hilaryRope commented on code in PR #38099:
URL: https://github.com/apache/beam/pull/38099#discussion_r3054451975
##########
sdks/go/pkg/beam/io/filesystem/gcs/gcs_test.go:
##########
@@ -271,6 +271,124 @@ func TestGCS_copy(t *testing.T) {
}
}
+func TestGlobToRegex(t *testing.T) {
+ tests := []struct {
+ pattern string
+ name string
+ want bool
+ }{
+ // Single * should NOT match / in object names
+ {"*.txt", "file.txt", true},
+ {"*.txt", "dir/file.txt", false},
+ {"prefix*", "prefix123", true},
+ {"prefix*", "prefix/subdir", false},
+
+ // ** should match any characters including /
+ {"**", "file.txt", true},
+ {"**", "dir/file.txt", true},
+ {"**", "dir/subdir/file.txt", true},
+ {"prefix/**", "prefix/file.txt", true},
+ {"prefix/**", "prefix/subdir/file.txt", true},
+ {"**/file.txt", "file.txt", true},
+ {"**/file.txt", "dir/file.txt", true},
+ {"**/file.txt", "dir/subdir/file.txt", true},
+
+ // Mixed patterns
+ {"dir/*.txt", "dir/file.txt", true},
+ {"dir/*.txt", "dir/subdir/file.txt", false},
+ {"dir/**/*.txt", "dir/file.txt", true},
+ {"dir/**/*.txt", "dir/subdir/file.txt", true},
+ {"dir/**/file.txt", "dir/file.txt", true},
+ {"dir/**/file.txt", "dir/a/b/c/file.txt", true},
+
+ // ? should match any single character including /
+ {"file?.txt", "file1.txt", true},
+ {"file?.txt", "file12.txt", false},
+
+ // Character classes
+ {"file[0-9].txt", "file1.txt", true},
+ {"file[0-9].txt", "filea.txt", false},
+ {"file[!0-9].txt", "filea.txt", true},
+ {"file[!0-9].txt", "file1.txt", false},
+
+ // Exact match (no wildcards)
+ {"exact.txt", "exact.txt", true},
+ {"exact.txt", "notexact.txt", false},
+
+ // Regex special characters should be escaped
+ {"file.txt", "file.txt", true},
+ {"file.txt", "fileXtxt", false},
+ {"file(1).txt", "file(1).txt", true},
+ }
+
+ for _, tt := range tests {
+ t.Run(tt.pattern+"_"+tt.name, func(t *testing.T) {
+ re, err := globToRegex(tt.pattern)
+ if err != nil {
+ t.Fatalf("globToRegex(%q) error = %v",
tt.pattern, err)
+ }
+ got := re.MatchString(tt.name)
+ if got != tt.want {
+ t.Errorf("globToRegex(%q).MatchString(%q) = %v,
want %v", tt.pattern, tt.name, got, tt.want)
+ }
+ })
+ }
+}
+
+func TestGCS_listWithSlashesInObjectNames(t *testing.T) {
+ ctx := context.Background()
+ bucket := "beamgogcsfilesystemtest"
+ dirPath := "gs://" + bucket
+
+ // Create server with objects that have / in their names
+ server := fakestorage.NewServer([]fakestorage.Object{
+ {ObjectAttrs: fakestorage.ObjectAttrs{BucketName: bucket, Name:
"file.txt"}, Content: []byte("")},
+ {ObjectAttrs: fakestorage.ObjectAttrs{BucketName: bucket, Name:
"dir/file.txt"}, Content: []byte("")},
+ {ObjectAttrs: fakestorage.ObjectAttrs{BucketName: bucket, Name:
"dir/subdir/file.txt"}, Content: []byte("")},
+ {ObjectAttrs: fakestorage.ObjectAttrs{BucketName: bucket, Name:
"other.txt"}, Content: []byte("")},
+ })
+ t.Cleanup(server.Stop)
+ c := &fs{client: server.Client()}
+
+ tests := []struct {
+ glob string
+ want []string
+ }{
+ // Single * should only match top-level files
+ {dirPath + "/*.txt", []string{dirPath + "/file.txt", dirPath +
"/other.txt"}},
Review Comment:
Added dirPath + "/dir/subdir/**" test case to verify deeply nested matching,
which is the core scenario from issue #38059
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
