fix: cloudscrape revert alla stable

This commit is contained in:
greko17
2019-12-03 00:14:29 +01:00
parent d0b02f327d
commit cc0685e84e
27 changed files with 11696 additions and 27181 deletions

61
lib/cloudscraper/interpreters/__init__.py Normal file → Executable file
View File

@@ -1,3 +1,4 @@
import re
import sys
import logging
import abc
@@ -7,24 +8,20 @@ if sys.version_info >= (3, 4):
else:
ABC = abc.ABCMeta('ABC', (), {})
# ------------------------------------------------------------------------------- #
##########################################################################################################################################################
interpreters = {}
BUG_REPORT = 'Cloudflare may have changed their technique, or there may be a bug in the script.'
# ------------------------------------------------------------------------------- #
##########################################################################################################################################################
interpreters = {}
class JavaScriptInterpreter(ABC):
# ------------------------------------------------------------------------------- #
@abc.abstractmethod
def __init__(self, name):
interpreters[name] = self
# ------------------------------------------------------------------------------- #
@classmethod
def dynamicImport(cls, name):
if name not in interpreters:
@@ -38,17 +35,55 @@ class JavaScriptInterpreter(ABC):
return interpreters[name]
# ------------------------------------------------------------------------------- #
@abc.abstractmethod
def eval(self, jsEnv, js):
pass
# ------------------------------------------------------------------------------- #
def solveChallenge(self, body, domain):
try:
return float(self.eval(body, domain))
js = re.search(
r'setTimeout\(function\(\){\s+(var s,t,o,p,b,r,e,a,k,i,n,g,f.+?\r?\n[\s\S]+?a\.value =.+?)\r?\n',
body
).group(1)
except Exception:
raise ValueError('Unable to identify Cloudflare IUAM Javascript on website. {}'.format(BUG_REPORT))
js = re.sub(r'\s{2,}', ' ', js, flags=re.MULTILINE | re.DOTALL).replace('\'; 121\'', '')
js += '\na.value;'
jsEnv = '''
String.prototype.italics=function(str) {{return "<i>" + this + "</i>";}};
var document = {{
createElement: function () {{
return {{ firstChild: {{ href: "https://{domain}/" }} }}
}},
getElementById: function () {{
return {{"innerHTML": "{innerHTML}"}};
}}
}};
'''
try:
innerHTML = re.search(
r'<div(?: [^<>]*)? id="([^<>]*?)">([^<>]*?)</div>',
body,
re.MULTILINE | re.DOTALL
)
innerHTML = innerHTML.group(2) if innerHTML else ''
except: # noqa
logging.error('Error extracting Cloudflare IUAM Javascript. {}'.format(BUG_REPORT))
raise
try:
result = self.eval(
re.sub(r'\s{2,}', ' ', jsEnv.format(domain=domain, innerHTML=innerHTML), flags=re.MULTILINE | re.DOTALL),
js
)
float(result)
except Exception:
logging.error('Error executing Cloudflare IUAM Javascript. {}'.format(BUG_REPORT))
raise
return result