fdb-record-layer
fdb-record-layer copied to clipboard
A record-oriented store built on FoundationDB
In #1200, we updated the SpotBugs version, which but at the same time, turned spotbugsTest off to avoid some new SpotBugs failures in test code from stopping the upgrade. While...
For some use cases, it would be useful to have an `ExecuteState` option that would allow cursors to stop when they hit a transaction timeout (at least) or perhaps any...
Since I new this would involve a lot of extracting of interfaces, and moving code I thought it might be easier to go commit-by-commit, and have some pretty simple commits...
The index build tests already largely randomly generated their data. This updates their source of seeds so that they take a random seed as a parameter, and it configures them...
There are cases where during index planning, we can match against the primary key components of a synthetic record type in a way that is incorrect. For example, suppose you...
The `RecordTypeKeyComparison` operator fails if it operates on a synthetic record type because when it gets the comparand, it calls `getRecordType`, which ignores synthetic types. This means that if there...
The `RecordTypeKeyComparison` operator fails if it operates on a synthetic record type because when it gets the comparand, it calls `getRecordType`, which ignores synthetic types. This means that if there...
Most of our index building tests already use a random number generator to create data. It should be fairly straightforward to use the new randomized test framework to run those...
Related to: #2574 If there is a failure during merge, the file lock won't be cleaned up. Lucene has logic to take care of this, but care needs to be...
When run with an agile context, PrimaryKeySegmentIndex may delete entries for segments that don't end up getting merged, forcing future updates to use a query to do updates, defeating the...