Skip to content
Closed
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
7 changes: 7 additions & 0 deletions cpp/src/arrow/dataset/partition.cc
Original file line number Diff line number Diff line change
Expand Up @@ -322,6 +322,13 @@ Result<std::vector<std::string>> KeyValuePartitioning::FormatPartitionSegments(
for (int i = 0; i < schema_->num_fields(); ++i) {
if (values[i] != nullptr && values[i]->is_valid) {
segments[i] = values[i]->ToString();
// The value is used verbatim as a path segment, so it must not be able
// to escape or add a directory level
if (segments[i] == "." || segments[i] == ".." ||
segments[i].find(fs::internal::kSep) != std::string::npos) {
return Status::Invalid("Partition value for ", schema_->field(i)->name(),
" cannot be used as a path segment: ", segments[i]);
}
continue;
}

Expand Down
16 changes: 16 additions & 0 deletions cpp/src/arrow/dataset/partition_test.cc
Original file line number Diff line number Diff line change
Expand Up @@ -310,6 +310,22 @@ TEST_F(TestPartitioning, DirectoryPartitioningFormat) {
equal(field_ref("beta"), literal("hello"))));
}

TEST_F(TestPartitioning, DirectoryPartitioningFormatInvalidSegment) {
partitioning_ =
std::make_shared<DirectoryPartitioning>(schema({field("alpha", utf8())}));
written_schema_ = partitioning_->schema();

AssertFormat(equal(field_ref("alpha"), literal("..a")), "..a");
AssertFormatError(equal(field_ref("alpha"), literal("..")));
AssertFormatError(equal(field_ref("alpha"), literal(".")));
AssertFormatError(equal(field_ref("alpha"), literal("../../escaped")));
AssertFormatError(equal(field_ref("alpha"), literal("a/b")));

partitioning_ =
std::make_shared<FilenamePartitioning>(schema({field("alpha", utf8())}));
AssertFormatError(equal(field_ref("alpha"), literal("../../escaped")));
}

TEST_F(TestPartitioning, DirectoryPartitioningFormatDictionary) {
auto dictionary = ArrayFromJSON(utf8(), R"(["hello", "world"])");
partitioning_ = std::make_shared<DirectoryPartitioning>(schema({DictStr("alpha")}),
Expand Down
Loading