Tuesday, January 17, 2006 8:14:45 AM
firefox, fieldset, float, bug
...
I really hate this regression bug in Firefox 1.5. It will really hurt many of web designers which decide to code pages in standard and accessible way.
That is -
Firefox 1.5 floating inside FIELDSET bug. You just can't make
any floating element inside FIELDSET.
Till now "Nobody's working on this, feel free to take it".
Update: Somebody now finally working on this. Now I hope that fix will be ready until next FF update.
Update: Fixed in Firefox 2.0 Alpha 2. In Firefox 1.5
many CSS form layouts still broken.
Tuesday, December 13, 2005 6:52:39 AM
firefox, CSS, bug, clear
...
Test
this page in different browsers.
Seems than in this particular case Firefox 1.5 doesn't get clear:left on the DL elements.Update: Ok. Clearing floats inside FIELDSET doesn't work in Firefox 1.5.
This bug was known before Firefox 1.5 release. They just left it in a hurry because:
Even though it is a regression, the impact is minor we are very close to lockdown for rc2 so it looks like it is not going to make it.
I hate regressions in CSS!
Thursday, October 27, 2005 6:26:17 PM
legend, float, overflow, bug
...
Just tested some of my intranet pages where I have fieldsets with applied CSS style overflow:hidden for clearing floating elements inside fieldset - and in Opera 9 TP1 it doesn't clear anything and do weired things.
Just one more thing to make it broken - add
position:relative in addition to
overflow:hidden.
This bug appears to be not connected to only fieldsets it also "works" with DIVs too.
I also create
separate test page for it.
Also notice the missing top border on the legends for fieldsets, and weird position of header "DIVs".
This test page also shows bug fixes in Opera 9 related to margins (there should be no bottom margin under first fieldset and first div).
Monday, October 24, 2005 4:43:27 PM
Opera, bug, marlin, opera 9
...
Test it yourself.
Update: Opera
aware of this "bug":
In a manner of speaking this is a bug. In a previous version of CSS2.1 "acronym {font-variant: small-caps;}" was suggested. This has later been reverted in the spec and there is a consensus that we should too.
Friday, October 21, 2005 6:45:11 PM
zoom, Opera, bug, GIF
...
With new impessive

Opera 9 TP1 we also get new GIF transparency bug. See this
test image in Opera 9 TP1 and any other browser including older Opera.
Also this test produce bug in Mac version of Opera 8 & 9 TP1 when you try to zoom (press [+] or [-] couple times).
I think this bug somehow connected with different bug which I named
Malevich Black Square.
Monday, September 26, 2005 7:53:18 PM
Opera, link, bug, anchor
I just accidentally found a regression bug in Opera 8.50.
Go to
this page and click to search link (on the left top corner of the page). Browser will scroll the page to the wery bottom (to the search section). Than scroll page to the top and try click on the search link one more time - in Opera 8.50 it doesn't work (I tried Windows and Mac versions). Opera 8.02 do this just fine along with all my browsers on desktop.
Will post this bug on Opera forum shortly.
Update: Didn't fixed in Opera 8.51, work fine in Opera 9 TP1.
Wednesday, September 14, 2005 1:50:10 AM
zoom, print, Opera, bug
...
Today I want to present one of my favourite Opera bugs in fun way. And because I see this bug during around 3 years (since first Opera 7 beta appears) I want to give this beatiful bug his own name - Malevich Black Square bug.


So, first of all
test page where we can see four Acid Smiles

Two of them (first and third) are images and other two are DIVs with image set as background.
Now funny part. Where Malevich Square here? Just change zoom on the page in your Opera browser or print out the page (needs Print Page Background option checked).
Last Acid Smile now will appears as black square!

I investigate this bug and found that it appears only if GIF image
- has only black color and transparency
- puts as background
- need to be smaller than 64x64 px
Have fun!

Also read
forum post about this bug.
Bug reports collection: #
198211, #
198161, #
126808, and one more I filled myself long time ago but forgot the number...
Test suites: 1,
2,
3Read more...