def call_only_one_page_for_getting_result_demo(self):
api_crawl_rules_link = self.service.api_crawl_rules_link
form_rules_link = self.service.form_rules_link
api_parameters = self.service.candidate[self.service.main_sec_id]
api_url = self.service.api_url
if api_crawl_rules_link and api_parameters:
try:
form_rules_link_ex = api_crawl_rules_link.split(
'/statics/', 1)[1]
strss = read_file_as_str("static/" + form_rules_link_ex)
api_crawl_rules_two = json.loads(strss)
# api_crawl_rules_two = requests.get(api_crawl_rules_link).json()
self.api_crawl_rules = api_crawl_rules_two[
self.service.main_sec_id]
except:
return "Error by service details"
try:
self.crawl = Crawler(app.logger)
self.crawl.max_page_set = 1
self.crawl.max_page_set_flag = True # 设置页数为1
self.set_required_parameter() # 设置必要的参数
self.wait_condition() # 设置加载网页的等待条件
if form_rules_link:
form_rules_link_ex = form_rules_link.split('/statics/',
1)[1]
strss = read_file_as_str("static/" + form_rules_link_ex)
self.crawl.form_list = json.loads(strss)
# self.crawl.form_list = requests.get(form_rules_link).json()
time_out_judge = self.crawl.segment(api_url)
if time_out_judge == "timeout":
return "Error by crawler timeout"
except Exception as e:
shutdown_crawl(self.crawl)
app.logger.error(repr(e))
return "Error by unknown crawler problem"
results = self.analyze_crawer_result()
try: ## track 1: dot cant be a key in mongodb,for example {"3.4":dsada}
numb = 3
if len(results) < 3:
numb = len(results)
result_example = results[:numb]
self.service.update(api_result_example=result_example)
return "Success"
except:
return "Error by update"
else:
return "Error by missing service message"
def get_static_resource_one(string):
if string.endswith("json"):
return read_file_as_str("static/" + string), 200, {'Content-Type': 'application/json;chaset=utf-8'}
elif string.endswith("png"):
with open("static/" + string, 'rb') as f:
image = f.read()
return Response(image, mimetype="image/png")
else:
return read_file_as_str("static/" + string)
def url_construct(api_id):
'''
服务调用的接口，调用示例：http://service.cheosgrid.org:8089/call_service/1?__max_page=2&time=20181023&link_2.describe=描述
:api http://service.cheosgrid.org:8089/call_service/<api_id>
:param api_id: int 是api的api_id信息
超参数:__max_page 用以设置爬取的最大页数，该值为整数，且最小为1，设置为小于1的值将不起作用；设置为其他类型将返回错误信息
返回属性中的参数：
- *类型1：* 作为一级key值存在的属性，可直接将该key值作为一个请求参数
- *类型2：* 作为多级下的key值存在的属性，将各级一次以“.”连接
对应上述调用示例,请求参数有三个：，time，link_2.describe；对应的返回类型数据示例为:
{
"time":"20181021",
"link_2":{
"describe":"描述",
"image":"http:XXX.png"
}
}
:return:
{'status': 200, 'msg': '请求成功,共处理x页,其中第2,5页处理失败.' 或者 '请求成功,共处理x页.', 'data': results}
{'status': 201, 'msg': '缺失api_crawl_rules_link文件或candidate项，无法处理网页', 'data': ""}
{'status': 403, 'msg': '请求失败，服务器请求爬虫规则失败', 'data': ''}
{'status': 410, 'msg': '爬虫失败，超时或请求错误', 'data': ''}
{'status': 411, 'msg': '解析失败，网页信息错误', 'data': ''}
{'status': 425, 'msg': '爬虫超时', 'data': ''}
{'status': 426, 'msg': '缺少必须的参数: __parameters', 'data': ''}
**ATTENTION - 注意**：
返回参数的key不能存在"."。 一方面是此处实现的要求，将"."作为级别的依据；另外一方面mongodb存储时要求key不能存在"."。
'''
request_parameters_ori = list(request.args.keys())
request_parameters_ori_c = []
#TODO remove ""'skeys
for request_parameter_ori_ev in request_parameters_ori:
if request.args.get(request_parameter_ori_ev) and request.args.get(
request_parameter_ori_ev) != "":
request_parameters_ori_c.append(request_parameter_ori_ev)
request_parameters_ori = request_parameters_ori_c
request_parameters_ori = [
request_parameter.split('.')
for request_parameter in request_parameters_ori
]
service = Api.objects(api_id=api_id).first()
# TODO
# judge the system parameters like frame and others.
api_crawl_rules_link = service.api_crawl_rules_link
form_rules_link = service.form_rules_link
api_url = service.api_url
static_exp_link = service.static_exp_link
candidate_parameters = service.api_request_parameters_candidate # 可以使用的参数集合
candidate_parameters = [
candidate_parameter["query_name"]
for candidate_parameter in candidate_parameters
]
candidate_parameters = [
candidate_parameter.split(".")
for candidate_parameter in candidate_parameters
]
candidate_required_parameters = [
para["query_name"] for para in service.api_request_parameters_candidate
if para["required"]
]
request_parameters_for_judge_required = list(request.args.keys())
if not set(candidate_required_parameters).issubset(
set(request_parameters_for_judge_required)):
not_have_required = ",".join(
list(
set(candidate_required_parameters).difference(
set(request_parameters_for_judge_required))))
return ReNum.PARAM_LACK.update('', ": " + not_have_required).value()
# 可使用的请求参数集合 实际请求的请求参数集合
request_parameters = [
i for i in request_parameters_ori if i in candidate_parameters
]
ret_cha = [
".".join(i) for i in request_parameters_ori
if i not in candidate_parameters
]
api_parameters_double_list = []
api_parameters = []
section_name = []
for sec_id in service.main_sec_id:
api_parameters = api_parameters + service.candidate[sec_id]["items"]
api_parameters_double_list.append(service.candidate[sec_id]["items"])
section_name.append(service.candidate[sec_id]["section_name"])
# api_parameters.sort(key=lambda k: k["id"])
api_crawl_rules_double_list = []
try:
if api_crawl_rules_link:
api_crawl_rules_two = requests.get(api_crawl_rules_link).json()
for sec_id in service.main_sec_id:
api_crawl_rules_double_list.append(api_crawl_rules_two[sec_id])
except:
return ReNum.REQUEST_ERROR.value()
if api_crawl_rules_link and api_parameters:
try:
crawl = Crawler(app.logger)
## level_0 parameters
if ["__max_page"] in request_parameters:
request_parameters.remove(["__max_page"]) #参数用完就删
if int(request.args.get("__max_page")) >= crawl.max_page_set:
crawl.max_page_set = int(request.args.get("__max_page"))
crawl.max_page_set_flag = True
## level_1 parameters
# 选出在request_parameters中属于level1的那些参数
level_1_parameters = []
for request_parameter in request_parameters:
if service.api_request_parameters_candidate[
candidate_parameters.index(
request_parameter)]["level"] == 1:
level_1_parameters.append(request_parameter)
request_parameters = [
i for i in request_parameters if i not in level_1_parameters
]
level_1_parameters = [".".join(i) for i in level_1_parameters]
crawl.input_key = []
crawl.input_value = []
for level_1_parameter in level_1_parameters:
crawl.input_key.append(level_1_parameter)
crawl.input_value.append(request.args.get(level_1_parameter))
# crawl.genNewRecord()
# Wait条件
if len(api_crawl_rules_double_list) > 0:
wait_rule = api_crawl_rules_double_list[0][0]
wait_images = wait_rule["images"]
wait_texts = wait_rule["texts"]
wait_links = wait_rule["links"]
if len(wait_texts) > 0:
crawl.css_selc = wait_texts[0]["css_selector"]
elif len(wait_images) > 0:
crawl.css_selc = wait_images[0]["css_selector"]
elif len(wait_links) > 0:
crawl.css_selc = wait_links[0]["css_selector"]
# 判断是否有iframe
if crawl.css_selc:
if crawl.css_selc.find(">f>") >= 0: # 对iframe中元素的特殊处理
crawl.iframe_addr = crawl.css_selc.split(">f>")[0]
crawl.css_selc = crawl.css_selc.split(">f>")[1]
crawl.iframe_ex = True
if form_rules_link:
form_rules_link_ex = form_rules_link.split('/statics/', 1)[1]
strss = read_file_as_str("static/" + form_rules_link_ex)
crawl.form_list = json.loads(strss)
# crawl.form_list = requests.get(form_rules_link).json()
time_out_judge = crawl.segment(api_url)
if time_out_judge == "timeout":
shutdown_crawl(crawl)
return ReNum.CRAW_TIME_OUT.value()
except Exception as e:
shutdown_crawl(crawl)
app.logger.error(repr(e))
return ReNum.CRAW_ERROR.update('', ":" + str(e)).value()
try:
results = []
message_error_return = [] # 将加载失败的页数记录在此，之后返回回去
crawl.css_selc = None
while crawl.judge_whether_final_page():
cram_click_next_page, data_error = crawl.click_next_page()
if cram_click_next_page:
page_pyquery = crawl.page ## 页面page
for indexxx in range(len(api_crawl_rules_double_list)):
api_crawl_rules = api_crawl_rules_double_list[indexxx]
api_parameters = api_parameters_double_list[indexxx]
result_one = []
for rule in api_crawl_rules:
images = rule["images"]
texts = rule["texts"]
links = rule["links"]
result = {"record_id": rule["record_id"]}
for image in images:
if crawl.iframe_ex and image[
"css_selector"].find('>f>') >= 0:
image["css_selector"] = image[
"css_selector"].split(">f>")[1]
if not page_pyquery(
image["css_selector"]) or len(
page_pyquery(
image["css_selector"])) == 0:
continue
if not crawl.css_selc:
crawl.css_selc = image["css_selector"]
res_value, parameter = image_filter(
image, page_pyquery, api_parameters,
api_url, crawl)
if parameter["select"] == 1:
result[parameter["name"]] = res_value
for text in texts:
if crawl.iframe_ex and text[
"css_selector"].find('>f>') >= 0:
text["css_selector"] = text[
"css_selector"].split(">f>")[1]
if not page_pyquery(
text["css_selector"]) or len(
page_pyquery(
text["css_selector"])) == 0:
continue
if not crawl.css_selc:
crawl.css_selc = text["css_selector"]
text_value, parameter = text_filter(
text, page_pyquery, api_parameters)
if parameter["select"] == 1:
result[parameter["name"]] = text_value
for link in links:
if crawl.iframe_ex and link[
"css_selector"].find('>f>') >= 0:
link["css_selector"] = link[
"css_selector"].split(">f>")[1]
if not page_pyquery(
link["css_selector"]) or len(
page_pyquery(
link["css_selector"])) == 0:
continue
if not crawl.css_selc:
crawl.css_selc = link["css_selector"]
css_link_result = pq(
page_pyquery(link["css_selector"])[0])
id = link["id"]
link_images = link["images"]
link_texts = link["texts"]
parameter = api_parameters[id]
result[parameter["name"]] = {}
for link_image in link_images:
if crawl.iframe_ex and link_image[
"css_selector"].find('>f>') >= 0:
link_image[
"css_selector"] = link_image[
"css_selector"].split(">f>")[1]
if not page_pyquery(
link_image["css_selector"]) or len(
page_pyquery(
link_image["css_selector"])
) == 0:
continue
link_image_res_value, link_image_parameter = image_filter(
link_image, page_pyquery,
api_parameters, api_url, crawl)
if link_image_parameter["select"] == 1:
result[parameter["name"]][
link_image_parameter[
"name"]] = link_image_res_value
for link_text in link_texts:
if crawl.iframe_ex and link_text[
"css_selector"].find('>f>') >= 0:
link_text["css_selector"] = link_text[
"css_selector"].split(">f>")[1]
if not page_pyquery(
link_text["css_selector"]) or len(
page_pyquery(
link_text["css_selector"])
) == 0:
continue
link_text_value, link_parameter = text_filter(
link_text, page_pyquery,
api_parameters)
if link_parameter["select"] == 1:
result[parameter["name"]][
link_parameter[
"name"]] = link_text_value
if parameter["select"] == 1:
if css_link_result.attr("href").startswith(
"http"):
result[parameter["name"]][
"href"] = css_link_result.attr(
'href')
else:
result[parameter["name"]][
"href"] = urljoin(
api_url, "/" +
css_link_result.attr('href'))
result_one.append(result)
result_double = {"section_name": section_name[indexxx]}
result_double["result"] = result_one
results.append(result_double)
else:
if data_error:
message_error_return.append(data_error)
try: ## TODO track 1: dot cant be a key in mongodb,for example {"3.4":dsada}
if not service.api_result_example:
numb = 3
if len(results) < 3:
numb = len(results)
result_example = results[:numb]
service.update(api_result_example=result_example)
except:
pass
service.update(api_network=list(crawl.api_request.keys(
))) ##getContentText的结果： 将可能有用的该网页的api network存入数据库的api_network字段
except Exception as e:
shutdown_crawl(crawl)
app.logger.error(e)
return ReNum.PAGE_PAESE_FAIL.update('', str(e)).value()
new_result = []
if len(results) > 0:
for result_one in results:
new_result_one = []
result_demo = result_one["result"][0]
section_name = result_one["section_name"]
parameters_unions, cha_set = select_result_parameter(
request_parameters, result_demo) # 请求参数和返回参数中求and的结果
cha_set = cha_set + ret_cha
for result in result_one["result"]:
__Flag = True
for parameters_union in parameters_unions:
_lik_val = "."
__request_value = request.args.get(
_lik_val.join(parameters_union))
if not iterator_judge(parameters_union, result,
__request_value):
__Flag = False
break
if __Flag:
new_result_one.append(result)
new_result_one_ = {"section_name": section_name}
new_result_one_["result"] = new_result_one
new_result.append(new_result_one_)
current_page = crawl.current_page - 1
shutdown_crawl(crawl)
msg = ""
if len(message_error_return) > 0:
msg = "，共处理" + str(current_page) + "页，其中第" + ",".join(
message_error_return) + "页处理失败."
else:
msg = "，共处理" + str(current_page) + "页."
if len(cha_set) > 0:
msg = msg + " 请求参数中 " + "，".join(
cha_set) + " 未起作用，请核实【tip:请按照query_name查询】."
return ReNum.SUCCESS.update(new_result, msg).value()
else:
return ReNum.INVOKE_LACK_DOC.value()
def add_service():
'''
POST: JSON
:parameter api_name : string
api_url : string
api_description : string
api_crawl_rules_link : string
candidate : list
main_sec_id: int
img_link : string
json_link : string
:example
{
"api_name":"weather_4",
"api_description":"api",
"candidate":[
{ "id":1,
"name":"text_1",
"description":"none",
"type":"text",
"example":"default",
"select":0
},{
"id":3,
"name":"text_1",
"description":"none",
"type":"text",
"example":"default",
"select":1
},{
"id":2,
"name":"text_1",
"description":"none",
"type":"text",
"example":"default",
"select":0
}],
"api_crawl_rules_link":"http://127.0.0.1:5000/weather1",
"img_link":"http://213.ds/com",
"json_link":"http://test.com",
"api_url":"http://test.cn", ## url and link 一定要是http:// 或者 https://
}
GET:
:parameter service : string
:exapmle
http://service.cheosgrid.org:8089/service?api_name=weather
http://service.cheosgrid.org:8089/service?api_id=1
PUT:JSON
:parameter
api_id: 2 ## 一定要有api_id
api_name : string
api_url : string
api_description : string
api_crawl_rules_link : string
api_parameters : list
img_link : string
main_sec_id: int
json_link : string
:example
{
"api_name":"weather_4",
"api_description":"api",
"api_parameters":[
{ "id":1,
"name":"text_1",
"description":"none",
"type":"text",
"example":"default",
"select":0
},{
"id":3,
"name":"text_1",
"description":"none",
"type":"text",
"example":"default",
"select":1
},{
"id":2,
"name":"text_1",
"description":"none",
"type":"text",
"example":"default",
"select":0
}],
"api_crawl_rules_link":"http://127.0.0.1:5000/weather1",
"img_link":"http://213.ds/com",
"json_link":"http://test.com",
"api_url":"http://test.cn",
"api_id":5
}
DELETE:
:parameter api_id : int
:example
http://service.cheosgrid.org:8089/service?api_id=1
:return:
ReNum.SUCCESS
ReNum.PARAM_ERR
ReNum.SERV_NO_EXIST
'''
if request.method == "POST":
# try:
# #####名字重复报错"服务已存在"
# service_exist = Api.objects(api_name=request.json['api_name']).first()
# if service_exist:
# app.logger.error("service - 已有该服务")
# return jsonify({'status': 402, 'msg': '已有该服务', 'data': ''})
# ##########################
service = Api(
api_name=request.json['api_name'],
api_description=request.json['api_description'],
candidate=request.json['candidate'],
main_sec_id=request.json['main_sec_id'],
form_rules_link=request.json['form_rules_link'],
api_url=request.json['api_url'],
api_crawl_rules_link=request.json['api_crawl_rules_link'],
img_link=request.json['img_link'],
json_link=request.json['json_link']).save()
service.update(url=setting.SERVER_ADDRESS + "call_service/" +
str(service.api_id))
candidates = service.candidate
## 针对service进行修改，对每个candidate增加query_name 字段
candidates = candidate_add_query_name(candidates)
api_request_parameters_candidate_level2 = get_api_request_parameters_candidate_Level2(
candidates, service.main_sec_id)
service.update(candidate=candidates)
api_request_parameters_candidate_level0 = [{
"type":
"int",
"query_name":
"__max_page",
"level":
0,
"required":
False,
"example":
3,
"description":
"[系统级别参数]: 最大页数限制，默认为5页，请设置较小的页数以避免返回时间过长。"
}]
api_request_parameters_candidate_level1 = []
if service.form_rules_link:
form_rules_link_ex = service.form_rules_link.split('/statics/',
1)[1]
strss = read_file_as_str("static/" + form_rules_link_ex)
form_list = json.loads(strss)
# form_list = requests.get(service.form_rules_link).json()
if form_list["form_check"] == 1: # 如果有参数
input_list = form_list["forms"][
form_list["main_form_index"]]["input_list"]
for input in input_list: # 模拟表单操作
if input["type"] != "hidden": # hidden一般不要动
api_request_parameters_candidate_level1.append({
"type":
input["type"],
"query_name":
input["query_name"],
"level":
1,
"required":
input["required"],
"example":
input["value"],
"description":
"[查询输入参数]: " + input["description"]
})
api_request_parameters_candidate = api_request_parameters_candidate_level0 + api_request_parameters_candidate_level1 + api_request_parameters_candidate_level2
service.update(
api_request_parameters_candidate=api_request_parameters_candidate)
service_all = Api.objects(api_id=service.api_id).first()
#########################
## To get the result_example
callservice = CallService(service_all)
resp = callservice.call_only_one_page_for_getting_result_demo()
callservice.shutDown()
call_result = ""
if resp.startswith("Error"):
call_result = ", " + resp
#########################
mysql = MYSQL()
mysql.insertOneToGrid_API(Api.objects(api_id=service.api_id).first())
return ReNum.SUCCESS.update(
Api.objects(api_id=service.api_id).first(), call_result).value()
# except:
# app.logger.error("service - 参数错误")
# return jsonify({'status': 400, 'msg': '参数错误', 'data': ''})
elif request.method == "GET":
if request.args.get('api_name'):
service_exist = Api.objects(
api_name=request.args.get('api_name')).first()
elif request.args.get('api_id'):
service_exist = Api.objects(
api_id=request.args.get('api_id')).first()
else:
app.logger.error("service - 参数错误")
return ReNum.PARAM_ERR.value()
return ReNum.SUCCESS.update(service_exist).value()
elif request.method == "DELETE":
if request.args.get('api_id'):
service_exist = Api.objects(
api_id=request.args.get('api_id')).first()
if service_exist:
mysql = MYSQL()
mysql.deleteOneToGrid_API(service_exist)
service_exist.delete()
return ReNum.SUCCESS.value()
else:
app.logger.error("service - 参数错误")
return ReNum.PARAM_ERR.value()
elif request.method == "PUT":
try:
service_exist = Api.objects(api_id=request.json['api_id']).first()
if not service_exist:
app.logger.error("service - 服务不存在")
return ReNum.SERV_NO_EXIST.value()
else:
service_exist.update(
api_name=request.json['api_name'],
api_description=request.json['api_description'],
candidate=request.json['candidate'],
api_url=request.json['api_url'],
form_rules_link=request.json['form_rules_link'],
main_sec_id=request.json['main_sec_id'],
api_crawl_rules_link=request.json['api_crawl_rules_link'],
img_link=request.json['img_link'],
json_link=request.json['json_link'])
candidates = service_exist.candidate
## 针对service进行修改，对每个candidate增加query_name 字段
candidates = candidate_add_query_name(candidates)
api_request_parameters_candidate_level2 = get_api_request_parameters_candidate_Level2(
candidates, service_exist.main_sec_id)
service_exist.update(candidate=candidates)
api_request_parameters_candidate_level0 = [{
"type":
"int",
"query_name":
"__max_page",
"level":
0,
"required":
False,
"example":
3,
"description":
"[系统级别参数]: 最大页数限制，默认为5页，请设置较小的页数以避免返回时间过长。"
}]
api_request_parameters_candidate_level1 = []
if service_exist.form_rules_link:
form_rules_link_ex = service_exist.form_rules_link.split(
'/statics/', 1)[1]
strss = read_file_as_str("static/" + form_rules_link_ex)
form_list = json.loads(strss)
# form_list = requests.get(service_exist.form_rules_link).json()
if form_list["form_check"] == 1: # 如果有参数
input_list = form_list["forms"][
form_list["main_form_index"]]["input_list"]
for input in input_list: # 模拟表单操作
if input["type"] != "hidden": # hidden一般不要动
api_request_parameters_candidate_level1.append(
{
"type":
input["type"],
"query_name":
input["query_name"],
"level":
1,
"required":
input["required"],
"example":
input["value"],
"description":
"[查询输入参数]: " + input["description"]
})
api_request_parameters_candidate = api_request_parameters_candidate_level0 + api_request_parameters_candidate_level1 + api_request_parameters_candidate_level2
service_exist.update(api_request_parameters_candidate=
api_request_parameters_candidate)
service_exist_all = Api.objects(
api_id=service_exist.api_id).first()
#########################
## To get the result_example
callservice = CallService(service_exist_all)
resp = callservice.call_only_one_page_for_getting_result_demo(
)
call_result = ""
if resp.startswith("Error"):
call_result = ", " + resp
#########################
mysql = MYSQL()
mysql.insertOneToGrid_API(service_exist_all, True)
return ReNum.SUCCESS.update(
Api.objects(api_id=service_exist.api_id).first(),
call_result).value()
except:
app.logger.error("service - 参数错误")
return ReNum.PARAM_ERR.value()
else:
return "503 wrong information"
