Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > comp.lang.java.programmer > #19263

Re: Glitch in Java Collections (No descendingMap in LinkedHashMap)

From Jim Janney <jjanney@shell.xmission.com>
Newsgroups comp.lang.java.programmer
Subject Re: Glitch in Java Collections (No descendingMap in LinkedHashMap)
Date 2012-10-12 09:12 -0600
Organization objurgating the centipede
Message-ID <ydnobk767xa.fsf@shell.xmission.com> (permalink)
References (4 earlier) <ydnvcei70lu.fsf@shell.xmission.com> <ydnr4p66x16.fsf@shell.xmission.com> <k54sau$htf$1@dont-email.me> <ydnfw5m6iw0.fsf@shell.xmission.com> <k592ob$8b4$1@news.albasani.net>

Show all headers | View raw


Jan Burse <janburse@fastmail.fm> writes:

> Any particular reason you use WeakHashMap?
> You have:
>
>   @Override
>   public V remove(Object key) {
>     insertionRanks.remove(key);
>     return super.remove(key);
>   }
>
> So I guess an ordinary HashMap would do.

I was worried about a potential memory leak, but I hadn't thought it
through.  The keys are held in the TreeMap anyway, so using WeakHashMap
doesn't accomplish anything.

> Further your implementation would not generate the
> order I desire. It generates an insert-order, but
> not a first-insert-order, but last-insert-order
> since you have:
>
>   @Override
>   public V put(K key, V value) {
>     insertionRanks.put(key, nextInsertionRank++);
>     return super.put(key, value);
>   };
>
> So when I do:
>
>     put("2","a")
>     put("1","b")
>     put("2","a")
>
> I get the order "1", "2".
> But result should be "2", "1".
>
> Bye

I missed that.  For first-insert order it should be

   @Override
   public V put(K key, V value) {
     if (!insertionRanks.containsKey(key)) {
        insertionRanks.put(key, nextInsertionRank++);
     }
     return super.put(key, value);
   };

And for last-insert order the original code is wrong: it should be

   @Override
   public V put(K key, V value) {
     super.remove(key);
     insertionRanks.put(key, nextInsertionRank++);
     return super.put(key, value);
   };

The tree structure is based on the ordering, so if you change the
ordering you have to modify the tree also.

>
> P.S.: It is also not access order as Eric Sosman
> claims, since the get() does not influence the order.

Right.  For access order you would have to remove the entry and add it
back on every get.

-- 
Jim Janney

Back to comp.lang.java.programmer | Previous | NextPrevious in thread | Next in thread | Find similar | Unroll thread


Thread

Glitch in Java Collections (No descendingMap in LinkedHashMap) Jan Burse <janburse@fastmail.fm> - 2012-10-04 14:09 +0200
  Re: Glitch in Java Collections (No descendingMap in LinkedHashMap) Robert Klemme <shortcutter@googlemail.com> - 2012-10-04 23:01 +0200
    Re: Glitch in Java Collections (No descendingMap in LinkedHashMap) Lew <lewbloch@gmail.com> - 2012-10-04 14:43 -0700
      Re: Glitch in Java Collections (No descendingMap in LinkedHashMap) Jan Burse <janburse@fastmail.fm> - 2012-10-05 00:17 +0200
      Re: Glitch in Java Collections (No descendingMap in LinkedHashMap) Robert Klemme <shortcutter@googlemail.com> - 2012-10-05 19:11 +0200
        Re: Glitch in Java Collections (No descendingMap in LinkedHashMap) Lew <lewbloch@gmail.com> - 2012-10-05 11:05 -0700
          Re: Glitch in Java Collections (No descendingMap in LinkedHashMap) Jan Burse <janburse@fastmail.fm> - 2012-10-05 20:18 +0200
          Re: Glitch in Java Collections (No descendingMap in LinkedHashMap) Daniel Pitts <newsgroup.nospam@virtualinfinity.net> - 2012-10-05 12:06 -0700
            Re: Glitch in Java Collections (No descendingMap in LinkedHashMap) Lew <lewbloch@gmail.com> - 2012-10-05 13:04 -0700
              Re: Glitch in Java Collections (No descendingMap in LinkedHashMap) Jan Burse <janburse@fastmail.fm> - 2012-10-05 23:44 +0200
                Re: Glitch in Java Collections (No descendingMap in LinkedHashMap) Jan Burse <janburse@fastmail.fm> - 2012-10-05 23:50 +0200
                Re: Glitch in Java Collections (No descendingMap in LinkedHashMap) Jan Burse <janburse@fastmail.fm> - 2012-10-06 00:00 +0200
                Re: Glitch in Java Collections (No descendingMap in LinkedHashMap) glen herrmannsfeldt <gah@ugcs.caltech.edu> - 2012-10-05 23:39 +0000
                Re: Glitch in Java Collections (No descendingMap in LinkedHashMap) Jan Burse <janburse@fastmail.fm> - 2012-10-06 12:18 +0200
                Re: Glitch in Java Collections (No descendingMap in LinkedHashMap) Jan Burse <janburse@fastmail.fm> - 2012-10-06 12:22 +0200
              Re: Glitch in Java Collections (No descendingMap in LinkedHashMap) Daniel Pitts <newsgroup.nospam@virtualinfinity.net> - 2012-10-05 15:51 -0700
          Re: Glitch in Java Collections (No descendingMap in LinkedHashMap) Robert Klemme <shortcutter@googlemail.com> - 2012-10-06 13:53 +0200
  Re: Glitch in Java Collections (No descendingMap in LinkedHashMap) Jim Janney <jjanney@shell.xmission.com> - 2012-10-09 14:16 -0600
    Re: Glitch in Java Collections (No descendingMap in LinkedHashMap) Jim Janney <jjanney@shell.xmission.com> - 2012-10-10 09:00 -0600
      Re: Glitch in Java Collections (No descendingMap in LinkedHashMap) Eric Sosman <esosman@comcast-dot-net.invalid> - 2012-10-10 11:34 -0400
        Re: Glitch in Java Collections (No descendingMap in LinkedHashMap) Jim Janney <jjanney@shell.xmission.com> - 2012-10-10 10:28 -0600
          Re: Glitch in Java Collections (No descendingMap in LinkedHashMap) Jim Janney <jjanney@shell.xmission.com> - 2012-10-10 11:45 -0600
            Re: Glitch in Java Collections (No descendingMap in LinkedHashMap) Lew <lewbloch@gmail.com> - 2012-10-10 10:52 -0700
              Re: Glitch in Java Collections (No descendingMap in LinkedHashMap) Jim Janney <jjanney@shell.xmission.com> - 2012-10-10 12:47 -0600
                Re: Glitch in Java Collections (No descendingMap in LinkedHashMap) Jim Janney <jjanney@shell.xmission.com> - 2012-10-11 14:17 -0600
            Re: Glitch in Java Collections (No descendingMap in LinkedHashMap) Eric Sosman <esosman@comcast-dot-net.invalid> - 2012-10-10 18:20 -0400
              Re: Glitch in Java Collections (No descendingMap in LinkedHashMap) Jim Janney <jjanney@shell.xmission.com> - 2012-10-10 16:51 -0600
                Re: Glitch in Java Collections (No descendingMap in LinkedHashMap) Jan Burse <janburse@fastmail.fm> - 2012-10-12 14:34 +0200
                Re: Glitch in Java Collections (No descendingMap in LinkedHashMap) Jim Janney <jjanney@shell.xmission.com> - 2012-10-12 09:12 -0600
                Re: Glitch in Java Collections (No descendingMap in LinkedHashMap) Eric Sosman <esosman@comcast-dot-net.invalid> - 2012-10-12 11:42 -0400
                Re: Glitch in Java Collections (No descendingMap in LinkedHashMap) Jan Burse <janburse@fastmail.fm> - 2012-10-12 21:02 +0200

csiph-web