gistfile1.py

#!/usr/bin/env python
# -*- coding: utf8 -*-
import re
from collections import namedtuple

import requests


BukkitPluginUser = namedtuple('BukkitPluginUser', [
    'name',
    'url'
])


_bukkit_plugin_category = namedtuple('BukkitPluginCategory', [
    'slug',
    'name',
    'icon_path',
    'icon_name'
])


class BukkitPluginCategory(_bukkit_plugin_category):
    def icon(self, width=32, height=32):
        return '{icon_path}{width}x{height}{icon_name}.-m1.png'.format(
            icon_path=self.icon_path,
            width=width,
            height=height,
            icon_name=self.icon_name
        )


class BukkitPlugin(object):
    def __init__(self, j):
        self._j = j

    @property
    def last_updated(self):
        return self._j.get('last_updated')

    @property
    def title(self):
        return self._j.get('title')

    @property
    def users(self):
        for user in self._j.get('users', []):
            yield BukkitPluginUser(**user)

    @property
    def curseforge_slug(self):
        return self._j.get('curseforge_slug')

    @property
    def categories(self):
        for category in self._j.get('categories', []):
            yield BukkitPluginCategory(*category)

    @property
    def url(self):
        return 'http://dev.bukkit.org/server-mods/{slug}'.format(
            slug=self.curseforge_slug
        )


class BukkitPBO(object):
    """
    Utility class for interacting with the Bukkit plugins website.
    """
    _TOKEN_R = re.compile(r'aJt[^=]*=[^"]"([^"]+)')
    _DATA_URL = 'http://plugins.bukkit.org/curseforge/data.php'

    @classmethod
    def token(cls):
        """
        Returns the token embedded in the page required to make the
        data.php request.
        """
        # Not entirely sure what the point of this is - some cheezy attempt
        # at "thwarting" crawlers?
        r = requests.get('http://plugins.bukkit.org/curseforge/')

        token = cls._TOKEN_R.search(r.content)
        if token:
            return token.group(1)

    @classmethod
    def plugins(cls, page=1, title='', tag='all', author=''):
        token = cls.token()
        max_page = 2

        while page < max_page:
            r = requests.post(cls._DATA_URL, params={
                'title': title,
                'tag': tag,
                'author': author,
                'pageno': page,
                'j': token
            }, headers={
                'Host': 'plugins.bukkit.org',
                'Referer': 'http://plugins.bukkit.org/curseforge/',
                'Origin': 'http://plugins.bukkit.org',
                'Content-Type': 'application/x-www-form-urlencoded',
                'X-Requested-With': 'XMLHttpRequest'
            })
            j = r.json

            max_page = j['subinfo']['pagemax']
            for result in j['realdata']:
                yield BukkitPlugin(result)

            page += 1


# Example:
for plugin in BukkitPBO.plugins():
    for category in plugin.categories:
        print category.icon()