database/postgres
1
0
Fork 0
mirror of https://github.com/postgres/postgres.git synced 2025-11-03 09:13:20 +03:00
Code Activity

Fix extraction of week and quarter fields from intervals.

Browse Source 
"EXTRACT(WEEK FROM interval_value)" formerly threw an error.
Define it as "tm->tm_mday / 7".  (With C99 division semantics,
this gives consistent results for negative intervals.)

"EXTRACT(QUARTER FROM interval_value)" has been implemented
all along, but it formerly gave extremely strange results for
negative intervals.  Fix it so that the output for -N months
is the negative of the output for N months.

Per bug #18348 from Michael Bondarenko and subsequent discussion.

Discussion: https://postgr.es/m/18348-b097a3587dfde8a4@postgresql.org
This commit is contained in:
Tom Lane
2024-08-16 12:35:50 -04:00
parent 108d2adb9e
commit 6be39d77a7
4 changed files with 94 additions and 16 deletions
Download Patch File Download Diff File Expand all files Collapse all files

18
src/backend/utils/adt/timestamp.c
View File

@@ -5899,6 +5899,7 @@ NonFiniteIntervalPart(int type, int unit, char *lowunits, bool isNegative)
case DTK_MILLISEC:
case DTK_SECOND:
case DTK_MINUTE:
case DTK_WEEK:
case DTK_MONTH:
case DTK_QUARTER:
return 0.0;
@@ -6018,12 +6019,27 @@ interval_part_common(PG_FUNCTION_ARGS, bool retnumeric)
intresult = tm->tm_mday;
break;
case DTK_WEEK:
intresult = tm->tm_mday / 7;
break;
case DTK_MONTH:
intresult = tm->tm_mon;
break;
case DTK_QUARTER:
intresult = (tm->tm_mon / 3) + 1;
/*
* We want to maintain the rule that a field extracted from a
* negative interval is the negative of the field's value for
* the sign-reversed interval. The broken-down tm_year and
* tm_mon aren't very helpful for that, so work from
* interval->month.
*/
if (interval->month >= 0)
intresult = (tm->tm_mon / 3) + 1;
else
intresult = -(((-interval->month % MONTHS_PER_YEAR) / 3) + 1);
break;
case DTK_YEAR: