add cat --csv --no-header option

README.md

@@ -5,7 +5,6 @@
 * Built using the Rust implementation of [Parquet](https://github.com/apache/arrow-rs/tree/master/parquet) and [Arrow](https://github.com/apache/arrow-rs/tree/master/arrow)
 * `pqrs` roughly means "parquet-tools in rust"
 
-
 ## Installation
 
 ### Recommended Method
@@ -21,6 +20,7 @@ For macOS users, `pqrs` is available as a homebrew tap.
 ```shell
 brew install manojkarthick/tap/pqrs
 ```
+
 NOTE: For users upgrading from v0.2 or prior, note that the location of the `pqrs` homebrew tap has been updated.
 To update to v0.2.1+, please uninstall using `brew uninstall pqrs` and use the above command to re-install.
 
@@ -73,8 +73,8 @@ SUBCOMMANDS:
 
 ### Subcommand: cat
 
-Prints the contents of the given files and folders. Recursively traverses and prints all the files if the input is a directory. 
-Supports json-like, json or CSV format. Use `--json` for JSON output and `--csv` for CSV output.
+Prints the contents of the given files and folders. Recursively traverses and prints all the files if the input is a directory.
+Supports json-like, json or CSV format. Use `--json` for JSON output, `--csv` for CSV output with column names in the first row, and `--csv-data-only` for CSV output without the column names row.
 
 ```shell
 ❯ pqrs cat data/cities.parquet
@@ -97,6 +97,12 @@ foo,bar
 10,20
 ```
 
+```shell
+❯ pqrs cat data/simple.parquet --csv --no-header
+1,2
+10,20
+```
+
 NOTE: CSV format is not supported for files that contain Struct or Byte fields.
 
 ### Subcommand: head
@@ -207,7 +213,7 @@ statistics: {min: [69, 117, 114, 111, 112, 101], max: [78, 111, 114, 116, 104, 3
 ```
 
 ```shell
-❯ pqrs schema --json data/cities.parquet                                                              
+❯ pqrs schema --json data/cities.parquet
 {"version":1,"num_rows":3,"created_by":"parquet-mr version 1.5.0-cdh5.7.0 (build ${buildNumber})","metadata":null,"columns":[{"optional":"true","physical_type":"BYTE_ARRAY","name":"continent","path":"continent","converted_type":"UTF8"},{"name":"name","converted_type":"UTF8","path":"country.name","physical_type":"BYTE_ARRAY","optional":"true"},{"optional":"true","name":"array_element","physical_type":"BYTE_ARRAY","path":"country.city.bag.array_element","converted_type":"UTF8"}],"message":"message hive_schema {\n  OPTIONAL BYTE_ARRAY continent (UTF8);\n  OPTIONAL group country {\n    OPTIONAL BYTE_ARRAY name (UTF8);\n    OPTIONAL group city (LIST) {\n      REPEATED group bag {\n        OPTIONAL BYTE_ARRAY array_element (UTF8);\n      }\n    }\n  }\n}\n"}
 
 ```
@@ -232,8 +238,6 @@ File Name: data/pems-1.snappy.parquet
 Compressed Size: 12 KiB
 ```
 
-
-
 ### TODO
 
 * [ ] Test on Windows

src/commands/cat.rs

@@ -16,6 +16,10 @@ pub struct CatCommandArgs {
     #[clap(short, long, conflicts_with = "json")]
     csv: bool,
 
+    /// Use CSV format without a header for printing
+    #[clap(long = "no-header", requires = "csv", conflicts_with = "json")]
+    csv_no_header: bool,
+
     /// Use JSON lines format for printing
     #[clap(short, long, conflicts_with = "csv")]
     json: bool,
@@ -27,6 +31,8 @@ pub struct CatCommandArgs {
 pub(crate) fn execute(opts: CatCommandArgs) -> Result<(), PQRSError> {
     let format = if opts.json {
         Formats::Json
+    } else if opts.csv_no_header {
+        Formats::CsvNoHeader
     } else if opts.csv {
         Formats::Csv
     } else {

src/utils.rs

@@ -26,6 +26,7 @@ static ONE_PI_B: i64 = ONE_TI_B * 1024;
 pub enum Formats {
     Default,
     Csv,
+    CsvNoHeader,
     Json,
 }
 
@@ -119,6 +120,31 @@ pub fn print_rows(
             let batch_reader = arrow_reader.get_record_reader(8192)?;
             let mut writer = arrow::csv::Writer::new(std::io::stdout());
 
+            for maybe_batch in batch_reader {
+                if left == Some(0) {
+                    break;
+                }
+
+                let mut batch = maybe_batch?;
+                if let Some(l) = left {
+                    if batch.num_rows() <= l {
+                        left = Some(l - batch.num_rows());
+                    } else {
+                        let n = min(batch.num_rows(), l);
+                        batch = batch.slice(0, n);
+                        left = Some(0);
+                    }
+                };
+
+                writer.write(&batch)?;
+            }
+        }
+        Formats::CsvNoHeader => {
+            let mut arrow_reader = ParquetFileArrowReader::new(parquet_reader);
+            let batch_reader = arrow_reader.get_record_reader(8192)?;
+            let writer_builder = arrow::csv::WriterBuilder::new().has_headers(false);
+            let mut writer = writer_builder.build(std::io::stdout());
+
             for maybe_batch in batch_reader {
                 if left == Some(0) {
                     break;
@@ -263,6 +289,7 @@ fn print_row(row: &Row, format: Formats) {
     match format {
         Formats::Default => println!("{}", row),
         Formats::Csv => println!("Unsupported! {}", row),
+        Formats::CsvNoHeader => println!("Unsupported! {}", row),
         Formats::Json => println!("{}", row.to_json_value()),
     }
 }

tests/integration.rs

@@ -14,6 +14,8 @@ static CAT_JSON_OUTPUT: &str = r#"{"continent":"Europe","country":{"name":"Franc
 static CAT_CSV_OUTPUT: &str = r#"foo,bar
 1,2
 10,20"#;
+static CAT_CSV_NO_HEADER_OUTPUT: &str = r#"1,2
+10,20"#;
 static SCHEMA_OUTPUT: &str = r#"message hive_schema {
   OPTIONAL BYTE_ARRAY continent (UTF8);
   OPTIONAL group country {
@@ -32,8 +34,8 @@ static SAMPLE_PARTIAL_OUTPUT_2: &str = "country: {name:";
 mod integration {
     // make sure any new commands added have a corresponding integration test here!
     use crate::{
-        CAT_CSV_OUTPUT, CAT_JSON_OUTPUT, CAT_OUTPUT, CITIES_PARQUET_PATH,
-        MERGED_FILE_NAME, PEMS_1_PARQUET_PATH, PEMS_2_PARQUET_PATH,
+        CAT_CSV_NO_HEADER_OUTPUT, CAT_CSV_OUTPUT, CAT_JSON_OUTPUT, CAT_OUTPUT,
+        CITIES_PARQUET_PATH, MERGED_FILE_NAME, PEMS_1_PARQUET_PATH, PEMS_2_PARQUET_PATH,
         SAMPLE_PARTIAL_OUTPUT_1, SAMPLE_PARTIAL_OUTPUT_2, SCHEMA_OUTPUT,
         SIMPLE_PARQUET_PATH,
     };
@@ -74,6 +76,20 @@ mod integration {
         Ok(())
     }
 
+    #[test]
+    fn validate_cat_csv_no_header() -> Result<(), Box<dyn std::error::Error>> {
+        let mut cmd = Command::cargo_bin("pqrs")?;
+        cmd.arg("cat")
+            .arg(SIMPLE_PARQUET_PATH)
+            .arg("--csv")
+            .arg("--no-header");
+        cmd.assert()
+            .success()
+            .stdout(predicate::str::starts_with(CAT_CSV_NO_HEADER_OUTPUT));
+
+        Ok(())
+    }
+
     #[test]
     fn validate_cat_directory() -> Result<(), Box<dyn std::error::Error>> {
         let mut cmd = Command::cargo_bin("pqrs")?;