Optimized pattern matching on deftype/record, POJOs

Question

(match [x] [({:a 0 :b 1} :type Foo)] :a0 ...)

:a and :b would be converted to field access, i.e. (.a x)

As with primitive array matching, we should do an instance check followed by hinted field access.

We should consider adopting the deftype/record syntax.

Answer 1

Comment made by: dnolen

(match [x] [^Foo {.-bar 5 .-baz 7}] :a0 [^Foo {.-bar _ .-woz 8}] :a1 :else :a2)

Given the recent ClojureScript property access discussions this seems like an acceptable syntax for high-performance property access.

Maybe:

(match [^Foo x] [{.-bar 5 .-baz 7}] :a0 [{.-bar _ .-woz 8}] :a1 :else :a2)

But probably not since then we should probably throw if someone tries to declare a different type in the row.

Answer 2

Comment made by: pron

Hi. Any plans on resolving this?

Answer 3

Comment made by: dnolen

I have no specific timeline, it will definitely be after 0.2.0, only focusing on bugs at the moment.

Answer 4

Comment made by: leonardoborges

Is it worth me picking this up?

I've implemented a very rudimentary version of this that I'm using here: https://github.com/leonardoborges/imminent#pattern-matching

Would love to see this extended and officially supported though. Thoughts?

Answer 5

Comment made by: dnolen

Yes happy to see someone pick this up.

Answer 6

Comment made by: leonardoborges

Excellent! Would you have any pointers/guidelines? If not, I'll just go from first principles using what's described in https://github.com/clojure/core.match/wiki/Advanced-usage#participating-in-pattern-matching

Answer 7

Comment made by: dnolen

Yeah the primary issue is just figuring out a good syntax for it. Once that's in place I don't think getting it working will be so hard.

Answer 8

Reference: https://clojure.atlassian.net/browse/MATCH-4 (reported by dnolen)