parquet-tools

Show schema

$ parquet-tools schema [file.parquet]

$ parquet-tools cat [file.parquet]

$ parquet-tools meta [file.parquet]

$ parquet-tools head -n [10] [file.parquet]

$ parquet-tools rowcount [file.parquet]

$ parquet-tools cat --json [file.parquet]

$ parquet-tools column-index [file.parquet]

parquet-tools command [options] file

parquet-tools inspects Apache Parquet files. Parquet is a columnar storage format used in big data systems.
Schema inspection shows column names, types, and nesting. This helps understand data structure without reading contents.
Cat and head commands display actual data. JSON output integrates with other tools.
Metadata shows compression, encoding, and statistics. Row groups and column chunks reveal physical layout.
Parquet files from Spark, Hive, and other systems can be examined. Useful for debugging data pipelines.

cat

Print file contents.

Print first rows.

Show schema.

Show file metadata.

Count rows.

Show column index.

Number of rows.

JSON output format.

Specific columns.

Large files may be slow to fully read. Some complex types display differently. Requires Java runtime.

Parquet format was developed by Twitter and Cloudera around 2013. parquet-tools provides command-line inspection for the widely-adopted columnar format.

parquet(1), avro-tools(1), orc-tools(1)