Checkout Tools
  • last updated 8 hours ago
Constraints: committers
Constraints: files
Constraints: dates
LUCENE-5973: fix index generation
fix CHANGES entry
SOLR-6130: Added com.uwyn:jhighlight dependency to, and removed asm:asm dependency from the extraction contrib - dependencies weren't fully upgraded with the Tika 1.4->1.5 upgrade (SOLR-5763) (merged trunk r1599663)
LUCENE-5686: fail if LUCENE/SOLR_NNNN (_ instead of -) appears in changes
Merged revision(s) 1595257 from lucene/dev/branches/branch_4x:

Merged revision(s) 1595255 from lucene/dev/trunk:

Disable our lock factory tests, if clover is enabled (the exact reason is unclear, but it looks like Clover gets a deadlock if instrumented code is run inside the ANT JVM)

LUCENE-5671: upgrade ICU to fix concurrency bug
LUCENE-5619: add back-compat index generation code for doc-values updates
use FSDirectory to avoid OOME
LUCENE-5668: fix the codec so segment sizes don't change on each flush
LUCENE-5668: er, actually fix the bug
LUCENE-5668: fix ob1 in TieredMergePolicy
add assert to maybe trick G1GC into not doing this (or at least make the fail explicit)
fix rare test fail
LUCENE-5644: favor an already initialized ThreadState
backport test fix (r1593425)
LUCENE-5662: add missing null checks to Field to avoid aborting exceptions
LUCENE-5660: add changes
throw IAE in OfflineSorter if incoming blob is too big
LUCENE-5661: add workaround for race conditions in the LiveIWC api
sync the buggy methods in LiveIWC, instead of all operations from all RIWs...
print number of active threads states in IW.infoStream, even when flushing due to RAM usage
fix test bug
don't use RIW in this test, its sensitive to random craziness
fix test to use IW not RIW; add verbosity when LTC changes live IWC settings
don't use RIW in this test, it depends on stable merging (and RIW does insane things)
randomize mergepolicy settings on the fly
invoke cms parameters randomly
live setters will be tested or die
add test