hashsum: treat hash values in sum file as case insensitive

Also warn duplicate file paths in sum files.

Fixes https://forum.rclone.org/t/rclone-check-sum/25566/45

cmd/checksum/checksum.go

@@ -33,6 +33,8 @@ don't match.  It doesn't alter the file system.
 If you supply the |--download| flag, it will download the data from remote
 and calculate the contents hash on the fly.  This can be useful for remotes
 that don't support hashes or if you really want to check all the data.
+
+Note that hash values in the SUM file are treated as case insensitive.
 `, "|", "`") + check.FlagsHelp,
 	RunE: func(command *cobra.Command, args []string) error {
 		cmd.CheckArgs(3, 3, command, args)

cmd/hashsum/hashsum.go

@@ -76,7 +76,7 @@ Then
 
     $ rclone hashsum MD5 remote:path
 
-Note that hash names are case insensitive.
+Note that hash names are case insensitive and values are output in lower case.
 `,
 	RunE: func(command *cobra.Command, args []string) error {
 		cmd.CheckArgs(0, 2, command, args)

docs/content/commands/rclone_checksum.md

@@ -20,6 +20,8 @@ If you supply the `--download` flag, it will download the data from remote
 and calculate the contents hash on the fly.  This can be useful for remotes
 that don't support hashes or if you really want to check all the data.
 
+Note that hash values in the SUM file are treated as case insensitive.
+
 If you supply the `--one-way` flag, it will only check that files in
 the source match the files in the destination, not the other way
 around. This means that extra files in the destination that are not in

docs/content/commands/rclone_hashsum.md

@@ -38,7 +38,7 @@ Then
 
     $ rclone hashsum MD5 remote:path
 
-Note that hash names are case insensitive.
+Note that hash names are case insensitive and values are output in lower case.
 
 
 ```

fs/operations/check.go

@@ -7,6 +7,7 @@ import (
 	"io"
 	"os"
 	"regexp"
+	"strings"
 	"sync"
 	"sync/atomic"
 
@@ -540,7 +541,7 @@ func ParseSumFile(ctx context.Context, sumFile fs.Object) (HashSums, error) {
 	}
 	parser := bufio.NewReader(rd)
 
-	const maxWarn = 4
+	const maxWarn = 3
 	numWarn := 0
 
 	re := regexp.MustCompile(`^([^ ]+) [ *](.+)$`)
@@ -558,19 +559,31 @@ func ParseSumFile(ctx context.Context, sumFile fs.Object) (HashSums, error) {
 			continue
 		}
 
-		if fields := re.FindStringSubmatch(line); fields != nil {
-			hashes[fields[2]] = fields[1]
+		fields := re.FindStringSubmatch(line)
+		if fields == nil {
+			numWarn++
+			if numWarn <= maxWarn {
+				fs.Logf(sumFile, "improperly formatted checksum line %d", lineNo)
+			}
 			continue
 		}
 
+		sum, file := fields[1], fields[2]
+		if hashes[file] != "" {
 			numWarn++
-		if numWarn < maxWarn {
-			fs.Logf(sumFile, "improperly formatted checksum line %d", lineNo)
-		} else if numWarn == maxWarn {
-			fs.Logf(sumFile, "more warnings suppressed...")
+			if numWarn <= maxWarn {
+				fs.Logf(sumFile, "duplicate file on checksum line %d", lineNo)
 			}
+			continue
 		}
 
+		// We've standardised on lower case checksums in rclone internals.
+		hashes[file] = strings.ToLower(sum)
+	}
+
+	if numWarn > maxWarn {
+		fs.Logf(sumFile, "%d warning(s) suppressed...", numWarn-maxWarn)
+	}
 	if err = rd.Close(); err != nil {
 		return nil, err
 	}

fs/operations/check_test.go

@@ -332,8 +332,10 @@ func testCheckSum(t *testing.T, download bool) {
 	const (
 		testString1      = "Hello, World!"
 		testDigest1      = "65a8e27d8879283831b664bd8b7f0ad4"
+		testDigest1Upper = "65A8E27D8879283831B664BD8B7F0AD4"
 		testString2      = "I am the walrus"
 		testDigest2      = "87396e030ef3f5b35bbf85c0a09a4fb3"
+		testDigest2Mixed = "87396e030EF3f5b35BBf85c0a09a4FB3"
 	)
 
 	type wantType map[string]string
@@ -428,7 +430,7 @@ func testCheckSum(t *testing.T, download bool) {
 	}
 
 	check := func(runNo, wantChecks, wantErrors int, wantResults wantType) {
-		runName := fmt.Sprintf("move%d", runNo)
+		runName := fmt.Sprintf("subtest%d", runNo)
 		t.Run(runName, func(t *testing.T) {
 			checkRun(runNo, wantChecks, wantErrors, wantResults)
 		})
@@ -519,6 +521,23 @@ func testCheckSum(t *testing.T, download bool) {
 		"differ":       "potato\n",
 		"error":        "",
 	})
+
+	// test mixed-case checksums
+	file1 = makeFile("banana", testString1)
+	file2 = makeFile("potato", testString2)
+	fcsums = makeSums(operations.HashSums{
+		"banana": testDigest1Upper,
+		"potato": testDigest2Mixed,
+	})
+	fstest.CheckItems(t, r.Fremote, fcsums, file1, file2)
+	check(7, 2, 0, wantType{
+		"combined":     "= banana\n= potato\n",
+		"missingonsrc": "",
+		"missingondst": "",
+		"match":        "banana\npotato\n",
+		"differ":       "",
+		"error":        "",
+	})
 }
 
 func TestCheckSum(t *testing.T) {