The check is made at read-time, by the time to form gets to the compiler it's already a hash-map and one of your forms will have been dropped. So the decision was made to make the reader check. One way you could solve your problem here is with tagged literals, as the literal would be created, and therefore the UUID would be unique before you even got to the creation of the hash-map.
http://clojure.org/reference/reader#_tagged_literals In short though...what should the reader do, it has no clue which of your two key/value pairs to drop. On Wed, Jan 4, 2017 at 9:05 PM, Tianxiang Xiong <tianxiang.xi...@gmail.com> wrote: > Using a Clojure 1.8.0 REPL, I get the following: > > user=> {(+ 1 2) 1 (+ 1 2) 2} > IllegalArgumentException Duplicate key: (+ 1 2) clojure.lang. > PersistentArrayMap.createWithCheck (PersistentArrayMap.java:71) > > It seems that the a check for key equality is made *before* the key is > evaluated, when they are still strings. > > PersistentArrayMap/createWithCheck is as below: > > static public PersistentArrayMap createWithCheck(Object[] init){ > for(int i=0;i< init.length;i += 2) > { > for(int j=i+2;j<init.length;j += 2) > { > if(equalKey(init[i],init[j])) > throw new IllegalArgumentException("Duplicate key: " + init[i]); > } > } > return new PersistentArrayMap(init); > } > > The MapReader in LispReader.java seems to pass an Object[] of Strings to > PersistentArrayMap/createWithCheck. > > public static class MapReader extends AFn{ > public Object invoke(Object reader, Object leftparen, Object opts, Object > pendingForms) { > PushbackReader r = (PushbackReader) reader; > Object[] a = readDelimitedList('}', r, true, opts, > ensurePending(pendingForms)).toArray(); > if((a.length & 1) == 1) > throw Util.runtimeException("Map literal must contain an even number of > forms"); > return RT.map(a); > } > > } > > > This behavior is surprising. Is there a reason the map creation process > works this way? > > The above example may seem trivial, since we would have duplicate keys > after evaluation as well, but consider the generation of random values: > > user=> {(java.util.UUID/randomUUID) 1 (java.util.UUID/randomUUID) 2} > IllegalArgumentException Duplicate key: (java.util.UUID/randomUUID) > clojure.lang.PersistentArrayMap.createWithCheck > (PersistentArrayMap.java:71) > > > -- > 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. > -- “One of the main causes of the fall of the Roman Empire was that–lacking zero–they had no way to indicate successful termination of their C programs.” (Robert Firth) -- 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.
