def test_a10_counter(self):
for i in range(30):
time.sleep(1)
if self.rpc.counter('5m', 'sum')\
.get('test_project', {}).get('success', 0) > 5:
break
rv = self.app.get('/counter?time=5m&type=sum')
self.assertEqual(rv.status_code, 200)
data = json.loads(utils.text(rv.data))
self.assertGreater(len(data), 0)
self.assertGreater(data['test_project']['success'], 3)
rv = self.app.get('/counter?time=1h&type=sum')
self.assertEqual(rv.status_code, 200)
data = json.loads(utils.text(rv.data))
self.assertGreater(len(data), 0)
self.assertGreater(data['test_project']['success'], 3)
rv = self.app.get('/counter?time=1d&type=sum')
self.assertEqual(rv.status_code, 200)
data = json.loads(utils.text(rv.data))
self.assertGreater(len(data), 0)
self.assertGreater(data['test_project']['success'], 3)
rv = self.app.get('/counter?time=all&type=sum')
self.assertEqual(rv.status_code, 200)
data = json.loads(utils.text(rv.data))
self.assertGreater(len(data), 0)
self.assertGreater(data['test_project']['success'], 3)
def test_a10_counter(self):
for i in range(30):
time.sleep(1)
if self.rpc.counter("5m", "sum").get("test_project", {}).get("success", 0) > 5:
break
rv = self.app.get("/counter?time=5m&type=sum")
self.assertEqual(rv.status_code, 200)
data = json.loads(utils.text(rv.data))
self.assertGreater(len(data), 0)
self.assertGreater(data["test_project"]["success"], 3)
rv = self.app.get("/counter?time=1h&type=sum")
self.assertEqual(rv.status_code, 200)
data = json.loads(utils.text(rv.data))
self.assertGreater(len(data), 0)
self.assertGreater(data["test_project"]["success"], 3)
rv = self.app.get("/counter?time=1d&type=sum")
self.assertEqual(rv.status_code, 200)
data = json.loads(utils.text(rv.data))
self.assertGreater(len(data), 0)
self.assertGreater(data["test_project"]["success"], 3)
rv = self.app.get("/counter?time=all&type=sum")
self.assertEqual(rv.status_code, 200)
data = json.loads(utils.text(rv.data))
self.assertGreater(len(data), 0)
self.assertGreater(data["test_project"]["success"], 3)
def _parse(data):
if six.PY3:
for key, value in list(six.iteritems(data)):
if isinstance(value, six.binary_type):
data[utils.text(key)] = utils.text(value)
else:
data[utils.text(key)] = value
return data
def test_50_get(self):
io = BytesIO()
self.webdav.download('handler.py', io)
self.assertEqual(utils.text(inspect.getsource(data_handler)), utils.text(io.getvalue()))
io.close()
io = BytesIO()
self.webdav.download('sample_handler.py', io)
self.assertEqual(utils.text(inspect.getsource(data_sample_handler)), utils.text(io.getvalue()))
io.close()
def test_50_get(self):
import easywebdav
with self.assertRaises(easywebdav.OperationFailed):
io = BytesIO()
self.webdav.download('handler.py', io)
io.close()
io = BytesIO()
self.webdav_up.download('handler.py', io)
self.assertEqual(utils.text(inspect.getsource(data_handler)), utils.text(io.getvalue()))
io.close()
def test_20_debug(self):
rv = self.app.get("/debug/test_project")
self.assertEqual(rv.status_code, 200)
self.assertIn(b"debugger", rv.data)
self.assertIn(b"var task_content = ", rv.data)
self.assertIn(b"var script_content = ", rv.data)
m = re.search(r"var task_content = (.*);\n", utils.text(rv.data))
self.assertIsNotNone(m)
self.__class__.task_content = json.loads(m.group(1))
m = re.search(r"var script_content = (.*);\n", utils.text(rv.data))
self.assertIsNotNone(m)
self.__class__.script_content = json.loads(m.group(1))
def test_20_debug(self):
rv = self.app.get('/debug/test_project')
self.assertEqual(rv.status_code, 200)
self.assertIn(b'debugger', rv.data)
self.assertIn(b'var task_content = ', rv.data)
self.assertIn(b'var script_content = ', rv.data)
m = re.search(r'var task_content = (.*);\n', utils.text(rv.data))
self.assertIsNotNone(m)
self.assertIn('test_project', json.loads(m.group(1)))
m = re.search(r'var script_content = (.*);\n', utils.text(rv.data))
self.assertIsNotNone(m)
self.assertIn('__START_URL__', json.loads(m.group(1)))
def test_20_debug(self):
rv = self.app.get('/debug/test_project')
self.assertEqual(rv.status_code, 200)
self.assertIn(b'debugger', rv.data)
self.assertIn(b'var task_content = ', rv.data)
self.assertIn(b'var script_content = ', rv.data)
m = re.search(r'var task_content = (.*);\n', utils.text(rv.data))
self.assertIsNotNone(m)
self.__class__.task_content = json.loads(m.group(1))
m = re.search(r'var script_content = (.*);\n', utils.text(rv.data))
self.assertIsNotNone(m)
self.__class__.script_content = (json.loads(m.group(1))
.replace('http://scrapy.org/',
'http://127.0.0.1:14887/pyspider/test.html'))
def test_20_debug(self):
rv = self.app.get("/debug/test_project")
self.assertEqual(rv.status_code, 200)
self.assertIn(b"debugger", rv.data)
self.assertIn(b"var task_content = ", rv.data)
self.assertIn(b"var script_content = ", rv.data)
m = re.search(r"var task_content = (.*);\n", utils.text(rv.data))
self.assertIsNotNone(m)
self.__class__.task_content = json.loads(m.group(1))
m = re.search(r"var script_content = (.*);\n", utils.text(rv.data))
self.assertIsNotNone(m)
self.__class__.script_content = json.loads(m.group(1)).replace(
"http://scrapy.org/", "http://127.0.0.1:14887/pyspider/test.html"
)
def load_tasks(self, status, project=None, fields=None):
if project is None:
project = self.projects
elif not isinstance(project, list):
project = [project, ]
if self.scan_available:
scan_method = self.redis.sscan_iter
else:
scan_method = self.redis.smembers
if fields:
def get_method(key):
obj = self.redis.hmget(key, fields)
if all(x is None for x in obj):
return None
return dict(zip(fields, obj))
else:
get_method = self.redis.hgetall
for p in project:
status_key = self._gen_status_key(p, status)
for taskid in scan_method(status_key):
obj = get_method(self._gen_key(p, utils.text(taskid)))
if not obj:
#self.redis.srem(status_key, taskid)
continue
else:
yield self._parse(obj)
def test_90_run(self):
time.sleep(0.5)
rv = self.app.post('/run', data={
'project': 'test_project',
})
self.assertEqual(rv.status_code, 200)
self.assertEqual(json.loads(utils.text(rv.data))['result'], True)
def _select2dic(self, tablename=None, what="*", where="", where_values=[],
order=None, offset=0, limit=None):
tablename = self.escape(tablename or self.__tablename__)
if isinstance(what, list) or isinstance(what, tuple) or what is None:
what = ','.join(self.escape(f) for f in what) if what else '*'
sql_query = "SELECT %s FROM %s" % (what, tablename)
if where:
sql_query += " WHERE %s" % where
if order:
sql_query += ' ORDER BY %s' % order
if limit:
sql_query += " LIMIT %d, %d" % (offset, limit)
elif offset:
sql_query += " LIMIT %d, %d" % (offset, self.maxlimit)
logger.debug("<sql: %s>", sql_query)
dbcur = self._execute(sql_query, where_values)
# f[0] may return bytes type
# https://github.com/mysql/mysql-connector-python/pull/37
fields = [utils.text(f[0]) for f in dbcur.description]
for row in dbcur:
yield dict(zip(fields, row))
def readonly(self):
projectdb = self.app.config["projectdb"]
if not projectdb:
return True
if (
"lock" in projectdb.split_group(self.project.get("group"))
and self.app.config.get("webui_username")
and self.app.config.get("webui_password")
):
authheader = self.environ.get("HTTP_AUTHORIZATION")
if not authheader:
return True
authheader = authheader[len("Basic ") :]
try:
username, password = text(base64.b64decode(authheader)).split(":", 1)
except Exception as e:
self.app.logger.error("wrong api key: %r, %r", authheader, e)
return True
if username == self.app.config["webui_username"] and password == self.app.config["webui_password"]:
return False
else:
return True
return False
def _parse(self, data):
for key, value in list(six.iteritems(data)):
if isinstance(value, (bytearray, six.binary_type)):
data[key] = utils.text(value)
if 'result' in data:
data['result'] = json.loads(data['result'])
return data
def crawl(self, url=None, track=None, **kwargs):
if url is None and kwargs.get('callback'):
url = dataurl.encode(utils.text(kwargs.get('callback')))
project_data = self.processor.project_manager.get(self.project_name)
assert project_data, "can't find project: %s" % self.project_name
instance = project_data['instance']
instance._reset()
task = instance.crawl(url, **kwargs)
if isinstance(task, list):
task = task[0]
task['track'] = track
result = self.fetcher.fetch(task)
self.processor.on_task(task, result)
status = None
while not self.status_queue.empty():
status = self.status_queue.get()
newtasks = []
while not self.newtask_queue.empty():
newtasks = self.newtask_queue.get()
result = None
while not self.result_queue.empty():
_, result = self.result_queue.get()
return status, newtasks, result
def doc(self):
"""Returns a PyQuery object of the response's content"""
if hasattr(self, '_doc'):
return self._doc
elements = self.etree
doc = self._doc = PyQuery(elements)
doc.make_links_absolute(utils.text(self.url))
return doc
def test_35_run_http_task(self):
rv = self.app.post('/debug/test_project/run', data={
'script': self.script_content,
'task': json.dumps(self.task_content2)
})
self.assertEqual(rv.status_code, 200)
data = json.loads(utils.text(rv.data))
self.assertIn(b'follows', rv.data)
def test_33_run_bad_script(self):
rv = self.app.post('/debug/test_project/run', data={
'script': self.script_content+'adfasfasdf',
'task': self.task_content
})
self.assertEqual(rv.status_code, 200)
data = json.loads(utils.text(rv.data))
self.assertGreater(len(data['logs']), 0)
self.assertEqual(len(data['follows']), 0)
def test_a20_tasks(self):
rv = self.app.get('/tasks')
self.assertEqual(rv.status_code, 200, rv.data)
self.assertIn(b'SUCCESS</span>', rv.data)
self.assertNotIn(b'>ERROR</span>', rv.data)
m = re.search(r'/task/test_project:[^"]+', utils.text(rv.data))
self.assertIsNotNone(m)
self.__class__.task_url = m.group(0)
self.assertIsNotNone(self.task_url)
m = re.search(r'/debug/test_project[^"]+', utils.text(rv.data))
self.assertIsNotNone(m)
self.__class__.debug_task_url = m.group(0)
self.assertIsNotNone(self.debug_task_url)
rv = self.app.get('/tasks?project=test_project')
self.assertEqual(rv.status_code, 200)
self.assertIn(b'SUCCESS</span>', rv.data)
self.assertNotIn(b'>ERROR</span>', rv.data)
def _parse(data):
for key, value in list(six.iteritems(data)):
if isinstance(value, six.binary_type):
data[key] = utils.text(value)
if "result" in data:
if isinstance(data["result"], bytearray):
data["result"] = str(data["result"])
data["result"] = json.loads(data["result"])
return data
def test_a15_queues(self):
rv = self.app.get('/queues')
self.assertEqual(rv.status_code, 200)
data = json.loads(utils.text(rv.data))
self.assertGreater(len(data), 0)
self.assertIn('scheduler2fetcher', data)
self.assertIn('fetcher2processor', data)
self.assertIn('processor2result', data)
self.assertIn('newtask_queue', data)
self.assertIn('status_queue', data)
def test_30_run(self):
rv = self.app.post('/debug/test_project/run', data={
'script': self.script_content,
'task': self.task_content
})
self.assertEqual(rv.status_code, 200)
data = json.loads(utils.text(rv.data))
self.assertIn(b'follows', rv.data)
self.assertGreater(len(data['follows']), 0)
self.__class__.task_content2 = data['follows'][0]
def __init__(self, path, environ, app, project=None):
super(ScriptResource, self).__init__(path, environ)
self.app = app
self.new_project = False
self._project = project
self.project_name = text(self.name)
self.writebuffer = None
if self.project_name.endswith('.py'):
self.project_name = self.project_name[:-len('.py')]
def _parse(self, data):
for key, value in list(six.iteritems(data)):
if isinstance(value, (bytearray, six.binary_type)):
data[key] = utils.text(value)
for each in ('schedule', 'fetch', 'process', 'track'):
if each in data:
if data[each]:
data[each] = json.loads(data[each])
else:
data[each] = {}
return data
def test_25_debug_post(self):
rv = self.app.post('/debug/test_project', data={
'project-name': 'other_project',
'start-urls': 'http://127.0.0.1:14887/pyspider/test.html',
'script-mode': 'script',
})
self.assertEqual(rv.status_code, 200)
self.assertIn(b'debugger', rv.data)
self.assertIn(b'var task_content = ', rv.data)
self.assertIn(b'var script_content = ', rv.data)
m = re.search(r'var task_content = (.*);\n', utils.text(rv.data))
self.assertIsNotNone(m)
self.assertIn('test_project', m.group(1))
self.__class__.task_content = json.loads(m.group(1))
m = re.search(r'var script_content = (.*);\n', utils.text(rv.data))
self.assertIsNotNone(m)
self.assertIn('127.0.0.1:14887', m.group(1))
self.__class__.script_content = json.loads(m.group(1))
def test_45_run_with_saved_script(self):
rv = self.app.post('/debug/test_project/run', data={
'webdav_mode': 'true',
'script': '',
'task': self.task_content
})
self.assertEqual(rv.status_code, 200)
data = json.loads(utils.text(rv.data))
self.assertIn(b'follows', rv.data)
self.assertGreater(len(data['follows']), 0)
self.__class__.task_content2 = data['follows'][0]
def _parse(self, data):
if six.PY3:
result = {}
for key, value in data.items():
if isinstance(value, bytes):
value = utils.text(value)
result[utils.text(key)] = value
data = result
for each in ('schedule', 'fetch', 'process', 'track'):
if each in data:
if data[each]:
data[each] = json.loads(data[each])
else:
data[each] = {}
if 'status' in data:
data['status'] = int(data['status'])
if 'lastcrawltime' in data:
data['lastcrawltime'] = float(data['lastcrawltime'] or 0)
if 'updatetime' in data:
data['updatetime'] = float(data['updatetime'] or 0)
return data
def _parse(data):
for key, value in list(six.iteritems(data)):
if isinstance(value, six.binary_type):
data[key] = utils.text(value)
for each in ("schedule", "fetch", "process", "track"):
if each in data:
if data[each]:
if isinstance(data[each], bytearray):
data[each] = str(data[each])
data[each] = json.loads(data[each])
else:
data[each] = {}
return data
def test_a60_fetch_via_cannot_connect_fetcher(self):
ctx = run.webui.make_context('webui', [
'--fetcher-rpc', 'http://localhost:20000/',
], self.ctx)
app = run.webui.invoke(ctx)
app = app.test_client()
rv = app.post('/debug/test_project/run', data={
'script': self.script_content,
'task': self.task_content
})
self.assertEqual(rv.status_code, 200)
data = json.loads(utils.text(rv.data))
self.assertGreater(len(data['logs']), 0)
self.assertEqual(len(data['follows']), 0)
def getMemberList(self):
members = []
for project in self.projectdb.get_all():
project_name = project['name']
if not project_name.endswith('.py'):
project_name += '.py'
native_path = os.path.join(self.path, project_name)
native_path = text(native_path) if six.PY3 else utf8(native_path)
members.append(ScriptResource(
native_path,
self.environ,
self.app,
project
))
return members
def test_a25_task_json(self):
rv = self.app.get(self.task_url + '.json')
self.assertEqual(rv.status_code, 200)
self.assertIn('status_string', json.loads(utils.text(rv.data)))
def search_blob_demo():
# 连接配置信息
mysql_config = {
'host': '192.168.1.244',
'port': 3306,
'user': '******',
'password': '******',
'db': 'resultdb',
'charset': 'utf8',
'cursorclass': pymysql.cursors.DictCursor,
}
# 创建连接
connection = pymysql.connect(**mysql_config)
# 执行sql语句
try:
with connection.cursor() as cursor:
# 执行sql语句，进行查询
sql = 'select * from boohee6'
# 获取查询结果
cursor.execute(sql)
data = cursor.fetchall()
wb = Workbook()
# 激活 worksheet
sheet = wb.create_sheet('薄荷网', 0)
# 可以附加行，从第一列开始附加
sheet.append(title_row_all)
num = 1
for j in range(len(SEARCH_LIST2)):
search_key = SEARCH_LIST2[j]
for i in range(len(data)):
for key, value in list(six.iteritems(data[i])):
if isinstance(value, (bytearray, six.binary_type)):
# data[key] = value.decode('utf8')
# 作用同上
data[i][key] = utils.text(value)
if 'result' in data[i]:
result = data[i]['result']
# 判断result类型为str
if isinstance(result, str):
data[i]['result'] = json.loads(data[i]['result'])
# 判断search_key 精准匹配name
if search_key != data[i]['result']['name']:
continue
else:
# 过滤type_row数据
if data[i]['result']['type'] in type_row:
continue
else:
search_key_title = []
search_key_title.append(search_key)
sheet.append(search_key_title)
num += 1
# 获取title行
title_cell = sheet.cell(None, num, 1)
title_cell.font = font1
# merge cells from num row
sheet.merge_cells(None, num, 1, num, 3)
contents = data[i]['result']['contents'].split(
'>>')[0].strip().split(' ')[2:]
content_all = []
for s in range(len(title_row)):
if title_row[s] in contents:
for z in range(len(contents)):
if title_row[s] == contents[z]:
content_all.append('' if contents[
z + 1] == '一' else contents[z +
1])
else:
content_all.append('')
content_all.insert(0, data[i]['result']['name'])
content_all.insert(1, data[i]['result']['type'])
sheet.append(content_all)
num += 1
# 保存文件
wb.save("薄荷网食物data-precise1.xlsx")
# 没有设置默认自动提交，需要主动提交，以保存所执行的语句
connection.commit()
finally:
connection.close()
def test_70_get(self):
io = BytesIO()
self.webdav.download('sample_handler.py', io)
self.assertEqual(utils.text(inspect.getsource(data_handler)),
utils.text(io.getvalue()))
io.close()
def on_task(self, task, response):
start_time = time.time()
try:
response = rebuild_response(response)
assert 'taskid' in task, 'need taskid in task'
project = task['project']
updatetime = task.get('updatetime', None)
project_data = self.project_manager.get(project, updatetime)
if not project_data:
logger.error("no such project: %s", project)
return False
ret = project_data['instance'].run(project_data['module'], task,
response)
except Exception as e:
logger.exception(e)
return False
process_time = time.time() - start_time
if not ret.extinfo.get('not_send_status', False):
if ret.exception:
track_headers = dict(response.headers)
else:
track_headers = {}
for name in ('etag', 'last-modified'):
if name not in response.headers:
continue
track_headers[name] = response.headers[name]
status_pack = {
'taskid': task['taskid'],
'project': task['project'],
'url': task.get('url'),
'track': {
'fetch': {
'ok':
response.isok(),
'redirect_url':
response.url
if response.url != response.orig_url else None,
'time':
response.time,
'error':
response.error,
'status_code':
response.status_code,
'encoding':
response.encoding,
'headers':
track_headers,
'content':
response.content[:500] if ret.exception else None,
},
'process': {
'ok':
not ret.exception,
'time':
process_time,
'follows':
len(ret.follows),
'result': (None if ret.result is None else utils.text(
ret.result)[:self.RESULT_RESULT_LIMIT]),
'logs':
ret.logstr()[-self.RESULT_LOGS_LIMIT:],
'exception':
ret.exception,
},
},
}
# FIXME: unicode_obj should used in scheduler before store to database
# it's used here for performance.
self.status_queue.put(utils.unicode_obj(status_pack))
# FIXME: unicode_obj should used in scheduler before store to database
# it's used here for performance.
if ret.follows:
self.newtask_queue.put(
[utils.unicode_obj(newtask) for newtask in ret.follows])
for project, msg, url in ret.messages:
self.inqueue.put(({
'taskid': utils.md5string(url),
'project': project,
'url': url,
'process': {
'callback': '_on_message',
}
}, {
'status_code': 200,
'url': url,
'save': (task['project'], msg),
}))
if response.error or ret.exception:
logger_func = logger.error
else:
logger_func = logger.info
logger_func(
'process %s:%s %s -> [%d] len:%d -> result:%.10r fol:%d msg:%d err:%r'
% (task['project'], task['taskid'], task.get('url'),
response.status_code, len(response.content), ret.result,
len(ret.follows), len(ret.messages), ret.exception))
return True
def test_x30_run(self):
rv = self.app.post('/run', data={
'project': 'test_project',
})
self.assertEqual(rv.status_code, 200)
self.assertEqual(json.loads(utils.text(rv.data))['result'], False)
def on_task(self, task, response):
'''Deal one task'''
start_time = time.time()
response = rebuild_response(response)
try:
assert 'taskid' in task, 'need taskid in task'
project = task['project']
updatetime = task.get('project_updatetime', None)
md5sum = task.get('project_md5sum', None)
project_data = self.project_manager.get(project, updatetime, md5sum)
assert project_data, "no such project!"
if project_data.get('exception'):
ret = ProcessorResult(logs=(project_data.get('exception_log'), ),
exception=project_data['exception'])
else:
ret = project_data['instance'].run_task(
project_data['module'], task, response)
except Exception as e:
logstr = traceback.format_exc()
ret = ProcessorResult(logs=(logstr, ), exception=e)
process_time = time.time() - start_time
if not ret.extinfo.get('not_send_status', False):
if ret.exception:
track_headers = dict(response.headers)
else:
track_headers = {}
for name in ('etag', 'last-modified'):
if name not in response.headers:
continue
track_headers[name] = response.headers[name]
status_pack = {
'taskid': task['taskid'],
'project': task['project'],
'url': task.get('url'),
'track': {
'fetch': {
'ok': response.isok(),
'redirect_url': response.url if response.url != response.orig_url else None,
'time': response.time,
'error': response.error,
'status_code': response.status_code,
'encoding': getattr(response, '_encoding', None),
'headers': track_headers,
'content': response.text[:500] if ret.exception else None,
},
'process': {
'ok': not ret.exception,
'time': process_time,
'follows': len(ret.follows),
'result': (
None if ret.result is None
else utils.text(ret.result)[:self.RESULT_RESULT_LIMIT]
),
'logs': ret.logstr()[-self.RESULT_LOGS_LIMIT:],
'exception': ret.exception,
},
'save': ret.save,
},
}
if 'schedule' in task:
status_pack['schedule'] = task['schedule']
# FIXME: unicode_obj should used in scheduler before store to database
# it's used here for performance.
self.status_queue.put(utils.unicode_obj(status_pack))
# FIXME: unicode_obj should used in scheduler before store to database
# it's used here for performance.
if ret.follows:
for each in (ret.follows[x:x + 1000] for x in range(0, len(ret.follows), 1000)):
self.newtask_queue.put([utils.unicode_obj(newtask) for newtask in each])
for project, msg, url in ret.messages:
try:
self.on_task({
'taskid': utils.md5string(url),
'project': project,
'url': url,
'process': {
'callback': '_on_message',
}
}, {
'status_code': 200,
'url': url,
'save': (task['project'], msg),
})
except Exception as e:
logger.exception('Sending message error.')
continue
if ret.exception:
logger_func = logger.error
else:
logger_func = logger.info
logger_func('process %s:%s %s -> [%d] len:%d -> result:%.10r fol:%d msg:%d err:%r' % (
task['project'], task['taskid'],
task.get('url'), response.status_code, len(response.content),
ret.result, len(ret.follows), len(ret.messages), ret.exception))
return True
def http_fetch(self, url, task):
'''HTTP fetcher'''
start_time = time.time()
self.on_fetch('http', task)
handle_error = lambda x: self.handle_error('http', url, task,
start_time, x)
# setup request parameters
fetch = self.pack_tornado_request_parameters(url, task)
task_fetch = task.get('fetch', {})
session = cookies.RequestsCookieJar()
# fix for tornado request obj
if 'Cookie' in fetch['headers']:
c = http_cookies.SimpleCookie()
try:
c.load(fetch['headers']['Cookie'])
except AttributeError:
c.load(utils.utf8(fetch['headers']['Cookie']))
for key in c:
session.set(key, c[key])
del fetch['headers']['Cookie']
if 'cookies' in fetch:
session.update(fetch['cookies'])
del fetch['cookies']
max_redirects = task_fetch.get('max_redirects', 5)
# we will handle redirects by hand to capture cookies
fetch['follow_redirects'] = False
# making requests
while True:
# robots.txt
if task_fetch.get('robots_txt', False):
can_fetch = yield self.can_fetch(
fetch['headers']['User-Agent'], fetch['url'])
if not can_fetch:
error = tornado.httpclient.HTTPError(
403, 'Disallowed by robots.txt')
raise gen.Return(handle_error(error))
try:
request = tornado.httpclient.HTTPRequest(**fetch)
# if cookie already in header, get_cookie_header wouldn't work
old_cookie_header = request.headers.get('Cookie')
if old_cookie_header:
del request.headers['Cookie']
cookie_header = cookies.get_cookie_header(session, request)
if cookie_header:
request.headers['Cookie'] = cookie_header
elif old_cookie_header:
request.headers['Cookie'] = old_cookie_header
except Exception as e:
logger.exception(fetch)
raise gen.Return(handle_error(e))
try:
response = yield gen.maybe_future(
self.http_client.fetch(request))
except tornado.httpclient.HTTPError as e:
if e.response:
response = e.response
else:
raise gen.Return(handle_error(e))
extract_cookies_to_jar(session, response.request, response.headers)
if (response.code in (301, 302, 303, 307)
and response.headers.get('Location')
and task_fetch.get('allow_redirects', True)):
if max_redirects <= 0:
error = tornado.httpclient.HTTPError(
599, 'Maximum (%d) redirects followed' %
task_fetch.get('max_redirects', 5), response)
raise gen.Return(handle_error(error))
if response.code in (302, 303):
fetch['method'] = 'GET'
if 'body' in fetch:
del fetch['body']
fetch['url'] = quote_chinese(
urljoin(fetch['url'], response.headers['Location']))
fetch['request_timeout'] -= time.time() - start_time
if fetch['request_timeout'] < 0:
fetch['request_timeout'] = 0.1
max_redirects -= 1
continue
result = {}
result['orig_url'] = url
result['content'] = response.body or ''
result['headers'] = dict(response.headers)
result['status_code'] = response.code
result['url'] = response.effective_url or url
result['time'] = time.time() - start_time
result['cookies'] = session.get_dict()
result['save'] = task_fetch.get('save')
if response.error:
result['error'] = utils.text(response.error)
if 200 <= response.code < 300:
logger.info("[%d] %s:%s %s %.2fs", response.code,
task.get('project'), task.get('taskid'), url,
result['time'])
else:
logger.warning("[%d] %s:%s %s %.2fs", response.code,
task.get('project'), task.get('taskid'), url,
result['time'])
raise gen.Return(result)
def puppeteer_fetch(self, url, task):
'''Fetch with puppeteer proxy'''
start_time = time.time()
self.on_fetch('puppeteer', task)
handle_error = lambda x: self.handle_error('puppeteer', url, task,
start_time, x)
# check puppeteer proxy is enabled
if not self.puppeteer_proxy:
result = {
"orig_url": url,
"content": "puppeteer is not enabled.",
"headers": {},
"status_code": 501,
"url": url,
"time": time.time() - start_time,
"cookies": {},
"save": task.get('fetch', {}).get('save')
}
logger.warning("[501] %s:%s %s 0s", task.get('project'),
task.get('taskid'), url)
raise gen.Return(result)
# setup request parameters
fetch = self.pack_tornado_request_parameters(url, task)
task_fetch = task.get('fetch', {})
for each in task_fetch:
if each not in fetch:
fetch[each] = task_fetch[each]
fetch['headless'] = "false" if "headless" not in fetch else fetch[
'headless']
# robots.txt
if task_fetch.get('robots_txt', False):
user_agent = fetch['headers']['User-Agent']
can_fetch = yield self.can_fetch(user_agent, url)
if not can_fetch:
error = tornado.httpclient.HTTPError(
403, 'Disallowed by robots.txt')
raise gen.Return(handle_error(error))
request_conf = {'follow_redirects': False}
request_conf['connect_timeout'] = fetch.get('connect_timeout', 20)
request_conf['request_timeout'] = fetch.get('request_timeout', 120) + 1
session = cookies.RequestsCookieJar()
if 'Cookie' in fetch['headers']:
c = http_cookies.SimpleCookie()
try:
c.load(fetch['headers']['Cookie'])
except AttributeError:
c.load(utils.utf8(fetch['headers']['Cookie']))
for key in c:
session.set(key, c[key])
del fetch['headers']['Cookie']
if 'cookies' in fetch:
session.update(fetch['cookies'])
del fetch['cookies']
request = tornado.httpclient.HTTPRequest(url=fetch['url'])
cookie_header = cookies.get_cookie_header(session, request)
if cookie_header:
fetch['headers']['Cookie'] = cookie_header
logger.info("%s", self.puppeteer_proxy)
# making requests
fetch['headers'] = dict(fetch['headers'])
headers = {}
headers['Content-Type'] = 'application/json; charset=UTF-8'
try:
request = tornado.httpclient.HTTPRequest(url=self.puppeteer_proxy,
method="POST",
headers=headers,
body=json.dumps(fetch),
**request_conf)
except Exception as e:
raise gen.Return(handle_error(e))
try:
response = yield gen.maybe_future(self.http_client.fetch(request))
except tornado.httpclient.HTTPError as e:
if e.response:
response = e.response
else:
raise gen.Return(handle_error(e))
if not response.body:
raise gen.Return(
handle_error(
Exception('no response from puppeteer: %r' % response)))
result = {}
try:
result = json.loads(utils.text(response.body))
assert 'status_code' in result, result
except Exception as e:
if response.error:
result['error'] = utils.text(response.error)
raise gen.Return(handle_error(e))
if result.get('status_code', 200):
logger.info("[%d] %s:%s %s %.2fs", result['status_code'],
task.get('project'), task.get('taskid'), url,
result['time'])
else:
logger.error("[%d] %s:%s %s, %r %.2fs", result['status_code'],
task.get('project'), task.get('taskid'), url,
result['content'], result['time'])
raise gen.Return(result)
def test_42_get(self):
rv = self.app.get('/debug/test_project/get')
self.assertEqual(rv.status_code, 200)
data = json.loads(utils.text(rv.data))
self.assertIn('script', data)
self.assertEqual(data['script'], self.script_content)
def test_x20_counter(self):
rv = self.app.get('/counter?time=5m&type=sum')
self.assertEqual(rv.status_code, 200)
self.assertEqual(json.loads(utils.text(rv.data)), {})
