def client_status(request, id):
if request.method == 'GET':
client = Client.objects.get(id=id)
scrapyd = ScrapydAPI(scrapyd_url(client.ip, client.port))
try:
scrapyd.list_projects()
return HttpResponse('1')
except:
return HttpResponse('0')
def ScrapydData(request):
from scrapyd_api import ScrapydAPI
scrapyd = ScrapydAPI('http://localhost:6800')
settings = {'DOWNLOAD_DELAY': 2}
# 运行一个爬虫
scrapyd.schedule('project_name', 'spider_name', settings=settings)
print(scrapyd.list_projects())
def get_scrapyd(client):
url = 'http://{ip}:{port}'.format(ip=client.ip, port=client.port)
try:
scrapyd = ScrapydAPI(url)
result = scrapyd.list_projects()
return scrapyd
except (ConnectionError, InvalidURL):
return False
def project_list(request, client_id):
"""
project deployed list on one client
:param request: request object
:param client_id: client id
:return: json
"""
if request.method == 'GET':
client = Client.objects.get(id=client_id)
scrapyd = ScrapydAPI(scrapyd_url(client.ip, client.port))
try:
projects = scrapyd.list_projects()
return JsonResponse(projects)
except ConnectionError:
return JsonResponse({'message': 'Connect Error'}, status=500)
class ScrapyAgent(object):
""" scrapy项目代理类 """
def __init__(self, server_url):
self.server_url = server_url
self.scrapyd_api = ScrapydAPI(server_url)
def __repr__(self):
return '<ScrapyAgent %s>' % self.server_url
@property
def server(self):
return self.server_url
def list_projects(self):
return self.scrapyd_api.list_projects()
def del_project(self, project_name):
try:
return self.scrapyd_api.delete_project(project_name)
except:
return False
def list_spiders(self, project_name):
return self.scrapyd_api.list_spiders(project_name)
def start_spider(self, project_name, spider_name):
return self.scrapyd_api.schedule(project_name, spider_name)
def cancel_spider(self, project_name, job_id):
return self.scrapyd_api.cancel(project_name, job_id)
def deploy(self, project_name: str, version: int,
egg_byte: BinaryIO) -> "Dict or bool":
spider_num = self.scrapyd_api.add_version(project_name, version,
egg_byte)
return {
'project': project_name,
'version': version,
'spiders': spider_num,
} if spider_num else False
def log_url(self, project_name, spider_name, job_id):
return '{}/logs/{}/{}/{}'\
.format(self.server_url, project_name, spider_name, job_id)
def job_status(self, project_name, job_id):
return self.scrapyd_api.job_status(project_name, job_id)
def project_list(request, node_id):
"""
project deployed list on one node
:param request: request object
:param node_id: node id
:return: json
"""
if request.method == 'GET':
node = Node.objects.get(id=node_id)
scrapyd = ScrapydAPI(scrapyd_url(node.node_ip, node.node_port))
try:
projects = scrapyd.list_projects()
lis = []
for project in projects:
lis.append({'spider_name': project})
return JsonResponse({'result': 1, 'lis': lis})
except ConnectionError:
return JsonResponse({'message': 'Connect Error'}, status=500)
class Spider(Resource):
def __init__(self, url):
self.scrapyd = ScrapydAPI(url)
def get(self):
data = []
for p in self.scrapyd.list_projects():
spiders = reduce_spiders(listjobs(self.scrapyd.list_jobs(p), p))
data.extend(spiders)
return data
def post(self, name):
p, s = name.split('.')
jobs = self.scrapyd.list_jobs(p)
for job in (jobs['running'] + jobs['pending']):
if job['spider'] == s:
return 'Already Running'
return self.scrapyd.schedule(p, s)
def start_hot_spider(self, request):
"""
启动每日热点的爬虫，需要验证admin权限
:param request: user_type:用户的身份权限
:return:
"""
data = request.data
user_type = data.get("user_type")
if user_type == "1": #admin权限,开启爬虫
# run_hot()
scrapyd = ScrapydAPI('http://localhost:6800') # 这里是去调用部署分布式爬虫
print(scrapyd.list_projects()) #获取爬虫项目名
# print(scrapyd.list_spiders('default'))#获取爬虫项目中的爬虫工程名
# print(scrapyd.list_jobs('default')) #获取爬虫项目中运行的爬虫工程信息
# print(scrapyd.list_versions('default'))# 获取爬虫项目中的版本
scrapyd.schedule('default', 'hotdaily') # 这里是启动爬虫
return Response("ok")
else:
return Response("failed")
def schedule():
scrapyd = ScrapydAPI("http://localhost:6800")
projects = scrapyd.list_projects()
print( projects)
list_start = []
list_end = []
while(True):
print ('start')
jobids = {}
for project in projects:
list_start.append(gevent.spawn(start, scrapyd, project))
donelist = gevent.joinall(list_start)
[jobids.update(i.value) for i in donelist]
print ('wait')
time.sleep(120)
print ('cancal')
for project,job in jobids.items():
list_end.append(gevent.spawn(end, scrapyd, project, job))
time.sleep(300)
print ('restart')
from scrapyd_api import ScrapydAPI
scrapyd = ScrapydAPI('http://139.224.233.92:6800')
#egg = open('some_egg.egg', 'rb')
#scrapyd.delete_version('zhihuuser','197ca468135511e9a5281dbc0d0a2d2c')
print(scrapyd.list_projects())
#print(scrapyd.list_spiders('zhihuuser'))
#print(scrapyd.job_status('zhihuuser','9d4d8078135511e9a5281dbc0d0a2d2c'))
#scrapyd.delete_version('zhihuuser','9d4d8078135511e9a5281dbc0d0a2d2c')
#scrapyd.schedule('zhihuuser', '9d4d8078135511e9a5281dbc0d0a2d2c')
class ScrapydProxy(SpiderServiceProxy):
'''
单个爬虫服务类
继承单个爬虫服务基类, 实现基类的功能
'''
def __init__(self, server):
self.spider_status_name_dict = {
SpiderStatus.PENDING: 'pending',
SpiderStatus.RUNNING: 'running',
SpiderStatus.FINISHED: 'finished'
}
super(ScrapydProxy, self).__init__(server) # super执行的是父类的方法
self.scrapyd_api = ScrapydAPI(self._scrapyd_url()) # 实例化ScrapydAPI
def _scrapyd_url(self):
return self.server # 得到scrapyd的url, 用到实现的get方法
def list_projects(self):
"""
获取指定scrapyd上的所有工程列表,返回工程名字符串列表,而get_project_list返回的是对象
:return:
"""
# 获取scrapyd上的所有工程列表
return self.scrapyd_api.list_projects()
def get_project_list(self):
"""
功能: 获取所有的爬虫工程列表
:return: 返回工程对象列表
"""
data = self.scrapyd_api.list_projects() # 获取scrapyd上的所有工程列表
result = []
if data:
for project_name in data:
project = Project() # 实例化工程对象
project.project_name = project_name
result.append(project)
return result
def delete_project(self, project_name):
"""
功能: scrapyd上删除指定工程
:param project_name: 工程名称
:return:
"""
try:
return self.scrapyd_api.delete_project(
project_name) # 返回状态, 工程存在, 删除后返回True
except:
return False
def get_slave_spider_list(self, project_name):
try:
data = self.scrapyd_api.list_spiders(
project_name) # 列出指定工程下所有的爬虫名称
return data if data else []
except:
return []
def get_spider_list(self, project_name):
"""
功能: 获取指定工程下的所有爬虫名称列表
:param project_name: 工程名称
:return: 返回爬虫实例对象列表
"""
try:
data = self.scrapyd_api.list_spiders(
project_name) # 列出指定工程下所有的爬虫名称
result = []
if data:
for spider_name in data:
spider_instance = SpiderInstance()
spider_instance.spider_name = spider_name
result.append(spider_instance)
return result
except:
return []
def get_daemon_status(self):
pass
def get_job_list(self, project_name, spider_status=None):
"""
从scrapyd中获取一个爬虫项目下面的所有蜘蛛任务状态
:param project_name: 爬虫项目名称
:param spider_status: 蜘蛛状态, 默认为None, 返回所有状态, 若传入状态值, 则返回某个状态
:return:
"""
result = {
SpiderStatus.PENDING: [],
SpiderStatus.RUNNING: [],
SpiderStatus.FINISHED: []
}
try:
data = self.scrapyd_api.list_jobs(project_name)
if data:
for _status in self.spider_status_name_dict.keys():
for item in data[self.spider_status_name_dict[_status]]:
start_time, end_time = None, None
if item.get('start_time'):
start_time = datetime.datetime.strptime(
item['start_time'], '%Y-%m-%d %H:%M:%S.%f')
if item.get('end_time'):
end_time = datetime.datetime.strptime(
item['end_time'], '%Y-%m-%d %H:%M:%S.%f')
result[_status].append(
dict(id=item['id'],
start_time=start_time,
end_time=end_time))
return result if not spider_status else result[spider_status]
except:
return result
def start_spider(self, project_name, spider_name):
"""
功能：启动指定工程下的指定爬虫
:param project_name: 工程名称
:param spider_name: 爬虫名称
:return: 返回启动的爬虫的id, 启动不成功, 返回None
"""
data = self.scrapyd_api.schedule(project_name,
spider_name,
pro_name=project_name)
return data if data else None
def cancel_spider(self, project_name, job_id):
"""
功能: 取消工程下的指定job
:param project_name: 工程名称 str
:param job_id: job_id str
:return: 成功取消, 返回True, 否则返回False
"""
data = self.scrapyd_api.cancel(project_name, job_id)
return data != None
def deploy(self, project_name, file_path):
"""
功能: 将上传的egg项目部署到scrapyd上
:param project_name: 工程名称 str
:param file_path: egg文件路径 str
:return: 成功返回字典型工程信息, 否则返回None
"""
egg = open(file_path, 'rb')
version = int(time.time())
spider_num = self.scrapyd_api.add_version(project_name,
int(time.time()), egg)
egg.close()
ret = {
'version': version,
'project': project_name,
'spiders': spider_num,
'node_name': socket.gethostname(),
'status': 'ok' if spider_num else 'error'
}
return str(ret) if spider_num else False
def log_url(self, project_name, spider_name, job_id):
"""
功能: 获取爬虫的日志
:param project_name: 工程名称 str
:param spider_name: 爬虫名称 str
:param job_id: job_id str
:return: 返回log日志文件的url str
"""
return self._scrapyd_url() + '/logs/%s/%s/%s.log' % (
project_name, spider_name, job_id)
def project_list(request, id):
if request.method == 'GET':
client = Client.objects.get(id=id)
scrapyd = ScrapydAPI(scrapyd_url(client.ip, client.port))
projects = scrapyd.list_projects()
return HttpResponse(json.dumps(projects))
# delversion.json: 此接口用来删除项目的某个版本
curl http://120.27.34.25:6800/delversion.json -d project=weibo -d version=v1
# 这里需要一个参数project，即项目名称,还需一个参数version，即项目的版本
# 返回结果：
{"status": "ok"}
# status代表请求执行情况，删除成功
# delproject.json：用来删除某个项目
curl http://120.27.34.25:6800/delproject.json -d project=weibo
# 这里需要一个参数project，即项目名称,
# 返回结果：
{"status": "ok"}
# status代表请求执行情况，删除成功
# 以上接口是Scrapyd所有的接口，可以直接请求HTTP接口，即可控制项目的部署、启动、运行等操作
# 5.ScrapyAPI的使用：Scrapyd API库对这些接口做了一层封装，核心原理和HTTP接口请求方式并无二致，只不过Python封装后的库更便捷
# 建立Scrapy API:
from scrapyd_api import ScrapydAPI
scrapyd = ScrapydAPI('http://120.27.34.25:6800')
# 调用它的方法来实现对应接口的操作，如部署的操作：
egg = open('weibo.egg', 'rb')
scrapyd.add_version('weibo', 'v1', egg)
# 这样就可以将项目打包为egg文件，然后把本地打包的Egg项目部署到远程Scrapyd
# Scrapyd API还实现了所有Scrapyd提供的API接口，名称都是相同的，参数也是相同的
# 调用list_projects方法即可列出Scrapyd中所有已部署的项目：
scrapyd.list_projects()
['weibo', 'zhihu']
# 详细操作可参考官方文档：http://python-scrapyd-api.readthedocs.io/
def scrapyd_api_test():
scrapyd = ScrapydAPI('http://localhost:6800')
print(scrapyd.list_projects())
class Scrapyd_Control(object):
def __init__(self):
scrapyd_url = input('请输入scrapyd地址： ')
project = input('请输入项目名称： ')
self.project = project
self.scrapyd = ScrapydAPI(scrapyd_url)
# 启动爬虫
def schedule(self):
spider = input('请输入爬虫名称： ')
return {
'project': self.project,
'spider': spider,
'jobid': self.scrapyd.schedule(self.project, spider)
}
start, run = schedule, schedule
# 取消爬虫
def cancel(self):
jobid = input('请粘贴要取消的爬虫jobid： ')
return self.scrapyd.cancel(self.project, jobid)
# 查看项目
def listprojects(self):
return self.scrapyd.list_projects()
# 查看爬虫
def listspiders(self):
return self.scrapyd.list_spiders(self.project)
# 列出所有jobs
def listjobs(self):
return self.scrapyd.list_jobs(self.project)
# 查看job状态
def jobstatus(self):
jobid = input('请粘贴要查看的jobid: ')
return self.scrapyd.job_status(self.project, jobid)
# 查看版本
def listversions(self):
return self.scrapyd.list_versions(self.project)
# 删除版本
def delversion(self):
version_name = input('请粘贴要删除的版本： ')
yes = input('是否确认删除该版本{}，请输yes否则回车跳过删除\n'.format(version_name))
if yes == 'yes':
return self.scrapyd.delete_version(self.project, version_name)
else:
pass
# 删除项目
def delproject(self):
yes = input('是否确认删除该项目{}，请输yes否则回车跳过删除\n'.format(self.project))
if yes == 'yes':
return self.scrapyd.delete_project(self.project)
else:
pass
# 列出所有命令
def help(self):
print("""
启动爬虫 schedule|start|run
取消爬虫 cancel
查看项目 listprojects
查看爬虫 listspiders
列出所有jobs listjobs
查看job状态 jobstatus
查看版本 listversions
删除版本 delversion
删除项目 deleproject
列出所有命令 help
""")
