Path: csiph.com!x330-a1.tempe.blueboxinc.net!usenet.pasdenom.info!weretis.net!feeder4.news.weretis.net!news.musoftware.de!wum.musoftware.de!fu-berlin.de!uni-berlin.de!individual.net!not-for-mail
From: blmblm@myrealbox.com <blmblm.myrealbox@gmail.com>
Newsgroups: comp.lang.java.programmer
Subject: Re: StringBuilder Difficulties
Date: 5 Jul 2011 21:16:44 GMT
Organization: None
Lines: 46
Message-ID: <97hd9sFa1jU2@mid.individual.net>
References: <c4tk07dv1pq7u8k32qd6b7o1brc2rj8r30@4ax.com> <ual3175uj4e5jb7u44q8jn7soo7k532heq@4ax.com> <97h616FhaeU3@mid.individual.net> <jsu6171vmasu6jfvsbroa052n8u7ukarr3@4ax.com>
X-Trace: individual.net F+hynIUAf3MJZWCAfcYIQg5dvrJLMLpSjdVi75BLNpxaa61Om0
X-Orig-Path: not-for-mail
Cancel-Lock: sha1:R+uHpK+NPqk1YQdfAvHoI6gQvrM=
X-Newsreader: trn 4.0-test76 (Apr 2, 2001)
Xref: x330-a1.tempe.blueboxinc.net comp.lang.java.programmer:5874

In article <jsu6171vmasu6jfvsbroa052n8u7ukarr3@4ax.com>,
Gene Wirchenko  <genew@ocis.net> wrote:
> On 5 Jul 2011 19:12:39 GMT, blmblm@myrealbox.com
> <blmblm.myrealbox@gmail.com> wrote:
> 
> [snip]
> 
> >What I found is that HashSet was noticeably faster on all the
> >systems where I ran the benchmarks.  Unless you need for the set
> >to be sorted (and it's not apparent from your code that you do),
> >why not .... ?  (I'm curious too about why you chose TreeSet in
> >the first place.  ? )
> 
>   1) I can output the set in order without having to do anything else.
> My real program has a lot of debugging info dumping.  (Read as "checks
> that I have not done something wrong".)


Ah.  Well, yes, then you probably do need a SortedSet, though
considering that you initially build the set from a string that's in
order, maybe you could use that (the string) instead.  Just sayin',
maybe.


>   2) When I read "hash", I think "collision", and I get nervous.
> Nothing I read reassured me that that could not happen.


Why would that make you nervous?  If you're worried about
correctness, um, as far as I know hash tables are supposed to
deal with collisions in some way that preserves the overall "map"
semantics.  Performance may suffer if there are a lot of collisions,
but -- benchmark on the system(s) of interest and check?


>   3) I had to pick something.  If it works, I can change it later.  If
> it does not, I have not solved my problem yet.  The former is safer.


Sure.  Then again, if you were only concerned about getting something
that works, why try various alternatives ....  <shrug>


-- 
B. L. Massingill
ObDisclaimer:  I don't speak for my employers; they return the favor.