Re: [Dynapi-Dev] set border bug

[Benoit M. <mar...@ma...>]

I agree it would be nice to have a solution that works the same in all 
browser. What we could do if we want the borders to be outside is 
basically have a Frame object. The frame object would use 4 sub layers 
for  the borders. Which is enough for color border. Now we could offer 
image border, and then we would add 4 corner layers on to of the 4 
borders, to be able to handle resize nicely with repeating or streching 
the borders without affecting the corner images.
So setting a border would add a frame parent to the receiver, or may be 
simpler position a frame on top of the layer you set the border to. We 
have the code that would move the frame when you move the layer already
I don't know which one of the parent or the positioned on top is better 
? But both could/should be transparent to the user, and it should work 
the same in every browser.

Benoit



On Friday, April 25, 2003, at 04:16  PM, Kevin wrote:

>
> "Dan Willemsen" <da...@wi...> wrote:
>
>> On Fri, 2003-04-25 at 17:05, Raymond Irving wrote:
>>> Hello Everyone,
>>>
>>> I'm currently trying to get the setBorder() feature to
>>> work in DOM browsers the way it works in IE. According
>>> to the CSS1 specs the box width includes padding,
>>> margin border and the element's width.
>>
>> Wrong... the elements width is only the width declared, the padding 
>> and
>> border are not added in.  In the IE 5(and 6 in quirks mode) box model,
>> they are added in, which makes it a real pain for css developers, I
>> know, i design css sites now for the most part. Javascript is just
>> extra.
>
> Raymond refered to box width to include margin, border, padding and
> elements width. So both right. I know of the broken IE5 box model
> being a real pain.
>
> Solution.
>
> 1) I like the way borders are done in dynapi for ns4. Layers for top,
> bottom, left and right. So if we implement this for all browsers and
> not mess with the style it would be consistent, though dynlayer width
> will include border.
>
>>>  It would seems
>>> that DynLayers were design to only work with the
>>> elements width. One quick solution would be to set the
>>> actual element width to be less than twice the size of
>>> the border.
>
> 2) If we mess with border style for dom browsers. For the
> dynlayer size to consistent with ns4 border implementation
> it''s the dynlayer width that should be set to the actual element
> width plus twice the size of the border.
>
> I have a setBorder method that fixes the bug and should
> work ok:
>
> p.setBorder=function(w,c,s) {
>   var width=w||0;
>   var bstyle=(typeof(s)=='string')?s:'solid';
>   var bcolor=(typeof(c)=='string')?c:'black';
>   this._border=width+'px '+bstyle+' '+bcolor;
>   this._cssBorder='border:'+this._border+'; ';
>   var s=this.css;
>   if(s) {
>     s.borderWidth=width+'px';
>     s.borderStyle=bstyle;
>     s.borderColor=bcolor;
>   }
> };
>
> Just need to work out what the dynlayer w & h should be
> as the w & h of a dynlayer should be the same as when
> using a border (internal border layers) in ns4.
>
> -
> Kevin
>
>
>>> For example in dom browsers we would something like:
>>>
>>> p.setSize = function(w,h) {
>>> .....
>>>    if (cw) this.css.width =
>>> (this.w-(this._border*2))||0;
>>>    if (ch) this.css.height =
>>> (this.h-(this._border*2))||0;
>>> ....
>>> }
>>>
>>>
>>> What do you think? Any other suggestions? Can we do
>>> this?
>>>
>>>
>>> --
>>> Raymond Irving
>>>
>>> __________________________________________________
>>> Do you Yahoo!?
>>> The New Yahoo! Search - Faster. Easier. Bingo
>>> http://search.yahoo.com
>>>
>>>
>>> -------------------------------------------------------
>>> This sf.net email is sponsored by:ThinkGeek
>>> Welcome to geek heaven.
>>> http://thinkgeek.com/sf
>>> _______________________________________________
>>> Dynapi-Dev mailing list
>>> Dyn...@li...
>>> http://www.mail-archive.com/dyn...@li.../
>> -- 
>> Dan Willemsen <da...@wi...>
>>
>>
>>
>> -------------------------------------------------------
>> This sf.net email is sponsored by:ThinkGeek
>> Welcome to geek heaven.
>> http://thinkgeek.com/sf
>> _______________________________________________
>> Dynapi-Dev mailing list
>> Dyn...@li...
>> http://www.mail-archive.com/dyn...@li.../
>
>
>
> -------------------------------------------------------
> This sf.net email is sponsored by:ThinkGeek
> Welcome to geek heaven.
> http://thinkgeek.com/sf
> _______________________________________________
> Dynapi-Dev mailing list
> Dyn...@li...
> http://www.mail-archive.com/dyn...@li.../
>