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

regex rewrite: fix js regex (dashes), add additional test case

Browse Source
This commit is contained in:
Ilya Kreymer 2014-11-01 15:39:51 -07:00
parent 841fd3f7b4
commit a3b931b45e
2 changed files with 5 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
pywb/rewrite/regex_rewriters.py
View File

@ -111,7 +111,8 @@ class JSLinkOnlyRewriter(RegexRewriter):
JS Rewriter which rewrites absolute http://, https:// and // urls
at the beginning of a string
"""
JS_HTTPX = r'(?<="|\'|;)(?:https?:)?\\{0,4}/\\{0,4}/[A-Za-z0-9:_@.-]+'
#JS_HTTPX = r'(?:(?:(?<=["\';])https?:)|(?<=["\']))\\{0,4}/\\{0,4}/[A-Za-z0-9:_@.-]+.*(?=["\s\';&\\])'
JS_HTTPX = r'(?<=["\';])(?:https?:)?\\{0,4}/\\{0,4}/[A-Za-z0-9:_@.\-/\\?&#]+(?=["\';&\\])'
def __init__(self, rewriter, rules=[]):
rules = rules + [

3
pywb/rewrite/test/test_regex_rewriters.py
View File

@ -61,6 +61,9 @@ r"""
>>> _test_js('&quot;http:\\/\\/www.example.com\\/some\\/path\\/?query=1&quot;')
'&quot;/web/20131010/http:\\/\\/www.example.com\\/some\\/path\\/?query=1&quot;'
>>> _test_js('"http:\/\/sub-site.example.com\/path-dashes\/path_other\/foo_bar.txt"')
'"/web/20131010/http:\\/\\/sub-site.example.com\\/path-dashes\\/path_other\\/foo_bar.txt"'
#=================================================================
# XML Rewriting