From e5a1a2a9742f58bcc9cc0bde0bad206f2fe7b9db Mon Sep 17 00:00:00 2001
From: Klaus Post <klauspost@gmail.com>
Date: Mon, 15 Mar 2021 22:15:52 +0100
Subject: [PATCH] s3 select: fix date_diff behavior (#11786)

Fixes #11785 and adds tests for samples given.
---
 pkg/s3select/select_test.go        | 41 ++++++++++++++++++++++++++++++
 pkg/s3select/sql/timestampfuncs.go | 34 +++++++------------------
 2 files changed, 50 insertions(+), 25 deletions(-)

diff --git a/pkg/s3select/select_test.go b/pkg/s3select/select_test.go
index 0f1a01b3f..cce521667 100644
--- a/pkg/s3select/select_test.go
+++ b/pkg/s3select/select_test.go
@@ -337,6 +337,47 @@ func TestJSONQueries(t *testing.T) {
 }`,
 			wantResult: `{"element_type":"__elem__merfu","element_id":"d868aefe-ef9a-4be2-b9b2-c9fd89cc43eb","attributes":{"__attr__image_dpi":300,"__attr__image_size":[2550,3299],"__attr__image_index":2,"__attr__image_format":"JPEG","__attr__file_extension":"jpg","__attr__data":null}}`,
 		},
+		{
+			name:       "date_diff_month",
+			query:      `SELECT date_diff(MONTH, '2019-10-20T', '2020-01-20T') FROM S3Object LIMIT 1`,
+			wantResult: `{"_1":3}`,
+		},
+		{
+			name:       "date_diff_month_neg",
+			query:      `SELECT date_diff(MONTH, '2020-01-20T', '2019-10-20T') FROM S3Object LIMIT 1`,
+			wantResult: `{"_1":-3}`,
+		},
+		// Examples from https://docs.aws.amazon.com/AmazonS3/latest/userguide/s3-glacier-select-sql-reference-date.html#s3-glacier-select-sql-reference-date-diff
+		{
+			name:       "date_diff_year",
+			query:      `SELECT date_diff(year, '2010-01-01T', '2011-01-01T') FROM S3Object LIMIT 1`,
+			wantResult: `{"_1":1}`,
+		},
+		{
+			name:       "date_diff_year",
+			query:      `SELECT date_diff(month, '2010-01-01T', '2010-05T') FROM S3Object LIMIT 1`,
+			wantResult: `{"_1":4}`,
+		},
+		{
+			name:       "date_diff_month_oney",
+			query:      `SELECT date_diff(month, '2010T', '2011T') FROM S3Object LIMIT 1`,
+			wantResult: `{"_1":12}`,
+		},
+		{
+			name:       "date_diff_month_neg",
+			query:      `SELECT date_diff(month, '2011T', '2010T') FROM S3Object LIMIT 1`,
+			wantResult: `{"_1":-12}`,
+		},
+		{
+			name:       "date_diff_days",
+			query:      `SELECT date_diff(day, '2010-01-01T23:00:00Z', '2010-01-02T01:00:00Z') FROM S3Object LIMIT 1`,
+			wantResult: `{"_1":0}`,
+		},
+		{
+			name:       "date_diff_days_one",
+			query:      `SELECT date_diff(day, '2010-01-01T23:00:00Z', '2010-01-02T23:00:00Z') FROM S3Object LIMIT 1`,
+			wantResult: `{"_1":1}`,
+		},
 	}
 
 	defRequest := `<?xml version="1.0" encoding="UTF-8"?>
diff --git a/pkg/s3select/sql/timestampfuncs.go b/pkg/s3select/sql/timestampfuncs.go
index d8c7f6351..df037a1ca 100644
--- a/pkg/s3select/sql/timestampfuncs.go
+++ b/pkg/s3select/sql/timestampfuncs.go
@@ -135,10 +135,6 @@ func dateAdd(timePart string, qty float64, t time.Time) (*Value, error) {
 	return FromTimestamp(t.Add(duration)), nil
 }
 
-const (
-	dayInNanoseconds = time.Hour * 24
-)
-
 // dateDiff computes the difference between two times in terms of the
 // `timePart` which can be years, months, days, hours, minutes or
 // seconds. For difference in years, months or days, the time part,
@@ -146,42 +142,30 @@ const (
 func dateDiff(timePart string, ts1, ts2 time.Time) (*Value, error) {
 	if ts2.Before(ts1) {
 		v, err := dateDiff(timePart, ts2, ts1)
-		v.negate()
+		if err == nil {
+			v.negate()
+		}
 		return v, err
 	}
 
 	duration := ts2.Sub(ts1)
 	y1, m1, d1 := ts1.Date()
 	y2, m2, d2 := ts2.Date()
-	dy, dm := int64(y2-y1), int64(m2-m1)
 
 	switch timePart {
 	case timePartYear:
+		dy := int64(y2 - y1)
 		if m2 > m1 || (m2 == m1 && d2 >= d1) {
 			return FromInt(dy), nil
 		}
 		return FromInt(dy - 1), nil
 	case timePartMonth:
-		months := 12 * dy
-		if m2 >= m1 {
-			months += dm
-		} else {
-			months += 12 + dm
-		}
-		if d2 < d1 {
-			months--
-		}
-		return FromInt(months), nil
+		m1 += time.Month(12 * y1)
+		m2 += time.Month(12 * y2)
+
+		return FromInt(int64(m2 - m1)), nil
 	case timePartDay:
-		// To compute the number of days between two times
-		// using the time package, zero out the time portions
-		// of the timestamps, compute the difference duration
-		// and then divide by the length of a day.
-		d1 := time.Date(y1, m1, d1, 0, 0, 0, 0, time.UTC)
-		d2 := time.Date(y2, m2, d2, 0, 0, 0, 0, time.UTC)
-		diff := d2.Sub(d1)
-		days := diff / dayInNanoseconds
-		return FromInt(int64(days)), nil
+		return FromInt(int64(duration / (24 * time.Hour))), nil
 	case timePartHour:
 		hours := duration / time.Hour
 		return FromInt(int64(hours)), nil