add parquet-fromcsv (#1)

parquet/Cargo.toml

@@ -49,6 +49,7 @@ serde_json = { version = "1.0", features = ["preserve_order"], optional = true }
 rand = "0.8"
 futures = { version = "0.3", optional = true }
 tokio = { version = "1.0", optional = true, default-features = false, features = ["macros", "fs", "rt", "io-util"] }
+anyhow = { version = "1.0.57", optional = true }
 
 [dev-dependencies]
 criterion = "0.3"
@@ -66,7 +67,7 @@ all-features = true
 
 [features]
 default = ["arrow", "snap", "brotli", "flate2", "lz4", "zstd", "base64"]
-cli = ["serde_json", "base64", "clap"]
+cli = ["serde_json", "base64", "clap", "anyhow", "arrow/csv"]
 test_common = []
 # Experimental, unstable functionality primarily used for testing
 experimental = []
@@ -85,6 +86,10 @@ required-features = ["cli"]
 name = "parquet-rowcount"
 required-features = ["cli"]
 
+[[bin]]
+name = "parquet-fromcsv"
+required-features = ["cli"]
+
 [[bench]]
 name = "arrow_writer"
 harness = false

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

@@ -0,0 +1,61 @@
+parquet 15.0.0
+Apache Arrow <dev@arrow.apache.org>
+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
+
+    -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