RESOLVED CONFIGURATION CHANGED 20251
REGRESSION(r29955 - r30001): Misrendering at direct tv page for HD
https://bugs.webkit.org/show_bug.cgi?id=20251
Summary REGRESSION(r29955 - r30001): Misrendering at direct tv page for HD
Gavin Sherlock
Reported 2008-08-01 05:48:09 PDT
Steps to reproduce: 1. Go to http://www.directv.com/DTVAPP/global/contentPageNR.jsp?assetId=3420012 Expected results: The Get DIRECTV HD image button should render correctly Actual results: It has a line through it. Tested with r35417 on 10.5.4 on a MacPro. Note, this is a regression from 3.1.2 (5525.20.1)
Attachments
screenshot showing oddity on bottom left (10.31 KB, image/png)
2008-08-01 05:49 PDT, Gavin Sherlock
no flags
partial reduction (1.32 KB, text/html)
2008-08-14 07:33 PDT, Gavin Sherlock
no flags
better reduction (8.01 KB, text/html)
2008-08-24 23:44 PDT, Gavin Sherlock
no flags
Gavin Sherlock
Comment 1 2008-08-01 05:49:03 PDT
Created attachment 22596 [details] screenshot showing oddity on bottom left Screenshot displaying problem
Gavin Sherlock
Comment 2 2008-08-01 05:49:56 PDT
The "screenshot showing oddity on bottom left" was autofilled in the last comment. It was just supposed to say "screenshot".
Gavin Sherlock
Comment 3 2008-08-09 11:49:34 PDT
Regressed between r29955 and r30001.
Gavin Sherlock
Comment 4 2008-08-14 07:33:17 PDT
Created attachment 22791 [details] partial reduction Here's a partial reduction, though it still relies on some external resources. Compare nightly rendering to release Safari
Gavin Sherlock
Comment 5 2008-08-24 23:44:54 PDT
Created attachment 22973 [details] better reduction This relies on no external resources, I managed to eliminate all javascript (which was a red herring), and have embedded the images as data urls. Compare r29955 to r30001 to see the difference.
Gavin Sherlock
Comment 6 2008-08-24 23:55:35 PDT
The: html*#prospect-promo { [margin: -5px 0px 0px 0px;margin: -5px 0px 0px 0px;] } is crucial - remove just the "html*" part, and tot and 3.1.2 render the same. I don't understand what the "html*", but it is required to observe the regression.
Gavin Sherlock
Comment 7 2008-08-25 08:12:49 PDT
Gavin Sherlock
Comment 8 2008-12-11 12:51:00 PST
This problem is still present in r39090.
Alin S
Comment 9 2009-03-02 15:09:16 PST
This bug is fixed. no longer occurs.
Gavin Sherlock
Comment 10 2009-03-02 15:14:51 PST
In what build do you think it's fixed? The problem is still present in r41348. Look at the latest nightly, and compare to Safari 3.2 - the reduction renders differently, at least on my Mac. Are you on a different platform?
Gavin Sherlock
Comment 11 2009-03-02 15:15:56 PST
Note, do not go to the bug url (their page changes regularly). Instead use the reduction I made, which is in comment #5
Alin S
Comment 12 2009-03-02 15:23:12 PST
Indeed, this bug still occurs. I do not know if this is a problem with the browser its self or the site's compatibility with the browser. tried the reduction in Firefox and it works fine. so it is Webkit specific.
Gavin Sherlock
Comment 14 2020-06-25 11:58:53 PDT
(In reply to THE_N4R4NT from comment #13) > Created attachment 402759 [details] > Oalah Asu This is spam - not sure if it can be deleted
Ahmad Saleem
Comment 15 2022-08-30 04:46:10 PDT
I am able to reproduce this bug in Safari Technology Preview 152 using attached "better reduction" test case and it has "line" in the button and it is same across all browsers (Chrome Canary 107 and Firefox Nightly 106). Since all browsers are matching, I am going to mark this as "RESOLVED CONFIGURATION CHANGED". Please reopen if you think all browsers are doing wrong here. I will add few people for oversight and comment.
Note You need to log in before you can comment on or make changes to this bug.