Add more checks in migration code (#21011) (#21050)

Backport #21011

When migrating add several more important sanity checks:

* SHAs must be SHAs
* Refs must be valid Refs
* URLs must be reasonable

Signed-off-by: Andrew Thornton <art27@cantab.net>

modules/git/ref.go

@@ -4,7 +4,10 @@
 
 package git
 
-import "strings"
+import (
+	"regexp"
+	"strings"
+)
 
 const (
 	// RemotePrefix is the base directory of the remotes information of git.
@@ -15,6 +18,29 @@ const (
 	pullLen = len(PullPrefix)
 )
 
+// refNamePatternInvalid is regular expression with unallowed characters in git reference name
+// They cannot have ASCII control characters (i.e. bytes whose values are lower than \040, or \177 DEL), space, tilde ~, caret ^, or colon : anywhere.
+// They cannot have question-mark ?, asterisk *, or open bracket [ anywhere
+var refNamePatternInvalid = regexp.MustCompile(
+	`[\000-\037\177 \\~^:?*[]|` + // No absolutely invalid characters
+		`(?:^[/.])|` + // Not HasPrefix("/") or "."
+		`(?:/\.)|` + // no "/."
+		`(?:\.lock$)|(?:\.lock/)|` + // No ".lock/"" or ".lock" at the end
+		`(?:\.\.)|` + // no ".." anywhere
+		`(?://)|` + // no "//" anywhere
+		`(?:@{)|` + // no "@{"
+		`(?:[/.]$)|` + // no terminal '/' or '.'
+		`(?:^@$)`) // Not "@"
+
+// IsValidRefPattern ensures that the provided string could be a valid reference
+func IsValidRefPattern(name string) bool {
+	return !refNamePatternInvalid.MatchString(name)
+}
+
+func SanitizeRefPattern(name string) string {
+	return refNamePatternInvalid.ReplaceAllString(name, "_")
+}
+
 // Reference represents a Git ref.
 type Reference struct {
 	Name   string