WebKit Bugzilla
New
Browse
Log In
×
Sign in with GitHub
or
Remember my login
Create Account
·
Forgot Password
Forgotten password account recovery
RESOLVED FIXED
16395
Inspector needs to use cached resources
https://bugs.webkit.org/show_bug.cgi?id=16395
Summary
Inspector needs to use cached resources
Gavin Sherlock
Reported
2007-12-10 23:09:00 PST
Go to:
http://gcep.stanford.edu/
Inspect the page. Note the image generated by:
http://cgi.stanford.edu/group/gcep/cgi-bin/random_image.cgi?path=images/home/rotating
The version showing on the page will be different than the image show in the web inspector - thus, the web inspector must re-interrogate that url instead of using the already rendered image.
Attachments
patch
(12.62 KB, patch)
2010-10-20 07:14 PDT
,
Andrey Kosyakov
no flags
Details
Formatted Diff
Diff
patch
(12.14 KB, patch)
2010-10-21 01:48 PDT
,
Andrey Kosyakov
yurys
: review+
Details
Formatted Diff
Diff
patch to land
(13.71 KB, patch)
2010-10-26 05:50 PDT
,
Andrey Kosyakov
no flags
Details
Formatted Diff
Diff
Show Obsolete
(2)
View All
Add attachment
proposed patch, testcase, etc.
Timothy Hatcher
Comment 1
2007-12-20 16:33:20 PST
Yep, we need to sue the cached version and not re-fetch the image.
Alexander Pavlov (apavlov)
Comment 2
2010-10-20 06:30:09 PDT
***
Bug 47982
has been marked as a duplicate of this bug. ***
Andrey Kosyakov
Comment 3
2010-10-20 07:14:44 PDT
Created
attachment 71291
[details]
patch - Added base64Encoded parameter to resourceContent() in Inspector.idl - Cache resource content in resource.getContent() - Report whether content is encoded along with content in callback - Handle resource.getContent() properly if content is not yet available - Use data: url for images if resource content is available
Andrey Kosyakov
Comment 4
2010-10-21 01:48:55 PDT
Created
attachment 71405
[details]
patch - Fixed an attempt to retrieve content before it's available - Dropped an accidental change to Settings.js
Yury Semikhatsky
Comment 5
2010-10-21 03:07:31 PDT
Comment on
attachment 71405
[details]
patch View in context:
https://bugs.webkit.org/attachment.cgi?id=71405&action=review
> WebCore/inspector/front-end/Resource.js:621 > + if (typeof this._contentEncoded !== "undefined")
May be just use a separate flag _contetRequested?
> WebCore/inspector/front-end/Resource.js:635 > + if (this.frameID) {
style: please remove the brackets
Pavel Feldman
Comment 6
2010-10-21 03:29:20 PDT
Comment on
attachment 71405
[details]
patch View in context:
https://bugs.webkit.org/attachment.cgi?id=71405&action=review
> WebCore/inspector/front-end/Resource.js:615 > + return "data:" + this.mimeType + (this._contentEncoded ? ";base64," : ",") + this._content;
Could you please make sure it does not regress performance prior to landing? Opening slashdot / techcrunch and looking at the speed the icons show up with would be a good test.
Andrey Kosyakov
Comment 7
2010-10-26 05:50:49 PDT
Created
attachment 71867
[details]
patch to land - rebased - picked nits - put usage of data url in resource tree thumbnails under flag (enabled by default)
Andrey Kosyakov
Comment 8
2010-10-26 05:56:19 PDT
(In reply to
comment #6
)
> (From update of
attachment 71405
[details]
) > View in context:
https://bugs.webkit.org/attachment.cgi?id=71405&action=review
> > > WebCore/inspector/front-end/Resource.js:615 > > + return "data:" + this.mimeType + (this._contentEncoded ? ";base64," : ",") + this._content; > > Could you please make sure it does not regress performance prior to landing? Opening slashdot / techcrunch and looking at the speed the icons show up with would be a good test.
There was no visual slow-down on techcrunch (with ~120 images, the average latency was ca. 500ms, 900ms worst; remote debugging mode slower by factor of ~1.3). As discussed, putting usage of data urls in resource tree thumbnails under a flag (useDataURLForResourceImageIcons).
Andrey Kosyakov
Comment 9
2010-10-26 08:08:08 PDT
Manually committed
r70519
:
http://trac.webkit.org/changeset/70519
Note
You need to
log in
before you can comment on or make changes to this bug.
Top of Page
Format For Printing
XML
Clone This Bug