Comments on: WebKit Page Cache I – The Basics http://www.webkit.org/blog/427/webkit-page-cache-i-the-basics/ All about WebKit development Thu, 09 Feb 2012 11:34:15 +0000 hourly 1 http://wordpress.org/?v=3.2.1 By: nhoel http://www.webkit.org/blog/427/webkit-page-cache-i-the-basics/comment-page-1/#comment-25209 nhoel Sun, 20 Sep 2009 11:50:04 +0000 http://webkit.org/blog/?p=427#comment-25209 @kangax THANKS FOR THAT LINK @adrian I THIRD DEMOTION thanks, -Nhoel of http://keywordspeak.com/?p=1234 @kangax THANKS FOR THAT LINK

@adrian I THIRD DEMOTION

thanks,
-Nhoel of http://keywordspeak.com/?p=1234

]]> By: kangax http://www.webkit.org/blog/427/webkit-page-cache-i-the-basics/comment-page-1/#comment-25208 kangax Sat, 19 Sep 2009 02:48:54 +0000 http://webkit.org/blog/?p=427#comment-25208 Link got stripped, here it is again - http://dev.rubyonrails.org/ticket/11430 Link got stripped, here it is again – http://dev.rubyonrails.org/ticket/11430

]]> By: kangax http://www.webkit.org/blog/427/webkit-page-cache-i-the-basics/comment-page-1/#comment-25207 kangax Sat, 19 Sep 2009 02:48:07 +0000 http://webkit.org/blog/?p=427#comment-25207 Unfortunately, in Prototype.js we always attach an "unload" handler in WebKit. This is due to a bug in some of the earlier versions of WebKit, where returning to a page via "page cache" would wipe-off all of the expandos on host objects (e.g. from the `document`). Since Prototype.js extends these host objects with custom methods, there was chaos all over the place : ) See this ticket for more info - What's even more disappointing is that we use browser sniffing as it seems impossible to feature test this quirk. I suppose disabling unload in newer WebKit wouldn't affect Prototype, as long as document state persists properly (and it does in newer versions). I would also love to hear if there's a saner work around for this issue (i.e. any kind of inference we can perform, to replace fragile `userAgent` parsing). Unfortunately, in Prototype.js we always attach an “unload” handler in WebKit. This is due to a bug in some of the earlier versions of WebKit, where returning to a page via “page cache” would wipe-off all of the expandos on host objects (e.g. from the `document`). Since Prototype.js extends these host objects with custom methods, there was chaos all over the place : )

See this ticket for more info –

What’s even more disappointing is that we use browser sniffing as it seems impossible to feature test this quirk. I suppose disabling unload in newer WebKit wouldn’t affect Prototype, as long as document state persists properly (and it does in newer versions).

I would also love to hear if there’s a saner work around for this issue (i.e. any kind of inference we can perform, to replace fragile `userAgent` parsing).

]]> By: Adrian Russell-Falla http://www.webkit.org/blog/427/webkit-page-cache-i-the-basics/comment-page-1/#comment-25206 Adrian Russell-Falla Fri, 18 Sep 2009 16:47:34 +0000 http://webkit.org/blog/?p=427#comment-25206 Brady, I fervently second jomohke's request! looking fwd to Part 2. Brady, I fervently second jomohke’s request!
looking fwd to Part 2.

]]> By: jomohke http://www.webkit.org/blog/427/webkit-page-cache-i-the-basics/comment-page-1/#comment-25205 jomohke Fri, 18 Sep 2009 11:55:27 +0000 http://webkit.org/blog/?p=427#comment-25205 It's fantastic to see the return of technical articles detailing the innards of webkit. They stopped appearing when this blog moved from just David Hyatt to an official team blog. Please give us more! It’s fantastic to see the return of technical articles detailing the innards of webkit. They stopped appearing when this blog moved from just David Hyatt to an official team blog. Please give us more!

]]> By: Brady Eidson http://www.webkit.org/blog/427/webkit-page-cache-i-the-basics/comment-page-1/#comment-25204 Brady Eidson Thu, 17 Sep 2009 20:03:48 +0000 http://webkit.org/blog/?p=427#comment-25204 CyberSkull, you're exactly right and I strongly recommend you keep your eyes out for part 2. CyberSkull, you’re exactly right and I strongly recommend you keep your eyes out for part 2.

]]> By: CyberSkull http://www.webkit.org/blog/427/webkit-page-cache-i-the-basics/comment-page-1/#comment-25203 CyberSkull Thu, 17 Sep 2009 08:51:54 +0000 http://webkit.org/blog/?p=427#comment-25203 Hmm, perhaps what we need is something like a cache handler for JS and plugins; Ideally to do whatever cleanup or state changes for going into cache and coming out of it… Hmm, perhaps what we need is something like a cache handler for JS and plugins; Ideally to do whatever cleanup or state changes for going into cache and coming out of it…

]]> By: Miles http://www.webkit.org/blog/427/webkit-page-cache-i-the-basics/comment-page-1/#comment-25202 Miles Thu, 17 Sep 2009 04:44:03 +0000 http://webkit.org/blog/?p=427#comment-25202 Spoilers: https://bugs.webkit.org/show_bug.cgi?id=29021 ;) Spoilers: https://bugs.webkit.org/show_bug.cgi?id=29021 ;)

]]>