new wombat! refactor of rewriting:

- use defineProperty overrides on element prototypes
- postMessage() rework: store actual origin with helper function __WB_pmw(window), from
server side rewrite
- Use window.URL (or external jsurl script) to override all properties of HTMLAnchorElement,
override getAttribute() to return original
- rename window -> $wbwindow

Author: ikreymer

pywb/rewrite/regex_rewriters.py

@@ -132,7 +132,9 @@ class JSLocationRewriterMixin(object):
     def __init__(self, rewriter, rules=[], prefix='WB_wombat_'):
         rules = rules + [
           #  (r'(?<![/$])\blocation\b(?!\":)', RegexRewriter.add_prefix(prefix), 0),
-          (r'(?<![/$\'"-])\blocation\b(?!(?:\":|=\d|-))', RegexRewriter.add_prefix(prefix), 0),
+          (r'(?<![/$\'"-])\blocation\b(?!(?:\":|:|=\d|-))', RegexRewriter.add_prefix(prefix), 0),
+
+          (r'(?<=\.)postMessage\b\(', RegexRewriter.add_prefix('__WB_pmw(window).'), 0),
 
           #  (r'(?<=document\.)domain', RegexRewriter.add_prefix(prefix), 0),
           #  (r'(?<=document\.)referrer', RegexRewriter.add_prefix(prefix), 0),

pywb/rules.yaml

@@ -1,34 +1,35 @@
 rules:
 
+
     # twitter rules
     #=================================================================
     - url_prefix: 'com,twitter)/i/profiles/show/'
 
       fuzzy_lookup: '/profiles/show/.*with_replies\?.*(max_id=[^&]+)'
 
-    - url_prefix: 'com,twimg,abs)/c/swift/en/bundle/boot'
+#    - url_prefix: 'comx,twimg,abs)/c/swift/en/bundle/boot'
 
-      rewrite:
-        js_rewrite_location: urls
-        js_regexs:
-            - match: 'window.location'
-              replace: 'window.WB_wombat_location'
+#      rewrite:
+#        js_rewrite_location: urls
+#        js_regexs:
+#            - match: 'window.location'
+#              replace: 'window.WB_wombat_location'
 
-            - match: 'document.location'
-              replace: 'document.WB_wombat_location'
+#            - match: 'document.location'
+#              replace: 'document.WB_wombat_location'
 
     #- url_prefix: 'com,twimg,ton)/'
 
     #  rewrite:
     #    js_rewrite_location: all
 
-    - url_prefix: 'com,twimg,amp)/amplify-web-player/prod/js/build.min.js'
+#    - url_prefix: 'comx,twimg,amp)/amplify-web-player/prod/js/build.min.js'
 
-      rewrite:
+#      rewrite:
         #js_rewrite_location: all
-        js_regexs:
-            - match: ';$'
-              replace: '; $(function() {{ var div = $(".poster-image")[0]; window._wb_wombat.watch_elem(div, function(elem) {{ if (elem.tagName == "IMG") {{ elem.setAttribute("width", $(div).width()); elem.setAttribute("height", $(div).height()); }}  }}) }}  );'
+#        js_regexs:
+#            - match: ';$'
+#              replace: '; $(function() {{ var div = $(".poster-image")[0]; window._wb_wombat.watch_elem(div, function(elem) {{ if (elem.tagName == "IMG") {{ elem.setAttribute("width", $(div).width()); elem.setAttribute("height", $(div).height()); }}  }}) }}  );'
 
 
     # facebook rules
@@ -168,17 +169,18 @@ rules:
     # vine
     - url_prefix: 'co,vine,cdn,'
 
-      rewrite:
-        js_rewrite_location: urls
-        js_regexs:
-            - match: 'window.location'
-              replace: 'WB_wombat_location'
+#      rewrite:
+#        js_rewrite_location: urls
+#        js_regexs:
+#            - match: 'window.location'
+#              replace: 'WB_wombat_location'
 
       fuzzy_lookup:
         replace: 'videos'
         match: 'videos(?:_[^/]+)?/([^?]+)'
 
     # disqus
+    #- url_prefix: ['com,disqus', 'com,disquscdn']
     - url_prefix: 'com,disqus'
 
       rewrite: