import logging
import os

from grab.spider import Spider, Task


class JavaGamesSpider(Spider):
    initial_urls = ['http://dc.sever.ru/?d=19']

    def task_initial(self, grab, task):
        for elem in grab.doc.select('/html/body/div[3]//dt/a'):
            name = ' '.join(elem.text().split()[:-1])

            dir_name = './data/' + name
            if not os.path.exists(dir_name):
                os.makedirs(dir_name)

            yield Task('dir', url='http://dc.sever.ru' + elem.attr('href'), dir=name)

    def task_dir(self, grab, task):
        for elem in grab.doc.select('/html/body/div[3]//dt/a'):
            name = ' '.join(elem.text().split()[:-1])

            dir_name = './data/' + task.dir + '/' + name
            if not os.path.exists(dir_name):
                os.makedirs(dir_name)

            yield Task('category', url='http://dc.sever.ru' + elem.attr('href'), dir=name, pre_dir=task.dir)

    def task_category(self, grab, task):
        for elem in grab.doc.select('/html/body/div[3]//dt/a'):
            name = ' '.join(elem.text().split()[:-1])

            dir_name = './data/' + task.pre_dir + '/' + task.dir + '/' + name
            if not os.path.exists(dir_name):
                os.makedirs(dir_name)

            yield Task('files', url='http://dc.sever.ru' + elem.attr('href'), path=dir_name)

    def task_files(self, grab, task):
        for elem in grab.doc.select('/html/body/div[4]/dl//a'):
            name = elem.text()

            dir_name = task.path + '/' + name
            if not os.path.exists(dir_name):
                os.makedirs(dir_name)

            yield Task('file', url='http://dc.sever.ru' + elem.attr('href'), dir=dir_name)

        url_next = grab.doc.select('/html/body/div[5]/a')
        if url_next.exists():
            for link in url_next:
                if link.text() == 'Далее->':
                    yield Task('files', url='http://dc.sever.ru' + link.attr('href'), path=task.path)

    def task_file(self, grab, task):
        images = grab.doc.select('/html/body/div[2]/a[starts-with(@href, "index")]')
        end_index = 1
        if images.exists():
            for item in images:
                end_index = item.text()

        image = grab.doc.select('/html/body/div[2]/img[2]')
        if image.exists():
            fid = task.url.split('=')[2].replace('&order', '')
            for index in range(1, int(end_index) + 1):
                n = index - 1
                path = task.dir + '/screenshot_%d.jpg' % n
                if not os.path.exists(path):
                    yield Task('screen', url='http://dc.sever.ru/scr.php?c=%s&s=%d' % (fid, n), path=task.dir, nid=n)

        files = grab.doc.select('/html/body/div[3]/a[starts-with(@href, "/get")]')
        if files.exists():
            for file in files:
                file_type = file.text().replace('Скачать(', '').replace(')', '').lower()
                path = task.dir + '/game.%s' % file_type
                if not os.path.exists(path):
                    yield Task('game', url='http://dc.sever.ru' + file.attr('href'), path=task.dir, type=file_type)

        description = grab.doc.select('/html/body/div[2]')
        description = description.text().split('Описание:')[1]

        stop_words = ['От Sever.Ru:', 'Рейтинг (max10):', 'Разработчик:', 'Версия:', 'Год:', 'Язык:', 'Добавлено:']
        for item in stop_words:
            description = description.split(item)[0]

        f = open(task.dir + '/description.txt', 'w')
        f.write(description.strip())
        f.close()

    def task_game(self, grab, task):
        path = task.path + '/game.%s' % task.type
        grab.response.save(path)

    def task_screen(self, grab, task):
        path = task.path + '/screenshot_%d.jpg' % task.nid
        grab.response.save(path)

if __name__ == '__main__':
    logging.basicConfig(level=logging.DEBUG)
    bot = JavaGamesSpider(thread_number=2)
    bot.run()