I found a small typo in a function comment: non-ASCII letter was
used. In my understanding, using non-ASCII letters in the program
files are not recommended. Patch attached.
Thank you for catching this! You are absolutely right. I apologize for
using non-ASCII characters in the code comments.
I've reviewed all the RPR-related files and found additional instances
of non-ASCII characters:
- Arrows "→" instead of "->"
- Superscript "²" instead of "^2"
- Greater-than-or-equal "≥" instead of ">="
Found in:
- src/backend/optimizer/plan/rpr.c (1 instance)
- src/backend/executor/nodeWindowAgg.c (5 instances)
- src/include/nodes/execnodes.h (4 instances)
- src/test/regress/sql/rpr.sql (7 instances)
- src/test/regress/sql/rpr_base.sql (15 instances)
- src/test/regress/expected/rpr.out (7 instances)
- src/test/regress/expected/rpr_base.out (15 instances)
Total: 54 instances across 7 files.
To ensure comprehensive coverage, I created a Python verification tool
that checks for non-ASCII characters and reports their Unicode code
points. I've verified all 46 files changed in the v42 patch series,
and confirmed no additional non-ASCII characters remain.
