app.py

import os
import socket
from pathlib import Path

import pdfplumber
from flask import Flask, render_template, request, redirect, send_file, jsonify
from werkzeug.utils import secure_filename

from title_focus_search_index import TitleFocusSearchIndex

app = Flask(__name__)

SLIDE_DIR = "slides"
IMAGE_DIR = "img_cache"

Index = TitleFocusSearchIndex


@app.route('/')
def index():
    ll = os.listdir(SLIDE_DIR)
    return render_template('index.html', ll=ll)


def allowed_file(filename):
    return '.' in filename and filename.rsplit('.', 1)[1].lower() in ["pdf"]


@app.route('/upload', methods=['POST'])
def upload():
    if 'file' in request.files:
        file = request.files['file']
        if file.filename != '':
            if file and allowed_file(file.filename):
                filename = secure_filename(file.filename)
                file_path = os.path.join(Path(SLIDE_DIR), filename)
                file.save(file_path)
                pdf = pdfplumber.open(file_path)
                index = Index()
                for i, page in enumerate(pdf.pages):
                    text = page.extract_text()
                    index.add(str(file_path), i, text)
                    img_name = str(file_path)[7:] + "_" + str(i) + ".jpg"
                    img_path = os.path.join(IMAGE_DIR, img_name)
                    page.to_image().save(img_path)
                del index
    return redirect('/')


@app.route("/search", methods=['POST'])
def query():
    try:
        index = Index()
        query = request.form.get("term")
        context = request.form.get("context")  # TODO: Use context to find better results
        result = index.search(query, context)
        img_name = result["path"][7:] + "_" + str(result["page"]) + ".jpg"
        return jsonify({
            "type": "image",
            "path": os.getenv('EXTERNAL_HOST', '<PLEASE_SET_EXTERNAL_HOST_ENV_VAR>') + "/slide/" + img_name
        })
    except IndexError:
        return jsonify({
            "type": "miss"
        })


@app.route("/slide/<img_name>")
def slide(img_name):
    path = os.path.join(IMAGE_DIR, img_name)
    if "/" in img_name or "\\" in img_name:
        return "bad image name", 400
    if not os.path.exists(path):
        return "image not present", 404
    return send_file(path, mimetype='image/jpg')


if __name__ == '__main__':
    app.debug = True
    app.run(host="0.0.0.0")