import os
import re
from bs4 import BeautifulSoup
import random
from genanki import Model
from genanki import Note
from genanki import Deck
from genanki import Package


mid = random.randrange(1 << 30, 1 << 31)
did = random.randrange(1 << 30, 1 << 31)

MY_MODEL = Model(
    mid,
    'Word Root Of The Day Archive',
    fields = [
        {'name': 'Text'},
        {'name': 'Supplement'},
        ],
    templates = [{
        'name': 'CARD',
        'qfmt': '{{Text}}',
        'afmt': '{{FrontSide}}{{Supplement}}',
        },]
    )

my_deck = Deck(did, 'Word Root Of The Day Archive')

for root, dirs, files in os.walk("wrotds"):
    for name in files:
        with open(os.path.join(root, name), "r") as frn:
            soup = BeautifulSoup(frn, "html5lib")

        for ps in soup.select("div#wrotd-quicksum p"):
            for pi in ps("img"):
                pi.decompose()

            for pt in ps(True):
                pt.attrs.clear()

            ps.attrs.clear()

            tp = re.sub(r' {2,}', ' ', str(ps)).replace('\n', '')
            fields = [tp, '']
            my_note = Note(model=MY_MODEL, fields=fields)
            my_deck.add_note(my_note)

        for pp in soup.select("div#wrotd-full-text p"):
            newp = re.sub(r'([.?!][”’]?) {2,}', r'\1</p><p>', str(pp))
            newp = re.sub(r' {2,}', ' ', newp).replace('\n', '')

            soup2 = BeautifulSoup(newp, "html5lib")

            for st in soup2(True):
                st.attrs.clear()

            for sp in soup2("p"):
                if not re.search(r'\w<strong>|</strong>\w', str(sp)):
                    continue

                fields = [str(sp), '']
                my_note = Note(model=MY_MODEL, fields=fields)
                my_deck.add_note(my_note)

Package(my_deck).write_to_file('Word_Root_Of_The_Day_Archive_2.apkg')
