def func(m):
''' 替换成编码的地址 '''
#"%(tag)s%(equals)s%(quote)s/%(url)s"),
## ( m.group('url').endswith("/") and number=="5" ) or \
endwith_slash_gt = \
( m.group('url').endswith(";") and (number in ["9" ,"6" ,"8"]) ) or \
( m.group('url').endswith("/") and number in ["1" ,"3","5"] and m.string.find('/>')>=0 \
and (mvalue(m,'tag')=='src' and m.end('url')== (m.string.find('/>')+1 ) ) \
)
#logging.error('No:%s tag:%s end of url:%s,findbegin:%s'%(number,mvalue(m,'tag'),m.end('url'), m.string.find('/>')+1))
last = m.group('url')[len(m.group('url'))-1:]
url_in = ''
if endwith_slash_gt:
url_in = m.group('url')[:len(m.group('url'))-1] and m.group('url')[:len(m.group('url'))-1] or ''
else:
url_in = m.group('url')
result_fullurl = scheme =''
#书签 , javascript
#No_b64_encoding = m.group('url').lstrip().startswith('#') or m.group('url').lstrip().lower().startswith('javascript')
No_b64_encoding = [x for x in ['#','javascript'] if m.group('url').lstrip().startswith(x)]
if No_b64_encoding:
result_fullurl = fullurl%{
"base": '',
#"netloc":url_obj.netloc,
"accessed_dir": '',
'url':m.group('url'),
'relative':'',
}
result_fullurl = result_fullurl[1:]
#logging.error('%s url:%s %s %s'%(number,m.group('url'),endwith_slash_gt ,len(m.group('url')) ))
else:
scheme = m.groupdict().has_key('scheme') and m.group('scheme') or url_obj.scheme
result_fullurl = '%s:/'%scheme.replace(":","") + fullurl%{
"base": base_url,
#"netloc":url_obj.netloc,
"accessed_dir": (accessed_dir),
'url':m.groupdict().has_key('url') and url_in or '',
'relative':m.groupdict().has_key('relative') and m.group('relative') or '',
}
#logging.error('\n\n %s result_fullurl result : %s '%(number,result_fullurl ))
#logging.error('scheme:"%s" base_url:%s accessed_dir:%s'%(scheme,base_url,accessed_dir) )
#logging.error('%s url:%s %s %s'%(number,m.group('url'),endwith_slash_gt ,len(m.group('url')) ))
#result_fullurl = result_fullurl[1:]
kk = result%{
'fullurl': No_b64_encoding and result_fullurl or '/'+b64.uri_b64encode(result_fullurl) + (endwith_slash_gt and last or ""),
'tag':m.groupdict().has_key('tag') and m.group('tag') or '',
'equals':m.groupdict().has_key('equals') and m.group('equals') or '',
'quote':m.groupdict().has_key('quote') and m.group('quote') or '',
'spacing':m.groupdict().has_key('spacing') and m.group('spacing') or '',
}
#logging.info('uri_b64encode:%s'%kk)
#logging.info('pattern %s end:%s '%(number,pattern))
#logging.info('%s result : %s '%(number,result_fullurl))
#logging.error('string : %s '%(m.string))
return kk
def post(self):
# Handle the input form to redirect the user to a relative url
form_url = self.request.get("url")
if form_url:
# Accept URLs that still have a leading 'http://'
inputted_url = form_url #urllib.quote(form_url.encode('utf-8')) #form_url #
#if inputted_url.startswith(HTTP_PREFIX):
# inputted_url = inputted_url[len(HTTP_PREFIX):]
#return self.redirect("/" + b64.uri_b64encode(inputted_url))
return self.redirect("/" + b64.uri_b64encode(inputted_url))
self.response.out.write(unicode(template.render("main.html", {})))
def post(self):
# Handle the input form to redirect the user to a relative url
form_url = self.request.get("url")
if form_url:
# Accept URLs that still have a leading 'http://'
inputted_url = form_url #urllib.quote(form_url.encode('utf-8')) #form_url #
#if inputted_url.startswith(HTTP_PREFIX):
# inputted_url = inputted_url[len(HTTP_PREFIX):]
#return self.redirect("/" + b64.uri_b64encode(inputted_url))
return self.redirect("/" + b64.uri_b64encode(inputted_url))
self.response.out.write(
template.render("main.html", {
'change_url': self.get_change_root(),
}))
def uri_b64encode(url):
url = url[1:]
return b64.uri_b64encode(url)
def _RunTransformTest(self, base_url, accessed_url, original, expected):
tag_tests = [
'<img src="%s"/>',
"<img src='%s'/>",
"<img src=%s/>",
"<img src=\"%s'/>",
"<img src='%s\"/>",
"<img src \t= '%s'/>",
"<img src \t= \t '%s'/>",
"<img src = '%s'/>",
'<a href="%s">',
"<a href='%s'>",
"<a href=%s>",
"<a href=\"%s'>",
"<a href='%s\">",
"<a href \t = \t'%s'>",
"<a href \t = '%s'>",
"<a href = \t'%s'>",
"<td background=%s>",
"<td background='%s'>",
'<td background="%s">',
'<form action="%s">',
"<form action='%s'>",
"<form action=%s>",
"<form action=\"%s'>",
"<form action='%s\">",
"<form action \t = \t'%s'>",
"<form action \t = '%s'>",
"<form action = \t'%s'>",
"@import '%s';",
"@import '%s'\nnext line here",
"@import \t '%s';",
"@import %s;",
"@import %s",
'@import "%s";',
'@import "%s"\nnext line here',
"@import url(%s)",
"@import url('%s')",
'@import url("%s")',
"background: transparent url(%s) repeat-x left;",
'background: transparent url("%s") repeat-x left;',
"background: transparent url('%s') repeat-x left;",
'<meta http-equiv="Refresh" content="0; URL=%s">',
'url(%s)',
'src="%s" ',
'style="background:url(%s)'
]
No_b64_encoding = expected.lstrip().startswith('#') or expected.lstrip().lower().startswith('javascript')
scheme = original.startswith('https') and 'https' or urlparse.urlparse(accessed_url).scheme
expected = No_b64_encoding and expected or ('/'+b64.uri_b64encode('%s:/'%scheme +expected))
for tag in tag_tests:
#logging.error("\n\n\n tag begin:%s "%tag)
test = tag % original
correct = tag % expected
result = transform_content.TransformContent(base_url, accessed_url, test)
logging.error("Test with\n"
"Accessed: %s\n"
"Input : %s\n"
"Received: %s\n"
"Expected: %s",
accessed_url, test, result, correct)
if result != correct:
logging.info("FAIL")
self.assertEquals(correct, result)
def uri_b64encode(url):
url = url[1:]
return b64.uri_b64encode(url)
def to_uri_b64encode(url):
url = url[1:]
return "/"+b64.uri_b64encode(url)
def to_uri_b64encode(url):
url = url[1:]
return "/" + b64.uri_b64encode(url)
