DotProcessing.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "authorship_tag": "ABX9TyNpRDL5XStFvgM0miAEZ5iO",
      "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/Amarantine-xiv/Amas-FF14-Combat-Sim/blob/main/DotProcessing.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 23,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "cellView": "form",
        "id": "SLqAVI_X--T3",
        "outputId": "2fa9ba6f-f369-4414-8a24-a3074aa27646"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Requirement already satisfied: coreapi==2.3.3 in /usr/local/lib/python3.10/dist-packages (2.3.3)\n",
            "Requirement already satisfied: python-graphql-client==0.4.3 in /usr/local/lib/python3.10/dist-packages (0.4.3)\n",
            "Requirement already satisfied: coreschema in /usr/local/lib/python3.10/dist-packages (from coreapi==2.3.3) (0.0.4)\n",
            "Requirement already satisfied: requests in /usr/local/lib/python3.10/dist-packages (from coreapi==2.3.3) (2.31.0)\n",
            "Requirement already satisfied: itypes in /usr/local/lib/python3.10/dist-packages (from coreapi==2.3.3) (1.2.0)\n",
            "Requirement already satisfied: uritemplate in /usr/local/lib/python3.10/dist-packages (from coreapi==2.3.3) (4.1.1)\n",
            "Requirement already satisfied: aiohttp~=3.0 in /usr/local/lib/python3.10/dist-packages (from python-graphql-client==0.4.3) (3.9.3)\n",
            "Requirement already satisfied: websockets>=5.0 in /usr/local/lib/python3.10/dist-packages (from python-graphql-client==0.4.3) (12.0)\n",
            "Requirement already satisfied: aiosignal>=1.1.2 in /usr/local/lib/python3.10/dist-packages (from aiohttp~=3.0->python-graphql-client==0.4.3) (1.3.1)\n",
            "Requirement already satisfied: attrs>=17.3.0 in /usr/local/lib/python3.10/dist-packages (from aiohttp~=3.0->python-graphql-client==0.4.3) (23.2.0)\n",
            "Requirement already satisfied: frozenlist>=1.1.1 in /usr/local/lib/python3.10/dist-packages (from aiohttp~=3.0->python-graphql-client==0.4.3) (1.4.1)\n",
            "Requirement already satisfied: multidict<7.0,>=4.5 in /usr/local/lib/python3.10/dist-packages (from aiohttp~=3.0->python-graphql-client==0.4.3) (6.0.5)\n",
            "Requirement already satisfied: yarl<2.0,>=1.0 in /usr/local/lib/python3.10/dist-packages (from aiohttp~=3.0->python-graphql-client==0.4.3) (1.9.4)\n",
            "Requirement already satisfied: async-timeout<5.0,>=4.0 in /usr/local/lib/python3.10/dist-packages (from aiohttp~=3.0->python-graphql-client==0.4.3) (4.0.3)\n",
            "Requirement already satisfied: charset-normalizer<4,>=2 in /usr/local/lib/python3.10/dist-packages (from requests->coreapi==2.3.3) (3.3.2)\n",
            "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.10/dist-packages (from requests->coreapi==2.3.3) (3.6)\n",
            "Requirement already satisfied: urllib3<3,>=1.21.1 in /usr/local/lib/python3.10/dist-packages (from requests->coreapi==2.3.3) (2.0.7)\n",
            "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.10/dist-packages (from requests->coreapi==2.3.3) (2024.2.2)\n",
            "Requirement already satisfied: jinja2 in /usr/local/lib/python3.10/dist-packages (from coreschema->coreapi==2.3.3) (3.1.3)\n",
            "Requirement already satisfied: MarkupSafe>=2.0 in /usr/local/lib/python3.10/dist-packages (from jinja2->coreschema->coreapi==2.3.3) (2.1.5)\n",
            "/content\n",
            "fatal: destination path 'Amas-FF14-Combat-Sim' already exists and is not an empty directory.\n",
            "/content/Amas-FF14-Combat-Sim\n",
            "remote: Enumerating objects: 5, done.\u001b[K\n",
            "remote: Counting objects: 100% (5/5), done.\u001b[K\n",
            "remote: Compressing objects: 100% (3/3), done.\u001b[K\n",
            "remote: Total 3 (delta 1), reused 0 (delta 0), pack-reused 0\u001b[K\n",
            "Unpacking objects: 100% (3/3), 112.91 KiB | 869.00 KiB/s, done.\n",
            "From https://github.com/Amarantine-xiv/Amas-FF14-Combat-Sim\n",
            "   b797769..929056f  main       -> origin/main\n",
            "Updating b797769..929056f\n",
            "Fast-forward\n",
            " DotProcessing.ipynb | 419 \u001b[32m+++++++++++++++++++++++++++++++++++++++++++++++++\u001b[m\u001b[31m-----------------------\u001b[m\n",
            " 1 file changed, 289 insertions(+), 130 deletions(-)\n",
            "Time taken: 2.6476292610168457\n",
            "Results: \n",
            "Average DPS: 14469.18\n",
            "Percentile 25: 14222.94\n",
            "Percentile 50: 14456.83\n",
            "Percentile 75: 14701.40\n",
            "Percentile 95: 15068.03\n",
            "Percentile 99: 15326.63\n",
            "\n",
            "Average Damage: 506421.46\n",
            "Percentile 25: 497803.00\n",
            "Percentile 50: 505989.00\n",
            "Percentile 75: 514549.00\n",
            "Percentile 95: 527381.00\n",
            "Percentile 99: 536432.01\n",
            "\n",
            "---Expected max damage over N runs---\n",
            "1 Runs: 14471.06\n",
            "5 Runs: 14886.88\n",
            "10 Runs: 15029.05\n",
            "20 Runs: 15155.01\n",
            "50 Runs: 15297.37\n",
            "100 Runs: 15398.05\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "---Times (in ms) when skills get used:---\n",
            "       0:               Tomahawk\n",
            "     650:              Infuriate\n",
            "     800:          Arcane Circle\n",
            "    2500:            Heavy Swing\n",
            "    5000:                   Maim\n",
            "    5650:       Grade 8 Tincture\n",
            "    6280:               Embolden\n",
            "    6300:        Chain Stratagem\n",
            "    7100:          Battle Litany\n",
            "    7500:            Storm's Eye\n",
            "    8150:          Inner Release\n",
            "   10000:            Inner Chaos\n",
            "   10650:               Upheaval\n",
            "   11300:              Onslaught\n",
            "   12500:            Primal Rend\n",
            "   15000:            Inner Chaos\n",
            "   15650:              Onslaught\n",
            "   17500:            Fell Cleave\n",
            "   18150:              Onslaught\n",
            "   20000:            Fell Cleave\n",
            "   22500:            Fell Cleave\n",
            "   25000:            Heavy Swing\n",
            "   27500:                   Maim\n",
            "   30000:           Storm's Path\n",
            "   32500:            Fell Cleave\n",
            "   35000:            Inner Chaos\n",
            "\n",
            "\n",
            "---Times (in ms) when damage lands:---\n",
            "time:      0.0, name:                   Auto, expected_damage:   3239.26, potency:   90, skill modifier:                       , damage variance: 838.98\n",
            "time:      0.0, name:               Tomahawk, expected_damage:   4838.96, potency:  150, skill modifier:                       , damage variance: 1253.33\n",
            "time:   2500.0, name:            Heavy Swing, expected_damage:   6645.47, potency:  200, skill modifier:                       , damage variance: 1718.94\n",
            "time:   3360.0, name:                   Auto, expected_damage:   3335.60, potency:   90, skill modifier:                       , damage variance: 862.79\n",
            "time:   5000.0, name:                   Maim, expected_damage:   9974.93, potency:  300, skill modifier:                       , damage variance: 2581.71\n",
            "time:   6720.0, name:                   Auto, expected_damage:   3611.00, potency:   90, skill modifier:                       , damage variance: 934.97\n",
            "time:   8200.0, name:           _Storm's Eye, expected_damage:  18399.12, potency:  440, skill modifier:        Surging Tempest, damage variance: 4776.60\n",
            "time:  10000.0, name:            Inner Chaos, expected_damage:  53244.66, potency:  660, skill modifier:        Surging Tempest, damage variance: 1539.19\n",
            "time:  10080.0, name:                   Auto, expected_damage:   4619.03, potency:   90, skill modifier:        Surging Tempest, damage variance: 1199.90\n",
            "time:  10650.0, name:               Upheaval, expected_damage:  18395.85, potency:  400, skill modifier:        Surging Tempest, damage variance: 4779.14\n",
            "time:  11300.0, name:              Onslaught, expected_damage:   6897.50, potency:  150, skill modifier:        Surging Tempest, damage variance: 1790.49\n",
            "time:  12500.0, name:            Primal Rend, expected_damage:  56473.41, potency:  700, skill modifier:        Surging Tempest, damage variance: 1629.82\n",
            "time:  13440.0, name:                   Auto, expected_damage:   4619.73, potency:   90, skill modifier:        Surging Tempest, damage variance: 1199.56\n",
            "time:  15000.0, name:            Inner Chaos, expected_damage:  53243.06, potency:  660, skill modifier:        Surging Tempest, damage variance: 1537.84\n",
            "time:  15650.0, name:              Onslaught, expected_damage:   6899.40, potency:  150, skill modifier:        Surging Tempest, damage variance: 1791.52\n",
            "time:  16800.0, name:                   Auto, expected_damage:   4620.62, potency:   90, skill modifier:        Surging Tempest, damage variance: 1200.72\n",
            "time:  17500.0, name:            Fell Cleave, expected_damage:  41944.48, potency:  520, skill modifier:        Surging Tempest, damage variance: 1211.25\n",
            "time:  18150.0, name:              Onslaught, expected_damage:   6896.13, potency:  150, skill modifier:        Surging Tempest, damage variance: 1790.27\n",
            "time:  20000.0, name:            Fell Cleave, expected_damage:  41944.34, potency:  520, skill modifier:        Surging Tempest, damage variance: 1211.40\n",
            "time:  20160.0, name:                   Auto, expected_damage:   4619.61, potency:   90, skill modifier:        Surging Tempest, damage variance: 1200.10\n",
            "time:  22500.0, name:            Fell Cleave, expected_damage:  36158.62, potency:  520, skill modifier:        Surging Tempest, damage variance: 1043.77\n",
            "time:  23520.0, name:                   Auto, expected_damage:   4049.80, potency:   90, skill modifier:        Surging Tempest, damage variance: 1047.37\n",
            "time:  25000.0, name:            Heavy Swing, expected_damage:   8065.99, potency:  200, skill modifier:        Surging Tempest, damage variance: 2088.13\n",
            "time:  26880.0, name:                   Auto, expected_damage:   3858.33, potency:   90, skill modifier:        Surging Tempest, damage variance: 998.88\n",
            "time:  27500.0, name:                   Maim, expected_damage:  11518.85, potency:  300, skill modifier:        Surging Tempest, damage variance: 2978.44\n",
            "time:  30000.0, name:           Storm's Path, expected_damage:  16900.39, potency:  440, skill modifier:        Surging Tempest, damage variance: 4373.25\n",
            "time:  30240.0, name:                   Auto, expected_damage:   3857.98, potency:   90, skill modifier:        Surging Tempest, damage variance: 998.95\n",
            "time:  32500.0, name:            Fell Cleave, expected_damage:  19977.17, potency:  520, skill modifier:        Surging Tempest, damage variance: 5170.84\n",
            "time:  33600.0, name:                   Auto, expected_damage:   3857.36, potency:   90, skill modifier:        Surging Tempest, damage variance: 997.46\n",
            "time:  35000.0, name:            Inner Chaos, expected_damage:  43714.80, potency:  660, skill modifier:        Surging Tempest, damage variance: 1262.21\n",
            "---Damage ranges and expectation:---\n",
            "    0: Auto                  , Potency:   90, Modifier: \"\". \n",
            "  Buffs: ||                      ||  crit: 0.00, dh: 0.00, damage_mult: 1.00, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "  Debuffs: ||                      ||  crit: 0.00, dh: 0.00, damage_mult: 1.00, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "   No crit, no DH ( 60.7%). expected_damage:   2648.50,   [low, high]: [2516.0, 2781.0]\n",
            "      Crit, no DH ( 23.7%). expected_damage:   4319.00,   [low, high]: [4103.0, 4535.0]\n",
            "      No crit, DH ( 11.2%). expected_damage:   3310.50,   [low, high]: [3145.0, 3476.0]\n",
            "         Crit, DH (  4.4%). expected_damage:   5398.00,   [low, high]: [5128.0, 5668.0]\n",
            "    0: Tomahawk              , Potency:  150, Modifier: \"\". \n",
            "  Buffs: ||                      ||  crit: 0.00, dh: 0.00, damage_mult: 1.00, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "  Debuffs: ||                      ||  crit: 0.00, dh: 0.00, damage_mult: 1.00, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "   No crit, no DH ( 60.7%). expected_damage:   3956.50,   [low, high]: [3759.0, 4154.0]\n",
            "      Crit, no DH ( 23.7%). expected_damage:   6452.50,   [low, high]: [6130.0, 6775.0]\n",
            "      No crit, DH ( 11.2%). expected_damage:   4945.00,   [low, high]: [4698.0, 5192.0]\n",
            "         Crit, DH (  4.4%). expected_damage:   8065.00,   [low, high]: [7662.0, 8468.0]\n",
            " 2500: Heavy Swing           , Potency:  200, Modifier: \"\". \n",
            "  Buffs: ||         Arcane Circle||  crit: 0.00, dh: 0.00, damage_mult: 1.03, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "  Debuffs: ||                      ||  crit: 0.00, dh: 0.00, damage_mult: 1.00, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "   No crit, no DH ( 60.7%). expected_damage:   5433.50,   [low, high]: [5162.0, 5705.0]\n",
            "      Crit, no DH ( 23.7%). expected_damage:   8862.00,   [low, high]: [8419.0, 9305.0]\n",
            "      No crit, DH ( 11.2%). expected_damage:   6791.00,   [low, high]: [6452.0, 7130.0]\n",
            "         Crit, DH (  4.4%). expected_damage:  11076.50,   [low, high]: [10523.0, 11630.0]\n",
            " 3360: Auto                  , Potency:   90, Modifier: \"\". \n",
            "  Buffs: ||         Arcane Circle||  crit: 0.00, dh: 0.00, damage_mult: 1.03, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "  Debuffs: ||                      ||  crit: 0.00, dh: 0.00, damage_mult: 1.00, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "   No crit, no DH ( 60.7%). expected_damage:   2727.50,   [low, high]: [2591.0, 2864.0]\n",
            "      Crit, no DH ( 23.7%). expected_damage:   4448.50,   [low, high]: [4226.0, 4671.0]\n",
            "      No crit, DH ( 11.2%). expected_damage:   3409.50,   [low, high]: [3239.0, 3580.0]\n",
            "         Crit, DH (  4.4%). expected_damage:   5559.50,   [low, high]: [5281.0, 5838.0]\n",
            " 5000: Maim                  , Potency:  300, Modifier: \"\". \n",
            "  Buffs: ||         Arcane Circle||  crit: 0.00, dh: 0.00, damage_mult: 1.03, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "  Debuffs: ||                      ||  crit: 0.00, dh: 0.00, damage_mult: 1.00, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "   No crit, no DH ( 60.7%). expected_damage:   8153.50,   [low, high]: [7746.0, 8561.0]\n",
            "      Crit, no DH ( 23.7%). expected_damage:  13297.50,   [low, high]: [12633.0, 13962.0]\n",
            "      No crit, DH ( 11.2%). expected_damage:  10192.00,   [low, high]: [9683.0, 10701.0]\n",
            "         Crit, DH (  4.4%). expected_damage:  16622.00,   [low, high]: [15791.0, 17453.0]\n",
            " 6720: Auto                  , Potency:   90, Modifier: \"\". \n",
            "  Buffs: ||Arcane Circle, Grade 8 Tincture||  crit: 0.00, dh: 0.00, damage_mult: 1.03, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "  Debuffs: ||                      ||  crit: 0.00, dh: 0.00, damage_mult: 1.00, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "   No crit, no DH ( 60.7%). expected_damage:   2953.00,   [low, high]: [2805.0, 3101.0]\n",
            "      Crit, no DH ( 23.7%). expected_damage:   4816.00,   [low, high]: [4575.0, 5057.0]\n",
            "      No crit, DH ( 11.2%). expected_damage:   3691.00,   [low, high]: [3507.0, 3875.0]\n",
            "         Crit, DH (  4.4%). expected_damage:   6019.50,   [low, high]: [5718.0, 6321.0]\n",
            " 8200: _Storm's Eye          , Potency:  440, Modifier: \"Surging Tempest\". \n",
            "  Buffs: ||Arcane Circle, Grade 8 Tincture, Embolden, Battle Litany||  crit: 0.10, dh: 0.00, damage_mult: 1.08, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "  Debuffs: ||       Chain Stratagem||  crit: 0.10, dh: 0.00, damage_mult: 1.00, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "   No crit, no DH ( 43.8%). expected_damage:  13584.50,   [low, high]: [12905.0, 14264.0]\n",
            "      Crit, no DH ( 40.6%). expected_damage:  22156.50,   [low, high]: [21048.0, 23265.0]\n",
            "      No crit, DH (  8.1%). expected_damage:  16980.50,   [low, high]: [16131.0, 17830.0]\n",
            "         Crit, DH (  7.5%). expected_damage:  27695.00,   [low, high]: [26309.0, 29081.0]\n",
            "10000: Inner Chaos           , Potency:  660, Modifier: \"Surging Tempest\". \n",
            "  Buffs: ||Arcane Circle, Grade 8 Tincture, Embolden, Battle Litany, Surging Tempest||  crit: 0.10, dh: 0.00, damage_mult: 1.19, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "  Debuffs: ||       Chain Stratagem||  crit: 0.10, dh: 0.00, damage_mult: 1.00, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "   No crit, no DH (  0.0%). expected_damage:  26116.50,   [low, high]: [24811.0, 27422.0]\n",
            "      Crit, no DH (  0.0%). expected_damage:  42596.50,   [low, high]: [40467.0, 44726.0]\n",
            "      No crit, DH (  0.0%). expected_damage:  32645.50,   [low, high]: [31014.0, 34277.0]\n",
            "         Crit, DH (100.0%). expected_damage:  53245.00,   [low, high]: [50583.0, 55907.0]\n",
            "10080: Auto                  , Potency:   90, Modifier: \"Surging Tempest\". \n",
            "  Buffs: ||Arcane Circle, Grade 8 Tincture, Embolden, Battle Litany, Surging Tempest||  crit: 0.10, dh: 0.00, damage_mult: 1.19, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "  Debuffs: ||       Chain Stratagem||  crit: 0.10, dh: 0.00, damage_mult: 1.00, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "   No crit, no DH ( 43.8%). expected_damage:   3411.00,   [low, high]: [3240.0, 3582.0]\n",
            "      Crit, no DH ( 40.6%). expected_damage:   5562.50,   [low, high]: [5284.0, 5841.0]\n",
            "      No crit, DH (  8.1%). expected_damage:   4263.00,   [low, high]: [4050.0, 4476.0]\n",
            "         Crit, DH (  7.5%). expected_damage:   6952.00,   [low, high]: [6604.0, 7300.0]\n",
            "10650: Upheaval              , Potency:  400, Modifier: \"Surging Tempest\". \n",
            "  Buffs: ||Arcane Circle, Grade 8 Tincture, Embolden, Battle Litany, Surging Tempest||  crit: 0.10, dh: 0.00, damage_mult: 1.19, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "  Debuffs: ||       Chain Stratagem||  crit: 0.10, dh: 0.00, damage_mult: 1.00, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "   No crit, no DH ( 43.8%). expected_damage:  13585.50,   [low, high]: [12906.0, 14265.0]\n",
            "      Crit, no DH ( 40.6%). expected_damage:  22157.00,   [low, high]: [21049.0, 23265.0]\n",
            "      No crit, DH (  8.1%). expected_damage:  16981.00,   [low, high]: [16132.0, 17830.0]\n",
            "         Crit, DH (  7.5%). expected_damage:  27696.50,   [low, high]: [26311.0, 29082.0]\n",
            "11300: Onslaught             , Potency:  150, Modifier: \"Surging Tempest\". \n",
            "  Buffs: ||Arcane Circle, Grade 8 Tincture, Embolden, Battle Litany, Surging Tempest||  crit: 0.10, dh: 0.00, damage_mult: 1.19, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "  Debuffs: ||       Chain Stratagem||  crit: 0.10, dh: 0.00, damage_mult: 1.00, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "   No crit, no DH ( 43.8%). expected_damage:   5093.00,   [low, high]: [4838.0, 5348.0]\n",
            "      Crit, no DH ( 40.6%). expected_damage:   8306.00,   [low, high]: [7890.0, 8722.0]\n",
            "      No crit, DH (  8.1%). expected_damage:   6365.50,   [low, high]: [6046.0, 6685.0]\n",
            "         Crit, DH (  7.5%). expected_damage:  10383.00,   [low, high]: [9863.0, 10903.0]\n",
            "12500: Primal Rend           , Potency:  700, Modifier: \"Surging Tempest\". \n",
            "  Buffs: ||Arcane Circle, Grade 8 Tincture, Embolden, Battle Litany, Surging Tempest||  crit: 0.10, dh: 0.00, damage_mult: 1.19, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "  Debuffs: ||       Chain Stratagem||  crit: 0.10, dh: 0.00, damage_mult: 1.00, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "   No crit, no DH (  0.0%). expected_damage:  27700.00,   [low, high]: [26315.0, 29085.0]\n",
            "      Crit, no DH (  0.0%). expected_damage:  45178.50,   [low, high]: [42919.0, 47438.0]\n",
            "      No crit, DH (  0.0%). expected_damage:  34624.50,   [low, high]: [32893.0, 36356.0]\n",
            "         Crit, DH (100.0%). expected_damage:  56473.00,   [low, high]: [53648.0, 59298.0]\n",
            "13440: Auto                  , Potency:   90, Modifier: \"Surging Tempest\". \n",
            "  Buffs: ||Arcane Circle, Grade 8 Tincture, Embolden, Battle Litany, Surging Tempest||  crit: 0.10, dh: 0.00, damage_mult: 1.19, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "  Debuffs: ||       Chain Stratagem||  crit: 0.10, dh: 0.00, damage_mult: 1.00, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "   No crit, no DH ( 43.8%). expected_damage:   3411.00,   [low, high]: [3240.0, 3582.0]\n",
            "      Crit, no DH ( 40.6%). expected_damage:   5562.50,   [low, high]: [5284.0, 5841.0]\n",
            "      No crit, DH (  8.1%). expected_damage:   4263.00,   [low, high]: [4050.0, 4476.0]\n",
            "         Crit, DH (  7.5%). expected_damage:   6952.00,   [low, high]: [6604.0, 7300.0]\n",
            "15000: Inner Chaos           , Potency:  660, Modifier: \"Surging Tempest\". \n",
            "  Buffs: ||Arcane Circle, Grade 8 Tincture, Embolden, Battle Litany, Surging Tempest||  crit: 0.10, dh: 0.00, damage_mult: 1.19, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "  Debuffs: ||       Chain Stratagem||  crit: 0.10, dh: 0.00, damage_mult: 1.00, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "   No crit, no DH (  0.0%). expected_damage:  26116.50,   [low, high]: [24811.0, 27422.0]\n",
            "      Crit, no DH (  0.0%). expected_damage:  42596.50,   [low, high]: [40467.0, 44726.0]\n",
            "      No crit, DH (  0.0%). expected_damage:  32645.50,   [low, high]: [31014.0, 34277.0]\n",
            "         Crit, DH (100.0%). expected_damage:  53245.00,   [low, high]: [50583.0, 55907.0]\n",
            "15650: Onslaught             , Potency:  150, Modifier: \"Surging Tempest\". \n",
            "  Buffs: ||Arcane Circle, Grade 8 Tincture, Embolden, Battle Litany, Surging Tempest||  crit: 0.10, dh: 0.00, damage_mult: 1.19, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "  Debuffs: ||       Chain Stratagem||  crit: 0.10, dh: 0.00, damage_mult: 1.00, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "   No crit, no DH ( 43.8%). expected_damage:   5093.00,   [low, high]: [4838.0, 5348.0]\n",
            "      Crit, no DH ( 40.6%). expected_damage:   8306.00,   [low, high]: [7890.0, 8722.0]\n",
            "      No crit, DH (  8.1%). expected_damage:   6365.50,   [low, high]: [6046.0, 6685.0]\n",
            "         Crit, DH (  7.5%). expected_damage:  10383.00,   [low, high]: [9863.0, 10903.0]\n",
            "16800: Auto                  , Potency:   90, Modifier: \"Surging Tempest\". \n",
            "  Buffs: ||Arcane Circle, Grade 8 Tincture, Embolden, Battle Litany, Surging Tempest||  crit: 0.10, dh: 0.00, damage_mult: 1.19, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "  Debuffs: ||       Chain Stratagem||  crit: 0.10, dh: 0.00, damage_mult: 1.00, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "   No crit, no DH ( 43.8%). expected_damage:   3411.00,   [low, high]: [3240.0, 3582.0]\n",
            "      Crit, no DH ( 40.6%). expected_damage:   5562.50,   [low, high]: [5284.0, 5841.0]\n",
            "      No crit, DH (  8.1%). expected_damage:   4263.00,   [low, high]: [4050.0, 4476.0]\n",
            "         Crit, DH (  7.5%). expected_damage:   6952.00,   [low, high]: [6604.0, 7300.0]\n",
            "17500: Fell Cleave           , Potency:  520, Modifier: \"Surging Tempest\". \n",
            "  Buffs: ||Arcane Circle, Grade 8 Tincture, Embolden, Battle Litany, Surging Tempest, Inner Release||  crit: 0.10, dh: 0.00, damage_mult: 1.19, guaranteed crit/dh: ForcedCritOrDH.FORCE_YES/ForcedCritOrDH.FORCE_YES:\n",
            "  Debuffs: ||       Chain Stratagem||  crit: 0.10, dh: 0.00, damage_mult: 1.00, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "   No crit, no DH (  0.0%). expected_damage:  20573.50,   [low, high]: [19545.0, 21602.0]\n",
            "      Crit, no DH (  0.0%). expected_damage:  33556.00,   [low, high]: [31879.0, 35233.0]\n",
            "      No crit, DH (  0.0%). expected_damage:  25716.50,   [low, high]: [24431.0, 27002.0]\n",
            "         Crit, DH (100.0%). expected_damage:  41945.00,   [low, high]: [39848.0, 44042.0]\n",
            "18150: Onslaught             , Potency:  150, Modifier: \"Surging Tempest\". \n",
            "  Buffs: ||Arcane Circle, Grade 8 Tincture, Embolden, Battle Litany, Surging Tempest||  crit: 0.10, dh: 0.00, damage_mult: 1.19, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "  Debuffs: ||       Chain Stratagem||  crit: 0.10, dh: 0.00, damage_mult: 1.00, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "   No crit, no DH ( 43.8%). expected_damage:   5093.00,   [low, high]: [4838.0, 5348.0]\n",
            "      Crit, no DH ( 40.6%). expected_damage:   8306.00,   [low, high]: [7890.0, 8722.0]\n",
            "      No crit, DH (  8.1%). expected_damage:   6365.50,   [low, high]: [6046.0, 6685.0]\n",
            "         Crit, DH (  7.5%). expected_damage:  10383.00,   [low, high]: [9863.0, 10903.0]\n",
            "20000: Fell Cleave           , Potency:  520, Modifier: \"Surging Tempest\". \n",
            "  Buffs: ||Arcane Circle, Grade 8 Tincture, Embolden, Battle Litany, Surging Tempest, Inner Release||  crit: 0.10, dh: 0.00, damage_mult: 1.19, guaranteed crit/dh: ForcedCritOrDH.FORCE_YES/ForcedCritOrDH.FORCE_YES:\n",
            "  Debuffs: ||       Chain Stratagem||  crit: 0.10, dh: 0.00, damage_mult: 1.00, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "   No crit, no DH (  0.0%). expected_damage:  20573.50,   [low, high]: [19545.0, 21602.0]\n",
            "      Crit, no DH (  0.0%). expected_damage:  33556.00,   [low, high]: [31879.0, 35233.0]\n",
            "      No crit, DH (  0.0%). expected_damage:  25716.50,   [low, high]: [24431.0, 27002.0]\n",
            "         Crit, DH (100.0%). expected_damage:  41945.00,   [low, high]: [39848.0, 44042.0]\n",
            "20160: Auto                  , Potency:   90, Modifier: \"Surging Tempest\". \n",
            "  Buffs: ||Arcane Circle, Grade 8 Tincture, Embolden, Battle Litany, Surging Tempest||  crit: 0.10, dh: 0.00, damage_mult: 1.19, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "  Debuffs: ||       Chain Stratagem||  crit: 0.10, dh: 0.00, damage_mult: 1.00, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "   No crit, no DH ( 43.8%). expected_damage:   3411.00,   [low, high]: [3240.0, 3582.0]\n",
            "      Crit, no DH ( 40.6%). expected_damage:   5562.50,   [low, high]: [5284.0, 5841.0]\n",
            "      No crit, DH (  8.1%). expected_damage:   4263.00,   [low, high]: [4050.0, 4476.0]\n",
            "         Crit, DH (  7.5%). expected_damage:   6952.00,   [low, high]: [6604.0, 7300.0]\n",
            "22500: Fell Cleave           , Potency:  520, Modifier: \"Surging Tempest\". \n",
            "  Buffs: ||Grade 8 Tincture, Embolden, Surging Tempest, Inner Release||  crit: 0.00, dh: 0.00, damage_mult: 1.16, guaranteed crit/dh: ForcedCritOrDH.FORCE_YES/ForcedCritOrDH.FORCE_YES:\n",
            "  Debuffs: ||                      ||  crit: 0.00, dh: 0.00, damage_mult: 1.00, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "   No crit, no DH (  0.0%). expected_damage:  17736.50,   [low, high]: [16850.0, 18623.0]\n",
            "      Crit, no DH (  0.0%). expected_damage:  28928.00,   [low, high]: [27482.0, 30374.0]\n",
            "      No crit, DH (  0.0%). expected_damage:  22170.50,   [low, high]: [21062.0, 23279.0]\n",
            "         Crit, DH (100.0%). expected_damage:  36159.50,   [low, high]: [34352.0, 37967.0]\n",
            "23520: Auto                  , Potency:   90, Modifier: \"Surging Tempest\". \n",
            "  Buffs: ||Grade 8 Tincture, Embolden, Surging Tempest||  crit: 0.00, dh: 0.00, damage_mult: 1.16, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "  Debuffs: ||                      ||  crit: 0.00, dh: 0.00, damage_mult: 1.00, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "   No crit, no DH ( 60.7%). expected_damage:   3311.50,   [low, high]: [3146.0, 3477.0]\n",
            "      Crit, no DH ( 23.7%). expected_damage:   5400.50,   [low, high]: [5130.0, 5671.0]\n",
            "      No crit, DH ( 11.2%). expected_damage:   4139.00,   [low, high]: [3932.0, 4346.0]\n",
            "         Crit, DH (  4.4%). expected_damage:   6750.00,   [low, high]: [6412.0, 7088.0]\n",
            "25000: Heavy Swing           , Potency:  200, Modifier: \"Surging Tempest\". \n",
            "  Buffs: ||Grade 8 Tincture, Embolden, Surging Tempest||  crit: 0.00, dh: 0.00, damage_mult: 1.16, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "  Debuffs: ||                      ||  crit: 0.00, dh: 0.00, damage_mult: 1.00, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "   No crit, no DH ( 60.7%). expected_damage:   6594.00,   [low, high]: [6264.0, 6924.0]\n",
            "      Crit, no DH ( 23.7%). expected_damage:  10754.50,   [low, high]: [10217.0, 11292.0]\n",
            "      No crit, DH ( 11.2%). expected_damage:   8242.00,   [low, high]: [7830.0, 8654.0]\n",
            "         Crit, DH (  4.4%). expected_damage:  13442.50,   [low, high]: [12770.0, 14115.0]\n",
            "26880: Auto                  , Potency:   90, Modifier: \"Surging Tempest\". \n",
            "  Buffs: ||Grade 8 Tincture, Surging Tempest||  crit: 0.00, dh: 0.00, damage_mult: 1.10, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "  Debuffs: ||                      ||  crit: 0.00, dh: 0.00, damage_mult: 1.00, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "   No crit, no DH ( 60.7%). expected_damage:   3154.00,   [low, high]: [2996.0, 3312.0]\n",
            "      Crit, no DH ( 23.7%). expected_damage:   5143.50,   [low, high]: [4886.0, 5401.0]\n",
            "      No crit, DH ( 11.2%). expected_damage:   3942.00,   [low, high]: [3745.0, 4139.0]\n",
            "         Crit, DH (  4.4%). expected_damage:   6428.50,   [low, high]: [6107.0, 6750.0]\n",
            "27500: Maim                  , Potency:  300, Modifier: \"Surging Tempest\". \n",
            "  Buffs: ||Grade 8 Tincture, Surging Tempest||  crit: 0.00, dh: 0.00, damage_mult: 1.10, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "  Debuffs: ||                      ||  crit: 0.00, dh: 0.00, damage_mult: 1.00, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "   No crit, no DH ( 60.7%). expected_damage:   9420.50,   [low, high]: [8949.0, 9892.0]\n",
            "      Crit, no DH ( 23.7%). expected_damage:  15364.00,   [low, high]: [14595.0, 16133.0]\n",
            "      No crit, DH ( 11.2%). expected_damage:  11776.00,   [low, high]: [11187.0, 12365.0]\n",
            "         Crit, DH (  4.4%). expected_damage:  19205.00,   [low, high]: [18244.0, 20166.0]\n",
            "30000: Storm's Path          , Potency:  440, Modifier: \"Surging Tempest\". \n",
            "  Buffs: ||Grade 8 Tincture, Surging Tempest||  crit: 0.00, dh: 0.00, damage_mult: 1.10, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "  Debuffs: ||                      ||  crit: 0.00, dh: 0.00, damage_mult: 1.00, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "   No crit, no DH ( 60.7%). expected_damage:  13817.50,   [low, high]: [13126.0, 14509.0]\n",
            "      Crit, no DH ( 23.7%). expected_damage:  22535.50,   [low, high]: [21408.0, 23663.0]\n",
            "      No crit, DH ( 11.2%). expected_damage:  17271.00,   [low, high]: [16407.0, 18135.0]\n",
            "         Crit, DH (  4.4%). expected_damage:  28169.00,   [low, high]: [26759.0, 29579.0]\n",
            "30240: Auto                  , Potency:   90, Modifier: \"Surging Tempest\". \n",
            "  Buffs: ||Grade 8 Tincture, Surging Tempest||  crit: 0.00, dh: 0.00, damage_mult: 1.10, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "  Debuffs: ||                      ||  crit: 0.00, dh: 0.00, damage_mult: 1.00, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "   No crit, no DH ( 60.7%). expected_damage:   3154.00,   [low, high]: [2996.0, 3312.0]\n",
            "      Crit, no DH ( 23.7%). expected_damage:   5143.50,   [low, high]: [4886.0, 5401.0]\n",
            "      No crit, DH ( 11.2%). expected_damage:   3942.00,   [low, high]: [3745.0, 4139.0]\n",
            "         Crit, DH (  4.4%). expected_damage:   6428.50,   [low, high]: [6107.0, 6750.0]\n",
            "32500: Fell Cleave           , Potency:  520, Modifier: \"Surging Tempest\". \n",
            "  Buffs: ||Grade 8 Tincture, Surging Tempest||  crit: 0.00, dh: 0.00, damage_mult: 1.10, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "  Debuffs: ||                      ||  crit: 0.00, dh: 0.00, damage_mult: 1.00, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "   No crit, no DH ( 60.7%). expected_damage:  16330.50,   [low, high]: [15514.0, 17147.0]\n",
            "      Crit, no DH ( 23.7%). expected_damage:  26635.00,   [low, high]: [25303.0, 27967.0]\n",
            "      No crit, DH ( 11.2%). expected_damage:  20413.50,   [low, high]: [19393.0, 21434.0]\n",
            "         Crit, DH (  4.4%). expected_damage:  33293.50,   [low, high]: [31628.0, 34959.0]\n",
            "33600: Auto                  , Potency:   90, Modifier: \"Surging Tempest\". \n",
            "  Buffs: ||Grade 8 Tincture, Surging Tempest||  crit: 0.00, dh: 0.00, damage_mult: 1.10, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "  Debuffs: ||                      ||  crit: 0.00, dh: 0.00, damage_mult: 1.00, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "   No crit, no DH ( 60.7%). expected_damage:   3154.00,   [low, high]: [2996.0, 3312.0]\n",
            "      Crit, no DH ( 23.7%). expected_damage:   5143.50,   [low, high]: [4886.0, 5401.0]\n",
            "      No crit, DH ( 11.2%). expected_damage:   3942.00,   [low, high]: [3745.0, 4139.0]\n",
            "         Crit, DH (  4.4%). expected_damage:   6428.50,   [low, high]: [6107.0, 6750.0]\n",
            "35000: Inner Chaos           , Potency:  660, Modifier: \"Surging Tempest\". \n",
            "  Buffs: ||Grade 8 Tincture, Surging Tempest||  crit: 0.00, dh: 0.00, damage_mult: 1.10, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "  Debuffs: ||                      ||  crit: 0.00, dh: 0.00, damage_mult: 1.00, guaranteed crit/dh: ForcedCritOrDH.DEFAULT/ForcedCritOrDH.DEFAULT:\n",
            "   No crit, no DH (  0.0%). expected_damage:  21442.00,   [low, high]: [20370.0, 22514.0]\n",
            "      Crit, no DH (  0.0%). expected_damage:  34972.50,   [low, high]: [33224.0, 36721.0]\n",
            "      No crit, DH (  0.0%). expected_damage:  26802.50,   [low, high]: [25462.0, 28143.0]\n",
            "         Crit, DH (100.0%). expected_damage:  43715.00,   [low, high]: [41530.0, 45900.0]\n",
            "Running: TimingSpecTest\n",
            "Running: UtilsTest\n",
            "Running: StatusEffectSpecTest\n",
            "Running: StatFnsTest\n",
            "Running: SkillTest\n",
            "Running: RotationBuilderTest\n",
            "Running: SnapshotAndApplicationEventsTest\n",
            "Running: StatusEffectTrackerTest\n",
            "Running: JobResourceTrackerTest\n",
            "Running: ComboTrackerTest\n",
            "Running: RotationBuilderAndDamageBuilderIntegrationTest\n",
            "Running: DamageBuilderTest\n",
            "Running: ComputeDamageUtilsTest\n",
            "Running: DamageSimulatorTest\n",
            "Running: EndToEndTest\n",
            "Running: JobsTest\n",
            "All tests done. Num passing: 189. Num failing: 0\n",
            "Tests completed in 5.908368825912476s.\n"
          ]
        }
      ],
      "source": [
        "#@title Imports and Setup\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "import pandas as pd\n",
        "import dataclasses\n",
        "import functools\n",
        "import requests\n",
        "import pprint\n",
        "import re\n",
        "import copy\n",
        "import math\n",
        "import logging\n",
        "\n",
        "from collections import namedtuple\n",
        "from google.colab import userdata\n",
        "from typing import *\n",
        "\n",
        "GITHUB_PTA = userdata.get('GITHUB_PTA')\n",
        "if not GITHUB_PTA:\n",
        "  raise Exception(f\"Please set Github Public Access Token (GITHUB_PTA) secret.\")\n",
        "\n",
        "# Ensure FFLOGS values are set when using this. Please set\n",
        "# as secrets and do not use them in plaintext.\n",
        "if not (userdata.get(\"FFLOGS_CLIENT_ID\") or userdata.get(\"FFLOGS_CLIENT_SECRET\")):\n",
        "  raise Exception(f\"Please set FFLOGs tokens to access logs. https://www.fflogs.com/api/clients/\")\n",
        "\n",
        "# TODO: move a lot of the implementation into python files on git\n",
        "# So it can also just be added onto drive\n",
        "\n",
        "# Pip req imports\n",
        "!pip install coreapi==2.3.3 python-graphql-client==0.4.3\n",
        "import coreapi\n",
        "import python_graphql_client\n",
        "\n",
        "%cd /content\n",
        "!git clone https://\"$GITHUB_PTA\"@github.com/Amarantine-xiv/Amas-FF14-Combat-Sim\n",
        "%cd /content/Amas-FF14-Combat-Sim\n",
        "!git pull\n",
        "\n",
        "%run '/content/Amas-FF14-Combat-Sim/CoreSimulator.ipynb'"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#@title Custom FFLOGs constants\n",
        "\"\"\"Defines and documents all custom columns in the data.\"\"\"\n",
        "\n",
        "# Not custom, but used a lot so recorded\n",
        "ABILITY_ID = \"abilityGameID\"\n",
        "TYPE = \"type\"\n",
        "CAST = \"cast\"\n",
        "DAMAGE = \"damage\"\n",
        "APPLY_BUFF = \"applybuff\"\n",
        "APPLY_DEBUFF = \"applydebuff\"\n",
        "REFRESH_BUFF = \"refreshbuff\"\n",
        "REFRESH_DEBUFF = \"refreshdebuff\"\n",
        "REMOVE_BUFF = \"removebuff\"\n",
        "REMOVE_DEBUFF = \"removedebuff\"\n",
        "PREPARE = \"calculateddamage\"\n",
        "\n",
        "# Human readable name for abilities\n",
        "ABILITY_NAME = \"ability_name\"\n",
        "\n",
        "# Human readable source name for actors\n",
        "SOURCE_NAME = \"source_name\"\n",
        "SOURCE_TYPE = \"source_type\"\n",
        "SOURCE_SUBTYPE = \"source_subtype\"\n",
        "SOURCE_PET_ACTOR = \"source_pet_actor\"\n",
        "SOURCE_ID = \"sourceID\"\n",
        "\n",
        "# Human readable target name for actors\n",
        "TARGET_NAME = \"target_name\"\n",
        "TARGET_TYPE = \"target_type\"\n",
        "TARGET_SUBTYPE = \"target_subtype\"\n",
        "TARGET_PET_ACTOR = \"target_pet_actor\"\n",
        "TARGET_ID = \"targetID\"\n",
        "\n",
        "# Fight time, based on lb update tick to reset gauge\n",
        "FIGHT_TIME = \"fight_time\"\n",
        "\n",
        "# Bonus percent, estimate the value of combos\n",
        "# 1 - basePotency/bonusPotency = bonusPercent\n",
        "BONUS_PERCENT = \"bonusPercent\"\n",
        "AMOUNT = \"amount\"\n",
        "LOG_DAMAGE = \"unmitigatedAmount\"  # amount column is affected by overkill\n",
        "LOG_HIT_TYPE = \"hitType\"\n",
        "LOG_DHIT = \"directHit\"\n",
        "LOG_MULTIPLIER = \"multiplier\"\n",
        "\n",
        "# Replaced names\n",
        "CRIT = \"crit\"\n",
        "DHIT = \"direct_hit\"\n",
        "\n",
        "# Special name added in for an assumed prepull apply\n",
        "ESTIMATED_APPLY_BUFF = \"estimatedapplybuff\"\n",
        "\n",
        "# Special name for \"correct\" card\n",
        "CARD_MATCHES_JOB = \"card_matches_job\"\n",
        "\n",
        "# Mappings (target_subtype)\n",
        "JOB = \"Job\"\n",
        "\n",
        "# Breaking melee vs ranged in seperate mapping\n",
        "TARGET_SUBTYPE_TO_JOB_MELEE = {\n",
        "    \"DarkKnight\": \"DRK\",\n",
        "    \"Gunbreaker\": \"GNB\",\n",
        "    \"Paladin\": \"PLD\",\n",
        "    \"Warrior\": \"WAR\",\n",
        "    \"Ninja\": \"NIN\",\n",
        "    \"Samurai\": \"SAM\",\n",
        "    \"Reaper\": \"RPR\",\n",
        "    \"Monk\": \"MNK\",\n",
        "    \"Dragoon\": \"DRG\",\n",
        "}\n",
        "\n",
        "TARGET_SUBTYPE_TO_JOB_RANGED = {\n",
        "    \"Dancer\": \"DNC\",\n",
        "    \"Bard\": \"BRD\",\n",
        "    \"Machinist\": \"MCH\",\n",
        "    \"RedMage\": \"RDM\",\n",
        "    \"BlackMage\": \"BLM\",\n",
        "    \"Summoner\": \"SMN\",\n",
        "    \"Astrologian\": \"AST\",\n",
        "    \"Scholar\": \"SCH\",\n",
        "    \"Sage\": \"SGE\",\n",
        "    \"WhiteMage\": \"WHM\",\n",
        "}\n",
        "\n",
        "TARGET_SUBTYPE_TO_JOB = {\n",
        "    **TARGET_SUBTYPE_TO_JOB_MELEE,\n",
        "    **TARGET_SUBTYPE_TO_JOB_RANGED,\n",
        "}\n"
      ],
      "metadata": {
        "cellView": "form",
        "id": "8Lbu4XTSBRm4"
      },
      "execution_count": 24,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#@title EventInfo, SnapshotAndApplicationInfo, FightProcessingUtils\n",
        "\n",
        "# NOTE: some of this code is copied from general logs processing, so it is a\n",
        "# bit more general than it needs to be.\n",
        "\n",
        "class EventInfo(namedtuple('EventInfo', [TYPE, 'timestamp', 'source_class', SOURCE_NAME, TARGET_NAME, 'skill_name', 'log_damage', 'direct_hit', 'critical_hit', 'bonus_percent', 'pet_name'])):\n",
        "  def __key(self):\n",
        "    return (self.timestamp, self.skill_name, self.pet_name)\n",
        "\n",
        "  def __eq__(self, other):\n",
        "    if isinstance(other, EventInfo):\n",
        "      return self.__key() == other.__key()\n",
        "\n",
        "  def __hash__(self):\n",
        "    return hash(self.__key())\n",
        "\n",
        "class SnapshotAndApplicationInfo(namedtuple('SnapshotAndDamageInfo', ['snapshot_time',\n",
        "                                                                      'application_time',\n",
        "                                                                      'source_class',\n",
        "                                                                      'skill_name',\n",
        "                                                                      'log_damage',\n",
        "                                                                      'direct_hit',\n",
        "                                                                      'critical_hit',\n",
        "                                                                      'bonus_percent',\n",
        "                                                                      'pet_name'])):\n",
        "  pass\n",
        "\n",
        "\n",
        "class FightProcessingUtils():\n",
        "  def __merge_snapshots_and_applications(snapshots, applications):\n",
        "    curr_snapshot_idx = 0\n",
        "    next_snapshot_idx = 1\n",
        "    res = []\n",
        "\n",
        "    for curr_application in applications:\n",
        "      next_snapshot_time = snapshots[next_snapshot_idx].timestamp if next_snapshot_idx < len(snapshots) else math.inf\n",
        "      while curr_application.timestamp > next_snapshot_time:\n",
        "        curr_snapshot_idx += 1\n",
        "        next_snapshot_idx += 1\n",
        "        if next_snapshot_idx >= len(snapshots):\n",
        "          next_snapshot_time = math.inf\n",
        "          break\n",
        "        else:\n",
        "          next_snapshot_time = snapshots[next_snapshot_idx].timestamp\n",
        "\n",
        "      snapshot_and_application_info = SnapshotAndApplicationInfo(snapshot_time= snapshots[curr_snapshot_idx].timestamp,\n",
        "                                                                                application_time= curr_application.timestamp,\n",
        "                                                                                skill_name = curr_application.skill_name,\n",
        "                                                                                source_class = curr_application.source_class,\n",
        "                                                                                log_damage = curr_application.log_damage,\n",
        "                                                                                direct_hit = curr_application.direct_hit,\n",
        "                                                                                critical_hit = curr_application.critical_hit,\n",
        "                                                                                bonus_percent = curr_application.bonus_percent,\n",
        "                                                                                pet_name = curr_application.pet_name)\n",
        "      res.append(snapshot_and_application_info)\n",
        "    return res\n",
        "\n",
        "  def merge_snapshots_and_applications(snapshots, applications):\n",
        "    snapshots_dict = {}\n",
        "    applications_dict = {}\n",
        "    for snapshot in snapshots:\n",
        "      if snapshot.skill_name not in snapshots_dict:\n",
        "        snapshots_dict[snapshot.skill_name] = []\n",
        "      snapshots_dict[snapshot.skill_name].append(snapshot)\n",
        "\n",
        "    for application in applications:\n",
        "      if application.skill_name not in applications_dict:\n",
        "        applications_dict[application.skill_name] = []\n",
        "      applications_dict[application.skill_name].append(application)\n",
        "\n",
        "    # Remap names here to line up, if needed\n",
        "    res = {}\n",
        "    for key in applications_dict.keys():\n",
        "      if key in snapshots_dict.keys():\n",
        "        res[key] = FightProcessingUtils.__merge_snapshots_and_applications(snapshots_dict[key], applications_dict[key])\n",
        "      else:\n",
        "        print('Skill not found in snapshots: {}'.format(key))\n",
        "      # verify this is not a bug if skill is not in snapshot?\n",
        "    return res\n",
        "\n",
        "  @staticmethod\n",
        "  def __get_data_for_event_types(data, ignore_pets):\n",
        "    if ignore_pets and (data.get('pet_name', None) is not None):\n",
        "      return None\n",
        "    critical_hit = False\n",
        "    if 'hitType' in data.keys():\n",
        "      critical_hit = True if data['hitType'] == 2 else False #wtf man?\n",
        "\n",
        "    info = EventInfo(type=data[TYPE],\n",
        "                     timestamp = data['timestamp'],\n",
        "                     source_class = data[SOURCE_SUBTYPE],\n",
        "                     source_name = data[SOURCE_NAME],\n",
        "                     target_name = data[TARGET_NAME],\n",
        "                     skill_name = data[ABILITY_NAME],\n",
        "                     log_damage = data.get('amount', None),\n",
        "                     direct_hit = data.get('directHit', False),\n",
        "                     critical_hit = critical_hit,\n",
        "                     bonus_percent=data.get('bonusPercent', None),\n",
        "                     pet_name = data.get('pet_name', None))\n",
        "    return info\n",
        "\n",
        "  @staticmethod\n",
        "  def get_data_for_event_types(event_data, event_types, ignore_pets=False):\n",
        "    # Output values of dicts are guaranteed to be sorted by timestamp.\n",
        "    # event_types is processed in priority order\n",
        "    res = {}\n",
        "    for event_type in event_types:\n",
        "      for i in range(0, len(event_data)):\n",
        "        data = event_data[i]\n",
        "        if data['type'] != event_type:\n",
        "          continue\n",
        "\n",
        "        player_name = data['source_name']    # check if player is target???\n",
        "        if player_name not in res.keys():\n",
        "          res[player_name] = []\n",
        "        info = FightProcessingUtils.__get_data_for_event_types(data, ignore_pets)\n",
        "        if info is not None and info not in res[player_name]:\n",
        "          res[player_name].append(info)\n",
        "\n",
        "      for _, tmp in res.items():\n",
        "        tmp.sort(key= lambda x: x.timestamp)\n",
        "    return res\n",
        "\n",
        "    # def remap_skill_names(e\n"
      ],
      "metadata": {
        "cellView": "form",
        "id": "TPU_rtWu_CY_"
      },
      "execution_count": 25,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#@title FFLOGS Api\n",
        "\n",
        "def get_auth_token() -> str:\n",
        "    \"\"\"Fetches an auth token via https://www.fflogs.com/oauth/token.\n",
        "\n",
        "    See https://www.fflogs.com/api/docs/\n",
        "\n",
        "    And https://www.fflogs.com/v2-api-docs/ff/report.doc.html\n",
        "\n",
        "    Requires FFLOGS_CLIENT_ID and FFLOGS_CLIENT_SECRET set as variables.\n",
        "    These can be obtained here: https://www.fflogs.com/api/clients/\n",
        "\n",
        "    Please keep these private to yourself and do not upload them to\n",
        "    github or share them.\n",
        "\n",
        "    :returns: Auth token for FFLogs APIv2\n",
        "    \"\"\"\n",
        "    x = requests.post(\n",
        "        url=\"https://www.fflogs.com/oauth/token\",\n",
        "        data={\"grant_type\": \"client_credentials\"},\n",
        "        auth=(\n",
        "            userdata.get(\"FFLOGS_CLIENT_ID\"),\n",
        "            userdata.get(\"FFLOGS_CLIENT_SECRET\"),\n",
        "        ),\n",
        "    )\n",
        "\n",
        "    x.raise_for_status()\n",
        "    return x.json()[\"access_token\"]"
      ],
      "metadata": {
        "cellView": "form",
        "id": "yctknpqmBcz1"
      },
      "execution_count": 26,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#@title LogsDataFetcher\n",
        "class LogsDataFetcher():\n",
        "  def __init__(self):\n",
        "    self.__code_and_fight_cache = {}\n",
        "    self.__ability_cache = {}\n",
        "    self.__actor_cache = {}\n",
        "    self.__events_cache = {}\n",
        "    self.__combatantinfo_cache = {}\n",
        "\n",
        "    # Constructs a client on init.\n",
        "    # TODO: Change this to a factory or an input instead to allow for mocking\n",
        "    # the API calls.\n",
        "    self.client = python_graphql_client.GraphqlClient(\n",
        "        endpoint=\"https://www.fflogs.com/api/v2/client\",\n",
        "        headers={\"Authorization\": f\"Bearer {get_auth_token()}\"})\n",
        "\n",
        "  @staticmethod\n",
        "  def parse_fflogs_url(url):\n",
        "    URL_REGEXP = \"https://www.fflogs.com/reports/(\\w+)\\#fight=(\\d+)\"\n",
        "    tmp =re.search(URL_REGEXP, url)\n",
        "    return tmp[1], int(tmp[2])\n",
        "\n",
        "  def __get_fight_data(self, code: str, fightID) -> List[Dict]:\n",
        "    \"\"\"Fetches the high level kills from a log.\"\"\"\n",
        "    return self.client.execute(\n",
        "        query=\"\"\"query reportData($code: String!, $fightID: Int!){\n",
        "              reportData\n",
        "              {\n",
        "                report(code: $code){\n",
        "                  fights(fightIDs: [$fightID]){\n",
        "                    encounterID\n",
        "                    startTime\n",
        "                    endTime\n",
        "                    name\n",
        "                    kill\n",
        "                    standardComposition\n",
        "                    combatTime\n",
        "                  }\n",
        "                }\n",
        "              }\n",
        "            }\n",
        "            \"\"\",\n",
        "        variables={\n",
        "            \"code\": code,\n",
        "            \"fightID\": fightID\n",
        "        },\n",
        "    )[\"data\"][\"reportData\"][\"report\"][\"fights\"][0]\n",
        "\n",
        "  def __get_abilities_data(self, code):\n",
        "    abilities = self.client.execute(\n",
        "              query=\"\"\"query reportData($code: String!){\n",
        "                            reportData\n",
        "                            {\n",
        "                              report(code: $code){\n",
        "                                masterData{\n",
        "                                  abilities {\n",
        "                                    gameID\n",
        "                                    name\n",
        "                                    type\n",
        "                                  }\n",
        "                                }\n",
        "                              }\n",
        "                            }\n",
        "                          }\n",
        "                          \"\"\",\n",
        "              variables={\"code\": code},\n",
        "          )[\"data\"][\"reportData\"][\"report\"][\"masterData\"][\"abilities\"]\n",
        "    res = {}\n",
        "    for tmp in abilities:\n",
        "      assert tmp['gameID'] not in res.keys(), \"Found duplicate abilities? : id/name: {}/{}\".format(tmp['gameID'], tmp['name'])\n",
        "      res[tmp['gameID']] = tmp\n",
        "    return res\n",
        "\n",
        "  def __get_actors_data(self, code):\n",
        "      actors = self.client.execute(\n",
        "              query=\"\"\"query reportData($code: String!){\n",
        "                            reportData\n",
        "                            {\n",
        "                              report(code: $code){\n",
        "                                masterData{\n",
        "                                  actors {\n",
        "                                    id\n",
        "                                    name\n",
        "                                    petOwner\n",
        "                                    subType\n",
        "                                    type\n",
        "                                  }\n",
        "                                }\n",
        "                              }\n",
        "                            }\n",
        "                          }\n",
        "                          \"\"\",\n",
        "              variables={\"code\": code},\n",
        "          )[\"data\"][\"reportData\"][\"report\"][\"masterData\"][\"actors\"]\n",
        "      res = {}\n",
        "      for tmp in actors:\n",
        "        assert tmp['id'] not in res.keys(), \"Found duplicate actors? : id/name: {}/{}\".format(tmp['id'], tmp['name'])\n",
        "        res[tmp['id']] = tmp\n",
        "      return res\n",
        "\n",
        "  @staticmethod\n",
        "  def __get_combatantinfo_data(events_data):\n",
        "    res = {}\n",
        "    for data in events_data:\n",
        "      if data['type'] == 'combatantinfo':\n",
        "        assert data['sourceID'] not in res.keys(), \"Found duplicate comabtants? : sourceID: {}\".format(data['sourceID'])\n",
        "        res[data['sourceID']] = data\n",
        "    return res\n",
        "\n",
        "  def get_fight_data_from_url(self, url):\n",
        "    code, fight_id = self.parse_fflogs_url(url)\n",
        "    code_and_fight_key = '{}-{}'.format(code, fight_id)\n",
        "\n",
        "    if code_and_fight_key in self.__code_and_fight_cache.keys():\n",
        "      fight_data = self.__code_and_fight_cache['{}-{}'.format(code, fight_id)]\n",
        "    else:\n",
        "      fight_data = self.__get_fight_data(code, fight_id)\n",
        "      fight_data['code'] = code\n",
        "      fight_data['fightID'] = fight_id\n",
        "      self.__code_and_fight_cache[code_and_fight_key] = fight_data\n",
        "    return copy.deepcopy(fight_data)\n",
        "\n",
        "  def get_ability_data(self, fight_data):\n",
        "    code = fight_data['code']\n",
        "    if code not in self.__ability_cache.keys():\n",
        "      self.__ability_cache[code] = self.__get_abilities_data(code)\n",
        "    return copy.deepcopy(self.__ability_cache[code])\n",
        "\n",
        "  def get_combatantinfo_data(self, fight_data, events_data):\n",
        "    code_and_fight_key = '{}-{}'.format(fight_data['code'], fight_data['fightID'])\n",
        "    if code_and_fight_key not in self.__combatantinfo_cache.keys():\n",
        "      self.__combatantinfo_cache[code_and_fight_key] = self.__get_combatantinfo_data(events_data)\n",
        "    return copy.deepcopy(self.__combatantinfo_cache[code_and_fight_key])\n",
        "\n",
        "  def get_actor_data(self, fight_data):\n",
        "    code = fight_data['code']\n",
        "    if code not in self.__actor_cache.keys():\n",
        "      self.__actor_cache[code] = self.__get_actors_data(code)\n",
        "    return copy.deepcopy(self.__actor_cache[code])\n",
        "\n",
        "  def get_events_data(\n",
        "        self,\n",
        "        fight_data: dict,\n",
        "    ) -> List:\n",
        "\n",
        "      res = []\n",
        "      start_time = fight_data['startTime']\n",
        "      code = fight_data['code']\n",
        "      fight_id = fight_data['fightID']\n",
        "\n",
        "      code_and_fight_key = '{}-{}'.format(code, fight_id)\n",
        "\n",
        "      if code_and_fight_key not in self.__events_cache.keys():\n",
        "        while start_time is not None and start_time < fight_data['endTime']:\n",
        "          \"\"\"Inner function for the timestamp management.\"\"\"\n",
        "          raw_data = self.client.execute(\n",
        "              # set events(startTime: $startTime, endTime: $endTime, limit: $limit, dataType: DamageDone) to\n",
        "              # get the buffs too\n",
        "              query=\"\"\"query reportData($code: String!, $startTime: Float!, $endTime: Float!, $limit: Int!){\n",
        "                            reportData\n",
        "                            {\n",
        "                              report(code: $code){\n",
        "                                events(startTime: $startTime, endTime: $endTime, limit: $limit){\n",
        "                                  data\n",
        "                                  nextPageTimestamp\n",
        "                                }\n",
        "                              }\n",
        "                            }\n",
        "                          }\n",
        "                          \"\"\",\n",
        "              variables={\n",
        "                  \"code\": fight_data['code'],\n",
        "                  \"startTime\": start_time,\n",
        "                  \"endTime\": fight_data['endTime'],\n",
        "                  \"limit\": 10000\n",
        "              },\n",
        "          )[\"data\"][\"reportData\"][\"report\"][\"events\"]\n",
        "          res.extend(raw_data[\"data\"])\n",
        "          start_time = raw_data[\"nextPageTimestamp\"]\n",
        "        self.__events_cache[code_and_fight_key] = res\n",
        "      return copy.deepcopy(self.__events_cache[code_and_fight_key])\n",
        "\n",
        "  @staticmethod\n",
        "  def __get_pet_name_and_owner(data, actor_data):\n",
        "    if data[TARGET_TYPE] == 'Pet':\n",
        "      pet_id = data[TARGET_ID ]\n",
        "      pet_name = actor_data[pet_id]['name']\n",
        "\n",
        "      owner_id = data[SOURCE_ID ]\n",
        "      owner_name = actor_data[owner_id]['name']\n",
        "    else:\n",
        "      pet_id = data[SOURCE_ID ]\n",
        "      pet_name = actor_data[pet_id]['name']\n",
        "\n",
        "      owner_id = actor_data[pet_id]['petOwner']\n",
        "      owner_name = actor_data[owner_id]['name']\n",
        "      owner_name = actor_data[owner_id]['name']\n",
        "    return owner_name, pet_name\n",
        "\n",
        "  @staticmethod\n",
        "  def __add_actor_and_ability_data_to_events_data(events_data, actor_data, ability_data):\n",
        "    # NOTE: this will modify event_data in-place. We do not create a defensive\n",
        "    # copy because that might be pretty expensive.\n",
        "    for data in events_data:\n",
        "\n",
        "      data[SOURCE_SUBTYPE] = actor_data[data[SOURCE_ID]]['subType'] if SOURCE_ID in data.keys() else None\n",
        "      data[TARGET_TYPE] = actor_data[data[TARGET_ID]]['subType'] if TARGET_ID in data.keys() else None\n",
        "      data[ABILITY_NAME] = ability_data[data[ABILITY_ID]]['name'] if ABILITY_ID in data.keys() else None\n",
        "\n",
        "      if data[SOURCE_SUBTYPE ] == 'Pet':\n",
        "        owner_name, pet_name = LogsDataFetcher.__get_pet_name_and_owner(data, actor_data)\n",
        "        data[SOURCE_NAME] = owner_name\n",
        "        data['pet_name'] = pet_name\n",
        "      else:\n",
        "        data[SOURCE_NAME] = actor_data[data[SOURCE_ID]]['name'] if SOURCE_ID in data.keys() else None\n",
        "\n",
        "      if data[TARGET_TYPE] == 'Pet':\n",
        "        owner_name, pet_name = LogsDataFetcher.__get_pet_name_and_owner(data, actor_data)\n",
        "        data['target_name'] = owner_name\n",
        "        data['pet_name'] = pet_name\n",
        "      else:\n",
        "        data['target_name'] = actor_data[data[TARGET_ID]]['name'] if TARGET_ID in data.keys() else None\n",
        "\n",
        "  def get_all_fight_data_from_url(self, url):\n",
        "    fight_data = self.get_fight_data_from_url(url)\n",
        "    ability_data = self.get_ability_data(fight_data)\n",
        "    actor_data = self.get_actor_data(fight_data)\n",
        "    events_data = self.get_events_data(fight_data)\n",
        "    self.__add_actor_and_ability_data_to_events_data(events_data, actor_data, ability_data)\n",
        "    combatant_info = self.get_combatantinfo_data(fight_data, events_data)\n",
        "    return fight_data, combatant_info, ability_data, actor_data, events_data"
      ],
      "metadata": {
        "cellView": "form",
        "id": "DMOdUtQ7BAge"
      },
      "execution_count": 27,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#@title Various processing fns\n",
        "def assign_center(dot_ticks, centres):\n",
        "  dot_ticks = np.reshape(np.asarray(dot_ticks),(-1,1))\n",
        "  centres = np.reshape(np.asarray(centres),(1,-1))\n",
        "  return np.argmin(np.abs(dot_ticks-centres), axis=1)\n",
        "\n",
        "def get_dot_ticks(all_events_data):\n",
        "  res = []\n",
        "  for tmp in all_events_data:\n",
        "    if tmp['ability_name'] == 'Combined DoTs':\n",
        "      res.append(tmp['unmitigatedAmount'])\n",
        "  return res\n",
        "\n",
        "def get_tick_ranges(all_ticks, assignments, hit_type):\n",
        "  res = {}\n",
        "  for i in range(0, 4):\n",
        "    assigned_ticks = np.asarray(all_ticks)[assignments==i]\n",
        "    if len(assigned_ticks) == 0:\n",
        "      res[hit_type[i]] = None\n",
        "    else:\n",
        "      res[hit_type[i]] = (min(assigned_ticks), max(assigned_ticks))\n",
        "  return res\n",
        "\n",
        "def get_roll_match(x, high_roll, low_roll):\n",
        "  return 0.1 * (x - low_roll) /(high_roll - low_roll)  + 0.95\n",
        "\n",
        "def get_ranges_dict(damage_ranges):\n",
        "  ranges_dict = {}\n",
        "  print('Skill: {}'.format(damage_ranges[0][0]))\n",
        "  for i in range(0, 4):\n",
        "    key = damage_ranges[2][i+1][0]\n",
        "    low, high = damage_ranges[1][i+1][1], damage_ranges[1][i+1][2]\n",
        "    ranges_dict[key] = (low, high)\n",
        "  return ranges_dict\n",
        "\n",
        "def get_roll_matches(all_ticks, assignments, damage_ranges, hit_type):\n",
        "  res = []\n",
        "  ranges_dict = get_ranges_dict(damage_ranges)\n",
        "  for i in range(0, 4):\n",
        "    damage = np.asarray(all_ticks)[assignments==i]\n",
        "    low_roll, high_roll = ranges_dict[hit_type[i]]\n",
        "    res.extend(get_roll_match(np.asarray(all_ticks)[assignments==i], low_roll, high_roll))\n",
        "  return np.asarray(res)"
      ],
      "metadata": {
        "id": "S_XQSQjSA-Jy"
      },
      "execution_count": 40,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#@title Dot logs: record all of them here.\n",
        "# Move to csv, if desired.\n",
        "ALL_DOT_LOGS = {}\n",
        "ALL_CENTRES = {}\n",
        "ALL_STATS = {}\n",
        "ALL_SKILL_NAMES = {}\n",
        "ALL_DOT_LOGS['SCH'] = ('https://www.fflogs.com/reports/PcBHhNY96FftTzqG#fight=39',\n",
        "                       'https://www.fflogs.com/reports/m2ctP4aYz3dNLVFZ#fight=1')\n",
        "ALL_CENTRES['SCH'] = [3000, 3750, 4700, 6000]\n",
        "ALL_STATS['SCH'] = Stats(wd=132, weapon_delay=3.12, main_stat=3377, det_stat=1806, crit_stat=2395, dh_stat=688, speed_stat=1126, job_class = 'SCH', healer_or_caster_strength=351, num_roles_in_party=0)\n",
        "ALL_SKILL_NAMES['SCH'] = 'Biolysis'\n",
        "\n",
        "ALL_DOT_LOGS['WHM'] = ('https://www.fflogs.com/reports/rcFHR4mVxCvBMkNd#fight=2',)\n",
        "ALL_CENTRES['WHM'] = [2800, 3350, 4350, 5500]\n",
        "ALL_STATS['WHM'] = Stats(wd=132, weapon_delay=3.44, main_stat=3367, det_stat=2047, crit_stat=2399, dh_stat=760, speed_stat=856, job_class='WHM', num_roles_in_party=0)\n",
        "ALL_SKILL_NAMES['WHM'] = 'Dia'\n",
        "\n",
        "ALL_DOT_LOGS['AST'] = ('https://www.fflogs.com/reports/9xFQVN8fM3CmXkcv#fight=3',)\n",
        "ALL_CENTRES['AST'] = [2100, 2700, 3500, 4350]\n",
        "ALL_STATS['AST'] = Stats(wd=128, weapon_delay=3.2, main_stat=3339, det_stat=1794, crit_stat=2455, dh_stat=760, speed_stat=718, job_class='AST', num_roles_in_party=0)\n",
        "ALL_SKILL_NAMES['AST'] = 'Combust III'\n",
        "\n",
        "ALL_DOT_LOGS['SGE'] = ('https://www.fflogs.com/reports/HzxDmX9h3Z1nMRKT#fight=12',)\n",
        "ALL_CENTRES['SGE'] = [3000, 3750, 4750, 5800]\n",
        "ALL_STATS['SGE'] = Stats(wd=128, weapon_delay=2.8, main_stat=3339, det_stat=1794, crit_stat=2362, dh_stat=760, speed_stat=811, job_class='SGE', num_roles_in_party=0)\n",
        "ALL_SKILL_NAMES['SGE'] = 'Eukrasian Dosis III'\n",
        "\n",
        "ALL_DOT_LOGS['PLD'] = ('https://www.fflogs.com/reports/HQnWmL9MXkRfVzh7#fight=6',)\n",
        "ALL_CENTRES['PLD'] = [500, 625, 800, 980]\n",
        "ALL_STATS['PLD'] = Stats(wd=105, weapon_delay=2.24, main_stat=2743, det_stat=1682, crit_stat=2113, dh_stat=796, speed_stat=592, tenacity=838, num_roles_in_party=0, job_class='PLD')\n",
        "ALL_SKILL_NAMES['PLD'] = 'Circle of Scorn'\n",
        "\n",
        "ALL_DOT_LOGS['GNB-SB'] = ('https://www.fflogs.com/reports/A28rPbcKnzQF7C3g#fight=10',)\n",
        "ALL_CENTRES['GNB-SB'] = [1450, 1750, 2300, 2800]\n",
        "ALL_STATS['GNB-SB'] = Stats(wd=129, weapon_delay=2.80, main_stat=3237, det_stat=1666, crit_stat=2417, dh_stat=724, speed_stat=1016, tenacity=529, num_roles_in_party=0, job_class='GNB')\n",
        "ALL_SKILL_NAMES['GNB-SB'] = 'Sonic Break'\n",
        "\n",
        "ALL_DOT_LOGS['GNB-Bow'] = ('https://www.fflogs.com/reports/r862WQCFX7La9Hpt#fight=1',)\n",
        "ALL_CENTRES['GNB-Bow'] = [1400, 1750, 2300, 2800]\n",
        "ALL_STATS['GNB-Bow'] = Stats(wd=129, weapon_delay=2.80, main_stat=3237, det_stat=1666, crit_stat=2417, dh_stat=724, speed_stat=1016, tenacity=529, num_roles_in_party=0, job_class='GNB')\n",
        "ALL_SKILL_NAMES['GNB-Bow'] = 'Bow Shock'\n",
        "\n",
        "ALL_DOT_LOGS['DRK'] = ('https://www.fflogs.com/reports/2MH4DJVLjgcdnPBX#fight=2',)\n",
        "ALL_CENTRES['DRK'] = [1200, 1500, 1900, 2400]\n",
        "ALL_STATS['DRK'] = Stats(wd=129, weapon_delay=2.96, main_stat=3256, det_stat=1666, crit_stat=2417, dh_stat=724, speed_stat=1016, tenacity=529, num_roles_in_party=0, job_class='DRK')\n",
        "ALL_SKILL_NAMES['DRK'] = 'Salted Earth'\n",
        "\n",
        "ALL_DOT_LOGS['SAM'] = ('https://www.fflogs.com/reports/Z3zKxFD1W7Hj8MGT#fight=16',)\n",
        "ALL_CENTRES['SAM'] = [1075, 1350, 1700, 2100]\n",
        "ALL_STATS['SAM'] = Stats(wd=122, weapon_delay=2.64, main_stat=2915, det_stat=910, crit_stat=2007, dh_stat=1572, speed_stat=898, num_roles_in_party=0, job_class='SAM')\n",
        "ALL_SKILL_NAMES['SAM'] = 'Higanbana'\n",
        "\n",
        "\n",
        "hit_type = ['No crit, no DH', 'No crit, DH', 'Crit, no DH', 'Crit, DH']"
      ],
      "metadata": {
        "id": "_HVTEyAJB1Xo"
      },
      "execution_count": 48,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "logs_data_fetcher = LogsDataFetcher()"
      ],
      "metadata": {
        "id": "xMn2tGO2D3n9"
      },
      "execution_count": 30,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# dot processing\n",
        "all_good_roll_matches = {}\n",
        "all_bad_roll_matches = {}\n",
        "\n",
        "for KEY in ALL_DOT_LOGS.keys():\n",
        "# for KEY in ['SAM', 'GNB-Bow', 'GNB-SB', 'PLD']:\n",
        "\n",
        "  dot_logs = ALL_DOT_LOGS[KEY]\n",
        "  stats = ALL_STATS[KEY]\n",
        "  centres = ALL_CENTRES[KEY]\n",
        "\n",
        "  all_events_data = []\n",
        "  for dot_log in dot_logs:\n",
        "    _, combatant_info, _, _, events_data = logs_data_fetcher.get_all_fight_data_from_url(dot_log)\n",
        "    all_events_data.extend(events_data)\n",
        "\n",
        "  if ALL_SKILL_NAMES[KEY] == 'Salted Earth':\n",
        "    all_ticks = []\n",
        "    for tmp in events_data:\n",
        "      if tmp['ability_name'] == 'Salted Earth' and 'unmitigatedAmount' in tmp.keys():\n",
        "        all_ticks.append(tmp['unmitigatedAmount'])\n",
        "  else:\n",
        "    all_ticks = get_dot_ticks(all_events_data)\n",
        "\n",
        "  assignments = assign_center(all_ticks, centres)\n",
        "  print(get_tick_ranges(all_ticks, assignments, hit_type))\n",
        "\n",
        "  rb = RotationBuilder(stats, SKILL_LIBRARY, ignore_trailing_dots=False, enable_autos=False, snap_dots_to_server_tick_starting_at=0)\n",
        "  rb.add_next(ALL_SKILL_NAMES[KEY])\n",
        "\n",
        "  _, _, _, damage_ranges, _ = execute_rotation(rb, num_samples=100000)\n",
        "  roll_matches = get_roll_matches(all_ticks, assignments, damage_ranges, hit_type)\n",
        "  out_of_range = (roll_matches < 0.95) | (roll_matches > 1.05)\n",
        "\n",
        "  all_good_roll_matches[KEY] = roll_matches[~out_of_range]\n",
        "  all_bad_roll_matches[KEY] = roll_matches[out_of_range]\n",
        "  print('Number of good matches: {}'.format(len(all_good_roll_matches[KEY])))\n",
        "  print('Number of bad matches: {}'.format(len(all_bad_roll_matches[KEY])))\n",
        "  print(all_bad_roll_matches[KEY])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "fhf_lyZzB3sK",
        "outputId": "7475d808-25eb-4406-9adb-34340ebda436"
      },
      "execution_count": 55,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "{'No crit, no DH': (2802, 3095), 'No crit, DH': (3503, 3853), 'Crit, no DH': (4509, 4978), 'Crit, DH': (5892, 6042)}\n",
            "Time taken: 0.08735895156860352\n",
            "Skill: _Biolysis dot\n",
            "Number of good matches: 389\n",
            "Number of bad matches: 0\n",
            "[]\n",
            "{'No crit, no DH': (2584, 2855), 'No crit, DH': (3236, 3560), 'Crit, no DH': (4161, 4596), 'Crit, DH': (5260, 5732)}\n",
            "Time taken: 0.09453701972961426\n",
            "Skill: Dia\n",
            "Number of good matches: 397\n",
            "Number of bad matches: 2\n",
            "[1.05036765 1.05022883]\n",
            "{'No crit, no DH': (2056, 2272), 'No crit, DH': (2570, 2836), 'Crit, no DH': (3322, 3671), 'Crit, DH': (4152, 4581)}\n",
            "Time taken: 0.11726522445678711\n",
            "Skill: _Combust III dot\n",
            "Number of good matches: 1190\n",
            "Number of bad matches: 1\n",
            "[1.05022831]\n",
            "{'No crit, no DH': (2844, 3137), 'No crit, DH': (3573, 3898), 'Crit, no DH': (4569, 5031), 'Crit, DH': (5752, 5890)}\n",
            "Time taken: 0.0542757511138916\n",
            "Skill: _Eukrasian Dosis III dot\n",
            "Number of good matches: 210\n",
            "Number of bad matches: 0\n",
            "[]\n",
            "{'No crit, no DH': (482, 533), 'No crit, DH': (602, 662), 'Crit, no DH': (767, 842), 'Crit, DH': (963, 991)}\n",
            "Time taken: 0.03241920471191406\n",
            "Skill: Circle of Scorn\n",
            "Number of good matches: 198\n",
            "Number of bad matches: 0\n",
            "[]\n",
            "{'No crit, no DH': (1355, 1498), 'No crit, DH': (1696, 1852), 'Crit, no DH': (2187, 2413), 'Crit, DH': (2777, 2921)}\n",
            "Time taken: 0.057923316955566406\n",
            "Skill: Sonic Break\n",
            "Number of good matches: 287\n",
            "Number of bad matches: 1\n",
            "[1.0513986]\n",
            "{'No crit, no DH': (1355, 1497), 'No crit, DH': (1697, 1871), 'Crit, no DH': (2192, 2411), 'Crit, DH': (2767, 3001)}\n",
            "Time taken: 0.038748979568481445\n",
            "Skill: Bow Shock\n",
            "Number of good matches: 221\n",
            "Number of bad matches: 4\n",
            "[1.0513986 1.0506993 1.0513986 1.0513986]\n",
            "{'No crit, no DH': (1145, 1266), 'No crit, DH': (1432, 1531), 'Crit, no DH': (1845, 2039), 'Crit, DH': None}\n",
            "Time taken: 0.029619216918945312\n",
            "Skill: _Salted Earth dot\n",
            "Number of good matches: 152\n",
            "Number of bad matches: 2\n",
            "[0.94834711 0.94897436]\n",
            "{'No crit, no DH': (1028, 1137), 'No crit, DH': (1285, 1421), 'Crit, no DH': (1614, 1782), 'Crit, DH': (2017, 2228)}\n",
            "Time taken: 0.12601375579833984\n",
            "Skill: Higanbana\n",
            "Number of good matches: 697\n",
            "Number of bad matches: 3\n",
            "[1.05091743 1.05091743 1.05073529]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "tot_good = 0\n",
        "tot_bad = 0\n",
        "for k,v in all_good_roll_matches.items():\n",
        "  tot_good += len(v)\n",
        "for k,v in all_bad_roll_matches.items():\n",
        "  tot_bad += len(v)\n",
        "print('Total good: {}. Total bad: {}'.format(tot_good, tot_bad))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1Pd9Ag4iJAYV",
        "outputId": "0749a1e4-8d78-4e14-c660-12266c62f4ab"
      },
      "execution_count": 56,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Total good: 3741. Total bad: 13\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#title Sanity checking dot hits\n",
        "plt.figure()\n",
        "plt.hist(all_ticks, bins=100)\n",
        "plt.title('Overall dot ticks')\n",
        "\n",
        "for i in range(0, 4):\n",
        "  plt.figure()\n",
        "  plt.hist(np.asarray(all_ticks)[assignments==i], bins=25)\n",
        "  plt.title(hit_type[i])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "mYbG3YnjC_HY",
        "outputId": "65c6b4bb-f989-4086-9923-191c3bb9e7ac"
      },
      "execution_count": 45,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    }
  ]
}