Menu
▾
▴
dynapi-dev
|
From: Raymond I. <xw...@ya...> - 2003-09-14 16:43:25
Attachments:
test_boxfix.zip
|
Hi Everyone, I've managed to get DynLayers to now support CSS borders in ie, opera and mozilla. All DynLayers will render the same way in all browsers. NS4 does not work well with CSS so I don't know it if makes sense to try and create a fix for that. Should I add this to the setCSSBorder() to BorderManager or should we just create a different libary (e.g. CSSBorder)? -- Raymond Irving __________________________________ Do you Yahoo!? Yahoo! SiteBuilder - Free, easy-to-use web site design software http://sitebuilder.yahoo.com |
|
From: Kevin <ke...@ke...> - 2003-09-14 21:45:17
|
"Raymond Irving" <xw...@ya...> wrote: > Hi Everyone, > > I've managed to get DynLayers to now support CSS > borders in ie, opera and mozilla. All DynLayers will > render the same way in all browsers. NS4 does not work > well with CSS so I don't know it if makes sense to try > and create a fix for that. The setInnerBorder code could be changed now to add the Highlighters as correct outer dimensions for NS4 (would the new auto size stuff take care of the parents increased size?). > Should I add this to the setCSSBorder() to > BorderManager or should we just create a different > libary (e.g. CSSBorder)? Add the above suggested NS4 code to new setCSSBorder (default solid style) and put it in the bordermanager.js file with the FixBoxModel code. If setCSSBorder is called on a layer do a dynapi.onLoad FixBoxModel for that layer if not already added? I'm assuming the FixBoxModel code is doing something similar to the previously posted: http://jscript.dk/Util/BoxModelFix/ so all browsers have outer borders? I haven't had much time to delv into this lately. - Kevin > > -- > Raymond Irving |
|
From: Raymond I. <xw...@ya...> - 2003-09-15 13:21:28
|
--- Kevin <ke...@ke...> wrote: > The setInnerBorder code could be changed now to add > the > Highlighters as correct outer dimensions for NS4 > (would > the new auto size stuff take care of the parents > increased > size?). I would not want to modify the setInnerBorder() function as it should only be used to create inner borders. > Add the above suggested NS4 code to new setCSSBorder > (default solid style) and put it in the > bordermanager.js file > with the FixBoxModel code. If setCSSBorder is called > on > a layer do a dynapi.onLoad FixBoxModel for that > layer if > not already added? I'm thinging about modifing setBorder() to support both css and layer style borders: example setBorder(1,'black','solid') // css, layers for ns4 setBorder(1,'black','dotted') // css, layers for ns4 setBorder(1,'black','double') // css, layers for ns4 setBorder(1,'black','groove') // css, layers for ns4 setBorder(1,'black','layer') // use dynlayers setBorder(1,'black','image') // use images where solid will be the default and we could still do things like setBorder([t,r,b,l],[t,r,b,l],style) What do you think? > I'm assuming the FixBoxModel code is doing something > similar to the previously posted: > http://jscript.dk/Util/BoxModelFix/ > so all browsers have outer borders? I haven't had > much > time to delv into this lately. Yes it is -- Raymond Irving > - > Kevin > > > > > -- > > Raymond Irving > > > > ------------------------------------------------------- > 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.../ __________________________________ Do you Yahoo!? Yahoo! SiteBuilder - Free, easy-to-use web site design software http://sitebuilder.yahoo.com |
|
From: Kevin <ke...@ke...> - 2003-09-15 17:07:41
|
"Raymond Irving" <xw...@ya...> wrote: > > --- Kevin <ke...@ke...> wrote: > > The setInnerBorder code could be changed now to add > > the > > Highlighters as correct outer dimensions for NS4 > > (would > > the new auto size stuff take care of the parents > > increased > > size?). > > I would not want to modify the setInnerBorder() > function as it should only be used to create inner > borders. I agree. I meant the technique setInnerBorder uses with Highlighters can be done in setOuterBorder for NS4 only (will the new auto size stuff fix parent layer's increased dimensions??) and use CSS and FixBoxModel for all other browsers. > > Add the above suggested NS4 code to new setCSSBorder > > (default solid style) and put it in the > > bordermanager.js file > > with the FixBoxModel code. If setCSSBorder is called > > on > > a layer do a dynapi.onLoad FixBoxModel for that > > layer if > > not already added? > > I'm thinging about modifing setBorder() to support > both css and layer style borders: > > example > > setBorder(1,'black','solid') // css, layers for ns4 > setBorder(1,'black','dotted') // css, layers for ns4 > setBorder(1,'black','double') // css, layers for ns4 > setBorder(1,'black','groove') // css, layers for ns4 > setBorder(1,'black','layer') // use dynlayers > setBorder(1,'black','image') // use images Mmm Currently the third param is used for corner images. Also using layers setBorder already does: setBorder([0,0,2],'dotted.gif') // dotted underline etc. setBorder(2,'double.gif',['tr-double.gif','br-double.gif','bl-double.gif','tl-double.gif']) // well you can do anything you like in ns4 using this but the gifs have to be made in the correct color! How were you going to get color and style in ns4? Without having a colored layer for each dot or dash :( > where solid will be the default and we could still do > things like setBorder([t,r,b,l],[t,r,b,l],style) > > What do you think? Perhaps change my setBorder to setOuterBorder (change examples) and work on a new all singing and dancing setBorder(([t,r,b,l],[t,r,b,l],*styleORcornerimagearray) Also my defaulting mechanism may break so I wouldn't want to go in there again :) Currently setBorder detects a '.' in the second or third param arrays to mean an image or solid color. *styleORcornerimagearray='solid' || 'inset' || 'outset' || ... || [ 'tr.gif','br.gif','bl.gif','tl.gif' ] eg. For ns4 'inset' would use layers and 'horizInset.gif' and 'vertInset.gif' and predetermined corner inset images. If not a known 'style' string use the array of corner images. - Kevin > > > I'm assuming the FixBoxModel code is doing something > > similar to the previously posted: > > http://jscript.dk/Util/BoxModelFix/ > > so all browsers have outer borders? I haven't had > > much > > time to delv into this lately. > > Yes it is > > -- > Raymond Irving > > - > > Kevin > > > > > > > > -- > > > Raymond Irving > > > > > > > > > ------------------------------------------------------- > > 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.../ > > > __________________________________ > Do you Yahoo!? > Yahoo! SiteBuilder - Free, easy-to-use web site design software > http://sitebuilder.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.../ |
|
From: Raymond I. <xw...@ya...> - 2003-09-15 19:33:55
|
--- Kevin <ke...@ke...> wrote: > > I agree. I meant the technique setInnerBorder uses > with Highlighters > can be done in setOuterBorder for NS4 only (will the > new auto size > stuff fix parent layer's increased dimensions??) and > use CSS and > FixBoxModel for all other browsers. When layer are used as borders they are anchored to their parent. The new autosize feature only adjust the size of the layer whenever its content has been modified. I think there is a problem with the Frame objects and the autosize function. This I think has something to do with the way in which parent/child relationships are altered and the setSize functions. > > Mmm Currently the third param is used for corner > images. > > Also using layers setBorder already does: > setBorder([0,0,2],'dotted.gif') // dotted underline > etc. > setBorder(2,'double.gif',['tr-double.gif','br-double.gif','bl-double.gif','tl-double.gif']) > // well > you can do anything you like in ns4 using this but > the gifs have to be made in the correct > color! > > How were you going to get color and style in ns4? > Without having a colored layer for > each dot or dash :( NS4 will only support solid style using dynlayers. For dotted lines in ns4 a user can use the image section. > > Perhaps change my setBorder to setOuterBorder > (change examples) and work > on a new all singing and dancing > setBorder(([t,r,b,l],[t,r,b,l],*styleORcornerimagearray) > Also my defaulting mechanism may break so I wouldn't > want to go in there again :) > Currently setBorder detects a '.' in the second or > third param arrays to mean an image or > solid color. > > *styleORcornerimagearray='solid' || 'inset' || > 'outset' || ... || [ > 'tr.gif','br.gif','bl.gif','tl.gif' ] > > eg. > For ns4 'inset' would use layers and > 'horizInset.gif' and 'vertInset.gif' and > predetermined > corner inset images. If not a known 'style' string > use the array of corner images. Sounds good to me. I will make the necessary changes to the BorderManager once I've fiqured out a way to neatly fix the FixBoxModel stuff inside DynAPI without having to wait for the layers to be created. -- Raymond Irving > - > Kevin > > > > > > I'm assuming the FixBoxModel code is doing > something > > > similar to the previously posted: > > > http://jscript.dk/Util/BoxModelFix/ > > > so all browsers have outer borders? I haven't > had > > > much > > > time to delv into this lately. > > > > Yes it is > > > > -- > > Raymond Irving > > > - > > > Kevin > > > > > > > > > > > -- > > > > Raymond Irving > > > > > > > > > > > > > > > ------------------------------------------------------- > > > 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.../ > > > > > > __________________________________ > > Do you Yahoo!? > > Yahoo! SiteBuilder - Free, easy-to-use web site > design software > > http://sitebuilder.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.../ > > > > ------------------------------------------------------- > 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.../ __________________________________ Do you Yahoo!? Yahoo! SiteBuilder - Free, easy-to-use web site design software http://sitebuilder.yahoo.com |
|
From: Michael B. <mi...@mi...> - 2003-09-14 21:53:10
|
Raymond,
I STRONGLEY suggest that we render the borders inside the layers...
All the widgets that we create can NOT have a border outside the layer...
If a layer is "100px in width" then it needs to be a "100px in width"
The DynAPI is a precise tool so the layers NEED to be that to.
Well that's what I think :-)
I wrote something for the IOS a while back (included in the IOS release)
when I got sick of using extra lay's for borders (to reduce the memory).
"borderbutton.js"
This also takes care of all the parent's layer childern.
This is a old version and I made some dynapi3 hacks but you get the point.
There are some minor drawbacks (resizing) because the layers a manipulated
and need to be realign.
Why no use shorthand for the CSS model
lyr.setBorder("10px solid purple")
I've uploaded a test for you to see
http://www.michaelbystrom.com/download/dynapi_border.zip
How big zip file can I send through to this list??
I've also inclued a css for a grid that I use now and then to test the
locations of layers. I've also included some screen shots of examples.
Rock'n Roll
Michael
> Hi Everyone,
>
> I've managed to get DynLayers to now support CSS
> borders in ie, opera and mozilla. All DynLayers will
> render the same way in all browsers. NS4 does not work
> well with CSS so I don't know it if makes sense to try
> and create a fix for that.
>
> Should I add this to the setCSSBorder() to
> BorderManager or should we just create a different
> libary (e.g. CSSBorder)?
>
>
> --
> Raymond Irving
|
|
From: Kevin <ke...@ke...> - 2003-09-14 22:47:20
|
See below.
Michael wrote:
> Raymond,
>
> I STRONGLEY suggest that we render the borders inside the layers...
>
> All the widgets that we create can NOT have a border outside the layer...
> If a layer is "100px in width" then it needs to be a "100px in width"
If you set a 1px border in CSS for 100px wide box it will be in total 102px
The same for adding margin and padding it is an outer dimension and that's
the way it is (well except for the broken box model in IE5). If I have got
this wrong please point me to the correct reference thanks!
It would be good to emulate correct CSS behaviour in NS4 with adding
outer borders. Or more precisely:
1) increasing the size of the layer (and content auto centers) then adding
the borders inside
2) creating a new larger container layer center anchoring the original layer
then adding the borders inside.
Something like this is in dynapi3x anyway!
> The DynAPI is a precise tool so the layers NEED to be that to.
>
> Well that's what I think :-)
It is precise. We just all need to sing from the same style sheet. This has
been a subject of discussion in the past. If personal projects have had the
development of DynAPI move in one direction then it would be difficult
to change back to the CSS standard (for complicated page layouts).
I don't care either way I'm just making the point about consistency and
CSS. We've all put something in the dynapi project for own needs I guess.
> I wrote something for the IOS a while back (included in the IOS release)
> when I got sick of using extra lay's for borders (to reduce the memory).
> "borderbutton.js"
>
> This also takes care of all the parent's layer childern.
>
> This is a old version and I made some dynapi3 hacks but you get the point.
>
> There are some minor drawbacks (resizing) because the layers a manipulated
> and need to be realign.
>
> Why no use shorthand for the CSS model
> lyr.setBorder("10px solid purple")
There is something like that:
lyr.setBorder(10,'red')
or
lyr.setBorder([1,2,3,4],['red','green','blue','yellow'])
or
lyr.setBorder([5,10],['top.gif','right.gif','bottom.gif','left.gif')
or
different side and corner images too.
> I've uploaded a test for you to see
> http://www.michaelbystrom.com/download/dynapi_border.zip
>
> How big zip file can I send through to this list??
Quite small I think about 30k.
_
Kevin
> I've also inclued a css for a grid that I use now and then to test the
> locations of layers. I've also included some screen shots of examples.
>
> Rock'n Roll
> Michael
>
>
>
> > Hi Everyone,
> >
> > I've managed to get DynLayers to now support CSS
> > borders in ie, opera and mozilla. All DynLayers will
> > render the same way in all browsers. NS4 does not work
> > well with CSS so I don't know it if makes sense to try
> > and create a fix for that.
> >
> > Should I add this to the setCSSBorder() to
> > BorderManager or should we just create a different
> > libary (e.g. CSSBorder)?
> >
> >
> > --
> > Raymond Irving
>
>
>
> -------------------------------------------------------
> 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.../
|
|
From: Michael B. <mi...@mi...> - 2003-09-15 07:12:36
|
Den 03-09-15 00.47, skrev "Kevin" <ke...@ke...>: > Michael wrote: > >> Raymond, >> >> I STRONGLEY suggest that we render the borders inside the layers... >> All the widgets that we create can NOT have a border outside the layer... >> If a layer is "100px in width" then it needs to be a "100px in width" > If you set a 1px border in CSS for 100px wide box it will be in total 102px > The same for adding margin and padding it is an outer dimension and that's > the way it is (well except for the broken box model in IE5). If I have got > this wrong please point me to the correct reference thanks! > Kevin Jepp! the borders are suppose to go on the outside http://www.w3.org/TR/REC-CSS1#formatting-model http://www.w3.org/TR/REC-CSS2/box.html#box-dimensions ... all boxes have a core content area with optional surrounding padding, border and margin areas. I just found it easier to work with borders if 100px is a 100px and not 98px using a 1px border Should stick to CSS2 recommendations that's perfectly fine with me Just a idea. :) Michael >>> Hi Everyone, >>> >>> I've managed to get DynLayers to now support CSS >>> borders in ie, opera and mozilla. All DynLayers will >>> render the same way in all browsers. NS4 does not work >>> well with CSS so I don't know it if makes sense to try >>> and create a fix for that. >>> >>> Should I add this to the setCSSBorder() to >>> BorderManager or should we just create a different >>> libary (e.g. CSSBorder)? >>> >>> >>> -- >>> Raymond Irving |
|
From: Raymond I. <xw...@ya...> - 2003-09-15 13:14:31
|
We will support both inner and outer borders in border manager. So which ever way the use wants it then they got it. -- Raymond Irving --- Michael Bystrom <mi...@mi...> wrote: > Den 03-09-15 00.47, skrev "Kevin" > <ke...@ke...>: > > > Michael wrote: > > > >> Raymond, > >> > >> I STRONGLEY suggest that we render the borders > inside the layers... > >> All the widgets that we create can NOT have a > border outside the layer... > >> If a layer is "100px in width" then it needs to > be a "100px in width" > > > If you set a 1px border in CSS for 100px wide box > it will be in total 102px > > The same for adding margin and padding it is an > outer dimension and that's > > the way it is (well except for the broken box > model in IE5). If I have got > > this wrong please point me to the correct > reference thanks! > > Kevin > > Jepp! the borders are suppose to go on the outside > http://www.w3.org/TR/REC-CSS1#formatting-model > http://www.w3.org/TR/REC-CSS2/box.html#box-dimensions > > ... all boxes have a core content area with optional > surrounding padding, > border and margin areas. > > I just found it easier to work with borders if 100px > is a 100px and not 98px > using a 1px border > > Should stick to CSS2 recommendations that's > perfectly fine with me > > Just a idea. > > :) > > Michael > > >>> Hi Everyone, > >>> > >>> I've managed to get DynLayers to now support CSS > >>> borders in ie, opera and mozilla. All DynLayers > will > >>> render the same way in all browsers. NS4 does > not work > >>> well with CSS so I don't know it if makes sense > to try > >>> and create a fix for that. > >>> > >>> Should I add this to the setCSSBorder() to > >>> BorderManager or should we just create a > different > >>> libary (e.g. CSSBorder)? > >>> > >>> > >>> -- > >>> Raymond Irving > > > > ------------------------------------------------------- > 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.../ __________________________________ Do you Yahoo!? Yahoo! SiteBuilder - Free, easy-to-use web site design software http://sitebuilder.yahoo.com |
|
From: Michael B. <mi...@mi...> - 2003-09-16 07:01:02
|
Den 03-09-15 15.14, skrev "Raymond Irving" <xw...@ya...>: > > We will support both inner and outer borders in border > manager. So which ever way the use wants it then they > got it. Perfectly prefect Michael |
×
Want the latest updates on software, tech news, and AI?
Get latest updates about software, tech news, and AI from SourceForge directly in your inbox once a month.
Thanks for helping keep SourceForge clean.
X