Lain_13 wrote:here is no way to block their advertisements by generic part in the request (instead of filename they have alphanumerical hash-like string right after the domain name with current sitename somewhere in the middle). There are no way to hide their ads by object names because they always random
Could not random naming be overcome with where in the page structure the annoyance lives? I'm not clear why XPath and CSS selectors cannot be employed interchangeably.
viewtopic.php?p=56535#p56535lewisje wrote:Maybe there could be a new syntax made for "XPath-based element-hiding rules" (maybe starting with ##$ instead of ##)...... (but Firefox has supported XPath since 1.5) or its uselessness on HTML (rather than XHTML) documents (but I also read that the XPath API allows traversal of HTML documents not explicitly serialized as XML).
This guy speaks of manual conversion so why not automated?
guest1337 wrote:Lool this freaking works!!!11oneeleven
- Code: Select all
site.net##html > body > table > tbody > tr > td:nth-of-type(2)
Actually, i don't care if it's XPath syntax or this CSS syntax as long as they mention the same. But you should freaking definitely add this feature to element hiding helper! Since attributes combined with object path make much less fake positives so i could now commit my rules if i want! Tools i used (Dom Inspector + XPather + manual convertion from XPath to CSS) and adding the rule after clicking through 5 settings windows are not very handily.
So, is it possible to implement such a thing? I think it is not complicated and could be implemented as an extension to the existing popup functionality without adding new $flag.
A board search for XPath turned up yet another term, "Advanced Selectors":
viewtopic.php?f=1&t=4702&p=30210&hilit=xpath#p30210And posts requesting XPath conversion to CSS selector
viewtopic.php?f=1&t=9413&p=55889#p55889 What kind of CSS selector is this
viewtopic.php?p=15477#p15477As I understand it the CSS selectors are applied "instantly" and XPath is after page rendering. But XPath can be more specific than CSS? So XPath use ought be used as last resort? The wiki XPath:talk is beyond comprehension at this point:
http://kb.mozillazine.org/Talk:XPath. Now fold in XLink and XPointer.
http://www.w3schools.com/xlink/xlink_intro.aspOr points and ranges?
http://www.w3.org/TR/xptr-xpointer/#dt-pointCould Clipping Region ,
http://www.w3.org/TR/css3-positioning/O ... g-clipping |
http://www.w3.org/TR/media-frags/ |
http://www.w3.org/TR/xptr-xpointer/ , be used to hide or not-render the images? Or floating interference?
At some point DOM inspector added a button to save as EHH filter. Could not the same be done for XPath?
Or could there be an ADP/EHH enhancement to specify [relative] page structure CSS location instead of using labels? disclaimer: contains wild speculation, unfamiliarity and outright ignorance