javanna commented on PR #13773: URL: https://github.com/apache/lucene/pull/13773#issuecomment-2346282034
Happy that you commented, thanks! Can you help me better understand why these are incorrect results? ``` Error Message: java.lang.AssertionError: Hit 0 docnumbers don't match Hits length1=25 length2=25 hit=0: doc0=1.0 shardIndex=-1, doc155=1.0 shardIndex=-1 hit=1: doc1=1.0 shardIndex=-1, doc156=1.0 shardIndex=-1 hit=2: doc2=1.0 shardIndex=-1, doc157=1.0 shardIndex=-1 hit=3: doc3=1.0 shardIndex=-1, doc158=1.0 shardIndex=-1 hit=4: doc4=1.0 shardIndex=-1, doc159=1.0 shardIndex=-1 hit=5: doc5=1.0 shardIndex=-1, doc160=1.0 shardIndex=-1 hit=6: doc6=1.0 shardIndex=-1, doc161=1.0 shardIndex=-1 hit=7: doc7=1.0 shardIndex=-1, doc162=1.0 shardIndex=-1 hit=8: doc8=1.0 shardIndex=-1, doc163=1.0 shardIndex=-1 hit=9: doc9=1.0 shardIndex=-1, doc164=1.0 shardIndex=-1 hit=10: doc10=1.0 shardIndex=-1, doc165=1.0 shardIndex=-1 hit=11: doc11=1.0 shardIndex=-1, doc166=1.0 shardIndex=-1 hit=12: doc12=1.0 shardIndex=-1, doc167=1.0 shardIndex=-1 hit=13: doc13=1.0 shardIndex=-1, doc168=1.0 shardIndex=-1 hit=14: doc14=1.0 shardIndex=-1, doc169=1.0 shardIndex=-1 hit=15: doc15=1.0 shardIndex=-1, doc170=1.0 shardIndex=-1 hit=16: doc16=1.0 shardIndex=-1, doc171=1.0 shardIndex=-1 hit=17: doc17=1.0 shardIndex=-1, doc172=1.0 shardIndex=-1 hit=18: doc18=1.0 shardIndex=-1, doc173=1.0 shardIndex=-1 hit=19: doc19=1.0 shardIndex=-1, doc174=1.0 shardIndex=-1 hit=20: doc20=1.0 shardIndex=-1, doc175=1.0 shardIndex=-1 hit=21: doc21=1.0 shardIndex=-1, doc176=1.0 shardIndex=-1 hit=22: doc22=1.0 shardIndex=-1, doc177=1.0 shardIndex=-1 hit=23: doc23=1.0 shardIndex=-1, doc178=1.0 shardIndex=-1 hit=24: doc24=1.0 shardIndex=-1, doc179=1.0 shardIndex=-1 for query:field:* ``` From the failure, I see all docs have the same score, but doc ids don't match. I assume that the column on the right targeted a leaf partition, which is why doc ids don't start from 0. I think that we are still early terminating though or you mean that we shouldn't have as there were docs with same score but lower doc ids? I do think that I am missing something though, or this test could have failed with inter-segment concurrency too, because it would otherwise rely on the first segment to be the first one being searched at all times? -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For additional commands, e-mail: issues-h...@lucene.apache.org
