Note

Expose per-shard "Internal cursor reads" in FT.PROFILE AGGREGATE and persist timeout state across cursor reads, with comprehensive RESP2/RESP3 tests.

• Profiling / Execution:
  • Add AREQ.cursor_reads to track number of internal cursor reads; increment in runCursor.
  • Print "Internal cursor reads" in profile output for cursor-based (internal) requests (src/profile.c).
  • Accumulate timeouts across chunks: set req->has_timedout |= has_timedout and use it in ProfilePrinterCtx.
• Aggregate Exec:
  • Use aggregated timeout flag in profile context for RESP2/RESP3 paths.
• Tests:
  • Add tests validating Internal cursor reads presence/count for AGGREGATE under RESP2/RESP3, including timeout scenarios.
  • Update RESP3 profile expectations to include "Internal cursor reads" in shard profiles.

^{Written by Cursor Bugbot for commit 08cf027. This will update automatically on new commits. Configure here.}