$ ./pq help
Usage: pq [-v] [COMMAND]
parquet query tool
  -v, --verbose   enable debug logs
Commands:
  count     print total number of rows in parquet file
  schema    print schema of parquet file
  read      print content of parquet file in json format
  metadata  print metadata of parquet file
  write     create a parquet file from a jsonl stream and a schema
  help      Display help information about the specified command.
Copyright(c) 2023 by @tonivade

$ ./pq help count
Usage: pq count [-v] [--filter=PREDICATE] FILE
print total number of rows in parquet file
      FILE                 parquet file
      --filter=PREDICATE   predicate to apply to the rows
  -v, --verbose            enable debug logs

$ ./pq count example.parquet
1000

$ ./pq count --filter 'gender == "Male"' example.parquet
451

$ ./pq help schema
Usage: pq schema [-v] [--select=COLUMN[,COLUMN...]]... FILE
print schema of parquet file
      FILE        parquet file
      --select=COLUMN[,COLUMN...]
                  list of columns to select
  -v, --verbose   enable debug logs

$ ./pq schema example.parquet
message spark_schema {
  optional int32 id;
  optional binary first_name (STRING);
  optional binary last_name (STRING);
  optional binary email (STRING);
  optional binary gender (STRING);
  optional binary ip_address (STRING);
  optional binary cc (STRING);
  optional binary country (STRING);
  optional binary birthdate (STRING);
  optional double salary;
  optional binary title (STRING);
  optional binary comments (STRING);
}

$ ./pq schema --select id,first_name,email example.parquet
message spark_schema {
  optional int32 id;
  optional binary first_name (STRING);
  optional binary email (STRING);
}

$ ./pq help read
Usage: pq read [-v] [--index] [--filter=PREDICATE] [--format=JSON|CSV]
               [--get=ROW] [--head=ROWS] [--skip=ROWS] [--tail=ROWS]
               [--select=COLUMN[,COLUMN...]]... FILE
print content of parquet file in json format
      FILE                 parquet file
      --filter=PREDICATE   predicate to apply to the rows
      --format=JSON|CSV    output format, json or csv
      --get=ROW            print just the row with given index
      --head=ROWS          get the first N number of rows
      --index              print row index
      --select=COLUMN[,COLUMN...]
                           list of columns to select
      --skip=ROWS          skip a number N of rows
      --tail=ROWS          get the last N number of rows
  -v, --verbose            enable debug logs

$ ./pq read example.parquet
{"id":1,"first_name":"Amanda","last_name":"Jordan","email":"[email protected]","gender":"Female","ip_address":null,"cc":"6759521864920116","country":"Indonesia","birthdate":"3/8/1971","salary":49756.53,"title":"Internal Auditor","comments":"1E+02"}
...

$ ./pq read --select id,first_name,email example.parquet
{"id":1,"first_name":"Amanda","email":"[email protected]"}
...

$ ./pq read --filter 'gender == "Male"' example.parquet
{"id":2,"first_name":"Albert","last_name":"Freeman","email":"[email protected]","gender":"Male","ip_address":"218.111.175.34","cc":"","country":"Canada","birthdate":"1/16/1968","salary":150280.17,"title":"Accountant IV","comments":""}
...

$ ./pq help metadata
Usage: pq metadata [-v] [--show-blocks] FILE
print metadata of parquet file
      FILE            parquet file
      --show-blocks   show block metadata info
  -v, --verbose       enable debug logs

$ ./pq metadata example.parquet
"createdBy":parquet-mr version 1.8.3 (build aef7230e114214b7cc962a8f3fc5aeed6ce80828)
"count":1000

$ ./pq help write
Usage: pq write [-v] [--format=JSON|CSV] [--schema=FILE] FILE
create a parquet file from a jsonl stream and a schema
      FILE                destination parquet file
      --format=JSON|CSV   input format, json or csv
      --schema=FILE       file with schema definition
  -v, --verbose           enable debug logs