(curse words)

March 25, 2009

I’ve been a busy boy lately.  Still.  Thankfully, I’ve met a deadline as of noon Tuesday March 24th, and I’ll have more time to field comments, and hopefully write some patches.  Sounds crazy, no?  I haven’t even downloaded IE8 public yet.

Need a few days off first though.  Sorta burnt out.

Advertisements

IE8 RC1: slightly better

January 28, 2009

I installed IE8 RC1 in a VM this evening, and looked at a yet-unreleased rough version of DD_roundies 0.0.3a.  Images still fail to load, but the onresize event works!  Rendering the VML also “feels” faster, like there isn’t so much of a browser freeze.


Drew hasn’t disappeared.

January 27, 2009

I know a lot of you are waiting for various bug fixes for both DD_roundies and DD_belatedPNG.  I have had very little free time lately.

I did see that IE8 RC1 is being offered, which has me eager to look at it in the context of DD_roundies (FOR THE LOVE OF GOD DID THEY FIX THE ONRESIZE EVENT?).

Hold tight folks, I’m still dealing with a healthy dose of Too Much Sh** Going On.


DD_roundies 0.0.2a available

January 2, 2009

Multiple border thickness support, multiple corner radii support, very INCOMPLETE support for IE8, full support for PNG background images in IE6:

http://www.dillerdesign.com/experiment/DD_roundies/


DD_belatedPNG 0.0.7a available (DD_roundies 0.0.2a “almost there”)

January 1, 2009

A host of performance improvements and glitch repairs.

http://www.dillerdesign.com/experiment/DD_belatedPNG/#release0.0.7a


DD_roundies 0.0.2a will have IE8 support

December 20, 2008

I have read these pages:

http://groups.google.com/group/google-excanvas/browse_thread/thread/8762ed1ede73509

https://connect.microsoft.com/IE/feedback/ViewFeedback.aspx?FeedbackID=333905

I’ll be a few extra steps to get everything to apply easily in IE8 (it might mean different function calls for a developer making use of the library), but the raw VML will work correctly in IE8 standards mode.  The tests from the above links work well in my virtual machine, and some rendered source code stolen from IE7 and plopped into raw HTML works in IE8 – so long as it is all encapsulated inside a relative-position DIV.  I’m not sure why that last part is important, but it appears to be a key ingredient!


DD_roundies 0.0.2a: coming soon

December 18, 2008

Sorry this is taking quite so long.  I have been working on DD_roundies as I have found available time every day this week.

New features to expect:

Same background-position and background-repeat support that is present in DD_belatedPNG.

Multiple border-widths: the VML is no longer simply one or two nodes, with one of them having a strokeWeight property.  It is now three VML nodes: one for the background color, one for the background image (if exists), and one for multiple border thicknesses.  A single stroke weight would not suffice.

Multiple border-radii, modeled after Firefox’s implementation.  You can do implied property mirroring, too, such as ’10px 20px’ would actually be interpreted as ’10px 20px 10px 20px’.


    DD_belatedPNG: 0.0.6a

    December 14, 2008

    The fuzzy edges are no more.  If you still see them on floated elements, let me know, but I think I’ve got ’em licked:

    http://www.dillerdesign.com/experiment/DD_belatedPNG/#release0.0.6a

    For giggles, try scrolling to about halfway in the document, and find the box that has a black background color.  Click on it.  The BG image almost has a lensing effect, which I find myself clicking on repeatedly like a cat playing with a ball of yarn.


    DD_belatedPNG: 0.0.3a released

    December 10, 2008

    Now with <IMG/> support =)

    No use of pixel.gif to be found.  You can be sloppy with width/height attributes as well.

    http://www.dillerdesign.com/experiment/DD_belatedPNG/#release0.0.3a


    DD_belatedPNG updated to 0.0.2a

    December 9, 2008

    Now under the MIT license per another developer’s helpful suggestion (I really don’t want to get sued over writing free code), some minor changes under the hood:

    + Each matched element z-index is being copied, if specified.

    + Each matched element (and its parentNode) is forced into taking on hasLayout.

    http://www.dillerdesign.com/experiment/DD_belatedPNG/