关于Python命令行的NBA文字直播小工具

找了一圈NBA文字直播网站，发现手机版直播吧有现成的接口，直接返回json格式数据。那就是它了，听我慢慢道来。
首先在电脑浏览器打开手机版直播吧，我用的是chrome浏览器，在Network中可以看到，它不停地用GET方式请求http://bifen4m.qiumibao.com/json/list.htm，这个地址会返回当前正在进行的所有类型比赛的基本情况，根据其中的type字段过滤掉非NBA比赛就OK了。其中最重要的是ID字段，之后的所有操作都需要用到。返回的数据如下所示：
{
"code": "2760624",
"second": "10",
"list": [
{
"id": "96233",
"sdate": "2017-04-20",
"time": "10:30",
"url": "/zhibo/nba/2017/042096233.htm",
"type": "basketball",
"start": "2017-04-20 10:30",
"home_team": "勇士",
"visit_team": "开拓者",
"home_score": "106",
"visit_score": "81",
"period_cn": "第4节\n01:30",
"from": "dc.live",
"code": "373",
"update": "13:13:37",
"big_score_1": "",
"big_score_2": ""
},
... # 省略了其它比赛的信息
]}12345678910111213141516171819202122232425

获得所有正在进行的比赛ID后，点击某一场比赛，进入文字直播页面。首先请求http://dingshi4pc.qiumibao.com/livetext/data/cache/max_sid/XXXX/0.htm页面，其中XXXX是上一步获取的id，它会返回一个数字，即max_sid。然后判断该max_sid是否大于上次获取的该值，如果大于，表示有新的直播文字，否则表示没有。
如果max_sid大于上一次的值，通过请求http://bifen4pc2.qiumibao.com/json/XXXX/YYYY.htm（其中XXXX是今天的日期，格式为2017-04-20，YYYY是第一步中获取的id），返回这场比赛的基本情况，比如比分，是第几节等，如下所示：
{ "id": "96233", "home_team": "勇士", "visit_team": "开拓者", "home_score": "110", "visit_score": "81", "period_cn": "第4节结束", ...}123456789

最后，就可以获取直播的文字了。请求http://dingshi4pc.qiumibao.com/livetext/data/cache/livetext/XXXX/0/lit_page_2/YYYY.htm（其中XXXX是比赛id，YYYY是max_sid），它会返回最新的直播文字，其中包括一条或多条直播文字，如下所示：
[
{ "live_id": "8769977", "live_text": "@仙女最库阿-：库里正负值最高32我去！！！！", "home_score": "110", "visit_score": "81", "pid_text": "比赛结束", ...
}, ... # 可能有多条直播文字]1234567891011

可以看到，该请求返回的信息中没有比赛剩余时间、主队和客队等信息，所以每次获取直播文字之前，需要多一次请求，获得比赛的基本信息。
基本流程就是这样，非常简单，一共就四个GET请求，返回四串json，用requests库请求，然后解析搞定。
先定义一个Match类，表示当前正在进行的每一场比赛。
# match.pyclass Match:
def __init__(self, **kwargs):
self.id = kwargs['id']
self.home_team = kwargs['home_team']
self.visit_team = kwargs['visit_team']
self.home_score = kwargs['home_score']
self.visit_score = kwargs['visit_score']
self.period_cn = kwargs['period_cn'].replace('\n', ' ') def __repr__(self):
return '{self.id} {self.home_team} {self.home_score} - {self.visit_score} {self.visit_team} {self.period_cn}'.format(self=self)12345678910111213

再定义一个TextLiving类，表示获取的每一条文字直播。
# text_living.pyclass TextLiving:
def __init__(self, match_info, **kwargs):
self.home_team = match_info['home_team']
self.visit_team = match_info['visit_team']
self.period_cn = match_info['period_cn']
self.live_text = kwargs['live_text']
self.home_score = kwargs['home_score']
self.visit_score = kwargs['visit_score'] def __repr__(self):
return '{self.home_team} {self.home_score} - {self.visit_score} {self.visit_team} {self.period_cn}\n{self.live_text}\n{sep}'.format(self=self, sep='*'*60)12345678910111213

接着创建zhibo8_api.py模块，用于获取相关数据。
# 当前正在进行的比赛Living_Matches_Url = 'http://bifen4m.qiumibao.com/json/list.htm'# 某一场比赛当前的max_sidMatch_Max_Sid_Url = 'http://dingshi4pc.qiumibao.com/livetext/data/cache/max_sid/%s/0.htm'# 某一场比赛最新文字直播Match_Living_Text_Url = 'http://dingshi4pc.qiumibao.com/livetext/data/cache/livetext/%s/0/lit_page_2/%d.htm'# 某一场比赛当前的基本情况Match_Info_Url = 'http://bifen4pc2.qiumibao.com/json/%s/%s.htm'def get_living_matches():
response = requests.get(Living_Matches_Url)
result = json.loads(response.text)
matches = [Match(**match) for match in result['list'] if match['type'] == 'basketball' and match['period_cn'] != '完赛'] return matchesdef get_match_max_sid(match_id):
response = requests.get(Match_Max_Sid_Url % match_id) if response.status_code == requests.codes.ok: return int(response.text)def get_match_living(match_id, max_sid):
# 先获取比赛的当前情况，再获取最新文字直播
match_info = get_match_info(match_id)

response = requests.get(Match_Living_Text_Url % (match_id, max_sid))

texts = [] if response.status_code == requests.codes.ok:
result = json.loads(response.text)
texts = [TextLiving(match_info, **living) for living in result] return textsdef get_match_info(match_id):
today = datetime.now().strftime('%Y-%m-%d')
response = requests.get(Match_Info_Url % (today, match_id))
match_info = json.loads(response.text) return match_info123456789101112131415161718192021222324252627282930313233343536373839404142

最后，在main.py模块中启动程序，开始直播！
def get_living_matches():
matches = zhibo8_api.get_living_matches() for match in matches:
print(match) return matchesdef get_watch_match(matches):
match_id = input('请输入比赛ID：') for match in matches: if match.id == match_id: return match else:
print('输入的ID不正确') return Nonedef main_loop():
matches = get_living_matches() if len(matches) == 0:
print('当前没有比赛！！！') return

match = get_watch_match(matches) if not match:
print('没去找到该比赛') return

current_match_max_sid = -1
while True:
match_max_sid = zhibo8_api.get_match_max_sid(match.id) if not match_max_sid:
print('没有直播数据') return

if current_match_max_sid == match_max_sid: continue

current_match_max_sid = match_max_sid
text_livings = zhibo8_api.get_match_living(match.id, current_match_max_sid) for text in text_livings:
print(text)if __name__ == '__main__':
main_loop()