Set a max size of 100mb for utils.http.get_url
This commit is contained in:
parent
68f5626189
commit
be75f72356
1 changed files with 7 additions and 3 deletions
|
@ -6,6 +6,8 @@ USER_AGENT = ("Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 "
|
||||||
"(KHTML, like Gecko) Chrome/49.0.2623.87 Safari/537.36")
|
"(KHTML, like Gecko) Chrome/49.0.2623.87 Safari/537.36")
|
||||||
REGEX_HTTP = re.compile("https?://", re.I)
|
REGEX_HTTP = re.compile("https?://", re.I)
|
||||||
|
|
||||||
|
RESPONSE_MAX = (1024*1024)*100
|
||||||
|
|
||||||
def get_url(url, method="GET", get_params={}, post_data=None, headers={},
|
def get_url(url, method="GET", get_params={}, post_data=None, headers={},
|
||||||
json_data=None, code=False, json=False, soup=False, parser="lxml"):
|
json_data=None, code=False, json=False, soup=False, parser="lxml"):
|
||||||
|
|
||||||
|
@ -23,16 +25,18 @@ def get_url(url, method="GET", get_params={}, post_data=None, headers={},
|
||||||
headers=headers,
|
headers=headers,
|
||||||
params=get_params,
|
params=get_params,
|
||||||
data=post_data,
|
data=post_data,
|
||||||
json=json_data
|
json=json_data,
|
||||||
|
stream=True
|
||||||
)
|
)
|
||||||
|
response_content = response.raw.read(RESPONSE_MAX, decode_content=True)
|
||||||
|
|
||||||
if soup:
|
if soup:
|
||||||
soup = bs4.BeautifulSoup(response.text, parser)
|
soup = bs4.BeautifulSoup(response_content, parser)
|
||||||
if code:
|
if code:
|
||||||
return response.code, soup
|
return response.code, soup
|
||||||
return soup
|
return soup
|
||||||
|
|
||||||
data = response.text
|
data = response_content
|
||||||
if json and data:
|
if json and data:
|
||||||
try:
|
try:
|
||||||
data = _json.loads(data)
|
data = _json.loads(data)
|
||||||
|
|
Loading…
Reference in a new issue