add parquet-fromcsv (#1) (#1798)

* add parquet-fromcsv (#1)

add command line tool for convert csv to parquet.

* add `text` for non-rust documentation text

* Update parquet/src/bin/parquet-fromcsv.rs

Co-authored-by: Raphael Taylor-Davies <[email protected]>

* Update parquet/src/bin/parquet-fromcsv.rs

Co-authored-by: Raphael Taylor-Davies <[email protected]>

* Update parquet/src/bin/parquet-fromcsv.rs

Co-authored-by: Raphael Taylor-Davies <[email protected]>

* Update parquet/src/bin/parquet-fromcsv.rs

Co-authored-by: Raphael Taylor-Davies <[email protected]>

* automate update help text

* remove anyhow

* add rat_exclude_files

* update test_command_help

* fix clippy warnings

* add writer-version, max-row-group-size arg

* fix cargo fmt lint

Co-authored-by: Raphael Taylor-Davies <[email protected]>

Author: kazuk

dev/release/rat_exclude_files.txt

@@ -20,3 +20,4 @@ conbench/.isort.cfg
 arrow-flight/src/arrow.flight.protocol.rs
 arrow-flight/src/sql/arrow.flight.protocol.sql.rs
 .github/*
+parquet/src/bin/parquet-fromcsv-help.txt

parquet/Cargo.toml

@@ -71,7 +71,7 @@ default = ["arrow", "snap", "brotli", "flate2", "lz4", "zstd", "base64"]
 # Enable arrow reader/writer APIs
 arrow = ["dep:arrow", "base64"]
 # Enable CLI tools
-cli = ["serde_json", "base64", "clap"]
+cli = ["serde_json", "base64", "clap","arrow/csv"]
 # Enable internal testing APIs
 test_common = []
 # Experimental, unstable functionality primarily used for testing
@@ -91,6 +91,10 @@ required-features = ["cli"]
 name = "parquet-rowcount"
 required-features = ["cli"]
 
+[[bin]]
+name = "parquet-fromcsv"
+required-features = ["cli"]
+
 [[bench]]
 name = "arrow_writer"
 required-features = ["arrow"]

parquet/src/bin/parquet-fromcsv-help.txt

@@ -0,0 +1,67 @@
+parquet 15.0.0
+Apache Arrow <[email protected]>
+Binary to convert csv to Parquet
+
+USAGE:
+    parquet [OPTIONS] --schema <SCHEMA> --input-file <INPUT_FILE> --output-file <OUTPUT_FILE>
+
+OPTIONS:
+    -b, --batch-size <BATCH_SIZE>
+            batch size
+            
+            [env: PARQUET_FROM_CSV_BATCHSIZE=]
+            [default: 1000]
+
+    -c, --parquet-compression <PARQUET_COMPRESSION>
+            compression mode
+            
+            [default: SNAPPY]
+
+    -d, --delimiter <DELIMITER>
+            field delimiter
+            
+            default value: when input_format==CSV: ',' when input_format==TSV: 'TAB'
+
+    -D, --double-quote <DOUBLE_QUOTE>
+            double quote
+
+    -e, --escape-char <ESCAPE_CHAR>
+            escape charactor
+
+    -f, --input-format <INPUT_FORMAT>
+            input file format
+            
+            [default: csv]
+            [possible values: csv, tsv]
+
+    -h, --has-header
+            has header
+
+        --help
+            Print help information
+
+    -i, --input-file <INPUT_FILE>
+            input CSV file
+
+    -m, --max-row-group-size <MAX_ROW_GROUP_SIZE>
+            max row group size
+
+    -o, --output-file <OUTPUT_FILE>
+            output Parquet file
+
+    -q, --quote-char <QUOTE_CHAR>
+            quate charactor
+
+    -r, --record-terminator <RECORD_TERMINATOR>
+            record terminator
+            
+            [possible values: lf, crlf, cr]
+
+    -s, --schema <SCHEMA>
+            message schema for output Parquet
+
+    -V, --version
+            Print version information
+
+    -w, --writer-version <WRITER_VERSION>
+            writer version