Skip to content

Commit 2f092c2

Browse files
committed
Add comments
1 parent 6700e75 commit 2f092c2

File tree

1 file changed

+4
-1
lines changed

1 file changed

+4
-1
lines changed

cpp/src/arrow/csv/fuzz.cc

Lines changed: 4 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -41,11 +41,14 @@ Status FuzzCsvReader(const uint8_t* data, int64_t size) {
4141
auto io_context = arrow::io::default_io_context();
4242

4343
auto read_options = ReadOptions::Defaults();
44-
// Make chunking more likely
44+
// Make chunking more likely to exercise chunked reading and optional parallelization.
45+
// Most files in the seed corpus are currently in the 4-10 kB range.
4546
read_options.block_size = 1000;
4647
auto parse_options = ParseOptions::Defaults();
4748
auto convert_options = ConvertOptions::Defaults();
4849
convert_options.auto_dict_encode = true;
50+
// This is the default value, but we might want to turn this knob to have a better
51+
// mix of dict-encoded and non-dict-encoded columns when reading.
4952
convert_options.auto_dict_max_cardinality = 50;
5053

5154
auto input_stream =

0 commit comments

Comments
 (0)