droidz/droidz.py

import argparse
import bs4
import datetime
import os
import re
import requests
import sqlite3
import subprocess
import sys
import types

from voussoirkit import betterhelp
from voussoirkit import pathclass
from voussoirkit import ratelimiter
from voussoirkit import sqlhelpers
from voussoirkit import threadpool
from voussoirkit import winwhich

CATEGORIES = [
    'stickmen',
    'stickpacks',
    'vehicles',
    'weapons',
    'objects',
    'random',
    'effects',
    'backgrounds',
]

DB_INIT = '''
BEGIN;
CREATE TABLE IF NOT EXISTS sticks(
    id TEXT PRIMARY KEY NOT NULL,
    name TEXT,
    description TEXT,
    date INT,
    author TEXT,
    download_link,
    category TEXT,
    downloads INT,
    version TEXT,
    vote_score INT,
    usage_rating TEXT,
    retrieved INT
);
CREATE INDEX IF NOT EXISTS index_sticks_id ON sticks(id);
COMMIT;
'''

SQL_COLUMNS = sqlhelpers.extract_table_column_map(DB_INIT)

sql = sqlite3.connect('sticks.db')
sql.executescript(DB_INIT)


USERAGENT = '''
Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko)
Chrome/79.0.3945.130 Safari/537.36
'''.replace('\n', ' ').strip()

HEADERS = {
    'User-Agent': USERAGENT
}

session = requests.Session()
session.headers.update(HEADERS)

DOWNLOAD_RATELIMITER = ratelimiter.Ratelimiter(allowance=1, period=5)

WINRAR = winwhich.which('winrar')

def get_now():
    return datetime.datetime.now(datetime.timezone.utc).timestamp()

def id_from_direct_url(direct_url):
    id = direct_url.split('/direct/')[-1]
    id = id.split('/')[0].split('?')[0]
    return id

# DB FUNCTIONS
################################################################################
def select_stick(id):
    cur = sql.cursor()
    cur.execute('SELECT * FROM sticks WHERE id == ?', [id])
    return cur.fetchone()

def insert_id(id, commit=True):
    cur = sql.cursor()
    cur.execute('SELECT 1 FROM sticks WHERE id == ?', [id])
    existing = cur.fetchone()
    if not existing:
        data = {'id': id}
        columns = SQL_COLUMNS['sticks']
        (qmarks, bindings) = sqlhelpers.insert_filler(columns, data, require_all=False)

        query = f'INSERT INTO sticks VALUES({qmarks})'
        cur.execute(query, bindings)

        if commit:
            sql.commit()

    status = types.SimpleNamespace(id=id, is_new=not existing)
    return status

def insert_ids(ids, commit=True):
    for id in ids:
        insert_id(id, commit=False)

    if commit:
        sql.commit()

def insert_stick(data, commit=True):
    cur = sql.cursor()

    cur.execute('SELECT 1 FROM sticks WHERE id == ?', [data['id']])
    existing = cur.fetchone()
    if existing:
        (qmarks, bindings) = sqlhelpers.update_filler(data, 'id')
        query = f'UPDATE sticks {qmarks}'
    else:
        (qmarks, bindings) = sqlhelpers.insert_filler(SQL_COLUMNS['sticks'], data)
        query = f'INSERT INTO sticks VALUES({qmarks})'

    cur.execute(query, bindings)

    if commit:
        sql.commit()

def insert_sticks(datas, commit=True):
    for data in datas:
        insert_stick(data, commit=False)

    if commit:
        sql.commit()

# SCRAPE
################################################################################
def request(url):
    print(url)
    response = session.get(url)
    response.raise_for_status()
    return response

def scrape_direct(id, commit=True):
    '''
    Return the dict of Stick data for this ID.
    '''
    url = f'http://droidz.org/direct/{id}'
    response = request(url)
    text = response.text

    # I had a weird issue where some brs were not self-closing and they
    # contained a bunch of other elements. This whitespace replacement fixed
    # the issue but I didn't quite understand why.
    text = re.sub(r'<\s*br\s*/\s*>', '<br/>', text)
    soup = bs4.BeautifulSoup(text, 'html.parser')

    for br in soup.find_all('br'):
        br.replace_with('\n')

    stick_info = soup.select('.content')[1].get_text()
    author = soup.find('a', href=re.compile(r'search\.php\?searchq=')).get_text()
    vote_score = int(re.search(r'Vote Score: ([-\d]+)\s*$', stick_info, flags=re.M).group(1))
    downloads = int(re.search(r'Downloads: (\d+)\s*$', stick_info, flags=re.M).group(1))
    category = re.search(r'Category: (.+?)\s*$', stick_info, flags=re.M).group(1)
    version = re.search(r'Version: (.+?)\s*$', stick_info, flags=re.M).group(1)
    usage_rating = re.search(r'Usage Rating: (.+?)\s*$', stick_info, flags=re.M).group(1)
    date = re.search(r'Date Submitted: (.+?)\s*$', stick_info, flags=re.M).group(1)
    date = datetime.datetime.strptime(date, '%B %d, %Y')
    date = date.timestamp()

    name = soup.select_one('.section .top h2').get_text().strip()
    description = soup.select_one('.section .content').get_text().strip()
    if description == f'{author}, has left no comments for this submission.':
        description = None
    else:
        description = description.replace(f'{author} says, ', '')
    download_link = soup.find('a', href=re.compile(r'/resources/grab\.php\?file='))['href']
    retrieved = int(get_now())

    data = {
        'id': id,
        'name': name,
        'description': description,
        'date': date,
        'author': author,
        'download_link': download_link,
        'category': category,
        'downloads': downloads,
        'version': version,
        'vote_score': vote_score,
        'usage_rating': usage_rating,
        'retrieved': retrieved,
    }

    return data

def scrape_directs(ids, threads=1, commit=True):
    '''
    Given many Stick IDs, yield Stick datas.
    '''
    if threads < 1:
        raise ValueError(threads)

    if threads == 1:
        for id in ids:
            yield scrape_direct(id)

    else:
        pool = threadpool.ThreadPool(size=threads)
        kwargss = [
            {'function': scrape_direct, 'args': [id], 'name': id}
            for id in ids
        ]
        pool.add_many(kwargss)
        for job in pool.result_generator():
            if job.exception:
                raise job.exception
            yield job.value

def scrape_category(category):
    '''
    Yield Stick IDs from all pages within this category. They are listed in
    alphabetical order by Stick name.
    '''
    page = 1
    all_directs = set()
    while True:
        url = f'http://droidz.org/stickmain/{category}.php?page={page}'
        response = request(url)
        soup = bs4.BeautifulSoup(response.text, 'html.parser')
        this_directs = soup.find_all('a', href=re.compile(r'/direct/\d+'))
        prev_count = len(all_directs)
        all_directs.update(this_directs)
        if len(all_directs) == prev_count:
            break
        page += 1
        for direct in this_directs:
            id = id_from_direct_url(direct['href'])
            yield id

def scrape_latest():
    '''
    Yield the latest Stick IDs from the /stickmain homepage, most recent first.
    '''
    url = 'http://droidz.org/stickmain/'
    response = request(url)
    soup = bs4.BeautifulSoup(response.text, 'html.parser')
    h2s = soup.find_all('h2')
    for h2 in h2s:
        if 'Latest 50 Accepted' in h2.get_text():
            latest_50_h2 = h2
            break

    div = latest_50_h2.parent
    directs = div.find_all('a', href=re.compile(r'/direct/\d+'))
    for direct in directs:
        id = id_from_direct_url(direct['href'])
        yield id

# UPDATE
################################################################################
def incremental_update(threads=1):
    latest_ids = scrape_latest()
    for id in latest_ids:
        status = insert_id(id, commit=False)

    if status.is_new:
        print('The Latest box didn\'t contain everything.')
        print('Need to check the categories for new sticks.')
        for category in CATEGORIES:
            ids = scrape_category(category)
            insert_ids(ids)
    else:
        print('No new sticks for incremental update.')

    cur = sql.execute('SELECT id FROM sticks WHERE retrieved IS NULL')
    ids = [row[0] for row in cur.fetchall()]

    sticks = scrape_directs(ids, threads=threads)
    try:
        insert_sticks(sticks)
    except KeyboardInterrupt:
        sql.commit()

def full_update(threads=1):
    for category in CATEGORIES:
        ids = scrape_category(category)
        insert_ids(ids)

    cur = sql.cursor()
    cur.execute('SELECT id FROM sticks ORDER BY retrieved ASC')
    ids = [row[0] for row in cur.fetchall()]

    sticks = scrape_directs(ids, threads=threads)
    try:
        insert_sticks(sticks)
    except KeyboardInterrupt:
        sql.commit()

# DOWNLOAD
################################################################################
def download_stick(id, overwrite=False, extract=False):
    directory = pathclass.Path('download').with_child(id)
    if directory.exists and not overwrite:
        return directory

    cur = sql.execute('SELECT download_link FROM sticks WHERE id == ?', [id])
    download_link = cur.fetchone()[0]
    filename = re.search(r'file=(.+)', download_link).group(1)
    filepath = directory.with_child(filename)

    DOWNLOAD_RATELIMITER.limit()
    print(f'Downloading {id}')
    response = request(download_link)

    directory.makedirs(exist_ok=True)
    with filepath.open('wb') as handle:
        handle.write(response.content)

    if extract and WINRAR is not None and filepath.extension == 'zip':
        # As much as I would like to use Python's zipfile module, I found that
        # some of the .zips on the site are actually rars.
        command = [
            WINRAR, 'x',
            '-o+', '-ibck',
            filepath.absolute_path,
            '*.*',
            directory.absolute_path + os.sep,
        ]
        subprocess.run(command)
        os.remove(filepath.absolute_path)

    return directory

def download_all(overwrite=False, extract=False):
    cur = sql.cursor()
    cur.execute('SELECT id FROM sticks')
    ids = [row[0] for row in cur.fetchall()]
    for id in ids:
        download_stick(id, overwrite=overwrite, extract=extract)

# COMMAND LINE
################################################################################

def update_argparse(args):
    if args.full:
        return full_update(threads=args.threads)
    else:
        return incremental_update(threads=args.threads)

def download_argparse(args):
    if args.extract and not WINRAR:
        raise Exception('The --extract flag requires you to have winrar on your path.')
    if len(args.ids) == 1 and args.ids[0] == 'all':
        return download_all(overwrite=args.overwrite, extract=args.extract)
    else:
        for id in args.ids:
            return download_stick(id, overwrite=args.overwrite, extract=args.extract)

def main(argv):
    parser = argparse.ArgumentParser(description='Scrape sticks from droidz.org.')
    subparsers = parser.add_subparsers()

    ################################################################################################

    p_update = subparsers.add_parser(
        'update',
        description='''
        Update the database with stick info.
        ''',
    )
    p_update.add_argument(
        '--full',
        dest='full',
        action='store_true',
        help='''
        Re-scrape all categories and all sticks to get fresh info.
        Otherwise, only new sticks will be scraped.
        ''',
    )
    p_update.add_argument(
        '--threads', dest='threads', type=int, default=1,
    )
    p_update.set_defaults(func=update_argparse)

    ################################################################################################

    p_download = subparsers.add_parser(
        'download',
        description='''
        Download the stick files.
        ''',
    )
    p_download.examples = [
        'all',
        '100 200 300 --overwrite',
    ]
    p_download.add_argument(
        'ids',
        nargs='+',
        default=None,
        help='''
        One or more stick IDs to download. You can use the word "all" to download
        all sticks.
        ''',
    )
    p_download.add_argument(
        '--overwrite',
        dest='overwrite',
        action='store_true',
        help='''
        Re-download any existing files. Otherwise they'll be skipped.
        ''',
    )
    p_download.add_argument(
        '--extract',
        dest='extract',
        action='store_true',
        help='''
        Extract downloaded zip files.
        NOTE: Some files on the site are labeled as .zip but are actually rars,
        so this extraction process requires you to have winrar on your PATH.
        Sorry.
        ''',
    )
    p_download.set_defaults(func=download_argparse)

    return betterhelp.go(parser, argv)

if __name__ == '__main__':
    raise SystemExit(main(sys.argv[1:]))