[ 
https://issues.apache.org/jira/browse/LUCENE-8249?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16435691#comment-16435691
 ] 

Alan Woodward commented on LUCENE-8249:
---------------------------------------

Here's the benchmark.  Looks like noise to me...

{code}
 TaskQPS baseline      StdDevQPS my_modified_version      StdDev                
Pct diff
                Wildcard     1714.40      (4.9%)     1622.18      (6.1%)   
-5.4% ( -15% -    5%)
                  Fuzzy2       76.05     (23.2%)       72.75     (20.2%)   
-4.3% ( -38% -   50%)
              HighPhrase      889.67      (6.3%)      862.68     (10.8%)   
-3.0% ( -18% -   15%)
   HighTermDayOfYearSort     1068.15      (6.0%)     1037.75      (8.1%)   
-2.8% ( -15% -   11%)
               OrHighLow     2594.37      (7.1%)     2524.49      (4.7%)   
-2.7% ( -13% -    9%)
         MedSloppyPhrase      902.87      (3.4%)      885.25      (4.1%)   
-2.0% (  -9% -    5%)
                HighTerm     4674.62      (6.2%)     4588.81      (7.0%)   
-1.8% ( -14% -   12%)
              AndHighMed     1068.36      (4.8%)     1050.54      (5.4%)   
-1.7% ( -11% -    8%)
                  IntNRQ     1422.34      (3.8%)     1404.61      (3.7%)   
-1.2% (  -8% -    6%)
              AndHighLow     2882.77      (8.0%)     2846.86      (8.7%)   
-1.2% ( -16% -   16%)
               OrHighMed     1306.57      (7.5%)     1290.96      (5.4%)   
-1.2% ( -13% -   12%)
               LowPhrase     1326.18      (4.7%)     1311.07      (4.4%)   
-1.1% (  -9% -    8%)
              OrHighHigh     1073.53      (4.8%)     1062.65      (3.7%)   
-1.0% (  -9% -    7%)
         LowSloppyPhrase     1018.38      (2.5%)     1009.19      (4.6%)   
-0.9% (  -7% -    6%)
                 Respell      366.92      (6.4%)      363.89      (9.0%)   
-0.8% ( -15% -   15%)
       HighTermMonthSort     3319.32      (6.1%)     3307.84      (6.2%)   
-0.3% ( -11% -   12%)
             AndHighHigh     1023.89      (5.3%)     1020.61      (5.4%)   
-0.3% ( -10% -   10%)
             LowSpanNear     1309.70      (5.7%)     1305.85      (8.1%)   
-0.3% ( -13% -   14%)
        HighSloppyPhrase      664.01      (3.7%)      663.67      (5.2%)   
-0.1% (  -8% -    9%)
                PKLookup      374.03      (5.5%)      373.96      (6.3%)   
-0.0% ( -11% -   12%)
                  Fuzzy1      386.18      (4.1%)      388.21      (5.2%)    
0.5% (  -8% -   10%)
             MedSpanNear     1217.07      (4.4%)     1226.70      (5.7%)    
0.8% (  -8% -   11%)
            HighSpanNear      641.59     (18.0%)      648.32     (22.7%)    
1.0% ( -33% -   50%)
               MedPhrase      999.35      (3.1%)     1011.72      (5.7%)    
1.2% (  -7% -   10%)
                 LowTerm     5747.36      (7.2%)     5835.87      (6.4%)    
1.5% ( -11% -   16%)
                 MedTerm     5195.74      (5.2%)     5290.94      (6.4%)    
1.8% (  -9% -   14%)
                 Prefix3      587.71      (8.6%)      601.34      (9.5%)    
2.3% ( -14% -   22%)
{code}

> Add matches to exact PhraseQuery and MultiPhraseQuery
> -----------------------------------------------------
>
>                 Key: LUCENE-8249
>                 URL: https://issues.apache.org/jira/browse/LUCENE-8249
>             Project: Lucene - Core
>          Issue Type: Improvement
>            Reporter: Alan Woodward
>            Assignee: Alan Woodward
>            Priority: Major
>         Attachments: LUCENE-8249.patch
>
>
> ExactPhraseScorer can be rejigged fairly easily to expose a MatchesIterator



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

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

Reply via email to