I'll be posting screenshots here.
Created attachment 51887 [details] [IMAGE] Firebug-alike version. (Computed style is a Styles' sibling below the screen).
Created attachment 51923 [details] [IMAGE] Screenshot while running with patch.
Created attachment 51948 [details] [IMAGE] Screenshot while running with patch.
Created attachment 51949 [details] [IMAGE] Screenshot while running with patch (2)
Created attachment 51950 [details] [PATCH] Proposed change.
Comment on attachment 51950 [details] [PATCH] Proposed change. We will redesign / land later.
I got this via email: Hi, thanks for commenting on my blog post about my Web Inspector wish list. I took a good look at your screenshot from bug #36747, and I really like it, but I'd go even further. I've Photoshopped your patch screenshot and attached it. I realize it is almost a carbon copy of Firebug, but it's with reason. Please follow along: - Moved the big sections to tabs. It makes a lot more sense to me, and they're much easier to find this way. We would however have to come up with a solution for the amount of tabs and the limited space. - Added a bit more spacing everywhere and removed the grey lines between selectors – allows everything to "breathe" a little more. - Removed all underlined links, they are all blue anyway, and this further reduces clutter. - Removed the element styles section, they should go on top of matching rules. I like the Firebug way of only showing them when there are any, and using an option in the contextual menu to add one. - Removed user agent styles, as per my request from https://bugs.webkit.org/show_bug.cgi?id=37766 Combine that with - putting the checkbox to disable/enable styles to the left of the property (https://bugs.webkit.org/show_bug.cgi?id=37770) - remembering split pane positions (https://bugs.webkit.org/show_bug.cgi?id=37767) - a slightly bigger font-size perhaps ... and I'd say it's a worthy competitor to Firebug. :) So, what do you think? Thanks a lot! Lennart
Created attachment 53651 [details] [IMAGE] Web Inspector mockup from Lennart
Created attachment 55583 [details] [IMAGE] Screenshot while running with patch (updated).
Created attachment 55584 [details] [PATCH] Rebaselined change.
(In reply to comment #10) > Created an attachment (id=55584) [details] > [PATCH] Rebaselined change. You need to use the --binary flag with git-diff: http://webkit-commit-queue.appspot.com/results/2214120
Created attachment 55594 [details] [PATCH] Same with binary diff. Are you saying CQ is now capable of landing git binary diffs as well? Nice!
(In reply to comment #12) > Created an attachment (id=55594) [details] > [PATCH] Same with binary diff. > > Are you saying CQ is now capable of landing git binary diffs as well? Nice! As of several months ago I think. I wasn't responsible for that addition though -- just the nicer error message. :)
(In reply to comment #12) > Created an attachment (id=55594) [details] > [PATCH] Same with binary diff. > > Are you saying CQ is now capable of landing git binary diffs as well? Nice! Oh, well, you tried. It doesn't completely support it yet: Failed to run "[u'/mnt/git/webkit-style-queue/WebKitTools/Scripts/svn-apply', u'--force']" exit_code: 9 Parsed 6 diffs from patch file(s). patching file LayoutTests/inspector/elements-panel-styles-expected.txt patching file WebCore/ChangeLog Hunk #1 succeeded at 1 with fuzz 3. only literal type is supported now at /mnt/git/webkit-style-queue/WebKitTools/Scripts/svn-apply line 256. https://webkit-commit-queue.appspot.com/results/2166128 It's a FIXME for svn-apply to support git binary diffs having the "delta" format. I'm not sure if there is a way to force the use of "literal".
> It's a FIXME for svn-apply to support git binary diffs having the "delta" format. > I'm not sure if there is a way to force the use of "literal". A little while ago I asked in #git and looked at the git source code. There currently is no way to force it to use literal. =/
> Oh, well, you tried. It doesn't completely support it yet: Heh. That's where I thought we were. That is why I am not cooking binary diffs and land binary-related stuff by hand.
(In reply to comment #16) > > Oh, well, you tried. It doesn't completely support it yet: > > Heh. That's where I thought we were. That is why I am not cooking binary diffs and land binary-related stuff by hand. I filed a bug for this here: https://bugs.webkit.org/show_bug.cgi?id=38864
Comment on attachment 55594 [details] [PATCH] Same with binary diff. WebCore/inspector/front-end/StylesSidebarPane.js:639 + collapse: function(dontRememberState) Maybe remove this too? The patch looks sane.
Committing to http://svn.webkit.org/repository/webkit/trunk ... M LayoutTests/inspector/elements-panel-styles-expected.txt M WebCore/ChangeLog M WebCore/English.lproj/localizedStrings.js M WebCore/inspector/InspectorDOMAgent.cpp M WebCore/inspector/front-end/Checkbox.js M WebCore/inspector/front-end/ElementsPanel.js M WebCore/inspector/front-end/StylesSidebarPane.js M WebCore/inspector/front-end/WorkersSidebarPane.js M WebCore/inspector/front-end/inspector.css M WebCore/inspector/front-end/treeoutline.js Committed r59150
Inspector is no longer showing any computed style for some elements for me now.
A patch like this could really use more detail in the ChangeLog. If not when it goes into Bugzilla (via an automated tool like webkit-patch) then most certainly before landing!