{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "! pip install -U \"transformers[torch]\" pypdf python-docx langdetect sentencepiece sacremoses"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import gradio as gr\n",
    "from transformers import pipeline\n",
    "from pypdf import PdfReader\n",
    "import docx\n",
    "import os\n",
    "from langdetect import detect"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "summarizer = pipeline(task=\"summarization\", model=\"facebook/bart-large-cnn\")\n",
    "translator_to_french = pipeline(\n",
    "    task=\"translation_en_to_fr\", model=\"Helsinki-NLP/opus-mt-en-fr\"\n",
    ")\n",
    "translator_to_english = pipeline(\n",
    "    task=\"translation_fr_to_en\", model=\"Helsinki-NLP/opus-mt-fr-en\"\n",
    ")\n",
    "\n",
    "MAX_FILE_SIZE = 10000000"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "class TextExtractor:\n",
    "    def __init__(self, doc_location: str):\n",
    "        if doc_location is None:\n",
    "            raise Exception(f\"Please select a PDF to summarize\")\n",
    "        self.doc_location = doc_location\n",
    "\n",
    "    def extract_text_from_pdf(self):\n",
    "        reader = PdfReader(self.doc_location)\n",
    "        text = \"\"\n",
    "\n",
    "        for page in reader.pages:\n",
    "            text += page.extract_text()\n",
    "\n",
    "        return text\n",
    "\n",
    "    def extract_text_from_doc(self):\n",
    "        doc = docx.Document(self.doc_location)\n",
    "        text = \"\"\n",
    "\n",
    "        for paragraph in doc.paragraphs:\n",
    "            text += paragraph.text + \"\\n\"\n",
    "        return text\n",
    "\n",
    "    def extract_text_from_txt(self):\n",
    "        with open(self.doc_location, \"r\", encoding=\"utf-8\") as file:\n",
    "            text = file.read()\n",
    "        return text\n",
    "\n",
    "    def extract_text_from_txt(self):\n",
    "        with open(self.doc_location, \"r\", encoding=\"utf-8\") as file:\n",
    "            text = file.read()\n",
    "        return text\n",
    "    \n",
    "    def text_length(self):\n",
    "        words = self.text.split()\n",
    "        num_words = len(words)\n",
    "        return num_words\n",
    "\n",
    "    def get_text(self) -> str:\n",
    "        file_extension = os.path.splitext(self.doc_location)[1]\n",
    "        if file_extension == \".pdf\":\n",
    "            self.text = self.extract_text_from_pdf()\n",
    "        elif file_extension == \".txt\":\n",
    "            self.text = self.extract_text_from_txt()\n",
    "        elif file_extension == \".docx\" or file_extension == \".doc\":\n",
    "            self.text = self.extract_text_from_doc()\n",
    "        else:\n",
    "            raise gr.Error(f\"We only support .pdf, .txt, .doc and .docx files\")\n",
    "\n",
    "        if len(self.text) > MAX_FILE_SIZE:\n",
    "            raise gr.Error(\n",
    "                f\"Document exceeds the maximum supported size of {MAX_FILE_SIZE} characters.\"\n",
    "            )\n",
    "\n",
    "        return self.text"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "text_extractor = TextExtractor(\"data/doc-file-example.docx\")\n",
    "text = text_extractor.get_text()\n",
    "\n",
    "text_length = text_extractor.text_length()\n",
    "summary_length = int(text_length / 2)\n",
    "\n",
    "summary = summarizer(text, max_length=summary_length, do_sample=False)[0][\"summary_text\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "summary"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "detected_lang = detect(summary)\n",
    "detected_lang"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "venv",
   "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": "3.11.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}