def remove_all_version(request, project, client_id):
node = Node.objects.get(id=client_id)
scrapyd = ScrapydAPI(scrapyd_url(node.node_ip, node.node_port))
try:
versions = scrapyd.delete_project(project)
return JsonResponse(versions)
except ConnectionError:
return JsonResponse({'message': 'Connet Error'}, status=500)
def delete_project(project, url=DEFAULT_URL):
"""
@param project: scrapy project name
@param spider: spider name
@param url: the url which target scrapyd daemon listens on
@param settings: the settings dictionary
To schedule a spider run:
curl http://localhost:6800/schedule.json -d project=myproject -d spider=spider2
"""
scrapyd = ScrapydAPI(url)
return scrapyd.delete_project(project)
def delete_project(project, url=DEFAULT_URL):
"""
@param project: scrapy project name
@param spider: spider name
@param url: the url which target scrapyd daemon listens on
@param settings: the settings dictionary
To schedule a spider run:
curl http://localhost:6800/schedule.json -d project=myproject -d spider=spider2
"""
scrapyd = ScrapydAPI(url)
return scrapyd.delete_project(project)
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)
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)
class Schedular:
def __init__(self):
self._scrapyd = None
try:
self._scrapyd = ScrapydAPI('http://{}:{}'.format(
config['Scrapyd']['host'], config['Scrapyd']['port']))
except KeyError as e:
logger.error("{}: No such key exists - {}".format(
class_fullname(e), str(e)))
except Exception as e:
logger.error("{}: Failed to create a scrapyd object - {}".format(
class_fullname(e), str(e)))
def addversion(self,
project,
version,
egg_filename='pricewatch_bot-0.0.1-py3.7.egg'):
""" Scrapyd API: addversion - https://scrapyd.readthedocs.io/en/stable/api.html#addversion-json
"""
if not self._scrapyd:
logger.error(
"No scrapyd object find. Unable to add a new version.")
return None
num_of_spiders = None
try:
with open(os.path.join(settings.APP_DIST_DIRPATH, egg_filename),
'rb') as egg:
num_of_spiders = self._scrapyd.add_version(
project, version, egg)
except FileNotFoundError as e:
logger.error("{}: {}".format(class_fullname(e), str(e)))
except ScrapydResponseError as e:
logger.error("{}: Response error - {}".format(
class_fullname(e), str(e)))
except Exception as e:
logger.error("{}: Failed to add a version - {}".format(
class_fullname(e), str(e)))
else:
logger.info(
"version '{}' for project '{}' added/updated - {} spider(s)".
format(project, version, num_of_spiders))
# call API to create a version
response = requests.post(
'http://{}:{}/api/schedule/version/'.format(
config['PriceWatchWeb']['host'],
config['PriceWatchWeb']['port']),
json={
'project': project,
'version': version,
'status': settings.SCHEDULES_VERSION_STATUS_ADDED,
'added_at': str(datetime.now()),
'deleted_at': None,
})
if not response.ok:
logger.error(
"{} HTTP Error: Failed to add a version - {} - {}".format(
response.status_code, response.reason, response.text))
finally:
return num_of_spiders
def schedule(self, project, spider, **kwargs):
if not self._scrapyd:
logger.error("No scrapyd object find. Unable to schedule a job.")
return None
_jobid = str(uuid.uuid4())
kwargs['jobid'] = _jobid # a scrapyd parameter
kwargs['job_id'] = _jobid # passing to a spider
try:
_s = None # scrapy settings in dict. eg {'DOWNLOAD_DELAY': 2}
jobid = self._scrapyd.schedule(project,
spider,
settings=_s,
**kwargs)
except ScrapydResponseError as e:
logger.error("{}: Response error - {}".format(
class_fullname(e), str(e)))
except Exception as e:
logger.error("{}: Failed to schedule a job - {}".format(
class_fullname(e), str(e)))
else:
if jobid != _jobid:
logger.error(
"{}: Invalid jobid [enteredid vs returnedid] [{} vs {}] - {}"
.format(class_fullname(e), _jobid, jobid, str(e)))
else:
logger.info(
"new scheduled job '{}' for project '{}', spider '{}' has been set"
.format(jobid, project, spider))
# call API to create a job
response = requests.post(
'http://{}:{}/api/schedule/job/'.format(
config['PriceWatchWeb']['host'],
config['PriceWatchWeb']['port']),
json={
'job_id': jobid,
'project': project,
'spider': spider,
'version': kwargs.pop('_version', None),
'settings': _s,
'other_params': kwargs,
'status': settings.SCHEDULES_JOB_STATUS_PENDING,
})
if not response.ok:
logger.error(
"{} HTTP Error: Failed to add a new job - {} - {}".
format(response.status_code, response.reason,
response.text))
finally:
return jobid
def listjobs(self, project):
if not self._scrapyd:
logger.error("No scrapyd object find. Unable to list jobs.")
return None
jobs = None
try:
jobs = self._scrapyd.list_jobs(project)
except ScrapydResponseError as e:
logger.error("{}: Response error - {}".format(
class_fullname(e), str(e)))
except Exception as e:
logger.error("{}: Failed to list jobs - {}".format(
class_fullname(e), str(e)))
else:
logger.info("list of jobs for project '{}' - {}".format(
project, str(jobs)))
self._store_jobs(project, jobs)
finally:
return jobs
def _store_jobs(self, project, jobs):
""" parse jobs and store information into db
"""
if all(_j in jobs for _j in ['running', 'finished']):
for x in jobs['running']:
# call API to update a running job
response = requests.put(
'http://{}:{}/api/schedule/job/{}/'.format(
config['PriceWatchWeb']['host'],
config['PriceWatchWeb']['port'], x['id']),
json={
'job_id': x['id'],
'project': project,
'spider': x['spider'],
'start_time': x['start_time'],
'status': settings.SCHEDULES_JOB_STATUS_RUNNING,
})
if not response.ok:
logger.error(
"{} HTTP Error: Failed to update a running job - {} - {}"
.format(response.status_code, response.reason,
response.text))
for x in jobs['finished']:
# call API to update a finished job
response = requests.put(
'http://{}:{}/api/schedule/job/{}/'.format(
config['PriceWatchWeb']['host'],
config['PriceWatchWeb']['port'], x['id']),
json={
'job_id': x['id'],
'project': project,
'spider': x['spider'],
'start_time': x['start_time'],
'end_time': x['end_time'],
'status': settings.SCHEDULES_JOB_STATUS_FINISHED,
})
if not response.ok:
logger.error(
"{} HTTP Error: Failed to update a finished job - {} - {}"
.format(response.status_code, response.reason,
response.text))
def delversion(self, project, version):
""" delversion
"""
if not self._scrapyd:
logger.error("No scrapyd object find. Unable to delete version.")
return False
deleted = False
try:
deleted = self._scrapyd.delete_version(project, version)
except ScrapydResponseError as e:
logger.error("{}: Response error - {}".format(
class_fullname(e), str(e)))
except Exception as e:
logger.error("{}: Failed to delete version - {}".format(
class_fullname(e), str(e)))
else:
logger.info(
"successfully deleted project '{}' version '{}'".format(
project, version))
# update deleted version
response = requests.put(
'http://{}:{}/api/schedule/version/'.format(
config['PriceWatchWeb']['host'],
config['PriceWatchWeb']['port']),
json={
'project': project,
'version': version,
'status': settings.SCHEDULES_VERSION_STATUS_DELETED,
'deleted_at': str(datetime.now()),
})
if not response.ok:
logger.error(
"{} HTTP Error: Failed to update a deleted version - {} - {}"
.format(response.status_code, response.reason,
response.text))
finally:
return deleted
def delproject(self, project):
""" delproject
"""
if not self._scrapyd:
logger.error("No scrapyd object find. Unable to delete version.")
return False
deleted = False
try:
deleted = self._scrapyd.delete_project(project)
except ScrapydResponseError as e:
logger.error("{}: Response error - {}".format(
class_fullname(e), str(e)))
except Exception as e:
logger.error("{}: Failed to delete project - {}".format(
class_fullname(e), str(e)))
else:
logger.info("successfully deleted project '{}'".format(project))
# update deleted project
response = requests.put(
'http://{}:{}/api/schedule/version/'.format(
config['PriceWatchWeb']['host'],
config['PriceWatchWeb']['port']),
json={
'project': project,
'status': settings.SCHEDULES_VERSION_STATUS_DELETED,
'deleted_at': str(datetime.now()),
})
if not response.ok:
logger.error(
"{} HTTP Error: Failed to update deleted project - {} - {}"
.format(response.status_code, response.reason,
response.text))
finally:
return deleted
def close(self):
self._scrapyd.client.close()
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
""")
