working_with_parquet.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "#r \"nuget:Parquet.Net\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "#r \"nuget:System.IO\" "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "using System.Collections.Generic;\n",
    "using System.IO;\n",
    "using System.Linq;\n",
    "using Parquet; \n",
    "using Parquet.Data;"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "//create data columns with schema metadata and the data you need\n",
    "var idColumn = new DataColumn(\n",
    "   new DataField<int>(\"id\"),\n",
    "   new int[] { 1, 2 });\n",
    "\n",
    "var cityColumn = new DataColumn(\n",
    "   new DataField<string>(\"city\"),\n",
    "   new string[] { \"London\", \"Derby\" });\n",
    "\n",
    "// create file schema\n",
    "var schema = new Schema(idColumn.Field, cityColumn.Field);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "Stream ms = new System.IO.MemoryStream();\n",
    "\n",
    "using (var parquetWriter = new ParquetWriter(schema, ms))\n",
    "{\n",
    "  // create a new row group in the file\n",
    "  using (ParquetRowGroupWriter groupWriter = parquetWriter.CreateRowGroup())\n",
    "  {\n",
    "     groupWriter.WriteColumn(idColumn);\n",
    "     groupWriter.WriteColumn(cityColumn);\n",
    "  }\n",
    "}\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "ms.Position = 0; "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1\n",
      "2\n"
     ]
    }
   ],
   "source": [
    "// open parquet file reader\n",
    "using (var parquetReader = new ParquetReader(ms))\n",
    "{\n",
    "  // get file schema (available straight after opening parquet reader)\n",
    "  // however, get only data fields as only they contain data values\n",
    "  DataField[] dataFields = parquetReader.Schema.GetDataFields();\n",
    "\n",
    "  // enumerate through row groups in this file\n",
    "  for(int i = 0; i < parquetReader.RowGroupCount; i++)\n",
    "  {\n",
    "     // create row group reader\n",
    "     using (ParquetRowGroupReader groupReader = parquetReader.OpenRowGroupReader(i))\n",
    "     {\n",
    "        // read all columns inside each row group (you have an option to read only\n",
    "        // required columns if you need to.\n",
    "        DataColumn[] columns = dataFields.Select(groupReader.ReadColumn).ToArray();\n",
    "\n",
    "        // get first column, for instance\n",
    "        DataColumn firstColumn = columns[0];\n",
    "\n",
    "        // .Data member contains a typed array of column data you can cast to the type of the column\n",
    "        Array data = firstColumn.Data;\n",
    "        int[] ids = (int[])data;\n",
    "        foreach (int id in ids)\n",
    "        { \n",
    "            Console.WriteLine(id);\n",
    "        }\n",
    "     }\n",
    "  }\n",
    "}\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": ".NET (C#)",
   "language": "C#",
   "name": ".net-csharp"
  },
  "language_info": {
   "file_extension": ".cs",
   "mimetype": "text/x-csharp",
   "name": "C#",
   "pygments_lexer": "csharp",
   "version": "8.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}