{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 💾 Save to Parquet\n",
    "\n",
    "The Parquet format is an efficient file format to save Synoptic's data.\n",
    "If I need to reuse data over many times (i.e., researching a case study) then I don't want to keep asking Synoptic for the data; I want to get the data and save it to local disk. Also, Synoptic restricts how much data you can retrieve in a single API request. If you need long a long time series then you will need to make multiple API calls. You should save the DataFrame information to a Parquet file to save disk space and most performant loading time.\n",
    "\n",
    "**_What are the benefits of saving Synoptic's data to Parquet instead of the raw JSON?_**\n",
    "\n",
    "To demonstrate the benefits of Parquet, let's collect a timeseries of 5 days of data for all the stations within 10 miles of WBB.\n",
    "\n",
    "1. Write the raw JSON to a JSON file.\n",
    "1. Write the Polars DataFrame to a Parquet file.\n",
    "\n",
    "- How large is the JSON file versus Parquet file? _Parquet is about 18x smaller than JSON, because it is efficiently compressed._\n",
    "- How long does it take to load a JSON file versus Parquet file? _Parquet is faster to load into memory, it's already organized in a clean table, you can read only select rows if you want, and it's easy to read multiple files at a time._\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "from datetime import timedelta\n",
    "\n",
    "import synoptic\n",
    "import polars as pl"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "🚚💨 Speedy delivery from Synoptic timeseries service.\n",
      "📦 Received data from 91 stations.\n",
      "Number of rows: 1,425,573\n"
     ]
    }
   ],
   "source": [
    "s = synoptic.TimeSeries(radius=\"wbb,10\", recent=timedelta(days=5))\n",
    "print(f\"Number of rows: {len(s.df()):,}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "JSON file size: 43.81 MB\n",
      "  Parquet size:  2.29 MB\n"
     ]
    }
   ],
   "source": [
    "import json\n",
    "from pathlib import Path\n",
    "\n",
    "filepath = Path(\"sample_timeseries.json\")\n",
    "parquet = filepath.with_suffix(\".parquet\")\n",
    "\n",
    "# Write raw data to JSON\n",
    "with open(filepath, \"w\") as f:\n",
    "    json.dump(s.json, f, indent=4)\n",
    "\n",
    "\n",
    "# Write DataFrame to Parquet\n",
    "s.df().write_parquet(parquet)\n",
    "\n",
    "print(f\"JSON file size: {filepath.stat().st_size / 1000 / 1000:>5.2f} MB\")\n",
    "print(f\"  Parquet size: {parquet.stat().st_size / 1000 / 1000:>5.2f} MB\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Wow, for 1.4 million observations Parquet is more than 19x smaller than the raw JSON. That's impressive.\n",
    "\n",
    "Reading Parquet is also fast, plus it's already in a DataFrame and we don't need to parse the JSON again."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 357 ms, sys: 200 ms, total: 557 ms\n",
      "Wall time: 551 ms\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "# Read the JSON file\n",
    "with open(filepath, \"r\") as json_file:\n",
    "    data = json.load(json_file)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 543 ms, sys: 531 ms, total: 1.07 s\n",
      "Wall time: 303 ms\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "_ = pl.read_parquet(parquet)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "synoptic2",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.12.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}