backup/pywb
1
0
Fork 0
mirror of https://github.com/webrecorder/pywb.git synced 2025-03-15 00:03:28 +01:00
Code Issues Projects Releases Activity

rewrite: properly rewrite scheme relative JS-escaped urls:

Browse Source 
'\/\/example.com', '\\/\\/example.com/', treat same as '//example.com'
adding http: prefix
This commit is contained in:
Ilya Kreymer 2014-11-23 18:56:49 -08:00
parent b8b8c30573
commit d7eb40af20
3 changed files with 29 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
pywb/rewrite/test/test_regex_rewriters.py
View File

@ -45,6 +45,16 @@ r"""
>>> _test_js('document_domain = "anotherdomain.com"; window.document.domain = "example.com"')
'document_domain = "anotherdomain.com"; window.document.WB_wombat_domain = "example.com"'
# protocol-rel escapes
>>> _test_js('"//example.com/"')
'"/web/20131010/http://example.com/"'
>>> _test_js(r'"\/\/example.com/"')
'"/web/20131010/http:\\/\\/example.com/"'
>>> _test_js(r'"\\/\\/example.com/"')
'"/web/20131010/http:\\\\/\\\\/example.com/"'
# custom rules added
>>> _test_js('window.location = "http://example.com/abc.html"; some_func(); ', [('some_func\(\).*', RegexRewriter.format('/*{0}*/'), 0)])
'window.WB_wombat_location = "/web/20131010/http://example.com/abc.html"; /*some_func(); */'

15
pywb/rewrite/test/test_url_rewriter.py
View File

@ -50,6 +50,21 @@
>>> do_rewrite(r'http:\/\/some-other-site.com', '20101226101112/http://example.com/index.html', 'localhost:8080/')
'localhost:8080/20101226101112/http:\\\\/\\\\/some-other-site.com'
>>> do_rewrite(r'//some-other-site.com', '20101226101112/http://example.com/index.html', 'localhost:8080/')
'localhost:8080/20101226101112/http://some-other-site.com'
>>> do_rewrite(r'\/\/some-other-site.com', '20101226101112/http://example.com/index.html', 'localhost:8080/')
'localhost:8080/20101226101112/http:\\\\/\\\\/some-other-site.com'
>>> do_rewrite(r'\\/\\/some-other-site.com', '20101226101112/http://example.com/index.html', 'localhost:8080/')
'localhost:8080/20101226101112/http:\\\\/\\\\/some-other-site.com'
>>> do_rewrite(r'http:\/\/some-other-site.com', '20101226101112/http://example.com/index.html', 'localhost:8080/')
'localhost:8080/20101226101112/http:\\\\/\\\\/some-other-site.com'
>>> do_rewrite(r'http:\/\/some-other-site.com', '20101226101112/http://example.com/index.html', 'localhost:8080/')
'localhost:8080/20101226101112/http:\\\\/\\\\/some-other-site.com'
>>> do_rewrite('../../other.html', '2020/http://example.com/index.html', '/')
'/2020/http://example.com/other.html'

6
pywb/rewrite/url_rewriter.py
View File

@ -17,7 +17,9 @@ class UrlRewriter(object):
PROTOCOLS = ['http:', 'https:', 'ftp:', 'mms:', 'rtsp:', 'wais:']
def __init__(self, wburl, prefix, full_prefix=None, rel_prefix=None,
REL_SCHEME = ('//', r'\/\/', r'\\/\\/')
def __init__(self, wburl, prefix, full_prefix=None, rel_prefix=None,
root_path=None, cookie_scope=None):
self.wburl = wburl if isinstance(wburl, WbUrl) else WbUrl(wburl)
self.prefix = prefix
@ -45,7 +47,7 @@ class UrlRewriter(object):
is_abs = any(url.startswith(x) for x in self.PROTOCOLS)
if url.startswith('//'):
if url.startswith(self.REL_SCHEME):
is_abs = True
url = 'http:' + url