This is likely happening due to the way retractions are processed out of temp_search_results and into the log. We check for change in values, and synthesize a retraction datom, but we also actually have a retraction datom as well, so we end up with a duplicate retraction when there should have been just one.
Some fun transactor behaviour:
{:test/ident "One"}
- say, it gets assigned e=1This is likely happening due to the way retractions are processed out of temp_search_results and into the log. We check for change in values, and synthesize a retraction datom, but we also actually have a retraction datom as well, so we end up with a duplicate retraction when there should have been just one.