctions[R] using an implicit class. Kind of round about, but safe for binary compatibility and makes source compatibility closer to binary compatibility.

[sritchie]

:boom: done @johnynek, and thanks for the suggested fix!

[thomas-stripe]

👍

[sritchie]

@thomas-stripe, can I trouble you for a merge?