Thanks for the ticket pointer. I didn't find it in my initial search because it "Affects Version/s: None" :-/, and I knew this behavior was new to 1.6.
For those not up for the medium-length comment-thread read: non-Clojure-aware Java Collections Framework interface instances are not considered values,* and 'Rich says "all bets should be off for hasheq/equiv of non-values."' That's a bit down in the details, but I think the upshot is that they're not currently supported as set members or hash keys. Unfortunately they used to be supported (unintentionally, I guess) due to Clojure following Java's lead, so I have some interop-heavy code to review and test very carefully. Though the above seems to frame non-support as based on principle, it seems the driver is keeping the new hasheq stuff highly performant. I would hope a performant implementation that supported the Collections interfaces would be accepted for some future version. *If there is one, I'd like to read a clear definition of what "values" means in Clojure. The headings on http://clojure.org/data_structures are almost an enumeration of them, except the "Collections" section talks a little about Java Collections classes, hashCode, and hasheq without making it clear that non-Clojure collections are not values. Unfortunately the current definition seems to admit only the Java primitives and near-primitives (String and primitive wrappers) and clojure-aware types. What "not supported" looks like via maps: user> *clojure-version* {:major 1, :minor 6, :incremental 0, :qualifier nil} user> (hash-map [1 2] "v" (java.util.ArrayList. [1 2]) "a") {[1 2] "a", [1 2] "v"} user> {[1 2] "v" (java.util.ArrayList. [1 2]) "a"} IllegalArgumentException Duplicate key: [1 2] clojure.lang.PersistentArrayMap.createWithCheck (PersistentArrayMap.java:70) user> (-> {} (assoc [1 2] "v") (assoc (java.util.ArrayList. [1 2]) "a")) {[1 2] "a"} On Friday, May 9, 2014 6:34:26 PM UTC-5, Andy Fingerhut wrote: > > Mike is correct. This change in behavior is due to the hashing changes in > Clojure 1.6. See the comments on ticket > http://dev.clojure.org/jira/browse/CLJ-1372 for some discussion of > whether this is considered a bug. It appears that perhaps the hash > consistency is not promised for mutable objects, only immutable ones. > > Below are a couple more REPL transcripts that illustrate the change in > behavior: > > user=> *clojure-version* > {:major 1, :minor 5, :incremental 1, :qualifier nil} > user=> (= [1 2] (java.util.ArrayList. [1 2])) > true > user=> (map hash [ [1 2] (java.util.ArrayList. [1 2]) ]) > (994 994) > user=> (hash-map [1 2] "vec" (java.util.ArrayList. [1 2]) "alist") > {[1 2] "alist"} > > > > user=> *clojure-version* > {:major 1, :minor 6, :incremental 0, :qualifier nil} > user=> (= [1 2] (java.util.ArrayList. [1 2])) > true > user=> (map hash [ [1 2] (java.util.ArrayList. [1 2]) ]) > (156247261 994) > user=> (hash-map [1 2] "vec" (java.util.ArrayList. [1 2]) "alist") > {[1 2] "alist", [1 2] "vec"} > > -- You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com Note that posts from new members are moderated - please be patient with your first post. To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en --- You received this message because you are subscribed to the Google Groups "Clojure" group. To unsubscribe from this group and stop receiving emails from it, send an email to clojure+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
