{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import os"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Access paths\n",
    "DATA_PATH = './data'\n",
    "btz_path = os.path.join(DATA_PATH, 'bretzel.csv')\n",
    "\n",
    "# You need to use the parser before to get the csv file.\n",
    "my_data = pd.read_csv(btz_path, index_col='date', dtype=str)\n",
    "my_data = my_data.replace(to_replace=[np.NaN], value='')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "btz_core_id = {\n",
    "            'Nathan':   'user502298447',\n",
    "            'Louison':  'user435924815',\n",
    "            'Clément':  'user451690413',\n",
    "            'Théo':     'user490556008',\n",
    "            'Léopold':  'user483543500',\n",
    "            'Arnaud':   'user469821944',\n",
    "            }"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def words_frequency(data):\n",
    "    words_freq = data['msg_text']\n",
    "    words_freq = words_freq.astype(str)\n",
    "    remove_list = [',', '.', '(', ')', '{', '}', '[', ']', '-']\n",
    "    for char in remove_list:\n",
    "        words_freq = words_freq.str.replace(char, '')\n",
    "\n",
    "    # Replace apostrophes with spaces for word counting purposes\n",
    "    words_freq = words_freq.astype(str).str.replace('\\'', ' ')\n",
    "    words_freq = words_freq.str.lower().str.split(expand=True).stack().value_counts()\n",
    "    return words_freq"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_message_count(data):\n",
    "    for name in btz_core_id:\n",
    "        idx = data['sender_id']==btz_core_id.get(name)\n",
    "        time_range = pd.to_datetime(data.index[idx], yearfirst=True).to_series().dt.date\n",
    "        cumulative_count = 1+np.arange(time_range.shape[0])\n",
    "        plt.plot(time_range, cumulative_count, label=name)\n",
    "        \n",
    "    plt.legend()\n",
    "    plt.xlabel('Date')\n",
    "    plt.ylabel('Nombre de messages')\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def message_length(data, mode='moving', freq='365D'):\n",
    "    text_message = data[['sender', 'sender_id', 'msg_text']]\n",
    "    text_message = text_message.sort_index()\n",
    "    for name in btz_core_id:\n",
    "        idx = text_message['sender_id'] == btz_core_id.get(name)\n",
    "        time_range = pd.to_datetime(data.index[idx], yearfirst=True).to_series().dt.date\n",
    "        user_messages = text_message.loc[idx]\n",
    "        user_messages_length = user_messages['msg_text'].str.len()\n",
    "        user_messages_length_no_spaces = user_messages['msg_text'].str.replace(' ', '').str.len()\n",
    "        print(f\"{name}, {user_messages_length.mean():.2f}, {user_messages_length_no_spaces.mean():.2f}\")\n",
    "        user_messages_length.index = pd.to_datetime(user_messages_length.index)\n",
    "\n",
    "        if mode == 'moving':\n",
    "            plt.plot(time_range, user_messages_length.rolling(freq).mean(), label=name)\n",
    "        if mode == 'cum':\n",
    "            plt.plot(time_range, user_messages_length.expanding(1).sum(), label=name) \n",
    "            \n",
    "    plt.legend()\n",
    "    plt.xlabel('Date')\n",
    "    plt.ylabel('Longueur (caractères)')\n",
    "    if mode == 'moving':\n",
    "        plt.title(f\"Moyenne glissante de fréquence {freq} du nombre de caractères par message\")\n",
    "    if mode == 'cum':\n",
    "        plt.title(f\"Somme cumulée du nombre de caractères\")\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "message_length(my_data, mode='moving', freq='365D')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def most_messages(data):\n",
    "    data.index = pd.to_datetime(data.index, yearfirst=True)\n",
    "    messages = data['msg_text']\n",
    "    top = messages.groupby(messages.index.date).count().sort_values(ascending=False)\n",
    "    top.index = pd.to_datetime(top.index, format='%Y-%m-%d').strftime('%d/%m/%Y')\n",
    "    return top"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "most_messages(my_data)[:30]"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.3"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "e7370f93d1d0cde622a1f8e1c04877d8463912d04d973331ad4851f04de6915a"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}