Alter scale selection for NUMERIC division and transcendental functions
Tom Lane <tgl@sss.pgh.pa.us>
Alter scale selection for NUMERIC division and transcendental functions so that precision of result is always at least as good as you'd get from float8 arithmetic (ie, always at least 16 digits of accuracy). Per pg_hackers discussion a few days ago.
Files
| Path | Change | +/− |
|---|---|---|
| src/backend/utils/adt/numeric.c | modified | +318 −109 |
| src/include/utils/numeric.h | modified | +21 −15 |
| src/test/regress/expected/aggregates.out | modified | +9 −9 |