[jira] [Commented] (LUCENE-3449) Fix FixedBitSet.nextSetBit/prevSetBit to support the common usage pattern in every programming book

["Michael McCandless (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/LUCENE-3449?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13112766#comment-13112766 ] 

Michael McCandless commented on LUCENE-3449:
--------------------------------------------

I don't like that this adds an additional if inside next/prevSetBit
that for some cases is not needed (eg BlockJoinQuery's usage).

I think performance trumps convenience here for these low level APIs.

Why not use a simple while loop?

{noformat}
int idx = 0;
while (idx < length) {
  idx = bits.nextSetBit(idx);
  if (idx == -1) {
    break;
  }
  // do something with idx
  idx++;
}
{noformat}

Nomatter what, this loop needs 2 ifs per cycle; I just think that 2nd
if belongs in the loop itself and not inside next/prevSetBit, since
not all use cases require it.


> Fix FixedBitSet.nextSetBit/prevSetBit to support the common usage pattern in every programming book
> ---------------------------------------------------------------------------------------------------
>
>                 Key: LUCENE-3449
>                 URL: https://issues.apache.org/jira/browse/LUCENE-3449
>             Project: Lucene - Java
>          Issue Type: Improvement
>          Components: core/other
>    Affects Versions: 3.4, 4.0
>            Reporter: Uwe Schindler
>            Priority: Minor
>         Attachments: LUCENE-3449.patch
>
>
> The usage pattern for nextSetBit/prevSetBit is the following:
> {code:java}
> for(int i=bs.nextSetBit(0); i>=0; i=bs.nextSetBit(i+1)) {
>  // operate on index i here
> }
> {code}
> The problem is that the i+1 at the end can be bs.length(), but the code in nextSetBit does not allow this (same applies to prevSetBit(0)). The above usage pattern is in every programming book, so it should really be supported. The check has to be done in all cases (with the current impl in the calling code).
> If the check is done inside xxxSetBit() it can also be optimized to be only called seldom and not all the time, like in the ugly looking replacement, thats currently needed:
> {code:java}
> for(int i=bs.nextSetBit(0); i>=0; i=(i<bs.length()-1) ? bs.nextSetBit(i+1) : -1) {
>  // operate on index i here
> }
> {code}
> We should change this and allow out-of bounds indexes for those two methods (they already do some checks in that direction). Enforcing this with an assert is unuseable on the client side.
> The test code for FixedBitSet also uses this, horrible. Please support the common usage pattern for BitSets.

--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org