fixing_a_big_test.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "toc_visible": true,
      "authorship_tag": "ABX9TyM8UHnfOKUWOy4wiS0j5OGO",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/jmasonlee/efficiently_testing_etl_pipelines/blob/main/fixing_a_big_test.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## STEP 0.A: Setup Notebook"
      ],
      "metadata": {
        "id": "aqYjZpizVD3i"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "!rm -rf efficiently_testing_etl_pipelines\n",
        "!git clone https://github.com/jmasonlee/efficiently_testing_etl_pipelines.git\n",
        "!cp -r /content/efficiently_testing_etl_pipelines/src/ .\n",
        "!cp -r /content/efficiently_testing_etl_pipelines/tests/ .\n",
        "!rm -rf efficiently_testing_etl_pipelines\n",
        "!rm -rf tests/diamond_pricing_test*\n",
        "!rm -rf tests/test_helpers/*verification_helpers.py\n",
        "!rm -rf tests/conftest.py\n",
        "!rm -rf sample_data\n"
      ],
      "metadata": {
        "id": "i4KRAp6vTEJm"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# STEP 0.B: Setup Tests"
      ],
      "metadata": {
        "id": "5cRRMTtEBYy0"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Install Dependencies\n",
        "\n",
        "For the exercise, we will need some special dependencies to allow us to run lots of tests in a notebook."
      ],
      "metadata": {
        "id": "_h199h_XEAwT"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "`ipytest` lets us run our tests in a notebook.\n",
        "\n"
      ],
      "metadata": {
        "id": "DVAcSonZmslt"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "!pip install ipytest"
      ],
      "metadata": {
        "id": "7v0kaWulDXi-"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "ipytest is what allows us to run our tests in a notebook. This next cell is not needed if you are writing tests in a separate pytest file."
      ],
      "metadata": {
        "id": "3_ILjU8goCNY"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import ipytest\n",
        "ipytest.autoconfig()"
      ],
      "metadata": {
        "id": "t7iwGQ4Le-Oe"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "We are installing `pyspark` because it doesn't come with the base colab environment"
      ],
      "metadata": {
        "id": "HzwfHFGmm2Ca"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "!pip install pyspark"
      ],
      "metadata": {
        "id": "QNnk84AsmmFQ"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "!pip install chispa"
      ],
      "metadata": {
        "id": "r_EYg-cC1cD5"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Create a local SparkSession\n",
        "\n",
        "Normally spark runs on a bunch of executors in the cloud. Since we want our tests to be able to run on a single dev machine, we make a fixture that gives us a local spark context."
      ],
      "metadata": {
        "id": "V_OJcnSiBgUb"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import pytest\n",
        "from _pytest.fixtures import FixtureRequest\n",
        "from pyspark import SparkConf\n",
        "from pyspark.sql import SparkSession\n",
        "\n",
        "@pytest.fixture(scope=\"session\")\n",
        "def spark(request: FixtureRequest):\n",
        "    conf = (SparkConf()\n",
        "        .setMaster(\"local\")\n",
        "        .setAppName(\"sample_pyspark_testing_starter\"))\n",
        "\n",
        "    spark = SparkSession \\\n",
        "        .builder \\\n",
        "        .config(conf=conf) \\\n",
        "        .getOrCreate()\n",
        "\n",
        "    request.addfinalizer(lambda: spark.stop())\n",
        "    return spark"
      ],
      "metadata": {
        "id": "xXPLpebJ6iI2"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Create Helpers\n",
        "\n",
        "This is a helper function that retrieves our test output from the expected.json file"
      ],
      "metadata": {
        "id": "actN5s7DCF9J"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import json\n",
        "from typing import List, Dict\n",
        "\n",
        "from pyspark.pandas import DataFrame\n",
        "\n",
        "\n",
        "def create_df_from_json(json_file, spark):\n",
        "    return spark.read.option(\"multiline\", \"true\").json(json_file)\n",
        "\n",
        "\n",
        "def data_frame_to_json(df: DataFrame) -> List:\n",
        "    output = [json.loads(item) for item in df.toJSON().collect()]\n",
        "    output.sort(key=lambda item: item[\"id\"])\n",
        "    print(output)\n",
        "    return output\n",
        "\n",
        "def expected_json(name: str) -> Dict:\n",
        "    with open(f\"tests/fixtures/{name}\") as f:\n",
        "        return json.loads(f.read())"
      ],
      "metadata": {
        "id": "VO2-3px25xEl"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import pyspark\n",
        "def build_indep_vars(df, independent_vars, string_indexer, encoder, idx, vec, keep_intermediate=False, summarizer=True):\n",
        "\n",
        "    from pyspark.ml import Pipeline\n",
        "    from pyspark.ml.feature import StringIndexer, OneHotEncoder, VectorAssembler\n",
        "    idx = 'index'\n",
        "    vec = 'vector'\n",
        "\n",
        "    assembler = VectorAssembler(inputCols=independent_vars,\n",
        "                                outputCol='indep_vars')\n",
        "    pipeline  = Pipeline(stages=string_indexer+encoder+[assembler])\n",
        "    model = pipeline.fit(df)\n",
        "    df = model.transform(df)\n",
        "\n",
        "    if not keep_intermediate:\n",
        "        fcols = [c for c in df.columns if f'_{idx}' not in c[-3:] and f'_{vec}' not in c[-7:]]\n",
        "        df = df[fcols]\n",
        "\n",
        "    return df\n",
        "\n",
        "def setup_categorical_vars(categorical_vars, idx, vec, independent_vars):\n",
        "      from pyspark.ml.feature import StringIndexer, OneHotEncoder\n",
        "      if categorical_vars:\n",
        "        string_indexer = [StringIndexer(inputCol=x,\n",
        "                                        outputCol=f\"{x}_{idx}\")\n",
        "                          for x in categorical_vars]\n",
        "\n",
        "        encoder        = [OneHotEncoder(dropLast=True,\n",
        "                                        inputCol =f'{x}_{idx}',\n",
        "                                        outputCol=f'{x}_{vec}')\n",
        "                          for x in categorical_vars]\n",
        "\n",
        "        new_independent_vars = ['{}_vector'.format(x) if x in categorical_vars else x for x in independent_vars]\n",
        "      else:\n",
        "        string_indexer, encoder = [], []\n",
        "        new_independent_vars = independent_vars\n",
        "\n",
        "      return string_indexer, encoder, new_independent_vars\n",
        "\n",
        "def check_input(categorical_vars, df, independent_vars):\n",
        "    assert (type(\n",
        "        df) is pyspark.sql.dataframe.DataFrame), 'pypark_glm: A pySpark dataframe is required as the first argument.'\n",
        "    assert (type(\n",
        "        independent_vars) is list), 'pyspark_glm: List of independent variable column names must be the third argument.'\n",
        "    for iv in independent_vars:\n",
        "        assert (type(iv) is str), 'pyspark_glm: Independent variables must be column name strings.'\n",
        "        assert (iv in df.columns), 'pyspark_glm: Independent variable name is not a dataframe column.'\n",
        "    if categorical_vars:\n",
        "        for cv in categorical_vars:\n",
        "            assert (type(cv) is str), 'pyspark_glm: Categorical variables must be column name strings.'\n",
        "            assert (cv in df.columns), 'pyspark_glm: Categorical variable name is not a dataframe column.'\n",
        "            assert (cv in independent_vars), 'pyspark_glm: Categorical variables must be independent variables.'\n"
      ],
      "metadata": {
        "id": "eBvHph1aCJiN"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from pyspark.sql import DataFrame, Window, Column\n",
        "from pyspark.sql.functions import log, when, mean, col\n",
        "\n",
        "def replace_null(orig: Column, average: Column):\n",
        "    return when(orig.isNull(), average).otherwise(orig)\n",
        "\n",
        "############################ THE CODE WE ARE TESTING ###########################\n",
        "def transform(df: DataFrame) -> DataFrame:\n",
        "\n",
        "    df = df.withColumn('lprice', log('price'))\n",
        "    window = Window.partitionBy('cut', 'clarity').orderBy('price').rowsBetween(-3, 3)\n",
        "    moving_avg = mean(df['price']).over(window)\n",
        "    df = df.withColumn('moving_avg', moving_avg)\n",
        "\n",
        "    df = df.withColumn('price', when(df.price.isNull(), df.moving_avg).otherwise(df.price))\n",
        "    df = df[['id', 'carat', 'clarity', 'color', 'price']]\n",
        "    check_input(['clarity', 'color'], df, ['carat', 'clarity', 'color'])\n",
        "    indexer, encoder, independent_vars = setup_categorical_vars(\n",
        "        ['clarity', 'color'], 'index', 'vector', ['carat', 'clarity', 'color']\n",
        "    )\n",
        "    df = build_indep_vars(df, independent_vars, indexer, encoder, \"index\", \"vector\",\n",
        "                                      keep_intermediate=False,\n",
        "                                      summarizer=True)\n",
        "    return df\n",
        "################################################################################\n"
      ],
      "metadata": {
        "id": "22lMaALmCTJm"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# STEP 1: Investigate The Test"
      ],
      "metadata": {
        "id": "MjwHttFB5TB7"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## STEP 1A - Run The Test\n",
        "\n",
        "- [ ] Run the test.\n",
        "  * Hover your mouse over the number in the square brackets on the upper left side of the cell.\n",
        "  * It should turn into a play button.\n",
        "- [ ] Click the play button. The test will start to run.\n",
        "  * When the run finishes, the test should pass. You can see the results at the bottom of your cell.\n",
        "  ![Screen Shot 2023-07-14 at 9.21.32 AM.png](data:image/png;base64,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)\n",
        "\n",
        "**How long did the first run of your test take?**"
      ],
      "metadata": {
        "id": "PA5KUFM0986z"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%%ipytest -qq\n",
        "from pyspark.sql import SparkSession\n",
        "\n",
        "\n",
        "def test_will_do_the_right_thing(spark: SparkSession):\n",
        "    #ARRANGE - SET UP TEST INPUTS\n",
        "    df = create_df_from_json(\"tests/fixtures/diamonds.json\", spark)\n",
        "\n",
        "    #ACT - RUN THE CODE WE ARE TESTING\n",
        "    actual_df = transform(df)\n",
        "\n",
        "    #ASSERT - CHECK THE OUTPUTS ARE CORRECT\n",
        "    assert data_frame_to_json(actual_df) == expected_json(\"expected.json\")\n"
      ],
      "metadata": {
        "id": "lH1x9ZOi58X7"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## STEP 1B - Investigate the inputs\n",
        "The `ARRANGE` part of the test sets up the inputs. In this case, it is reading in a json file, and converting that file to a dataframe called `df`.\n",
        "\n",
        "Let's take a look at it!\n",
        "\n",
        "- [ ] Add `assert False` to the end of your test so that we can see outputs\n",
        "- [ ] Immediately after we create df, let's print out the whole dataframe, by adding:\n",
        "```\n",
        "df.show()\n",
        "```\n",
        "- [ ] Run your test.\n",
        "\n",
        "The test will fail because we used `assert False`.\n",
        "You will see an image of the dataframe printed out after this line:\n",
        "```\n",
        "--------------------------------------- Captured stdout call ---------------------------------------\n",
        "```\n"
      ],
      "metadata": {
        "id": "ISdGWlBRCJYE"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%%ipytest -qq\n",
        "from pyspark.sql import SparkSession\n",
        "\n",
        "\n",
        "def test_will_do_the_right_thing(spark: SparkSession):\n",
        "    #ARRANGE - SET UP TEST INPUTS\n",
        "    df = create_df_from_json(\"tests/fixtures/diamonds.json\", spark)\n",
        "    # df.show() # <-UNCOMMENT THIS LINE\n",
        "\n",
        "    #ACT - RUN THE CODE WE ARE TESTING\n",
        "    actual_df = transform(df)\n",
        "\n",
        "    #ASSERT - CHECK THE OUTPUTS ARE CORRECT\n",
        "    assert data_frame_to_json(actual_df) == expected_json(\"expected.json\")\n",
        "    # assert False # <-UNCOMMENT THIS LINE"
      ],
      "metadata": {
        "id": "Mm9ClXJgEdpc"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## STEP 1C - Investigate the outputs\n",
        "The assert block checks that our code produces the output we expect when we provide it with a predictable input.\n",
        "\n",
        "We expect the output of our test to match what's in the `expected.json` file.\n",
        "\n",
        "- [ ] Let's find the file.\n",
        "  * Expand the tests directory\n",
        "  * Expand the fixtures directory\n",
        "- [ ] Let's open the file.\n",
        "  * Double-click on `expected.json`.\n",
        "  \n",
        "  \n",
        "  **What is the price of the DI-26-null-price diamond in expected.json?**\n",
        "\n"
      ],
      "metadata": {
        "id": "3tq01K2lrBGc"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%%ipytest -qq\n",
        "from pyspark.sql import SparkSession\n",
        "\n",
        "\n",
        "def test_will_do_the_right_thing(spark: SparkSession):\n",
        "    #ARRANGE - SET UP TEST INPUTS\n",
        "    df = create_df_from_json(\"tests/fixtures/diamonds.json\", spark)\n",
        "\n",
        "    #ACT - RUN THE CODE WE ARE TESTING\n",
        "    actual_df = transform(df)\n",
        "\n",
        "    #ASSERT - CHECK THE OUTPUTS ARE CORRECT\n",
        "    assert data_frame_to_json(actual_df) == expected_json(\"expected.json\")"
      ],
      "metadata": {
        "id": "ZTnpwXhWsTh8"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Step 2: Setup For the Saff Squeeze"
      ],
      "metadata": {
        "id": "95eTfqjbZfGV"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Let's get ready to improve the test.\n",
        "\n",
        "**our bug**: Diamonds of the same cut and clarity are influencing the calculated price of diamonds with a different color. Only diamonds with the same cut, clarity _**and color**_ should be influencing the calculated price for diamonds with a null price."
      ],
      "metadata": {
        "id": "oDLkPTvpnKGS"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## STEP 2A - Test for the desired behaviour\n",
        "- [ ] There is a second json file (`expected_correct.json`) where the expected price for the unpriced diamond has been updated to the correct value. Use that file name as the argument passed to `expected_json`\n",
        "- [ ] Run the test. It should fail.\n",
        "- [ ] Look at the left margin of your test. Find the red $\\color{red}{\\text{E}}$'s.\n",
        "They mark the specific reason your test failed.\n",
        "\n",
        "**How many E's are there?**"
      ],
      "metadata": {
        "id": "ToSNXfVOxzz8"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%%ipytest -qq\n",
        "from pyspark.sql import SparkSession\n",
        "\n",
        "\n",
        "def test_will_do_the_right_thing(spark: SparkSession):\n",
        "    df = create_df_from_json(\"tests/fixtures/diamonds.json\", spark)\n",
        "\n",
        "    actual_df = transform(df)\n",
        "\n",
        "    ########### Rename the \"expected.json\" file below to \"expected_correct.json\"\n",
        "    assert data_frame_to_json(actual_df) == expected_json(\"expected.json\")"
      ],
      "metadata": {
        "id": "_VKLXAMic975"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## STEP 2B - Duplicate the test\n",
        "\n",
        "- [ ] Duplicate the test. Now you should have two copies of the same test.\n",
        "  * One copy will stay the same, so we can make sure that nothing is broken.\n",
        "  * The second copy is what we will change in the next steps.\n",
        "- [ ] Name the second test\n",
        "```\n",
        "test_null_price_is_replaced_based_on_cut_clarity_and_color\n",
        "```\n",
        "- [ ] Run the tests. They should both fail.\n",
        "- [ ] Look at the first line of your test run output. Find the capital $\\color{red}{\\text{F}}$'s. Those indicate failing tests.\n",
        "\n",
        "**How many F's do you see?**"
      ],
      "metadata": {
        "id": "t32-FfGUyLA4"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%%ipytest -qq\n",
        "from pyspark.sql import SparkSession\n",
        "\n",
        "\n",
        "# Duplicate the test below and rename the duplicate to:\n",
        "# test_null_price_is_replaced_based_on_cut_clarity_and_color\n",
        "\n",
        "########################## START OF THE ORIGINAL TEST ##########################\n",
        "def test_will_do_the_right_thing(spark: SparkSession):\n",
        "    df = create_df_from_json(\"tests/fixtures/diamonds.json\", spark)\n",
        "\n",
        "    actual_df = transform(df)\n",
        "\n",
        "    assert data_frame_to_json(actual_df) == expected_json(\"expected_correct.json\")\n",
        "########################### END OF THE ORIGINAL TEST ###########################"
      ],
      "metadata": {
        "id": "n7tUmVtwzSBN"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "##STEP 2C - Temporarily disable the original test\n",
        "- [ ] Mark the failing test with `@pytest.mark.skip()` to make our output less confusing.\n",
        "- [ ] Run the tests. You should see this in your cell output:\n",
        "![Screen Shot 2023-07-14 at 8.44.24 AM.png](data:image/png;base64,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)\n",
        "\n",
        "The lowercase $\\color{goldenrod}{\\text{s}}$$\\color{red}{\\text{F}}$\n",
        "indicates that two tests were run. One test ($\\color{goldenrod}{\\text{s}}$) was skipped. The other test ($\\color{red}{\\text{F}}$) failed.\n",
        "\n",
        "Immediately below the FAILURES line, you should see the name of the test that failed, in red. It is the test we are changing:\n",
        "$\\color{red}{\\text{____________________ test_null_price_is_replaced_based_on_cut_clarity_and_color ____________________}}$\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "id": "zUTmZZoM18kE"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%%ipytest -qq\n",
        "# import pytest # <- UNCOMMENT THIS LINE\n",
        "\n",
        "from pyspark.sql import SparkSession\n",
        "\n",
        "\n",
        "# @pytest.mark.skip() # <- UNCOMMENT THIS LINE\n",
        "def test_will_do_the_right_thing(spark: SparkSession):\n",
        "    df = create_df_from_json(\"tests/fixtures/diamonds.json\", spark)\n",
        "\n",
        "    actual_df = transform(df)\n",
        "\n",
        "    assert data_frame_to_json(actual_df) == expected_json(\"expected_correct.json\")\n",
        "\n",
        "def test_null_price_is_replaced_based_on_cut_clarity_and_color(spark: SparkSession):\n",
        "    df = create_df_from_json(\"tests/fixtures/diamonds.json\", spark)\n",
        "\n",
        "    actual_df = transform(df)\n",
        "\n",
        "    assert data_frame_to_json(actual_df) == expected_json(\"expected_correct.json\")"
      ],
      "metadata": {
        "id": "yOq-T54i14tl"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Step 3: Make The Assert Specific"
      ],
      "metadata": {
        "id": "cbCuKHDpso8C"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Right now, our test compares everything in the output dataframe to everything in a large json file. That's a lot of rows to compare and we can't tell which ones are important!\n",
        "\n",
        "Let's make this test assert on the thing we actually care about - the output price of the diamond!"
      ],
      "metadata": {
        "id": "X-xfpXMwstwN"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "##STEP 3A - Show the actual dataframe and remove the old assert\n",
        "\n",
        "We have added `actual_df.show()` and `assert False` to the bottom of the test we want to change, so that we can see how the dataframe changes over the next few steps.\n",
        "\n",
        "- [ ] Uncomment them and run the test to see what the dataframe looks like initially.\n",
        "- [ ] Find the red $\\color{red}{\\text{E}}$'s.\n",
        "  * **What is the error message next to them?**\n",
        "- [ ] Delete the old assert.\n",
        "- [ ] Run the test. It should fail.\n",
        "- [ ] Find the red $\\color{red}{\\text{E}}$'s.\n",
        "  * **What is the error message next to them now?**"
      ],
      "metadata": {
        "id": "8ez_6F2SYxHQ"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%%ipytest -qq\n",
        "import pytest\n",
        "\n",
        "from pyspark.sql import SparkSession\n",
        "\n",
        "\n",
        "@pytest.mark.skip()\n",
        "def test_will_do_the_right_thing(spark: SparkSession):\n",
        "    df = create_df_from_json(\"tests/fixtures/diamonds.json\", spark)\n",
        "\n",
        "    actual_df = transform(df)\n",
        "\n",
        "    assert data_frame_to_json(actual_df) == expected_json(\"expected_correct.json\")\n",
        "\n",
        "def test_null_price_is_replaced_based_on_cut_clarity_and_color(spark: SparkSession):\n",
        "    df = create_df_from_json(\"tests/fixtures/diamonds.json\", spark)\n",
        "\n",
        "    actual_df = transform(df)\n",
        "\n",
        "    ########################### DELETE THE LINE BELOW ##########################\n",
        "    assert data_frame_to_json(actual_df) == expected_json(\"expected_correct.json\")\n",
        "    # actual_df.show() # <- Uncomment this line\n",
        "    # assert False # <- Uncomment this line"
      ],
      "metadata": {
        "id": "txWJJ1FBdnAf"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## STEP 3B - Filter the dataframe down to one row.\n",
        "\n",
        "- [ ] Filter the dataframe for the unique id of the diamond we care about:  \n",
        "`DI-26-null-price`\n",
        "- [ ] Run the test. It should fail. You should see a DataFrame with only one row printed out below\n",
        "\n",
        "```\n",
        "--------------------------------------- Captured stdout call ---------------------------------------\n",
        "```\n"
      ],
      "metadata": {
        "id": "jz3ygb6GxcYG"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%%ipytest -qq\n",
        "import pytest\n",
        "\n",
        "from pyspark.sql import SparkSession\n",
        "\n",
        "\n",
        "@pytest.mark.skip()\n",
        "def test_will_do_the_right_thing(spark: SparkSession):\n",
        "    df = create_df_from_json(\"tests/fixtures/diamonds.json\", spark)\n",
        "\n",
        "    actual_df = transform(df)\n",
        "\n",
        "    assert data_frame_to_json(actual_df) == expected_json(\"expected_correct.json\")\n",
        "\n",
        "def test_null_price_is_replaced_based_on_cut_clarity_and_color(spark: SparkSession):\n",
        "    df = create_df_from_json(\"tests/fixtures/diamonds.json\", spark)\n",
        "\n",
        "    actual_df = transform(df)\n",
        "    actual_df=actual_df.filter(actual_df.id == \"PUT THE DIAMOND ID HERE\")\n",
        "\n",
        "    actual_df.show()\n",
        "    assert False"
      ],
      "metadata": {
        "id": "QqPx3UP9fX4i"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## STEP 3C - Add expected values\n",
        "\n",
        "\n",
        "\n",
        "- [ ] Create a new column in our dataframe that contains our expected price:  \n",
        "`actual_df=actual_df.withColumn('expected_price', lit(3333.0))`\n",
        "- [ ] Run the test. It should fail. You should see a new column added to your dataframe called `expected_price`, with a single row containing the value 3333.0\n"
      ],
      "metadata": {
        "id": "iBv3uIA73Ws1"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%%ipytest -qq\n",
        "import pytest\n",
        "\n",
        "from pyspark.sql import SparkSession\n",
        "from pyspark.sql.functions import lit\n",
        "\n",
        "\n",
        "@pytest.mark.skip()\n",
        "def test_will_do_the_right_thing(spark: SparkSession):\n",
        "    df = create_df_from_json(\"tests/fixtures/diamonds.json\", spark)\n",
        "\n",
        "    actual_df = transform(df)\n",
        "\n",
        "    assert data_frame_to_json(actual_df) == expected_json(\"expected_correct.json\")\n",
        "\n",
        "def test_null_price_is_replaced_based_on_cut_clarity_and_color(spark: SparkSession):\n",
        "    df = create_df_from_json(\"tests/fixtures/diamonds.json\", spark)\n",
        "\n",
        "    actual_df = transform(df)\n",
        "    actual_df=actual_df.filter(actual_df.id == \"DI-26-null-price\")\n",
        "\n",
        "    ###################### Replace the_expected_column with expected_price\n",
        "    ###################### Replace 0 with the expected price.\n",
        "    actual_df=actual_df.withColumn('the_expected_column', lit(0))\n",
        "\n",
        "    actual_df.show()\n",
        "    assert False"
      ],
      "metadata": {
        "id": "ibJrFcYahTlx"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## STEP 3D - Assert the expected value equals the actual values\n",
        "\n",
        "- [ ] Add this imports to the top of the cell, below the `%%ipytest -qq` line:  \n",
        "`from chispa import assert_column_equality`  \n",
        "- [ ] Assert the value in the price column matches the value we want:  \n",
        "`assert_column_equality(actual_df, 'price', 'expected_price')`\n",
        "- [ ] Run your test. It will fail. Look for the $\\color{red}{\\text{E}}$'s. The error should look like this:\n",
        "\n",
        "![Screen Shot 2023-07-15 at 6.53.48 AM.png](data:image/png;base64,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)"
      ],
      "metadata": {
        "id": "6a47fmgB3hYL"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%%ipytest -qq\n",
        "from chispa import assert_column_equality\n",
        "\n",
        "import pytest\n",
        "\n",
        "from pyspark.sql import SparkSession\n",
        "from pyspark.sql.functions import lit\n",
        "\n",
        "\n",
        "@pytest.mark.skip()\n",
        "def test_will_do_the_right_thing(spark: SparkSession):\n",
        "    df = create_df_from_json(\"tests/fixtures/diamonds.json\", spark)\n",
        "\n",
        "    actual_df = transform(df)\n",
        "\n",
        "    assert data_frame_to_json(actual_df) == expected_json(\"expected_correct.json\")\n",
        "\n",
        "def test_null_price_is_replaced_based_on_cut_clarity_and_color(spark: SparkSession):\n",
        "    df = create_df_from_json(\"tests/fixtures/diamonds.json\", spark)\n",
        "\n",
        "    actual_df = transform(df)\n",
        "    actual_df=actual_df.filter(actual_df.id == \"DI-26-null-price\")\n",
        "    actual_df=actual_df.withColumn('expected_price', lit(3333.0))\n",
        "\n",
        "    #### IN THE LINE BELOW, REPLACE 'the_actual_column' WITH 'price'.\n",
        "    #### IN THE LINE BELOW, REPLACE 'the_expected_column' WITH 'expected_price'.\n",
        "    assert_column_equality(actual_df, 'the_actual_column', 'the_expected_column')\n",
        "\n",
        "    actual_df.show()\n",
        "    assert False"
      ],
      "metadata": {
        "id": "v1abxIrelFD_"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## STEP 3E - What's wrong with our test?\n",
        "\n",
        "- [ ] Make a typo in your id, so no rows are picked by our filter\n",
        "- [ ] Comment out the assert False\n",
        "- [ ] Run your tests. They Pass? 🙀\n",
        "- [ ] Uncomment your assert False. Run your tests. You should be able to see the results of .show().\n",
        "\n",
        "**`assert_column_equality` compares the values in two columns to see if they are identical. Why is our test passing?**\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "id": "dYs9vKwA2qfJ"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%%ipytest -qq\n",
        "import pytest\n",
        "\n",
        "from pyspark.sql import SparkSession\n",
        "\n",
        "\n",
        "@pytest.mark.skip()\n",
        "def test_will_do_the_right_thing(spark: SparkSession):\n",
        "    df = create_df_from_json(\"tests/fixtures/diamonds.json\", spark)\n",
        "\n",
        "    actual_df = transform(df)\n",
        "\n",
        "    assert data_frame_to_json(actual_df) == expected_json(\"expected_correct.json\")\n",
        "\n",
        "def test_null_price_is_replaced_based_on_cut_clarity_and_color(spark: SparkSession):\n",
        "    df = create_df_from_json(\"tests/fixtures/diamonds.json\", spark)\n",
        "\n",
        "    actual_df = transform(df)\n",
        "\n",
        "    actual_df=actual_df.filter(actual_df.id == \"DI-26-null-price\") # <- MAKE A TYPO HERE\n",
        "    actual_df=actual_df.withColumn('expected_price', lit(3333.0))\n",
        "    assert_column_equality(actual_df, 'price', 'expected_price')\n",
        "\n",
        "    actual_df.show()\n",
        "    assert False # <-COMMENT THIS OUT FOR YOUR TESTS TO PASS. UNCOMMENT IT TO SEE THE OUTPUT FROM ACTUAL_DF.SHOW()"
      ],
      "metadata": {
        "id": "azCBzGOPgfPk"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## STEP 3F - Check we have exactly one row\n",
        "\n",
        "There is a risk that our dataframe is empty, so we want to make sure it isn't.\n",
        "\n",
        "- [ ] After you filter, add:\n",
        "```\n",
        "assert actual_df.count() == 1\n",
        "```\n",
        "- [ ] Run your tests. They should fail exactly the same way they did before.\n"
      ],
      "metadata": {
        "id": "MzonDDW6Ig0-"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%%ipytest -qq\n",
        "import pytest\n",
        "\n",
        "from pyspark.sql import SparkSession\n",
        "\n",
        "\n",
        "@pytest.mark.skip()\n",
        "def test_will_do_the_right_thing(spark: SparkSession):\n",
        "    df = create_df_from_json(\"tests/fixtures/diamonds.json\", spark)\n",
        "\n",
        "    actual_df = transform(df)\n",
        "\n",
        "    assert data_frame_to_json(actual_df) == expected_json(\"expected_correct.json\")\n",
        "\n",
        "def test_null_price_is_replaced_based_on_cut_clarity_and_color(spark: SparkSession):\n",
        "    df = create_df_from_json(\"tests/fixtures/diamonds.json\", spark)\n",
        "\n",
        "    actual_df = transform(df)\n",
        "\n",
        "    actual_df=actual_df.filter(actual_df.id == \"DI-26-null-price\")\n",
        "    # assert actual_df.count() == 1 # <- Uncomment this line\n",
        "    actual_df=actual_df.withColumn('expected_price', lit(3333.0))\n",
        "    assert_column_equality(actual_df, 'price', 'expected_price')\n",
        "\n",
        "    actual_df.show()\n",
        "    assert False"
      ],
      "metadata": {
        "id": "20A9J1n_I1wO"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## STEP 3G - Make the assert logic easier to work with\n",
        "\n",
        "Let's extract a function so the assert is easier to work with.\n",
        "\n",
        "- [ ] Copy the lines in between the comments and paste them into the body of the `assert_diamond_has_expected_price` function.\n",
        "- [ ] Delete the lines you copied, and replace them with a single method call:\n",
        "```\n",
        "assert_diamond_has_expected_price()\n",
        "```\n",
        "- [ ] Run your test, it should fail with the chispa error:\n",
        "![Screen Shot 2023-07-15 at 6.53.48 AM.png](data:image/png;base64,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)"
      ],
      "metadata": {
        "id": "Kh1yL2EkmnaN"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%%ipytest -qq\n",
        "from chispa import assert_column_equality\n",
        "\n",
        "import pytest\n",
        "\n",
        "from pyspark.sql import SparkSession\n",
        "from pyspark.sql.functions import lit\n",
        "\n",
        "\n",
        "def assert_diamond_has_expected_price(actual_df: DataFrame) -> None:\n",
        "    pass # <- REPLACE THIS LINE WITH THE COPIED LINES\n",
        "\n",
        "\n",
        "@pytest.mark.skip()\n",
        "def test_will_do_the_right_thing(spark: SparkSession):\n",
        "    df = create_df_from_json(\"tests/fixtures/diamonds.json\", spark)\n",
        "\n",
        "    actual_df = transform(df)\n",
        "\n",
        "    assert data_frame_to_json(actual_df) == expected_json(\"expected_correct.json\")\n",
        "\n",
        "def test_null_price_is_replaced_based_on_cut_clarity_and_color(spark: SparkSession):\n",
        "    df = create_df_from_json(\"tests/fixtures/diamonds.json\", spark)\n",
        "\n",
        "    actual_df = transform(df)\n",
        "\n",
        "    ######################## COPY THESE LINES ####################\n",
        "    actual_df=actual_df.filter(actual_df.id == \"DI-26-null-price\")\n",
        "    assert actual_df.count() == 1\n",
        "    actual_df=actual_df.withColumn('expected_price', lit(3333.0))\n",
        "    assert_column_equality(actual_df, 'price', 'expected_price')\n",
        "    ##############################################################\n",
        "    # Replace the 4 lines above with a single call to the method:\n",
        "    # assert_diamond_has_expected_price(actual_df)\n",
        "\n",
        "    actual_df.show()\n",
        "    assert False"
      ],
      "metadata": {
        "id": "n77-SLnzoEgd"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## STEP 3H - Clean up\n",
        "\n",
        "- [ ] Delete `assert False` from the bottom of your test.\n",
        "- [ ] Delete `actual_df.show()` from the bottom of your test.\n",
        "- [ ] Run your test, it should fail with the chispa error:\n",
        "![Screen Shot 2023-07-15 at 6.53.48 AM.png](data:image/png;base64,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)"
      ],
      "metadata": {
        "id": "jz52ONCaprDf"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%%ipytest -qq\n",
        "from chispa import assert_column_equality\n",
        "\n",
        "import pytest\n",
        "\n",
        "from pyspark.sql.functions import lit\n",
        "\n",
        "\n",
        "def assert_diamond_has_expected_price(actual_df: DataFrame) -> None:\n",
        "    actual_df=actual_df.filter(actual_df.id == \"DI-26-null-price\")\n",
        "    assert actual_df.count() == 1\n",
        "    actual_df=actual_df.withColumn('expected_price', lit(3333.0))\n",
        "    assert_column_equality(actual_df, 'price', 'expected_price')\n",
        "\n",
        "\n",
        "@pytest.mark.skip()\n",
        "def test_will_do_the_right_thing(spark: SparkSession):\n",
        "    df = create_df_from_json(\"tests/fixtures/diamonds.json\", spark)\n",
        "\n",
        "    actual_df = transform(df)\n",
        "\n",
        "    assert data_frame_to_json(actual_df) == expected_json(\"expected_correct.json\")\n",
        "\n",
        "def test_null_price_is_replaced_based_on_cut_clarity_and_color(spark: SparkSession):\n",
        "    df = create_df_from_json(\"tests/fixtures/diamonds.json\", spark)\n",
        "\n",
        "    actual_df = transform(df)\n",
        "\n",
        "    assert_diamond_has_expected_price(actual_df)\n",
        "\n",
        "    actual_df.show() # <- DELETE THIS LINE\n",
        "    assert False # <- DELETE THIS LINE"
      ],
      "metadata": {
        "id": "yzavIJb-ZLgi"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Step 4: Reduce Duplicate Coverage by Squeezing the bottom\n",
        "\n",
        "Right now, our test is running the entire transform function. Because there are multiple tests in `diamonds.json`, each test is running the same large block of code over and over again."
      ],
      "metadata": {
        "id": "sw4mvQYw5pkE"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## STEP 4A - Put the transform function where you can work with it\n",
        "\n",
        "- [ ] Run the test. It should be failing with this error.\n",
        "![Screen Shot 2023-07-15 at 6.53.48 AM.png](data:image/png;base64,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)\n",
        "- [ ] Uncomment the line that maps `df` to `actual_df`\n",
        "- [ ] Copy the body of the tranform function into the test in between the two lines of comments\n",
        "- [ ] Delete the line that calls the transform function.\n",
        "- [ ] Run the test. It should still be failing with the same error.\n"
      ],
      "metadata": {
        "id": "-2pcPdIhf6ad"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%%ipytest -qq\n",
        "from pyspark.sql import DataFrame, SparkSession\n",
        "from pyspark.sql.functions import lit, log\n",
        "from chispa import assert_column_equality\n",
        "\n",
        "############################ THE CODE WE ARE TESTING ###########################\n",
        "def transform(df: DataFrame) -> DataFrame:\n",
        "\n",
        "### THE BODY OF THE TRANSFORM FUNCTION STARTS HERE\n",
        "    df = df.withColumn('lprice', log('price'))\n",
        "    window = Window.partitionBy('cut', 'clarity').orderBy('price').rowsBetween(-3, 3)\n",
        "    moving_avg = mean(df['price']).over(window)\n",
        "    df = df.withColumn('moving_avg', moving_avg)\n",
        "\n",
        "    df = df.withColumn('price', when(df.price.isNull(), df.moving_avg).otherwise(df.price))\n",
        "    df = df[['id', 'carat', 'clarity', 'color', 'price']]\n",
        "    check_input(['clarity', 'color'], df, ['carat', 'clarity', 'color'])\n",
        "    indexer, encoder, independent_vars = setup_categorical_vars(\n",
        "        ['clarity', 'color'], 'index', 'vector', ['carat', 'clarity', 'color']\n",
        "    )\n",
        "    df = build_indep_vars(df, independent_vars, indexer, encoder, \"index\", \"vector\",\n",
        "                                      keep_intermediate=False,\n",
        "                                      summarizer=True)\n",
        "### THE BODY OF THE TRANSFORM FUNCTION ENDS HERE\n",
        "    return df\n",
        "################################################################################\n",
        "\n",
        "\n",
        "\n",
        "def assert_diamond_has_expected_price(actual_df: DataFrame) -> None:\n",
        "    actual_df=actual_df.filter(actual_df.id == \"DI-26-null-price\")\n",
        "    assert actual_df.count() == 1\n",
        "    actual_df=actual_df.withColumn('expected_price', lit(3333.0))\n",
        "    assert_column_equality(actual_df, 'price', 'expected_price')\n",
        "\n",
        "@pytest.mark.skip()\n",
        "def test_will_do_the_right_thing(spark: SparkSession):\n",
        "    df = create_df_from_json(\"tests/fixtures/diamonds.json\", spark)\n",
        "\n",
        "    actual_df = transform(df)\n",
        "\n",
        "    assert data_frame_to_json(actual_df) == expected_json(\"expected_correct.json\")\n",
        "\n",
        "def test_null_price_is_replaced_based_on_cut_clarity_and_color(spark: SparkSession):\n",
        "    df = create_df_from_json(\"tests/fixtures/diamonds.json\", spark)\n",
        "\n",
        "    ###################### PASTE THE BODY OF THE TRANSFORM HERE ################\n",
        "    ############################################################################\n",
        "    # actual_df = df  # <- UNCOMMENT THIS LINE\n",
        "    actual_df = transform(df)# <-- DELETE THIS LINE.\n",
        "    assert_diamond_has_expected_price(actual_df)"
      ],
      "metadata": {
        "id": "Ci_qnCgB57k9"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "\n",
        "## STEP 4B - Squeeze the bottom\n",
        "\n",
        "**Squeeze the bottom until you find the bug**\n",
        "- [ ] Move your assert up one line at a time.  \n",
        "- [ ] After each move, run your test.  \n",
        "- [ ] If it fails for a different reason, figure out why it's failing.\n",
        "- [ ] If the test fails for the same reason, the line wasn't important for the bug you wanted to catch. Delete it.\n",
        "- [ ] Continue until your failure reason changes.\n",
        "\n",
        "If your error doesn't look _**exactly**_ like this one, it's time to STOP 🛑\n",
        "\n",
        "![Screen Shot 2023-07-15 at 6.53.48 AM.png](data:image/png;base64,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)\n",
        "\n",
        "**Where does the failure reason change? When do you need to stop?**"
      ],
      "metadata": {
        "id": "PhriAy0jfsTj"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%%ipytest -qq\n",
        "from pyspark.sql import DataFrame, SparkSession\n",
        "from pyspark.sql.functions import lit, log\n",
        "from chispa import assert_column_equality\n",
        "\n",
        "\n",
        "############################ THE CODE WE ARE TESTING ###########################\n",
        "def transform(df: DataFrame) -> DataFrame:\n",
        "    df = df.withColumn('lprice', log('price'))\n",
        "    window = Window.partitionBy('cut', 'clarity').orderBy('price').rowsBetween(-3, 3)\n",
        "    moving_avg = mean(df['price']).over(window)\n",
        "    df = df.withColumn('moving_avg', moving_avg)\n",
        "\n",
        "    df = df.withColumn('price', when(df.price.isNull(), df.moving_avg).otherwise(df.price))\n",
        "    df = df[['id', 'carat', 'clarity', 'color', 'price']]\n",
        "    check_input(['clarity', 'color'], df, ['carat', 'clarity', 'color'])\n",
        "    indexer, encoder, independent_vars = setup_categorical_vars(\n",
        "        ['clarity', 'color'], 'index', 'vector', ['carat', 'clarity', 'color']\n",
        "    )\n",
        "    df = build_indep_vars(df, independent_vars, indexer, encoder, \"index\", \"vector\",\n",
        "                                      keep_intermediate=False,\n",
        "                                      summarizer=True)\n",
        "    return df\n",
        "################################################################################\n",
        "\n",
        "\n",
        "\n",
        "def assert_diamond_has_expected_price(actual_df: DataFrame) -> None:\n",
        "    actual_df=actual_df.filter(actual_df.id == \"DI-26-null-price\")\n",
        "    assert actual_df.count() == 1\n",
        "    actual_df=actual_df.withColumn('expected_price', lit(3333.0))\n",
        "    assert_column_equality(actual_df, 'price', 'expected_price')\n",
        "\n",
        "@pytest.mark.skip()\n",
        "def test_will_do_the_right_thing(spark: SparkSession):\n",
        "    df = create_df_from_json(\"tests/fixtures/diamonds.json\", spark)\n",
        "\n",
        "    actual_df = transform(df)\n",
        "\n",
        "    assert data_frame_to_json(actual_df) == expected_json(\"expected_correct.json\")\n",
        "\n",
        "def test_null_price_is_replaced_based_on_cut_clarity_and_color(spark: SparkSession):\n",
        "    df = create_df_from_json(\"tests/fixtures/diamonds.json\", spark)\n",
        "\n",
        "    df = df.withColumn('lprice', log('price'))\n",
        "    window = Window.partitionBy('cut', 'clarity').orderBy('price').rowsBetween(-3, 3)\n",
        "    moving_avg = mean(df['price']).over(window)\n",
        "    df = df.withColumn('moving_avg', moving_avg)\n",
        "\n",
        "    df = df.withColumn('price', when(df.price.isNull(), df.moving_avg).otherwise(df.price))\n",
        "    df = df[['id', 'carat', 'clarity', 'color', 'price']]\n",
        "    check_input(['clarity', 'color'], df, ['carat', 'clarity', 'color'])\n",
        "    indexer, encoder, independent_vars = setup_categorical_vars(\n",
        "        ['clarity', 'color'], 'index', 'vector', ['carat', 'clarity', 'color']\n",
        "    )\n",
        "    df = build_indep_vars(df, independent_vars, indexer, encoder, \"index\", \"vector\",\n",
        "                                      keep_intermediate=False,\n",
        "                                      summarizer=True)\n",
        "    actual_df = df\n",
        "\n",
        "    assert_diamond_has_expected_price(actual_df)"
      ],
      "metadata": {
        "id": "LE1V0QDVp44x"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Step 5: Simplify the Top"
      ],
      "metadata": {
        "id": "5K0RWKPdr4fn"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## DEMO 5A - Create the minimum inputs needed to reproduce the bug\n",
        "- [ ] Delete the line that reads in from diamonds.json\n",
        "- [ ] Replace it with two lines, one that parses the json text file to a string, and the other that converts the json string to spark\n",
        "- [ ] Run your test. It should fail with the `ColumnsNotEqualError`.\n",
        "\n",
        "![Screen Shot 2023-07-15 at 6.53.48 AM.png](data:image/png;base64,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)\n"
      ],
      "metadata": {
        "id": "MTcFz54o2_i0"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%%ipytest -qq\n",
        "from pyspark.sql import DataFrame, SparkSession\n",
        "from pyspark.sql.functions import lit, log\n",
        "from chispa import assert_column_equality\n",
        "\n",
        "from pathlib import Path\n",
        "\n",
        "############################ THE CODE WE ARE TESTING ###########################\n",
        "def transform(df: DataFrame) -> DataFrame:\n",
        "    df = df.withColumn('lprice', log('price'))\n",
        "    window = Window.partitionBy('cut', 'clarity').orderBy('price').rowsBetween(-3, 3)\n",
        "    moving_avg = mean(df['price']).over(window)\n",
        "    df = df.withColumn('moving_avg', moving_avg)\n",
        "\n",
        "    df = df.withColumn('price', when(df.price.isNull(), df.moving_avg).otherwise(df.price))\n",
        "    df = df[['id', 'carat', 'clarity', 'color', 'price']]\n",
        "    check_input(['clarity', 'color'], df, ['carat', 'clarity', 'color'])\n",
        "    indexer, encoder, independent_vars = setup_categorical_vars(\n",
        "        ['clarity', 'color'], 'index', 'vector', ['carat', 'clarity', 'color']\n",
        "    )\n",
        "    df = build_indep_vars(df, independent_vars, indexer, encoder, \"index\", \"vector\",\n",
        "                                      keep_intermediate=False,\n",
        "                                      summarizer=True)\n",
        "    return df\n",
        "################################################################################\n",
        "\n",
        "\n",
        "def assert_diamond_has_expected_price(actual_df: DataFrame) -> None:\n",
        "    actual_df=actual_df.filter(actual_df.id == \"DI-26-null-price\")\n",
        "    assert actual_df.count() == 1\n",
        "    actual_df=actual_df.withColumn('expected_price', lit(3333.0))\n",
        "    assert_column_equality(actual_df, 'price', 'expected_price')\n",
        "\n",
        "\n",
        "@pytest.mark.skip()\n",
        "def test_will_do_the_right_thing(spark: SparkSession):\n",
        "    df = create_df_from_json(\"tests/fixtures/diamonds.json\", spark)\n",
        "\n",
        "    actual_df = transform(df)\n",
        "\n",
        "    assert data_frame_to_json(actual_df) == expected_json(\"expected_correct.json\")\n",
        "\n",
        "\n",
        "def test_null_price_is_replaced_based_on_cut_clarity_and_color(spark: SparkSession):\n",
        "    ############################# DELETE THE LINE BELOW ########################\n",
        "    df = create_df_from_json(\"tests/fixtures/diamonds.json\", spark)\n",
        "    #json_text = Path(\"./tests/fixtures/diamonds.json\").read_text() # <- UNCOMMENT THIS LINE\n",
        "    #df = spark.read.option(\"multiline\", \"true\").json(spark.sparkContext.parallelize([json_text])) # <- UNCOMMENT THIS LINE\n",
        "\n",
        "    df = df.withColumn('lprice', log('price'))\n",
        "    window = Window.partitionBy('cut', 'clarity').orderBy('price').rowsBetween(-3, 3)\n",
        "    moving_avg = mean(df['price']).over(window)\n",
        "    df = df.withColumn('moving_avg', moving_avg)\n",
        "\n",
        "    df = df.withColumn('price', when(df.price.isNull(), df.moving_avg).otherwise(df.price))\n",
        "    assert_diamond_has_expected_price(df)\n",
        "\n",
        "\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "id": "Gah6-OpWq48Y"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## DEMO 5B - Create the minimum Inputs needed to reproduce the bug\n",
        "- [ ] Copy the contents of `tests/fixtures/diamonds.json` to a multiline string.\n",
        "- [ ] Delete the line that reads the input data from a file\n",
        "- [ ] Run your test. It should fail with the `ColumnsNotEqualError`."
      ],
      "metadata": {
        "id": "KGy1pOc4-_S8"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%%ipytest -qq\n",
        "from pyspark.sql import DataFrame, SparkSession\n",
        "from pyspark.sql.functions import lit, log\n",
        "from chispa import assert_column_equality\n",
        "\n",
        "from pathlib import Path\n",
        "\n",
        "############################ THE CODE WE ARE TESTING ###########################\n",
        "def transform(df: DataFrame) -> DataFrame:\n",
        "    df = df.withColumn('lprice', log('price'))\n",
        "    window = Window.partitionBy('cut', 'clarity').orderBy('price').rowsBetween(-3, 3)\n",
        "    moving_avg = mean(df['price']).over(window)\n",
        "    df = df.withColumn('moving_avg', moving_avg)\n",
        "\n",
        "    df = df.withColumn('price', when(df.price.isNull(), df.moving_avg).otherwise(df.price))\n",
        "    df = df[['id', 'carat', 'clarity', 'color', 'price']]\n",
        "    check_input(['clarity', 'color'], df, ['carat', 'clarity', 'color'])\n",
        "    indexer, encoder, independent_vars = setup_categorical_vars(\n",
        "        ['clarity', 'color'], 'index', 'vector', ['carat', 'clarity', 'color']\n",
        "    )\n",
        "    df = build_indep_vars(df, independent_vars, indexer, encoder, \"index\", \"vector\",\n",
        "                                      keep_intermediate=False,\n",
        "                                      summarizer=True)\n",
        "    return df\n",
        "################################################################################\n",
        "\n",
        "\n",
        "def assert_diamond_has_expected_price(actual_df: DataFrame) -> None:\n",
        "    actual_df=actual_df.filter(actual_df.id == \"DI-26-null-price\")\n",
        "    assert actual_df.count() == 1\n",
        "    actual_df=actual_df.withColumn('expected_price', lit(3333.0))\n",
        "    assert_column_equality(actual_df, 'price', 'expected_price')\n",
        "\n",
        "\n",
        "@pytest.mark.skip()\n",
        "def test_will_do_the_right_thing(spark: SparkSession):\n",
        "    df = create_df_from_json(\"tests/fixtures/diamonds.json\", spark)\n",
        "\n",
        "    actual_df = transform(df)\n",
        "\n",
        "    assert data_frame_to_json(actual_df) == expected_json(\"expected_correct.json\")\n",
        "\n",
        "\n",
        "def test_null_price_is_replaced_based_on_cut_clarity_and_color(spark: SparkSession):\n",
        "    json_text = '''\n",
        "      PASTE CONTENTS OF tests/fixtures/diamonds.json HERE\n",
        "    '''\n",
        "    ############################# DELETE THE LINE BELOW ########################\n",
        "    json_text = Path(\"./tests/fixtures/diamonds.json\").read_text()\n",
        "    df = spark.read.option(\"multiline\", \"true\").json(spark.sparkContext.parallelize([json_text]))\n",
        "\n",
        "    df = df.withColumn('lprice', log('price'))\n",
        "    window = Window.partitionBy('cut', 'clarity').orderBy('price').rowsBetween(-3, 3)\n",
        "    moving_avg = mean(df['price']).over(window)\n",
        "    df = df.withColumn('moving_avg', moving_avg)\n",
        "\n",
        "    df = df.withColumn('price', when(df.price.isNull(), df.moving_avg).otherwise(df.price))\n",
        "    assert_diamond_has_expected_price(df)\n",
        "\n",
        "\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "id": "9s13wcV2tWDR"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## DEMO 5C - Squeeze the inputs\n",
        "- [ ] Run your test. It should fail with the `ColumnsNotEqualError`.\n",
        "- [ ] Delete the objects from the JSON array one at a time.\n",
        "- [ ] After each deletion, run your test. If it fails with a different error, undo the delete.\n",
        "- [ ] Move onto the next object\n",
        "- [ ] At the end, you should have the minimum objects needed for our test to test the behaviour\n",
        "- [ ] Run your test. It should fail with the `ColumnsNotEqualError`.\n",
        "\n",
        "This is the error you're looking for:\n",
        "![Screen Shot 2023-07-15 at 6.53.48 AM.png](data:image/png;base64,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)\n",
        "\n",
        "**How many inputs are the minimum number of inputs you need?**\n",
        "\n",
        "**VERY IMPORTANT:** Your test will fail for the wrong reasons if the JSON array ends in a comma.\n",
        "\n",
        "Make sure the last 4 lines of the array always look like this:\n",
        "```\n",
        "    \"z\": null\n",
        "  }\n",
        "]\n",
        "    '''\n",
        " ```\n",
        "\n",
        "NOT this:\n",
        "```\n",
        "    \"z\": null\n",
        "  },\n",
        "]\n",
        "    '''\n",
        "```"
      ],
      "metadata": {
        "id": "AwLe7ZUx_8wR"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%%ipytest -qq\n",
        "from pyspark.sql import DataFrame, SparkSession\n",
        "from pyspark.sql.functions import lit, log\n",
        "from chispa import assert_column_equality\n",
        "\n",
        "from pathlib import Path\n",
        "\n",
        "############################ THE CODE WE ARE TESTING ###########################\n",
        "def transform(df: DataFrame) -> DataFrame:\n",
        "    df = df.withColumn('lprice', log('price'))\n",
        "    window = Window.partitionBy('cut', 'clarity').orderBy('price').rowsBetween(-3, 3)\n",
        "    moving_avg = mean(df['price']).over(window)\n",
        "    df = df.withColumn('moving_avg', moving_avg)\n",
        "\n",
        "    df = df.withColumn('price', when(df.price.isNull(), df.moving_avg).otherwise(df.price))\n",
        "    df = df[['id', 'carat', 'clarity', 'color', 'price']]\n",
        "    check_input(['clarity', 'color'], df, ['carat', 'clarity', 'color'])\n",
        "    indexer, encoder, independent_vars = setup_categorical_vars(\n",
        "        ['clarity', 'color'], 'index', 'vector', ['carat', 'clarity', 'color']\n",
        "    )\n",
        "    df = build_indep_vars(df, independent_vars, indexer, encoder, \"index\", \"vector\",\n",
        "                                      keep_intermediate=False,\n",
        "                                      summarizer=True)\n",
        "    return df\n",
        "################################################################################\n",
        "\n",
        "\n",
        "def assert_diamond_has_expected_price(actual_df: DataFrame) -> None:\n",
        "    actual_df=actual_df.filter(actual_df.id == \"DI-26-null-price\")\n",
        "    assert actual_df.count() == 1\n",
        "    actual_df=actual_df.withColumn('expected_price', lit(3333.0))\n",
        "    assert_column_equality(actual_df, 'price', 'expected_price')\n",
        "\n",
        "\n",
        "@pytest.mark.skip()\n",
        "def test_will_do_the_right_thing(spark: SparkSession):\n",
        "    df = create_df_from_json(\"tests/fixtures/diamonds.json\", spark)\n",
        "\n",
        "    actual_df = transform(df)\n",
        "\n",
        "    assert data_frame_to_json(actual_df) == expected_json(\"expected_correct.json\")\n",
        "\n",
        "\n",
        "def test_null_price_is_replaced_based_on_cut_clarity_and_color(spark: SparkSession):\n",
        "    json_text = '''\n",
        "      [\n",
        "  {\n",
        "    \"id\": 1,\n",
        "    \"carat\": 0.23,\n",
        "    \"cut\": \"Ideal\",\n",
        "    \"color\": \"E\",\n",
        "    \"clarity\": \"SI2\",\n",
        "    \"depth\": 61.5,\n",
        "    \"table\": 55,\n",
        "    \"price\": 326,\n",
        "    \"x\": 3.95,\n",
        "    \"y\": 3.98,\n",
        "    \"z\": 2.43\n",
        "  },\n",
        "  {\n",
        "    \"id\": \"minimum_inputs\",\n",
        "    \"carat\": 0.23,\n",
        "    \"cut\": \"Good\",\n",
        "    \"color\": \"F\",\n",
        "    \"clarity\": \"SI1\",\n",
        "    \"depth\": null,\n",
        "    \"table\": null,\n",
        "    \"price\": null,\n",
        "    \"x\": null,\n",
        "    \"y\": null,\n",
        "    \"z\": null\n",
        "  },\n",
        "  {\n",
        "    \"id\": \"DI-26-null-price\",\n",
        "    \"carat\": 0.21,\n",
        "    \"cut\": \"Good\",\n",
        "    \"color\": \"D\",\n",
        "    \"clarity\": \"VVS1\",\n",
        "    \"depth\": null,\n",
        "    \"table\": null,\n",
        "    \"price\": null,\n",
        "    \"x\": null,\n",
        "    \"y\": null,\n",
        "    \"z\": null\n",
        "  },\n",
        "  {\n",
        "    \"id\": \"DI-26\",\n",
        "    \"carat\": 0.21,\n",
        "    \"cut\": \"Good\",\n",
        "    \"color\": \"D\",\n",
        "    \"clarity\": \"VVS1\",\n",
        "    \"depth\": null,\n",
        "    \"table\": null,\n",
        "    \"price\": 3332,\n",
        "    \"x\": null,\n",
        "    \"y\": null,\n",
        "    \"z\": null\n",
        "  },\n",
        "  {\n",
        "    \"id\": \"DI-26\",\n",
        "    \"carat\": 0.21,\n",
        "    \"cut\": \"Good\",\n",
        "    \"color\": \"D\",\n",
        "    \"clarity\": \"VVS1\",\n",
        "    \"depth\": null,\n",
        "    \"table\": null,\n",
        "    \"price\": 3334,\n",
        "    \"x\": null,\n",
        "    \"y\": null,\n",
        "    \"z\": null\n",
        "  },\n",
        "  {\n",
        "    \"id\": \"DI-27\",\n",
        "    \"carat\": 0.21,\n",
        "    \"cut\": \"Very Good\",\n",
        "    \"color\": \"D\",\n",
        "    \"clarity\": \"VVS1\",\n",
        "    \"depth\": null,\n",
        "    \"table\": null,\n",
        "    \"price\": 2692,\n",
        "    \"x\": null,\n",
        "    \"y\": null,\n",
        "    \"z\": null\n",
        "  },\n",
        "  {\n",
        "    \"id\": \"DI-30\",\n",
        "    \"carat\": 0.32,\n",
        "    \"cut\": \"Good\",\n",
        "    \"color\": \"D\",\n",
        "    \"clarity\": \"I1\",\n",
        "    \"depth\": 60.9,\n",
        "    \"table\": 58,\n",
        "    \"price\": 345,\n",
        "    \"x\": 4.38,\n",
        "    \"y\": 4.42,\n",
        "    \"z\": 2.68\n",
        "  },\n",
        "  {\n",
        "    \"id\": \"DI-28\",\n",
        "    \"carat\": 0.21,\n",
        "    \"cut\": \"Good\",\n",
        "    \"color\": \"G\",\n",
        "    \"clarity\": \"VVS1\",\n",
        "    \"depth\": null,\n",
        "    \"table\": null,\n",
        "    \"price\": 1665,\n",
        "    \"x\": null,\n",
        "    \"y\": null,\n",
        "    \"z\": null\n",
        "  }\n",
        "]\n",
        "\n",
        "    '''\n",
        "\n",
        "    df = spark.read.option(\"multiline\", \"true\").json(spark.sparkContext.parallelize([json_text]))\n",
        "\n",
        "    df = df.withColumn('lprice', log('price'))\n",
        "    window = Window.partitionBy('cut', 'clarity').orderBy('price').rowsBetween(-3, 3)\n",
        "    moving_avg = mean(df['price']).over(window)\n",
        "    df = df.withColumn('moving_avg', moving_avg)\n",
        "\n",
        "    df = df.withColumn('price', when(df.price.isNull(), df.moving_avg).otherwise(df.price))\n",
        "    assert_diamond_has_expected_price(df)\n",
        "\n",
        "\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "id": "lJ3zFaYKvwgh"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## DEMO 5D - Squeeze the inputs\n",
        "- [ ] Run your test. It should fail with the `ColumnsNotEqualError`.\n",
        "- [ ] Delete the properties from each JSON object one at a time.\n",
        "- [ ] After each deletion, run your test. If it fails with any error that is not the `ColumnsNotEqualError`, undo the delete.\n",
        "- [ ] Move onto the next property\n",
        "- [ ] At the end, you should have the minimum objects and properties needed for our test to test the behaviour\n",
        "- [ ] Run your test. It should fail with the same `ColumnsNotEqualError`.\n",
        "\n",
        "This is the error you're looking for:\n",
        "![Screen Shot 2023-07-15 at 6.53.48 AM.png](data:image/png;base64,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)\n",
        "\n",
        "**At the end, Which properties are you left with?**\n",
        "\n",
        "**VERY IMPORTANT:** Your test will fail for the wrong reasons if the last property in any JSON object ends in a comma.\n",
        "\n",
        "Make sure the last property of each object always looks like this:\n",
        "```\n",
        "    \"z\": null\n",
        "  }\n",
        " ```\n",
        "\n",
        "NOT this:\n",
        "```\n",
        "    \"z\": null,\n",
        "  }\n",
        "```"
      ],
      "metadata": {
        "id": "pDKcNZHkBspK"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%%ipytest -qq\n",
        "from pyspark.sql import DataFrame, SparkSession\n",
        "from pyspark.sql.functions import lit, log\n",
        "from chispa import assert_column_equality\n",
        "\n",
        "from pathlib import Path\n",
        "\n",
        "############################ THE CODE WE ARE TESTING ###########################\n",
        "def transform(df: DataFrame) -> DataFrame:\n",
        "    df = df.withColumn('lprice', log('price'))\n",
        "    window = Window.partitionBy('cut', 'clarity').orderBy('price').rowsBetween(-3, 3)\n",
        "    moving_avg = mean(df['price']).over(window)\n",
        "    df = df.withColumn('moving_avg', moving_avg)\n",
        "\n",
        "    df = df.withColumn('price', when(df.price.isNull(), df.moving_avg).otherwise(df.price))\n",
        "    df = df[['id', 'carat', 'clarity', 'color', 'price']]\n",
        "    check_input(['clarity', 'color'], df, ['carat', 'clarity', 'color'])\n",
        "    indexer, encoder, independent_vars = setup_categorical_vars(\n",
        "        ['clarity', 'color'], 'index', 'vector', ['carat', 'clarity', 'color']\n",
        "    )\n",
        "    df = build_indep_vars(df, independent_vars, indexer, encoder, \"index\", \"vector\",\n",
        "                                      keep_intermediate=False,\n",
        "                                      summarizer=True)\n",
        "    return df\n",
        "################################################################################\n",
        "\n",
        "\n",
        "\n",
        "def assert_diamond_has_expected_price(actual_df: DataFrame) -> None:\n",
        "    actual_df=actual_df.filter(actual_df.id == \"DI-26-null-price\")\n",
        "    assert actual_df.count() == 1\n",
        "    actual_df=actual_df.withColumn('expected_price', lit(3333.0))\n",
        "    assert_column_equality(actual_df, 'price', 'expected_price')\n",
        "\n",
        "\n",
        "@pytest.mark.skip()\n",
        "def test_will_do_the_right_thing(spark: SparkSession):\n",
        "    df = create_df_from_json(\"tests/fixtures/diamonds.json\", spark)\n",
        "\n",
        "    actual_df = transform(df)\n",
        "\n",
        "    assert data_frame_to_json(actual_df) == expected_json(\"expected_correct.json\")\n",
        "\n",
        "\n",
        "def test_null_price_is_replaced_based_on_cut_clarity_and_color(spark: SparkSession):\n",
        "    json_text = '''\n",
        "      [\n",
        "  {\n",
        "    \"id\": \"DI-26-null-price\",\n",
        "    \"carat\": 0.21,\n",
        "    \"cut\": \"Good\",\n",
        "    \"color\": \"D\",\n",
        "    \"clarity\": \"VVS1\",\n",
        "    \"depth\": null,\n",
        "    \"table\": null,\n",
        "    \"price\": null,\n",
        "    \"x\": null,\n",
        "    \"y\": null,\n",
        "    \"z\": null\n",
        "  },\n",
        "  {\n",
        "    \"id\": \"DI-26\",\n",
        "    \"carat\": 0.21,\n",
        "    \"cut\": \"Good\",\n",
        "    \"color\": \"D\",\n",
        "    \"clarity\": \"VVS1\",\n",
        "    \"depth\": null,\n",
        "    \"table\": null,\n",
        "    \"price\": 3332,\n",
        "    \"x\": null,\n",
        "    \"y\": null,\n",
        "    \"z\": null\n",
        "  },\n",
        "  {\n",
        "    \"id\": \"DI-26\",\n",
        "    \"carat\": 0.21,\n",
        "    \"cut\": \"Good\",\n",
        "    \"color\": \"D\",\n",
        "    \"clarity\": \"VVS1\",\n",
        "    \"depth\": null,\n",
        "    \"table\": null,\n",
        "    \"price\": 3334,\n",
        "    \"x\": null,\n",
        "    \"y\": null,\n",
        "    \"z\": null\n",
        "  },\n",
        "  {\n",
        "    \"id\": \"DI-28\",\n",
        "    \"carat\": 0.21,\n",
        "    \"cut\": \"Good\",\n",
        "    \"color\": \"G\",\n",
        "    \"clarity\": \"VVS1\",\n",
        "    \"depth\": null,\n",
        "    \"table\": null,\n",
        "    \"price\": 1665,\n",
        "    \"x\": null,\n",
        "    \"y\": null,\n",
        "    \"z\": null\n",
        "  }\n",
        "]\n",
        "\n",
        "    '''\n",
        "\n",
        "    df = spark.read.option(\"multiline\", \"true\").json(spark.sparkContext.parallelize([json_text]))\n",
        "\n",
        "    df = df.withColumn('lprice', log('price'))\n",
        "    window = Window.partitionBy('cut', 'clarity').orderBy('price').rowsBetween(-3, 3)\n",
        "    moving_avg = mean(df['price']).over(window)\n",
        "    df = df.withColumn('moving_avg', moving_avg)\n",
        "\n",
        "    df = df.withColumn('price', when(df.price.isNull(), df.moving_avg).otherwise(df.price))\n",
        "    assert_diamond_has_expected_price(df)\n",
        "\n",
        "\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "id": "na7p6g77wmUh"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## STEP 5E - Squeeze the top\n",
        "\n",
        "- [ ] Run your test. It should fail with the `ColumnsNotEqualError`.\n",
        "- [ ] Comment out:\n",
        "\n",
        "```\n",
        "actual_df = df.withColumn('lprice', log('price'))\n",
        "```\n",
        "- [ ] Run your test. It fails with the `ColumnsNotEqualError`.\n",
        "- [ ] This means that line isn't important to the behaviour we are testing. Delete it.\n"
      ],
      "metadata": {
        "id": "GM1rYTV824Z-"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%%ipytest -qq\n",
        "from pyspark.sql import DataFrame, SparkSession\n",
        "from pyspark.sql.functions import lit, log\n",
        "from chispa import assert_column_equality\n",
        "\n",
        "from pathlib import Path\n",
        "\n",
        "############################ THE CODE WE ARE TESTING ###########################\n",
        "def transform(df: DataFrame) -> DataFrame:\n",
        "    df = df.withColumn('lprice', log('price'))\n",
        "    window = Window.partitionBy('cut', 'clarity').orderBy('price').rowsBetween(-3, 3)\n",
        "    moving_avg = mean(df['price']).over(window)\n",
        "    df = df.withColumn('moving_avg', moving_avg)\n",
        "\n",
        "    df = df.withColumn('price', when(df.price.isNull(), df.moving_avg).otherwise(df.price))\n",
        "    df = df[['id', 'carat', 'clarity', 'color', 'price']]\n",
        "    check_input(['clarity', 'color'], df, ['carat', 'clarity', 'color'])\n",
        "    indexer, encoder, independent_vars = setup_categorical_vars(\n",
        "        ['clarity', 'color'], 'index', 'vector', ['carat', 'clarity', 'color']\n",
        "    )\n",
        "    df = build_indep_vars(df, independent_vars, indexer, encoder, \"index\", \"vector\",\n",
        "                                      keep_intermediate=False,\n",
        "                                      summarizer=True)\n",
        "    return df\n",
        "################################################################################\n",
        "\n",
        "\n",
        "\n",
        "def assert_diamond_has_expected_price(actual_df: DataFrame) -> None:\n",
        "    actual_df=actual_df.filter(actual_df.id == \"DI-26-null-price\")\n",
        "    assert actual_df.count() == 1\n",
        "    actual_df=actual_df.withColumn('expected_price', lit(3333.0))\n",
        "    assert_column_equality(actual_df, 'price', 'expected_price')\n",
        "\n",
        "\n",
        "@pytest.mark.skip()\n",
        "def test_will_do_the_right_thing(spark: SparkSession):\n",
        "    df = create_df_from_json(\"tests/fixtures/diamonds.json\", spark)\n",
        "\n",
        "    actual_df = transform(df)\n",
        "\n",
        "    assert data_frame_to_json(actual_df) == expected_json(\"expected_correct.json\")\n",
        "\n",
        "\n",
        "def test_null_price_is_replaced_based_on_cut_clarity_and_color(spark: SparkSession):\n",
        "    json_text = '''\n",
        "      [\n",
        "  {\n",
        "    \"id\": \"DI-26-null-price\",\n",
        "    \"cut\": \"Good\",\n",
        "    \"color\": \"D\",\n",
        "    \"clarity\": \"VVS1\",\n",
        "    \"price\": null\n",
        "  },\n",
        "  {\n",
        "    \"id\": \"DI-26\",\n",
        "    \"cut\": \"Good\",\n",
        "    \"color\": \"D\",\n",
        "    \"clarity\": \"VVS1\",\n",
        "    \"price\": 3332\n",
        "  },\n",
        "  {\n",
        "    \"id\": \"DI-26\",\n",
        "    \"cut\": \"Good\",\n",
        "    \"color\": \"D\",\n",
        "    \"clarity\": \"VVS1\",\n",
        "    \"price\": 3334\n",
        "  },\n",
        "  {\n",
        "    \"id\": \"DI-28\",\n",
        "    \"cut\": \"Good\",\n",
        "    \"color\": \"G\",\n",
        "    \"clarity\": \"VVS1\",\n",
        "    \"price\": 1665\n",
        "  }\n",
        "]\n",
        "\n",
        "    '''\n",
        "\n",
        "    df = spark.read.option(\"multiline\", \"true\").json(spark.sparkContext.parallelize([json_text]))\n",
        "\n",
        "    df = df.withColumn('lprice', log('price')) # <- Comment out this line. Run your tests and check the error. If it is the right error, delete this line.\n",
        "    window = Window.partitionBy('cut', 'clarity').orderBy('price').rowsBetween(-3, 3)\n",
        "    moving_avg = mean(df['price']).over(window)\n",
        "    df = df.withColumn('moving_avg', moving_avg)\n",
        "\n",
        "    df = df.withColumn('price', when(df.price.isNull(), df.moving_avg).otherwise(df.price))\n",
        "    assert_diamond_has_expected_price(df)\n",
        "\n",
        "\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "id": "hh40g2AUyAT9"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## STEP 5F - Fix the bug\n",
        "\n",
        "- [ ] Run your test. It should fail with the `ColumnsNotEqualError`\n",
        "- [ ] Try commenting out the next line from the top.\n",
        "- [ ] Run your test. It should fail for a different reason. This means that line is important for the behaviour you are testing.\n",
        "- [ ] Read the line. Can you find the bug?\n",
        "- [ ] Add 'color' to the list of args passes into the window function.\n",
        "- [ ] Run your test. It should pass.\n"
      ],
      "metadata": {
        "id": "d_rny-zd06BQ"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%%ipytest -qq\n",
        "from pyspark.sql import DataFrame, SparkSession\n",
        "from pyspark.sql.functions import lit, log\n",
        "from chispa import assert_column_equality\n",
        "\n",
        "from pathlib import Path\n",
        "\n",
        "############################ THE CODE WE ARE TESTING ###########################\n",
        "def transform(df: DataFrame) -> DataFrame:\n",
        "    df = df.withColumn('lprice', log('price'))\n",
        "    window = Window.partitionBy('cut', 'clarity').orderBy('price').rowsBetween(-3, 3)\n",
        "    moving_avg = mean(df['price']).over(window)\n",
        "    df = df.withColumn('moving_avg', moving_avg)\n",
        "\n",
        "    df = df.withColumn('price', when(df.price.isNull(), df.moving_avg).otherwise(df.price))\n",
        "    df = df[['id', 'carat', 'clarity', 'color', 'price']]\n",
        "    check_input(['clarity', 'color'], df, ['carat', 'clarity', 'color'])\n",
        "    indexer, encoder, independent_vars = setup_categorical_vars(\n",
        "        ['clarity', 'color'], 'index', 'vector', ['carat', 'clarity', 'color']\n",
        "    )\n",
        "    df = build_indep_vars(df, independent_vars, indexer, encoder, \"index\", \"vector\",\n",
        "                                      keep_intermediate=False,\n",
        "                                      summarizer=True)\n",
        "    return df\n",
        "################################################################################\n",
        "\n",
        "\n",
        "def assert_diamond_has_expected_price(actual_df: DataFrame) -> None:\n",
        "    actual_df=actual_df.filter(actual_df.id == \"DI-26-null-price\")\n",
        "    assert actual_df.count() == 1\n",
        "    actual_df=actual_df.withColumn('expected_price', lit(3333.0))\n",
        "    assert_column_equality(actual_df, 'price', 'expected_price')\n",
        "\n",
        "\n",
        "@pytest.mark.skip()\n",
        "def test_will_do_the_right_thing(spark: SparkSession):\n",
        "    df = create_df_from_json(\"tests/fixtures/diamonds.json\", spark)\n",
        "\n",
        "    actual_df = transform(df)\n",
        "\n",
        "    assert data_frame_to_json(actual_df) == expected_json(\"expected_correct.json\")\n",
        "\n",
        "\n",
        "def test_null_price_is_replaced_based_on_cut_clarity_and_color(spark: SparkSession):\n",
        "    json_text = '''\n",
        "      [\n",
        "  {\n",
        "    \"id\": \"DI-26-null-price\",\n",
        "    \"cut\": \"Good\",\n",
        "    \"color\": \"D\",\n",
        "    \"clarity\": \"VVS1\",\n",
        "    \"price\": null\n",
        "  },\n",
        "  {\n",
        "    \"id\": \"DI-26\",\n",
        "    \"cut\": \"Good\",\n",
        "    \"color\": \"D\",\n",
        "    \"clarity\": \"VVS1\",\n",
        "    \"price\": 3332\n",
        "  },\n",
        "  {\n",
        "    \"id\": \"DI-26\",\n",
        "    \"cut\": \"Good\",\n",
        "    \"color\": \"D\",\n",
        "    \"clarity\": \"VVS1\",\n",
        "    \"price\": 3334\n",
        "  },\n",
        "  {\n",
        "    \"id\": \"DI-28\",\n",
        "    \"cut\": \"Good\",\n",
        "    \"color\": \"G\",\n",
        "    \"clarity\": \"VVS1\",\n",
        "    \"price\": 1665\n",
        "  }\n",
        "]\n",
        "\n",
        "    '''\n",
        "\n",
        "    df = spark.read.option(\"multiline\", \"true\").json(spark.sparkContext.parallelize([json_text]))\n",
        "\n",
        "    window = Window.partitionBy('cut', 'clarity').orderBy('price').rowsBetween(-3, 3)\n",
        "    moving_avg = mean(df['price']).over(window)\n",
        "    df = df.withColumn('moving_avg', moving_avg)\n",
        "\n",
        "    df = df.withColumn('price', when(df.price.isNull(), df.moving_avg).otherwise(df.price))\n",
        "    assert_diamond_has_expected_price(df)\n",
        "\n",
        "\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "id": "VXQ6fWtx0r6c"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## STEP 5G - Run all the tests.\n",
        "\n",
        "- [ ] Run your test. It should pass. The old test should still be skipped\n",
        "- [ ] Delete `pytest.mark.skip()` from the top of the old test.\n",
        "- [ ] Run your tests. Both should pass.\n"
      ],
      "metadata": {
        "id": "5Nn-MzIJTRGw"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%%ipytest -qq\n",
        "from pyspark.sql import DataFrame, SparkSession\n",
        "from pyspark.sql.functions import lit, log\n",
        "from chispa import assert_column_equality\n",
        "\n",
        "from pathlib import Path\n",
        "\n",
        "############################ THE CODE WE ARE TESTING ###########################\n",
        "def calculate_avg_price_for_similar_diamonds(df: DataFrame) -> DataFrame:\n",
        "    window = Window.partitionBy('cut', 'clarity', 'color').orderBy('price').rowsBetween(-3, 3)\n",
        "    moving_avg = mean(df['price']).over(window)\n",
        "    df = df.withColumn('moving_avg', moving_avg)\n",
        "    df = df.withColumn('price', when(df.price.isNull(), df.moving_avg).otherwise(df.price))\n",
        "\n",
        "    return df\n",
        "\n",
        "def transform(df: DataFrame) -> DataFrame:\n",
        "    df = df.withColumn('lprice', log('price'))\n",
        "    df = calculate_avg_price_for_similar_diamonds(df)\n",
        "\n",
        "    df = df[['id', 'carat', 'clarity', 'color', 'price']]\n",
        "    check_input(['clarity', 'color'], df, ['carat', 'clarity', 'color'])\n",
        "    indexer, encoder, independent_vars = setup_categorical_vars(\n",
        "        ['clarity', 'color'], 'index', 'vector', ['carat', 'clarity', 'color']\n",
        "    )\n",
        "    df = build_indep_vars(df, independent_vars, indexer, encoder, \"index\", \"vector\",\n",
        "                                      keep_intermediate=False,\n",
        "                                      summarizer=True)\n",
        "    return df\n",
        "################################################################################\n",
        "\n",
        "\n",
        "def assert_diamond_has_expected_price(actual_df: DataFrame) -> None:\n",
        "    actual_df=actual_df.filter(actual_df.id == \"DI-26-null-price\")\n",
        "    assert actual_df.count() == 1\n",
        "    actual_df=actual_df.withColumn('expected_price', lit(3333.0))\n",
        "    assert_column_equality(actual_df, 'price', 'expected_price')\n",
        "\n",
        "\n",
        "@pytest.mark.skip() # <- DELETE THIS LINE\n",
        "def test_will_do_the_right_thing(spark: SparkSession):\n",
        "    df = create_df_from_json(\"tests/fixtures/diamonds.json\", spark)\n",
        "\n",
        "    actual_df = transform(df)\n",
        "\n",
        "    assert data_frame_to_json(actual_df) == expected_json(\"expected_correct.json\")\n",
        "\n",
        "\n",
        "def test_null_price_is_replaced_based_on_cut_clarity_and_color(spark: SparkSession):\n",
        "    json_text = '''\n",
        "      [\n",
        "  {\n",
        "    \"id\": \"DI-26-null-price\",\n",
        "    \"cut\": \"Good\",\n",
        "    \"color\": \"D\",\n",
        "    \"clarity\": \"VVS1\",\n",
        "    \"price\": null\n",
        "  },\n",
        "  {\n",
        "    \"id\": \"DI-26\",\n",
        "    \"cut\": \"Good\",\n",
        "    \"color\": \"D\",\n",
        "    \"clarity\": \"VVS1\",\n",
        "    \"price\": 3332\n",
        "  },\n",
        "  {\n",
        "    \"id\": \"DI-26\",\n",
        "    \"cut\": \"Good\",\n",
        "    \"color\": \"D\",\n",
        "    \"clarity\": \"VVS1\",\n",
        "    \"price\": 3334\n",
        "  },\n",
        "  {\n",
        "    \"id\": \"DI-28\",\n",
        "    \"cut\": \"Good\",\n",
        "    \"color\": \"G\",\n",
        "    \"clarity\": \"VVS1\",\n",
        "    \"price\": 1665\n",
        "  }\n",
        "]\n",
        "\n",
        "    '''\n",
        "\n",
        "    df = spark.read.option(\"multiline\", \"true\").json(spark.sparkContext.parallelize([json_text]))\n",
        "\n",
        "\n",
        "    window = Window.partitionBy('cut', 'clarity', 'color').orderBy('price').rowsBetween(-3, 3)\n",
        "    moving_avg = mean(df['price']).over(window)\n",
        "    df = df.withColumn('moving_avg', moving_avg)\n",
        "    df = df.withColumn('price', when(df.price.isNull(), df.moving_avg).otherwise(df.price))\n",
        "\n",
        "\n",
        "    assert_diamond_has_expected_price(df)\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "id": "EjIwGavaTuCe"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## STEP 5H - Clean up - extract a method\n",
        "\n",
        "The four lines in the test work together to give us the behaviour we care about. Let's extract them into a method.\n",
        "- [ ] Run the test. It should pass.\n",
        "- [ ] Copy the four lines into the body of the `calculate_avg_price_for_similar_diamonds` method.\n",
        "- [ ] Delete them from your test and call that method instead.\n",
        "- [ ] Run the test. It should pass."
      ],
      "metadata": {
        "id": "tyJn6Ksu2KZk"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%%ipytest -qq\n",
        "from pyspark.sql import DataFrame, SparkSession\n",
        "from pyspark.sql.functions import lit, log\n",
        "from chispa import assert_column_equality\n",
        "\n",
        "from pathlib import Path\n",
        "\n",
        "############################ THE CODE WE ARE TESTING ###########################\n",
        "def calculate_avg_price_for_similar_diamonds(df: DataFrame) -> DataFrame:\n",
        "    #PASTE THE LINES YOU COPIED FROM BELOW\n",
        "    return df\n",
        "\n",
        "\n",
        "def transform(df: DataFrame) -> DataFrame:\n",
        "    df = df.withColumn('lprice', log('price'))\n",
        "    window = Window.partitionBy('cut', 'clarity').orderBy('price').rowsBetween(-3, 3)\n",
        "    moving_avg = mean(df['price']).over(window)\n",
        "    df = df.withColumn('moving_avg', moving_avg)\n",
        "\n",
        "    df = df.withColumn('price', when(df.price.isNull(), df.moving_avg).otherwise(df.price))\n",
        "    df = df[['id', 'carat', 'clarity', 'color', 'price']]\n",
        "    check_input(['clarity', 'color'], df, ['carat', 'clarity', 'color'])\n",
        "    indexer, encoder, independent_vars = setup_categorical_vars(\n",
        "        ['clarity', 'color'], 'index', 'vector', ['carat', 'clarity', 'color']\n",
        "    )\n",
        "    df = build_indep_vars(df, independent_vars, indexer, encoder, \"index\", \"vector\",\n",
        "                                      keep_intermediate=False,\n",
        "                                      summarizer=True)\n",
        "    return df\n",
        "################################################################################\n",
        "\n",
        "def assert_diamond_has_expected_price(actual_df: DataFrame) -> None:\n",
        "    actual_df=actual_df.filter(actual_df.id == \"DI-26-null-price\")\n",
        "    assert actual_df.count() == 1\n",
        "    actual_df=actual_df.withColumn('expected_price', lit(3333.0))\n",
        "    assert_column_equality(actual_df, 'price', 'expected_price')\n",
        "\n",
        "\n",
        "def test_will_do_the_right_thing(spark: SparkSession):\n",
        "    df = create_df_from_json(\"tests/fixtures/diamonds.json\", spark)\n",
        "\n",
        "    actual_df = transform(df)\n",
        "\n",
        "    assert data_frame_to_json(actual_df) == expected_json(\"expected_correct.json\")\n",
        "\n",
        "\n",
        "def test_null_price_is_replaced_based_on_cut_clarity_and_color(spark: SparkSession):\n",
        "    json_text = '''\n",
        "      [\n",
        "  {\n",
        "    \"id\": \"DI-26-null-price\",\n",
        "    \"cut\": \"Good\",\n",
        "    \"color\": \"D\",\n",
        "    \"clarity\": \"VVS1\",\n",
        "    \"price\": null\n",
        "  },\n",
        "  {\n",
        "    \"id\": \"DI-26\",\n",
        "    \"cut\": \"Good\",\n",
        "    \"color\": \"D\",\n",
        "    \"clarity\": \"VVS1\",\n",
        "    \"price\": 3332\n",
        "  },\n",
        "  {\n",
        "    \"id\": \"DI-26\",\n",
        "    \"cut\": \"Good\",\n",
        "    \"color\": \"D\",\n",
        "    \"clarity\": \"VVS1\",\n",
        "    \"price\": 3334\n",
        "  },\n",
        "  {\n",
        "    \"id\": \"DI-28\",\n",
        "    \"cut\": \"Good\",\n",
        "    \"color\": \"G\",\n",
        "    \"clarity\": \"VVS1\",\n",
        "    \"price\": 1665\n",
        "  }\n",
        "]\n",
        "\n",
        "    '''\n",
        "\n",
        "    df = spark.read.option(\"multiline\", \"true\").json(spark.sparkContext.parallelize([json_text]))\n",
        "\n",
        "    ############################## CUT THESE LINES #############################\n",
        "    window = Window.partitionBy('cut', 'clarity', 'color').orderBy('price').rowsBetween(-3, 3)\n",
        "    moving_avg = mean(df['price']).over(window)\n",
        "    df = df.withColumn('moving_avg', moving_avg)\n",
        "    df = df.withColumn('price', when(df.price.isNull(), df.moving_avg).otherwise(df.price))\n",
        "    ############################################################################\n",
        "    # df = calculate_avg_price_for_similar_diamonds(df) # <- UNCOMMENT THIS LINE\n",
        "\n",
        "    assert_diamond_has_expected_price(df)\n",
        "\n",
        "\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "id": "lL00FWIS3RHy"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## STEP 5I - Use the new method.\n",
        "\n",
        "- [ ] Run your test. It should pass.\n",
        "- [ ] Replace the same lines in the actual code with a call to the new method.\n",
        "- [ ] Run your tests. Both should pass.\n"
      ],
      "metadata": {
        "id": "AzHjbKAw4QmB"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%%ipytest -qq\n",
        "from pyspark.sql import DataFrame, SparkSession\n",
        "from pyspark.sql.functions import lit, log\n",
        "from chispa import assert_column_equality\n",
        "\n",
        "from pathlib import Path\n",
        "\n",
        "############################ THE CODE WE ARE TESTING ###########################\n",
        "def calculate_avg_price_for_similar_diamonds(df: DataFrame) -> DataFrame:\n",
        "    window = Window.partitionBy('cut', 'clarity', 'color').orderBy('price').rowsBetween(-3, 3)\n",
        "    moving_avg = mean(df['price']).over(window)\n",
        "    df = df.withColumn('moving_avg', moving_avg)\n",
        "    df = df.withColumn('price', when(df.price.isNull(), df.moving_avg).otherwise(df.price))\n",
        "\n",
        "    return df\n",
        "\n",
        "def transform(df: DataFrame) -> DataFrame:\n",
        "    df = df.withColumn('lprice', log('price'))\n",
        "    ############################ DELETE THESE LINES ############################\n",
        "    window = Window.partitionBy('cut', 'clarity').orderBy('price').rowsBetween(-3, 3)\n",
        "    moving_avg = mean(df['price']).over(window)\n",
        "    df = df.withColumn('moving_avg', moving_avg)\n",
        "    df = df.withColumn('price', when(df.price.isNull(), df.moving_avg).otherwise(df.price))\n",
        "    ############################################################################\n",
        "    # df = calculate_avg_price_for_similar_diamonds(df) # <- UNCOMMENT THIS LINE\n",
        "\n",
        "    df = df[['id', 'carat', 'clarity', 'color', 'price']]\n",
        "    check_input(['clarity', 'color'], df, ['carat', 'clarity', 'color'])\n",
        "    indexer, encoder, independent_vars = setup_categorical_vars(\n",
        "        ['clarity', 'color'], 'index', 'vector', ['carat', 'clarity', 'color']\n",
        "    )\n",
        "    df = build_indep_vars(df, independent_vars, indexer, encoder, \"index\", \"vector\",\n",
        "                                      keep_intermediate=False,\n",
        "                                      summarizer=True)\n",
        "    return df\n",
        "################################################################################\n",
        "\n",
        "\n",
        "def assert_diamond_has_expected_price(actual_df: DataFrame) -> None:\n",
        "    actual_df=actual_df.filter(actual_df.id == \"DI-26-null-price\")\n",
        "    assert actual_df.count() == 1\n",
        "    actual_df=actual_df.withColumn('expected_price', lit(3333.0))\n",
        "    assert_column_equality(actual_df, 'price', 'expected_price')\n",
        "\n",
        "\n",
        "def test_will_do_the_right_thing(spark: SparkSession):\n",
        "    df = create_df_from_json(\"tests/fixtures/diamonds.json\", spark)\n",
        "\n",
        "    actual_df = transform(df)\n",
        "\n",
        "    assert data_frame_to_json(actual_df) == expected_json(\"expected_correct.json\")\n",
        "\n",
        "\n",
        "def test_null_price_is_replaced_based_on_cut_clarity_and_color(spark: SparkSession):\n",
        "    json_text = '''\n",
        "      [\n",
        "  {\n",
        "    \"id\": \"DI-26-null-price\",\n",
        "    \"cut\": \"Good\",\n",
        "    \"color\": \"D\",\n",
        "    \"clarity\": \"VVS1\",\n",
        "    \"price\": null\n",
        "  },\n",
        "  {\n",
        "    \"id\": \"DI-26\",\n",
        "    \"cut\": \"Good\",\n",
        "    \"color\": \"D\",\n",
        "    \"clarity\": \"VVS1\",\n",
        "    \"price\": 3332\n",
        "  },\n",
        "  {\n",
        "    \"id\": \"DI-26\",\n",
        "    \"cut\": \"Good\",\n",
        "    \"color\": \"D\",\n",
        "    \"clarity\": \"VVS1\",\n",
        "    \"price\": 3334\n",
        "  },\n",
        "  {\n",
        "    \"id\": \"DI-28\",\n",
        "    \"cut\": \"Good\",\n",
        "    \"color\": \"G\",\n",
        "    \"clarity\": \"VVS1\",\n",
        "    \"price\": 1665\n",
        "  }\n",
        "]\n",
        "\n",
        "    '''\n",
        "\n",
        "    df = spark.read.option(\"multiline\", \"true\").json(spark.sparkContext.parallelize([json_text]))\n",
        "\n",
        "    df = calculate_avg_price_for_similar_diamonds(df)\n",
        "\n",
        "    assert_diamond_has_expected_price(df)\n",
        "\n",
        "\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "id": "r4NANzuF5Tvm"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Step 6: Work with clean tests!"
      ],
      "metadata": {
        "id": "u73U7To0xBvK"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## STEP 6A - Make tests correct for new behaviour\n",
        "\n",
        "Having only two other diamonds with the same cut, clarity and color is not enough to accurately predict the price of a third based on mean alone. We have a new requirement - we should only make that prediction if we have 3 matching diamonds or more. Otherwise we should drop the row\n",
        "\n",
        "Let's update our test case to be correct for the new behaviour.\n",
        "\n",
        "- [ ] Run your tests, they should pass.\n",
        "- [ ] Add 3 more matching diamonds with a price to the dataframe.\n",
        "- [ ] Run your tests. They should pass."
      ],
      "metadata": {
        "id": "wjYANsWB5gnF"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%%ipytest -qq\n",
        "from pyspark.sql import DataFrame\n",
        "from pyspark.sql.functions import lit\n",
        "from chispa import assert_column_equality\n",
        "from typing import Dict\n",
        "\n",
        "############################ THE CODE WE ARE TESTING ###########################\n",
        "def calculate_avg_price_for_similar_diamonds(df: DataFrame) -> DataFrame:\n",
        "    window = Window.partitionBy('cut', 'clarity', 'color').orderBy('price').rowsBetween(-3, 3)\n",
        "    moving_avg = mean(df['price']).over(window)\n",
        "    df = df.withColumn('moving_avg', moving_avg)\n",
        "    df = df.withColumn('price', when(df.price.isNull(), df.moving_avg).otherwise(df.price))\n",
        "\n",
        "    return df\n",
        "\n",
        "def transform(df: DataFrame) -> DataFrame:\n",
        "    df = df.withColumn('lprice', log('price'))\n",
        "    df = calculate_avg_price_for_similar_diamonds(df)\n",
        "\n",
        "    df = df[['id', 'carat', 'clarity', 'color', 'price']]\n",
        "    check_input(['clarity', 'color'], df, ['carat', 'clarity', 'color'])\n",
        "    indexer, encoder, independent_vars = setup_categorical_vars(\n",
        "        ['clarity', 'color'], 'index', 'vector', ['carat', 'clarity', 'color']\n",
        "    )\n",
        "    df = build_indep_vars(df, independent_vars, indexer, encoder, \"index\", \"vector\",\n",
        "                                      keep_intermediate=False,\n",
        "                                      summarizer=True)\n",
        "    return df\n",
        "################################################################################\n",
        "\n",
        "def unpriced_diamond() -> Dict:\n",
        "  return {\"id\": \"DI-26-null-price\", \"cut\": \"Good\", \"color\": \"D\", \"clarity\": \"VVS1\", \"price\": None}\n",
        "\n",
        "def diamond_with_price(price: float, different: Dict={}) -> Dict:\n",
        "  diamond = unpriced_diamond()\n",
        "  diamond[\"id\"] = \"with-price\"\n",
        "  diamond[\"price\"] = price\n",
        "\n",
        "  if different:\n",
        "    for key in different.keys():\n",
        "      diamond[\"id\"] = f\"{diamond['id']}-wrong-{key}\"\n",
        "\n",
        "  return diamond | different\n",
        "\n",
        "\n",
        "def assert_unpriced_diamond_has_expected_price(actual_df: DataFrame, price: float) -> None:\n",
        "    actual_df=actual_df.filter(actual_df.id == \"DI-26-null-price\")\n",
        "    assert actual_df.count() == 1\n",
        "    actual_df=actual_df.withColumn('expected_price', lit(price))\n",
        "    assert_column_equality(actual_df, 'price', 'expected_price')\n",
        "\n",
        "\n",
        "def test_will_do_the_right_thing(spark: SparkSession):\n",
        "    df = create_df_from_json(\"tests/fixtures/diamonds.json\", spark)\n",
        "\n",
        "    actual_df = transform(df)\n",
        "    actual_df.show()\n",
        "\n",
        "    assert data_frame_to_json(actual_df) == expected_json(\"expected_correct.json\")\n",
        "\n",
        "\n",
        "def test_null_price_is_replaced_based_on_cut_clarity_and_color(spark: SparkSession):\n",
        "    df = spark.createDataFrame([\n",
        "      unpriced_diamond(),\n",
        "\n",
        "      diamond_with_price(price=3333.0),\n",
        "      diamond_with_price(price=3333.0),\n",
        "      # <- Add 1 more diamond with a price here\n",
        "\n",
        "      diamond_with_price(price=1665.0, different={\"color\": \"G\"}),\n",
        "      diamond_with_price(price=1665.0, different={\"clarity\": \"S1\"}),\n",
        "      diamond_with_price(price=1665.0, different={\"cut\": \"Very Good\"})\n",
        "    ])\n",
        "\n",
        "    actual_df = calculate_avg_price_for_similar_diamonds(df)\n",
        "\n",
        "    assert_unpriced_diamond_has_expected_price(actual_df, price=3333.0)"
      ],
      "metadata": {
        "id": "kD57M41R3z0h"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## STEP 6B - Test for the new behaviour.\n",
        "\n",
        "We need to check that a diamond with no price, and less than 3 matching diamonds gets dropped from the dataframe.\n",
        "\n",
        "We haven't written the code yet, so the test will fail once it's written.\n",
        "\n",
        "- [ ] Run your tests. Two will pass, one will be skipped.\n",
        "- [ ] Remove the line to skip the test.\n",
        "- [ ] Construct the input dataset inside of the spark.createDataFrame function. (Use the helper methods!)\n",
        "- [ ] Select the 5 columns we care about\n",
        "- [ ] Construct the expected dataset using the helper methods\n",
        "- [ ] Uncomment the assertion at the end. In this case, we are looking for a whole row that is _missing_ so we will use `assert_df_equality` from chispa.\n",
        "- [ ] Run your tests. They will fail with a new error: $\\color{red}{\\text{chispa.dataframe_comparer.DataFramesNotEqualError}}$"
      ],
      "metadata": {
        "id": "WlZQd0QC4ufV"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%%ipytest -qq\n",
        "from pyspark.sql import DataFrame\n",
        "from pyspark.sql.functions import lit\n",
        "from chispa import assert_column_equality, assert_df_equality\n",
        "from typing import Dict\n",
        "\n",
        "############################ THE CODE WE ARE TESTING ###########################\n",
        "def calculate_avg_price_for_similar_diamonds(df: DataFrame) -> DataFrame:\n",
        "    window = Window.partitionBy('cut', 'clarity', 'color').orderBy('price').rowsBetween(-3, 3)\n",
        "    moving_avg = mean(df['price']).over(window)\n",
        "    df = df.withColumn('moving_avg', moving_avg)\n",
        "    df = df.withColumn('price', when(df.price.isNull(), df.moving_avg).otherwise(df.price))\n",
        "\n",
        "    return df\n",
        "\n",
        "def transform(df: DataFrame) -> DataFrame:\n",
        "    df = df.withColumn('lprice', log('price'))\n",
        "    df = calculate_avg_price_for_similar_diamonds(df)\n",
        "\n",
        "    df = df[['id', 'carat', 'clarity', 'color', 'price']]\n",
        "    check_input(['clarity', 'color'], df, ['carat', 'clarity', 'color'])\n",
        "    indexer, encoder, independent_vars = setup_categorical_vars(\n",
        "        ['clarity', 'color'], 'index', 'vector', ['carat', 'clarity', 'color']\n",
        "    )\n",
        "    df = build_indep_vars(df, independent_vars, indexer, encoder, \"index\", \"vector\",\n",
        "                                      keep_intermediate=False,\n",
        "                                      summarizer=True)\n",
        "    return df\n",
        "################################################################################\n",
        "\n",
        "\n",
        "def assert_unpriced_diamond_has_expected_price(actual_df: DataFrame, price: float) -> None:\n",
        "    actual_df=actual_df.filter(actual_df.id == \"DI-26-null-price\")\n",
        "    assert actual_df.count() == 1\n",
        "    actual_df=actual_df.withColumn('expected_price', lit(price))\n",
        "    assert_column_equality(actual_df, 'price', 'expected_price')\n",
        "\n",
        "\n",
        "def unpriced_diamond() -> Dict:\n",
        "  return {\"id\": \"DI-26-null-price\", \"cut\": \"Good\", \"color\": \"D\", \"clarity\": \"VVS1\", \"price\": None}\n",
        "\n",
        "def diamond_with_price(price: float, different: Dict={}) -> Dict:\n",
        "  diamond = unpriced_diamond()\n",
        "  diamond[\"id\"] = \"with-price\"\n",
        "  diamond[\"price\"] = price\n",
        "\n",
        "  if different:\n",
        "    for key in different.keys():\n",
        "      diamond[\"id\"] = f\"{diamond['id']}-wrong-{key}\"\n",
        "\n",
        "  return diamond | different\n",
        "\n",
        "\n",
        "def test_will_do_the_right_thing(spark: SparkSession):\n",
        "    df = create_df_from_json(\"tests/fixtures/diamonds.json\", spark)\n",
        "\n",
        "    actual_df = transform(df)\n",
        "    actual_df.show()\n",
        "\n",
        "    assert data_frame_to_json(actual_df) == expected_json(\"expected_correct.json\")\n",
        "\n",
        "\n",
        "def test_null_price_is_replaced_based_on_cut_clarity_and_color(spark: SparkSession):\n",
        "    df = spark.createDataFrame([\n",
        "      unpriced_diamond(),\n",
        "\n",
        "      diamond_with_price(price=3333.0),\n",
        "      diamond_with_price(price=3333.0),\n",
        "      diamond_with_price(price=3333.0),\n",
        "\n",
        "      diamond_with_price(price=1665.0, different={\"color\": \"G\"}),\n",
        "      diamond_with_price(price=1665.0, different={\"clarity\": \"S1\"}),\n",
        "      diamond_with_price(price=1665.0, different={\"cut\": \"Very Good\"})\n",
        "    ])\n",
        "\n",
        "    actual_df = calculate_avg_price_for_similar_diamonds(df)\n",
        "\n",
        "    assert_unpriced_diamond_has_expected_price(actual_df, price=3333.0)\n",
        "\n",
        "@pytest.mark.skip() # <- DELETE THIS LINE\n",
        "def test_null_price_is_dropped_with_too_few_examples(spark: SparkSession):\n",
        "    df = spark.createDataFrame([\n",
        "      unpriced_diamond(),\n",
        "\n",
        "      # <- Add 2 diamonds with a price of 3333.0 here\n",
        "\n",
        "      # We don't need these diamonds. Remove them\n",
        "      diamond_with_price(price=1665.0, different={\"color\": \"G\"}),\n",
        "      diamond_with_price(price=1665.0, different={\"clarity\": \"S1\"}),\n",
        "      diamond_with_price(price=1665.0, different={\"cut\": \"Very Good\"})\n",
        "    ])\n",
        "\n",
        "    actual_df = calculate_avg_price_for_similar_diamonds(df).select(\n",
        "        # WE ONLY CARE ABOUT 5 COLUMNS FOR THIS TEST. SELECT ['id', 'color', 'clarity', 'cut', 'price']\n",
        "    )\n",
        "\n",
        "    expected_df = spark.createDataFrame([\n",
        "        #OUR OUTPUT SHOULD BE TWO DIAMONDS WITH A PRICE OF 3333.0\n",
        "    ])\n",
        "\n",
        "    #UNCOMMENT THE ASSERTION\n",
        "    # assert_df_equality( actual_df, expected_df, ignore_column_order=True, ignore_row_order=True )"
      ],
      "metadata": {
        "id": "PSKoUzBY6BWH"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## STEP 6C - See all your tests pass with the new behaviour added\n",
        "\n",
        "We have added a function with the new behaviour.\n",
        "\n",
        "- [ ] Uncomment `drop_null_rows_with_too_few_examples` and it's invocation.\n",
        "- [ ] Run your tests. They should pass."
      ],
      "metadata": {
        "id": "2jP-dY0sA-yx"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%%ipytest -qq\n",
        "from pyspark.sql import DataFrame\n",
        "from pyspark.sql.functions import lit, count\n",
        "from chispa import assert_column_equality, assert_df_equality\n",
        "from typing import Dict\n",
        "\n",
        "############################ THE CODE WE ARE TESTING ###########################\n",
        "\n",
        "############################ UNCOMMENT THE LINES BELOW #########################\n",
        "# def drop_null_rows_with_too_few_examples(df: DataFrame) -> DataFrame:\n",
        "#   counts_of_cut_clarity_color_groups_df = df.groupby(\n",
        "#       'cut', 'clarity', 'color').agg(count('*').alias('num_matches'))\n",
        "#   df = df.join(counts_of_cut_clarity_color_groups_df, on=['cut','clarity', 'color'])\n",
        "#   df.show()\n",
        "#   df = df.filter((df.price.isNotNull()) | (3 < df.num_matches))\n",
        "#   return df\n",
        "\n",
        "def calculate_avg_price_for_similar_diamonds(df: DataFrame) -> DataFrame:\n",
        "    # df = drop_null_rows_with_too_few_examples(df) # <- Uncomment this line\n",
        "    window = Window.partitionBy('cut', 'clarity', 'color').orderBy('price').rowsBetween(-3, 3)\n",
        "    moving_avg = mean(df['price']).over(window)\n",
        "    df = df.withColumn('moving_avg', moving_avg)\n",
        "    df = df.withColumn('price', when(df.price.isNull(), df.moving_avg).otherwise(df.price))\n",
        "\n",
        "    return df\n",
        "\n",
        "def transform(df: DataFrame) -> DataFrame:\n",
        "    df = df.withColumn('lprice', log('price'))\n",
        "    df = calculate_avg_price_for_similar_diamonds(df)\n",
        "\n",
        "    df = df[['id', 'carat', 'clarity', 'color', 'price']]\n",
        "    check_input(['clarity', 'color'], df, ['carat', 'clarity', 'color'])\n",
        "    indexer, encoder, independent_vars = setup_categorical_vars(\n",
        "        ['clarity', 'color'], 'index', 'vector', ['carat', 'clarity', 'color']\n",
        "    )\n",
        "    df = build_indep_vars(df, independent_vars, indexer, encoder, \"index\", \"vector\",\n",
        "                                      keep_intermediate=False,\n",
        "                                      summarizer=True)\n",
        "    return df\n",
        "################################################################################\n",
        "\n",
        "def unpriced_diamond() -> Dict:\n",
        "  return {\"id\": \"DI-26-null-price\", \"cut\": \"Good\", \"color\": \"D\", \"clarity\": \"VVS1\", \"price\": None}\n",
        "\n",
        "def diamond_with_price(price: float, different: Dict={}) -> Dict:\n",
        "  diamond = unpriced_diamond()\n",
        "  diamond[\"id\"] = \"with-price\"\n",
        "  diamond[\"price\"] = price\n",
        "\n",
        "  if different:\n",
        "    for key in different.keys():\n",
        "      diamond[\"id\"] = f\"{diamond['id']}-wrong-{key}\"\n",
        "\n",
        "  return diamond | different\n",
        "\n",
        "def assert_unpriced_diamond_has_expected_price(actual_df: DataFrame, price: float) -> None:\n",
        "    actual_df=actual_df.filter(actual_df.id == \"DI-26-null-price\")\n",
        "    assert actual_df.count() == 1\n",
        "    actual_df=actual_df.withColumn('expected_price', lit(price))\n",
        "    assert_column_equality(actual_df, 'price', 'expected_price')\n",
        "\n",
        "\n",
        "def test_will_do_the_right_thing(spark: SparkSession):\n",
        "    df = create_df_from_json(\"tests/fixtures/diamonds.json\", spark)\n",
        "\n",
        "    actual_df = transform(df)\n",
        "    actual_df.show()\n",
        "\n",
        "    assert data_frame_to_json(actual_df) == expected_json(\"expected_correct_with_filter.json\")\n",
        "\n",
        "\n",
        "def test_null_price_is_replaced_based_on_cut_clarity_and_color(spark: SparkSession):\n",
        "    df = spark.createDataFrame([\n",
        "      unpriced_diamond(),\n",
        "\n",
        "      diamond_with_price(price=3333.0),\n",
        "      diamond_with_price(price=3333.0),\n",
        "      diamond_with_price(price=3333.0),\n",
        "\n",
        "      diamond_with_price(price=1665.0, different={\"color\": \"G\"}),\n",
        "      diamond_with_price(price=1665.0, different={\"clarity\": \"S1\"}),\n",
        "      diamond_with_price(price=1665.0, different={\"cut\": \"Very Good\"})\n",
        "    ])\n",
        "\n",
        "    actual_df = calculate_avg_price_for_similar_diamonds(df)\n",
        "\n",
        "    assert_unpriced_diamond_has_expected_price(actual_df, price=3333.0)\n",
        "\n",
        "\n",
        "def test_null_price_is_dropped_with_too_few_examples(spark: SparkSession):\n",
        "    df = spark.createDataFrame([\n",
        "      unpriced_diamond(),\n",
        "      diamond_with_price(price=3333.0),\n",
        "      diamond_with_price(price=3333.0)\n",
        "    ])\n",
        "\n",
        "    actual_df = calculate_avg_price_for_similar_diamonds(df).select(\n",
        "        ['id', 'color', 'clarity', 'cut', 'price']\n",
        "    )\n",
        "\n",
        "    expected_df = spark.createDataFrame([\n",
        "        diamond_with_price(price=3333.0),\n",
        "        diamond_with_price(price=3333.0)\n",
        "    ])\n",
        "\n",
        "    assert_df_equality(actual_df, expected_df, ignore_column_order=True, ignore_row_order=True)"
      ],
      "metadata": {
        "id": "fngB1hMdBhus"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Extras"
      ],
      "metadata": {
        "id": "5LGdjV0ek442"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Instructions - Specific Assert with Pandas\n",
        "- [ ] import pandas:  \n",
        "`import pandas as pd`\n",
        "- [ ] Filter the dataframe for the unique id of the diamond we care about:  \n",
        "  `actual_df=actual_df.filter(actual_df.id == 'DI-26-null-price')`\n",
        "- [ ] Create your expected dataframe using Pandas:  \n",
        " `expected = pd.DataFrame(({'id': [\"DI-26-null-price\"], 'price':[3333.0] }))`\n",
        "- [ ] Select the column you care about:  \n",
        "  `actual_df=actual_df.select(['id', 'price'])\n",
        "- [ ] Assert for dataframe equality using pandas:  \n",
        "  `pd.testing.assert_frame_equal(actual_df, expected)`"
      ],
      "metadata": {
        "id": "wW0OfrWo8Qn4"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Instructions - Specific Assert with JSON properties\n",
        "\n",
        "- [ ] Filter the dataframe for the unique id of the diamond we care about:  \n",
        "  `actual_df=actual_df.filter(actual_df.id == 'DI-26-null-price')`\n",
        "- [ ] Convert your dataframe to JSON:  \n",
        "`actual_df_json = data_frame_to_json(actual_df)`\n",
        "- [ ] Assert the price property of the first object matches your expected price:  \n",
        "`assert actual_df_json[0]['price'] == 3333.0`"
      ],
      "metadata": {
        "id": "lD-3MVVv9D9d"
      }
    }
  ]
}