init

.gitignore

@@ -0,0 +1 @@
+!offline_data/README.md

README.md

@@ -0,0 +1,17 @@
+## 北京实时公交
+
+该项目是北京实时公交查询接口的Python绑定，接口逆向自其[安卓客户端](http://www.bjjtw.gov.cn/topic/bjssgj/)
+
+### Quick Start
+
+* `pip install -r requirements.txt` 安装依赖
+* `python manage.py build_cache` 获取离线数据，建立本地缓存
+
+项目自带了一个终端中的查询工具作为例子，运行： `python manage.py query`
+
+### Roadmap
+
+- [x] 实现 beijing_bus 模块，提供需要的Python接口
+- [x] 终端中的查询工具
+- [ ] webapp
+- [ ] 微信公众号

beijing_bus/__init__.py

@@ -0,0 +1,20 @@
+#-*-coding:utf-8-*-
+
+from .cache import cache
+from .line import BusLine
+
+
+class BeijingBus(object):
+
+    @classmethod
+    def build_cache(cls):
+        cache.invalidate(hard=True)
+        cls.get_all_lines()
+
+    @classmethod
+    def get_all_lines(cls):
+        return BusLine.get_all_lines()
+
+    @classmethod
+    def search_lines(cls, keyword):
+        return list(BusLine.search(keyword))

beijing_bus/api.py

@@ -0,0 +1,38 @@
+#-*-coding:utf-8-*-
+
+import logging
+import requests
+import xmltodict 
+
+
+API_ENDPOINT = 'http://mc.aibang.com/aiguang/bjgj.c'
+REALTIME_ENDPOINT = 'http://bjgj.aibang.com:8899/bus.php'
+
+
+def request_api(url, params):
+    r = requests.get(url, params=params, headers={'cid':1024})
+    return xmltodict.parse(r.text)
+
+
+def get_line_update_state():
+    logging.debug('Getting all lines')
+    params = {'m': 'checkUpdate', 'version': '1'}
+    return request_api(API_ENDPOINT, params)
+
+
+def get_bus_offline_data(line_id):
+    logging.debug('Fetching line: %s' % line_id)
+    params = {'m': 'update', 'id': line_id}
+    return request_api(API_ENDPOINT, params)
+
+
+def get_realtime_data(line_id, station_num):
+    params = {
+        'city': '北京',
+        'id': line_id,
+        'no': station_num,
+        'type': 2,
+        'encrpt': 1,
+        'versionid': 2
+    }
+    return request_api(REALTIME_ENDPOINT, params)

beijing_bus/cache.py

@@ -0,0 +1,23 @@
+#-*-coding:utf-8-*-
+
+"""
+为了支持多种离线方案，使用了 dogpile.cache，支持 本地文件(默认), memcached, redis 等
+参见文档：http://dogpilecache.readthedocs.org/en/latest/api.html#module-dogpile.cache.backends.memory
+"""
+
+from dogpile.cache import make_region
+
+
+cache_config_null = {
+    'backend': 'dogpile.cache.null'
+}
+
+cache_config_file = {
+    'backend': 'dogpile.cache.dbm',
+    'arguments': {
+        'filename': 'offline_data/bus_offline_data.dbm'
+    }
+}
+
+
+cache = make_region().configure(**cache_config_file)

beijing_bus/cipher.py

@@ -0,0 +1,57 @@
+#-*-coding:utf-8-*-
+
+"""
+
+加密算法的实现 (实际上是RC4)
+
+http://zh.wikipedia.org/wiki/RC4
+
+>>> cipher = Cipher('aibang127')
+>>> print cipher.decrypt('N+j17kGHx8Mj')
+运通122
+
+"""
+
+import copy
+import base64
+import hashlib
+
+
+class Cipher(object):
+
+    def __init__(self, key):
+        k = bytearray(self._md5(key))
+        self.s_box = self._get_s_box(k)
+
+    def _md5(self, data):
+        return hashlib.md5(data).hexdigest()
+
+    def _get_s_box(self, key):
+        s_box = range(256)
+        j = 0
+        for i in range(256):
+            j = (j + s_box[i] + key[i % len(key)]) % 256
+            s_box[i], s_box[j] = s_box[j], s_box[i]
+        return s_box
+
+    def calc(self, data):
+        s_box = copy.copy(self.s_box)
+        results = range(len(data))
+        j = 0
+        for i in range(len(data)):
+            k = (i + 1) % 256
+            j = (j + s_box[k]) % 256
+            s_box[j], s_box[k] = s_box[k], s_box[j]
+            n = (s_box[j] + s_box[k]) % 256
+            results[i] = data[i] ^ s_box[n]
+        return results
+
+    def decrypt(self, message, decode=base64.b64decode):
+        message = decode(message)
+        data = bytearray(message)
+        return str(bytearray(self.calc(data)))
+
+    def encrypt(self, message, encode=base64.b64encode):
+        message = encode(message)
+        data = bytearray(message)
+        return str(bytearray(self.calc(data)))

beijing_bus/line.py

@@ -0,0 +1,97 @@
+#-*-coding:utf-8-*-
+
+from datetime import datetime
+
+from . import api
+from .cache import cache
+from .cipher import Cipher
+from .station import BusStation
+
+
+class BusLine(object):
+
+    def __init__(self, **kwargs):
+        self.stations = []
+        for name, value in kwargs.iteritems():
+            setattr(self, name, value)
+
+    def __repr__(self):
+        return '<Line: %s>' % self.name
+
+    @classmethod
+    @cache.cache_on_arguments()
+    def get(cls, id):
+        resp_doc = api.get_bus_offline_data(id)
+        busline = resp_doc['root']['busline']
+
+        key = 'aibang%s' % id
+        cipher = Cipher(key)
+        d = cipher.decrypt  # for convenience
+
+        line = cls(**{
+            'id': int(busline['lineid']),
+            'name': d(busline['linename']),
+            'short_name': d(busline['shotname']),
+            'distince': float(busline['distince']),
+            'ticket_type': busline['ticket'],
+            'price': float(busline['totalPrice']),
+            'running_time': busline['time'],
+            'line_type': busline['type'],
+            'coords': d(busline['coord']),
+            'status': busline['status'],
+            'version': busline['version']
+        })
+
+        for station_data in busline['stations']['station']:
+            name, no, lon, lat = map(d, station_data.values())
+            station = BusStation(name, float(lat), float(lon))
+            line.stations.append(station)
+
+        return line
+
+    @classmethod
+    def gets(cls, ids):
+        return [cls.get(id) for id in ids]
+
+    @classmethod
+    @cache.cache_on_arguments()
+    def get_all_line_ids(cls):
+        resp_doc = api.get_line_update_state()
+        root = resp_doc['root']
+        line_ids = [line['id'] for line in root['lines']['line']]
+        return line_ids
+
+    @classmethod
+    def get_all_lines(cls):
+        line_ids = cls.get_all_line_ids()
+        return cls.gets(line_ids)
+
+    @classmethod
+    def search(cls, name):
+        for line in cls.get_all_lines():
+            if name in line.name:
+                yield line
+
+    def get_realtime_data(self, station_num):
+
+        def t(ts):
+            if float(ts) > -1:
+                return datetime.fromtimestamp(float(ts))
+
+        data = []
+        resp_doc = api.get_realtime_data(self.id, station_num)
+        for bus_data in resp_doc['root']['data']['bus']:
+            key = 'aibang%s' % bus_data['gt']
+            d = Cipher(key).decrypt
+            data.append({
+                'id': int(bus_data['id']),
+                'lat': float(d(bus_data['x'])),
+                'lon': float(d(bus_data['y'])),
+                'next_station_name': d(bus_data['ns']),
+                'next_station_num': int(d(bus_data['nsn'])),
+                'next_station_distance': float(bus_data['nsd']),
+                'next_station_arriving_time': t(bus_data['nst']),
+                'station_distance': float(d(bus_data['sd'])),
+                'station_arriving_time': t(d(bus_data['st'])),
+            })
+        return data

beijing_bus/station.py

@@ -0,0 +1,21 @@
+#-*-coding:utf-8-*-
+
+
+class BusStation(object):
+
+    def __init__(self, name, lat, lon):
+        self.name = name
+        self.lat = lat
+        self.lon = lon
+
+    def __eq__(self, other):
+        return self.name == other.name
+
+    def __repr__(self):
+        return '<Station %s>' % self.name
+
+    def get_num_in_a_line(self, line):
+        try:
+            return line.stations.index(self) + 1
+        except ValueError:
+            return 0

manage.py

@@ -0,0 +1,73 @@
+#-*-coding:utf-8-*-
+
+import time
+import click
+from datetime import datetime
+from beijing_bus import BeijingBus
+
+
+@click.group()
+def cli():
+    pass
+
+
+@click.command(help='build or re-build the cache')
+def build_cache():
+    import logging
+    logging.basicConfig(level=logging.DEBUG)
+    BeijingBus.build_cache()
+    click.secho('Done!', fg='green')
+
+
+def echo_realtime_data(line, station_num):
+    station = line.stations[station_num-1]
+    realtime_data = line.get_realtime_data(station_num)
+    click.clear()
+    now = datetime.now().strftime('%H:%M:%S')
+    title = '实时数据 [%s]  线路：%s  (每5秒自动刷新，更新于%s)' % (station.name, line.name, now)
+    click.secho(title, fg='green', bold=True)
+    click.echo()
+    realtime_data = filter(lambda d: d['station_arriving_time'], realtime_data)
+    realtime_data.sort(key=lambda d: d['station_arriving_time'])
+    for i, data in enumerate(realtime_data):
+        click.secho('公交%s：' % (i+1), bold=True, underline=True)
+        click.echo('距离 %s 还有%s米' % (station.name, data['station_distance']))
+        click.echo('预计 %s 到达' % data['station_arriving_time'].strftime('%H:%M'))
+        click.echo()
+
+
+@click.command()
+def query():
+    q = click.prompt('请输入线路名', value_proc=str)
+    lines = BeijingBus.search_lines(q)
+    for index, line in enumerate(lines):
+        click.echo()
+        click.secho('[%s] %s' % (index+1, line.name), bold=True, underline=True)
+        station_names = [s.name for s in line.stations]
+        click.echo()
+        click.echo('站点列表：%s' % ','.join(station_names))
+
+    click.echo()
+    q = click.prompt('请从结果中选择线路编号', type=int)
+
+    line = lines[q-1]
+    click.clear()
+    click.echo('你选择了 %s，下面请选择站点' % line.name)
+    click.echo()
+    for index, station in enumerate(line.stations):
+        click.echo('[%s] %s' % (index+1, station.name))
+
+    click.echo()
+    q = click.prompt('请从结果中选择线路编号', type=int)
+
+    while True:
+        echo_realtime_data(line, q)
+        time.sleep(5)
+
+
+cli.add_command(build_cache)
+cli.add_command(query)
+
+
+if __name__ == '__main__':
+    cli()

requirements.txt

@@ -0,0 +1,4 @@
+click
+requests
+xmltodict
+dogpile.cache

webapp/__init__.py

@@ -0,0 +1,2 @@
+#-*-coding:utf-8-*-
+