Menu
▾
▴
dynapi-dev — List related to the current DynAPI Core development
You can subscribe to this list here.
| 2000 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
(75) |
Nov
(252) |
Dec
(418) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2001 |
Jan
(659) |
Feb
(1039) |
Mar
(870) |
Apr
(235) |
May
(329) |
Jun
(251) |
Jul
(123) |
Aug
(119) |
Sep
(67) |
Oct
(194) |
Nov
(535) |
Dec
(133) |
| 2002 |
Jan
(122) |
Feb
(24) |
Mar
(29) |
Apr
(28) |
May
(16) |
Jun
(20) |
Jul
(11) |
Aug
(12) |
Sep
(13) |
Oct
(14) |
Nov
(23) |
Dec
(19) |
| 2003 |
Jan
(28) |
Feb
(170) |
Mar
(288) |
Apr
(211) |
May
(126) |
Jun
(166) |
Jul
(131) |
Aug
(102) |
Sep
(211) |
Oct
(301) |
Nov
(22) |
Dec
(6) |
| 2004 |
Jan
(14) |
Feb
(16) |
Mar
(7) |
Apr
|
May
(8) |
Jun
(25) |
Jul
(21) |
Aug
(2) |
Sep
(7) |
Oct
|
Nov
(2) |
Dec
(1) |
| 2005 |
Jan
(4) |
Feb
(2) |
Mar
(14) |
Apr
(24) |
May
(3) |
Jun
(7) |
Jul
(30) |
Aug
(5) |
Sep
(1) |
Oct
(3) |
Nov
|
Dec
(1) |
| 2006 |
Jan
|
Feb
|
Mar
|
Apr
(1) |
May
|
Jun
|
Jul
|
Aug
|
Sep
(4) |
Oct
|
Nov
|
Dec
|
|
From: Michael P. <mp...@ph...> - 2001-01-10 00:27:05
|
<!doctype html public "-//w3c//dtd html 4.0 transitional//en"> <html> I've found that there are two sets of arrays that store DynDocument references. <p>If the line in DynDocument is changed from: <br><tt> DynDocument.dyndocs[DynAPI.documents.length]=this;</tt> <br>If the line in DynDocument is changed from: <br><tt> DynDocument.dyndocs[DynDocument.dyndocs.length]=this;</tt><tt></tt> <p>Then there is no need for the following lines in DynAPI.js: <br><tt> documents : [],</tt> <br><tt> documentsID : [],</tt> <br><tt></tt> <p>-- <br>Michael Pemberton <br>mp...@ph... <br>ICQ: 12107010 <br> </html> |
|
From: Dann <da...@to...> - 2001-01-10 00:20:58
|
Great. I myself would be greatful if we all would just get this 'version' to work 100%. I can sympathize with everybody here doing their fancy walks on IDE's, OOP and what not. The problem is, everybody seems to forget that this effort is remarkable in that it attempts to close a gap under very difficult circumstances. Anyone with javascript programming in her/his fingers knows that it is a far cry from a solid base to build on - and I'm just talking about one single instance browser/platform. Bring on the incompatibilities, browser/platform dependencies, implementation differences and the complete bug circus that comes with it, and you end up with something of a nightmare. Snap out of it, people : there will never be anything even close to the extent of what you are dreaming about here, given the circumstances. What *is* feasible is being done here already as much as possible. Build a stable core (keep it simple - keep it simple - keep it simple) API to raise the bar on what you can do in the dreary HTML/CSS universe. And let's face it, while we're doing it, we're already overkilling it so much that there's nothing left of what the basic idea of hypertext was al about. Must sleep now. CU, Dann Eytan Heidingsfeld wrote: > Like I have mentioned many times before I don't like how DynAPI is used. It > is an API not a language, but I'm getting ahead of myself. Let's start at > the beginning. In the beginning there is JS. Problem number one is that JS > is class oriented and not object oriented therefore the "weird" inheriting. > That problem was later solved by my Inherit method. Problem Number two is > that although it is a program language that is class oriented it doesn't > give you the necesary building blocks to use it correctly. Because of this > reason I decided to write the OOJS (Object Oriented Java Script). Before I > explain my ideas for how when and where (which you are free and are > encouraged to criticize) I will warn you that unlike JS which is a scrambled > mix of Java, C and some other languages this OOJS is Delphi like in syntax. > The reason for this is that when I started I was doing this for a certain > project which I would be working on with Delphi programmers. Now for the > OOJS. > > First of all you have the most base object TObject. All objects inherit from > it! It includes a small amount of properties and methods including a create > and a destroy (and there events). Then you have TComponent. TComponent is > the base class used for visual objects. How does it work. It uses the DynAPI > as I think it is supposed to be used, as an API. That means that I have a > DynLayer as a private variable. Then I have methods as such DrawComponent, > Draw and events such as onPaint or onRefresh. Then if you want to create > your visual component all you have to do is inherit from TComponent and call > the correct methods. As you may have noticed the event structure is > different too. It is like in Delphi (big surprise) you define a type which > is basically a definition of a procedure then you implement it and set a > property of your object with this procedure and when the object wants to > invoke it calls its property for the event. > > This is just the beginning and I know it is very jumbled. By tomorrow I will > send out the basic code I have till now and anyone who is interested in > helping please email me. > > Work needed to be done: > A. Finish fixing up base types > B. Rewrite DynDocument to a new type called TApplication (no biggie since > DynDocument is 80 lines) > C. Start converting objects from the library to the new OOJS model (easy > enough once model is done) > > Any Takers, > > Thanx, > 8an > > _______________________________________________ > Dynapi-Dev mailing list > Dyn...@li... > http://lists.sourceforge.net/mailman/listinfo/dynapi-dev |
|
From: Raymond S. <dst...@or...> - 2001-01-09 22:29:31
|
These are remaining issues, that I know of. 1) Event problems. None of my widgets 'on mouseover' work right now in IE or NS since precreation was introduced. If you test ButtonImage you can see what I mean. Also, this widget is throwing an error at line 49 (I believe) in NS. Line 49 is related to setImage with a boolean for checkbox vs not a check box. This might be related to removal of get.component methods. Worked in IE til I updated them to the new methods. But I am not sure what replaced this, so I have to go back and read up on alot of code I am unfamiliar with. 2) Scrollpane currently doesn't render knobs and bars in IE5.5 on Inw ME. Works fine in NS. 3) Select text Label issues (ie) in child elements. Pascal mentioned he was working on this last night. 4) Hard to explain this one but it might provide clues as to where to look. I had a viewport widget with an oversized image element (content). I used a series of moveTo's to window a portion of the element in a viewport. All worked wonderous til recent updates, now moveTo does nothing. I fixed this by dumping the viewport and using a clipped dynlayer with moveTo's. Might be related to auto-size detection, just a guess. Later |
|
From: Bart B. <ba...@ho...> - 2001-01-09 22:26:08
|
This is infact interesting.. I am interested to see how it goes :)) I have created(still working on it) a SuperClass object which enables = "true" OO inheritance for any function... (I used Brandon Myers = SuperClass as a starting point.) It includes: - Platform/browser compatability (currently works on IE4+, Netscape = 4.05+, Opera 5 ; PC and Macintosh)=20 - Abstract Classes (currently working on private and protected methods = and variables) - proper inheritance, i.e. prototype and Object method reference = inheritance, with Java like syntax, i.e. theClass.Extends(theSuperClass) - inheriting specific functions from any Object - Multiple Inheritance - a Class Tree and access to the entire ClassTree from any instance or = Class - ability to call Super(), up the entire Class tree, in both methods and = and constructors, no argument evaling=20 - correct construction of Classes, with a call to SuperClass and = instance id's set implicitly.=20 - subClassOf(), instanceOf(), real destroy() =20 - possibility to add inherited Object methods - Instance tracking - possibility to clone instances And probably more to come :) -----Ursprungligt meddelande----- Fr=E5n: Eytan Heidingsfeld <ey...@tr...> Till: Dynapi-Dev <dyn...@li...> Datum: den 9 januari 2001 20:56 =C4mne: [Dynapi-Dev] New OOJS >Like I have mentioned many times before I don't like how DynAPI is = used. It >is an API not a language, but I'm getting ahead of myself. Let's start = at >the beginning. In the beginning there is JS. Problem number one is that = JS >is class oriented and not object oriented therefore the "weird" = inheriting. >That problem was later solved by my Inherit method. Problem Number two = is >that although it is a program language that is class oriented it = doesn't >give you the necesary building blocks to use it correctly. Because of = this >reason I decided to write the OOJS (Object Oriented Java Script). = Before I >explain my ideas for how when and where (which you are free and are >encouraged to criticize) I will warn you that unlike JS which is a = scrambled >mix of Java, C and some other languages this OOJS is Delphi like in = syntax. >The reason for this is that when I started I was doing this for a = certain >project which I would be working on with Delphi programmers. Now for = the >OOJS. > >First of all you have the most base object TObject. All objects inherit = from >it! It includes a small amount of properties and methods including a = create >and a destroy (and there events). Then you have TComponent. TComponent = is >the base class used for visual objects. How does it work. It uses the = DynAPI >as I think it is supposed to be used, as an API. That means that I have = a >DynLayer as a private variable. Then I have methods as such = DrawComponent, >Draw and events such as onPaint or onRefresh. Then if you want to = create >your visual component all you have to do is inherit from TComponent and = call >the correct methods. As you may have noticed the event structure is >different too. It is like in Delphi (big surprise) you define a type = which >is basically a definition of a procedure then you implement it and set = a >property of your object with this procedure and when the object wants = to >invoke it calls its property for the event. > >This is just the beginning and I know it is very jumbled. By tomorrow I = will >send out the basic code I have till now and anyone who is interested in >helping please email me. > >Work needed to be done: >A. Finish fixing up base types >B. Rewrite DynDocument to a new type called TApplication (no biggie = since >DynDocument is 80 lines) >C. Start converting objects from the library to the new OOJS model = (easy >enough once model is done) > >Any Takers, > >Thanx, >8an > > >_______________________________________________ >Dynapi-Dev mailing list >Dyn...@li... >http://lists.sourceforge.net/mailman/listinfo/dynapi-dev > |
|
From: Raymond S. <dst...@or...> - 2001-01-09 21:55:51
|
I was beginning to wonder where Dan went off to, now we know. It does appear that we've created a bit of a two headed hydra here, as others mentioned might happen. Dan's off working on a DynBuilder while other pontificate about the holy-grail of RAD tools to speed up "the still bugged version of DynAPI2" that hasn't had a refresh since November of 2000. While I do my best to muddle thru this code and help Robert debug this stuff I fear I submit about as much confusion as contribution. But I get the feeling that others are not really 'actively debugging' the current CVS. Dan's working on a GUI which he will 'share' with us, thus allowing us to further defocus our efforts while other do grand debate on what this "thing" should be. It would be nice to, at a minimum, clean up the current base code in the CVS before we defocus the limited masses even more. Sorry for the rant. But ALOT of change was introduced with precreate and inline creation integration that still needs to be cleaned up. It seems the only really active debugger right now is Robert and Pascal (on a limited basis). |
|
From: Doug M. <do...@cr...> - 2001-01-09 21:32:51
|
----- Original Message ----- From: "Dan Steinman" <dy...@fu...> To: <dyn...@li...> Sent: Tuesday, January 09, 2001 6:58 PM Subject: [Dynapi-Dev] Making progress on a Java gui tool > Just wanted to let you guys know I'm taking on the task of getting a basic foundation laid down for a DynAPI authoring tool in Java. I'm not a professional Java programmer but capable of getting a good start to the app. Hopefully after I get the initial code working, others can pick away at my code and get it set up for more functionality to be added. As well to see if this is something worthwhile putting more effort into. Be Sure to comment and document your code thorughly.. That's the one thing that REALLY bugs me about working with the DynAPI - the lack of internal documentation (comments) > > I might not get to the point of saving/loading files before releasing the code, perhaps someone else can handle that part while I concentrate on the interface. > > It will initally just be a like a drawing program that only draws rectangles, selects between them, resizes, and colors them. In the back of my mind I've thought of >a few ways to handle nested layers. I know that I didn't like how Dreamwaver did it and sure I can come up with something more suitable for this app. It should >be quite easy to add support for adding image and text support (althought I don't know how to add wrapping text using the 2D API). get the font width (not sure how) myRect.width / FontWidth = NumberOfCharsWide so let's say the above returned 15 then we insert a '\n' after each 15 chars... This will simulate wrapping text.. (I hope) > I also don't know how adding widgets should be done yet either. > > I'm reusing a lot of code that I wrote for my imagemap program, so up to this point it's been going very quickly. I could probably get something out the door in a day or 2, I'll be away for all next week so I want to have something that others can look at and play with before then. > > Just wanted to update ya. > > Dan > > _______________________________________________ > Dynapi-Dev mailing list > Dyn...@li... > http://lists.sourceforge.net/mailman/listinfo/dynapi-dev > |
|
From: Dan S. <dy...@fu...> - 2001-01-09 21:16:29
|
Just wanted to let you guys know I'm taking on the task of getting a basic foundation laid down for a DynAPI authoring tool in Java. I'm not a professional Java programmer but capable of getting a good start to the app. Hopefully after I get the initial code working, others can pick away at my code and get it set up for more functionality to be added. As well to see if this is something worthwhile putting more effort into. I might not get to the point of saving/loading files before releasing the code, perhaps someone else can handle that part while I concentrate on the interface. It will initally just be a like a drawing program that only draws rectangles, selects between them, resizes, and colors them. In the back of my mind I've thought of a few ways to handle nested layers. I know that I didn't like how Dreamwaver did it and sure I can come up with something more suitable for this app. It should be quite easy to add support for adding image and text support (althought I don't know how to add wrapping text using the 2D API). I also don't know how adding widgets should be done yet either. I'm reusing a lot of code that I wrote for my imagemap program, so up to this point it's been going very quickly. I could probably get something out the door in a day or 2, I'll be away for all next week so I want to have something that others can look at and play with before then. Just wanted to update ya. Dan |
|
From: Eytan H. <ey...@tr...> - 2001-01-09 19:57:56
|
Like I have mentioned many times before I don't like how DynAPI is used. It is an API not a language, but I'm getting ahead of myself. Let's start at the beginning. In the beginning there is JS. Problem number one is that JS is class oriented and not object oriented therefore the "weird" inheriting. That problem was later solved by my Inherit method. Problem Number two is that although it is a program language that is class oriented it doesn't give you the necesary building blocks to use it correctly. Because of this reason I decided to write the OOJS (Object Oriented Java Script). Before I explain my ideas for how when and where (which you are free and are encouraged to criticize) I will warn you that unlike JS which is a scrambled mix of Java, C and some other languages this OOJS is Delphi like in syntax. The reason for this is that when I started I was doing this for a certain project which I would be working on with Delphi programmers. Now for the OOJS. First of all you have the most base object TObject. All objects inherit from it! It includes a small amount of properties and methods including a create and a destroy (and there events). Then you have TComponent. TComponent is the base class used for visual objects. How does it work. It uses the DynAPI as I think it is supposed to be used, as an API. That means that I have a DynLayer as a private variable. Then I have methods as such DrawComponent, Draw and events such as onPaint or onRefresh. Then if you want to create your visual component all you have to do is inherit from TComponent and call the correct methods. As you may have noticed the event structure is different too. It is like in Delphi (big surprise) you define a type which is basically a definition of a procedure then you implement it and set a property of your object with this procedure and when the object wants to invoke it calls its property for the event. This is just the beginning and I know it is very jumbled. By tomorrow I will send out the basic code I have till now and anyone who is interested in helping please email me. Work needed to be done: A. Finish fixing up base types B. Rewrite DynDocument to a new type called TApplication (no biggie since DynDocument is 80 lines) C. Start converting objects from the library to the new OOJS model (easy enough once model is done) Any Takers, Thanx, 8an |
|
From: <no...@so...> - 2001-01-09 19:39:37
|
Bug #128201, was updated on 2001-Jan-09 11:39
Here is a current snapshot of the bug.
Project: DynAPI 2
Category: None
Status: Open
Resolution: None
Bug Group: None
Priority: 5
Submitted by: nobody
Assigned to : nobody
Summary: Duplicating layers in case of NS4 when resizing the browser
Details: We should avoid calling invokeEvent('create') in
DynLayer.createElement when
it is called from DynDocument.recreateAll. (The latter is called after the
NS 4.x browser window is resized.)
The 'create' event duplicates all the layers which are created in custom
widgets' onCreate event handlers
when we resize the NS 4.x browser.
For detailed info, follow this link:
http://sourceforge.net/bugs/?func=detailbug&bug_id=128201&group_id=5757
|
|
From: Richard :o <ma...@ri...> - 2001-01-09 19:19:49
|
Hi, It's nice to see you're updating your site, it's been a great reference point. (and blue's a lot better than green :o) > PS: How is it that a machine capable to run Q3A, Messiah, Carmageddon 2000, Colin > McRae 2 You know all these programming languages??? Cheers, Richard :o ma...@ri... www.richardinfo.com (Everything running on, and ported to the 19/12/2000 snapshot of DynAPI2) ----- Original Message ----- From: "Jordi 'IlMaestro' Ministral" <jmi...@or...> To: <dyn...@li...> Sent: Tuesday, January 09, 2001 11:57 AM Subject: Re: [Dynapi-Dev] Latest Issues > I am currently working on an scalable menu interface that should provide a > solution for most of these. I hope to release it as soon as it is fully tested an > I can write some related docs and tutorials. Releasing a standalone bunch of > cross-related js files would only add more confusion to the current situation. > > You can check my site for news and info about my progress. > > www.cantir.com/dynapi > > PS: How is it that a machine capable to run Q3A, Messiah, Carmageddon 2000, Colin > McRae 2, etc... can't create and move several layers in a reasonable time ? > Browser developers might as well go back to college and try to learn something > this time. > > Robert Rainwater wrote: > > > I updated CVS and the snapshot with some changes to the label and list > > widget that should get the list widget working. However, the event > > problems in IE are still causing problems. Also, the onmouseover event > > is never called for the ListItems. This could be related to the Label > > problems. > > > > Also, the list widget is very slow and the events are lagging behind. > > I think part of this is the constant rewriting of the html. Anyway to > > improve this? > > > > -- > > // Robert Rainwater > > > > On 1/8/2001, 12:13:02 AM EST, Robert wrote about "[Dynapi-Dev] Latest Issues": > > > > > I want to release a new version of the DynAPI soon, but there are a > > > few things preventing it. > > > > > 1) In IE 5 the events inside of text in the label widget are broke > > > again. > > > > > 2) The list widget does not work with the updated dynlayer creation > > > system. > > > > > 3) Events in NS 6 are firing twice. There is a patch for this at > > > sourceforge. Has this patch been tested by anyone? > > > > > If we can resolve these issues, then I think we can release a new > > > version. > > > > _______________________________________________ > > Dynapi-Dev mailing list > > Dyn...@li... > > http://lists.sourceforge.net/mailman/listinfo/dynapi-dev > > > _______________________________________________ > Dynapi-Dev mailing list > Dyn...@li... > http://lists.sourceforge.net/mailman/listinfo/dynapi-dev > ____________________________________________________________ > Get your free domain name and domain-based e-mail from > Namezero.com. New! Namezero Plus domains now available. > Find out more at: http://www.namezero.com > |
|
From: Doug M. <do...@cr...> - 2001-01-09 18:02:43
|
As if it where that simple!!!
I spent a good month trying to do just that..
Shows you waht can happen when you gather many minds into a forum such =
as this eh?
Good job.
Doug Melvin
----- Original Message -----=20
From: Michael Pemberton=20
To: dyn...@li...=20
Sent: Tuesday, January 09, 2001 6:18 AM
Subject: [Dynapi-Dev] NS4 and document size adjustment when adding =
children
I've found a oneway method of allowing NS4 to increase the document =
size if the child is outside the current dimensions. Unfortunately, =
once the size is increased, it can't be decreased.=20
Just add the following at the end of the Dynlayer.prototype.addChild:=20
if (this.isDynDocument && is.ns4) {=20
if (child.h+child.y>this.h) =
document.height=3Dthis.h=3Dchild.h+child.y;=20
if (child.w+child.x>this.w) =
document.width=3Dthis.w=3Dchild.w+child.x;=20
};=20
=20
--=20
Michael Pemberton=20
mp...@ph...=20
ICQ: 12107010=20
_______________________________________________ Dynapi-Dev mailing =
list Dyn...@li... =
http://lists.sourceforge.net/mailman/listinfo/dynapi-dev
|
|
From: Robert R. <rra...@ya...> - 2001-01-09 16:31:55
|
A possible fix is to use:
if (dlyr.clip)
That should fix the problem I think.
--
// Robert Rainwater
On 1/9/2001, 3:50:22 AM EST, noreply wrote about "[Dynapi-Dev] [Bug #128146] netscape addChild() bug":
> Bug #128146, was updated on 2001-Jan-09 00:50
> Here is a current snapshot of the bug.
> Project: DynAPI 2
> Category: Core API
> Status: Open
> Resolution: None
> Bug Group: None
> Priority: 5
> Submitted by: nobody
> Assigned to : nobody
> Summary: netscape addChild() bug
> Details: this basic code:
> DynAPI.onLoad=function() {
> lay = new DynLayer(null,0,0,280,12,'red');
> DynAPI.document.addChild(lay);
> };
> doesn't work under netscape 4.76.
> ("clip is not defined. ")
> this is because the clip-check on line 75 (im using
> the latest files from cvs):
> if (dlyr.clip!=null) {
> isn't correct, dlyr.clip is undefined when i check the value.
> line 272 has the same problems, dlyr.clip is undefined
> but it only checks for null
> maybe a possible fix would be:
> if (dlyr.clip!=null && typeof dlyr.clip!='undefined') {
> For detailed info, follow this link:
> http://sourceforge.net/bugs/?func=detailbug&bug_id=128146&group_id=5757
> _______________________________________________
> Dynapi-Dev mailing list
> Dyn...@li...
> http://lists.sourceforge.net/mailman/listinfo/dynapi-dev
|
|
From: Michael P. <mp...@ph...> - 2001-01-09 14:18:34
|
<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
I've found a oneway method of allowing NS4 to increase the document size
if the child is outside the current dimensions. Unfortunately, once
the size is increased, it can't be decreased.
<p>Just add the following at the end of the Dynlayer.prototype.addChild:
<p><tt> if (this.isDynDocument && is.ns4) {</tt>
<br><tt> if (child.h+child.y>this.h)
document.height=this.h=child.h+child.y;</tt>
<br><tt> if (child.w+child.x>this.w)
document.width=this.w=child.w+child.x;</tt>
<br><tt> };</tt>
<br><tt></tt>
<p>--
<br>Michael Pemberton
<br>mp...@ph...
<br>ICQ: 12107010
<br> </html>
|
|
From: Eytan H. <ey...@tr...> - 2001-01-09 13:43:36
|
First of all I agree you should check for created. Also you might leave = an old reference to the old setHTML (call it oldsetHTML or some other = creative name) then if you ever need to use it (let's say that IE needs = it then just perofrm a check if not created call the original. 8an |
|
From: Michael P. <mp...@ph...> - 2001-01-09 12:40:38
|
<!doctype html public "-//w3c//dtd html 4.0 transitional//en"> <html> Can you give an example of your custom method. If you place a check for "this.created==true", it should stop it from executing during the initial creation stage. <p>no...@so... wrote: <blockquote TYPE=CITE>Bug #128155, was updated on 2001-Jan-09 03:49 <br>Here is a current snapshot of the bug. <p>Project: DynAPI 2 <br>Category: Core API <br>Status: Open <br>Resolution: None <br>Bug Group: None <br>Priority: 5 <br>Submitted by: camhart <br>Assigned to : nobody <br>Summary: Overriding setHTML method in sub class of DynLayer problem <p>Details: I've made a widget that is a sub class of DynLayer. This widget <br>overrides the behaviour of setHTML so that a call to setHTML changes the <br>HTML of a child layer of the widget, rather than the widget itself. This is <br>causing the Layer creation to break in IE5 and Mozilla (but not NS4) <br>presumably because when DynLayer calls setHTML during layer creation in <br>createElement, it been replaced with my method. NS4 is not using setHTML in <br>createElement. <p>This is only a bug if you think people should be allowed to override the <br>setHTML method when creating sub classes of dynlayer. <p>However this did used to work in earlier versions of the API. <p>The line that causes this to break is line 91 in dynlayer.js: <p>dlyr.setHTML(dlyr.getInnerHTML(),false); <p>For detailed info, follow this link: <br><a href="http://sourceforge.net/bugs/?func=detailbug&bug_id=128155&group_id=5757">http://sourceforge.net/bugs/?func=detailbug&bug_id=128155&group_id=5757</a> <p>_______________________________________________ <br>Dynapi-Dev mailing list <br>Dyn...@li... <br><a href="http://lists.sourceforge.net/mailman/listinfo/dynapi-dev">http://lists.sourceforge.net/mailman/listinfo/dynapi-dev</a></blockquote> <p>-- <br>Michael Pemberton <br>mp...@ph... <br>ICQ: 12107010 <br> </html> |
|
From: <ma...@ab...> - 2001-01-09 12:40:11
|
oh i'm sorry, it was me posting the bug, thought i already was logged in.. well /martin > -----Original Message----- > From: dyn...@li... > [mailto:dyn...@li...]On Behalf Of Pascal > Sent: den 9 januari 2001 10:27 > To: dyn...@li... > Subject: RE: [Dynapi-Dev] [Bug #128146] netscape addChild() bug > > > small request, but if you submit a bug could you atleast specify your name > somewhere, if your not loged on to sourceforge, please type it in the > description or something. > > thanks, > > Pascal Bestebroer (pb...@oi...) > Software ontwikkelaar > Oberon Informatiesystemen b.v. > http://www.oibv.com > > > -----Oorspronkelijk bericht----- > > Van: dyn...@li... > > [mailto:dyn...@li...]Namens > > no...@so... > > Verzonden: dinsdag 9 januari 2001 9:50 > > Aan: no...@so...; no...@so...; > > dyn...@so... > > Onderwerp: [Dynapi-Dev] [Bug #128146] netscape addChild() bug > > > > > > Bug #128146, was updated on 2001-Jan-09 00:50 > > Here is a current snapshot of the bug. > > > > Project: DynAPI 2 > > Category: Core API > > Status: Open > > Resolution: None > > Bug Group: None > > Priority: 5 > > Submitted by: nobody > > Assigned to : nobody > > Summary: netscape addChild() bug > > > > Details: this basic code: > > > > DynAPI.onLoad=function() { > > lay = new DynLayer(null,0,0,280,12,'red'); > > DynAPI.document.addChild(lay); > > }; > > > > doesn't work under netscape 4.76. > > ("clip is not defined. ") > > > > this is because the clip-check on line 75 (im using > > the latest files from cvs): > > > > if (dlyr.clip!=null) { > > > > isn't correct, dlyr.clip is undefined when i check the value. > > > > line 272 has the same problems, dlyr.clip is undefined > > but it only checks for null > > > > maybe a possible fix would be: > > > > if (dlyr.clip!=null && typeof dlyr.clip!='undefined') { > > > > > > > > > > For detailed info, follow this link: > > http://sourceforge.net/bugs/?func=detailbug&bug_id=128146&grou > p_id=5757 > > _______________________________________________ > Dynapi-Dev mailing list > Dyn...@li... > http://lists.sourceforge.net/mailman/listinfo/dynapi-dev > > > _______________________________________________ > Dynapi-Dev mailing list > Dyn...@li... > http://lists.sourceforge.net/mailman/listinfo/dynapi-dev |
|
From: <no...@so...> - 2001-01-09 11:49:42
|
Bug #128155, was updated on 2001-Jan-09 03:49 Here is a current snapshot of the bug. Project: DynAPI 2 Category: Core API Status: Open Resolution: None Bug Group: None Priority: 5 Submitted by: camhart Assigned to : nobody Summary: Overriding setHTML method in sub class of DynLayer problem Details: I've made a widget that is a sub class of DynLayer. This widget overrides the behaviour of setHTML so that a call to setHTML changes the HTML of a child layer of the widget, rather than the widget itself. This is causing the Layer creation to break in IE5 and Mozilla (but not NS4) presumably because when DynLayer calls setHTML during layer creation in createElement, it been replaced with my method. NS4 is not using setHTML in createElement. This is only a bug if you think people should be allowed to override the setHTML method when creating sub classes of dynlayer. However this did used to work in earlier versions of the API. The line that causes this to break is line 91 in dynlayer.js: dlyr.setHTML(dlyr.getInnerHTML(),false); For detailed info, follow this link: http://sourceforge.net/bugs/?func=detailbug&bug_id=128155&group_id=5757 |
|
From: Pascal <pb...@oi...> - 2001-01-09 11:05:04
|
> PS: How is it that a machine capable to run Q3A, Messiah, > Carmageddon 2000, Colin > McRae 2, etc... can't create and move several layers in a > reasonable time ? > Browser developers might as well go back to college and try > to learn something > this time. run them at once.. that would be impressive! :-) but your right, it's strange. Pascal Bestebroer (pb...@oi...) Software ontwikkelaar Oberon Informatiesystemen b.v. http://www.oibv.com |
|
From: Jordi 'I. M. <jmi...@or...> - 2001-01-09 10:59:34
|
I am currently working on an scalable menu interface that should provide a solution for most of these. I hope to release it as soon as it is fully tested an I can write some related docs and tutorials. Releasing a standalone bunch of cross-related js files would only add more confusion to the current situation. You can check my site for news and info about my progress. www.cantir.com/dynapi PS: How is it that a machine capable to run Q3A, Messiah, Carmageddon 2000, Colin McRae 2, etc... can't create and move several layers in a reasonable time ? Browser developers might as well go back to college and try to learn something this time. Robert Rainwater wrote: > I updated CVS and the snapshot with some changes to the label and list > widget that should get the list widget working. However, the event > problems in IE are still causing problems. Also, the onmouseover event > is never called for the ListItems. This could be related to the Label > problems. > > Also, the list widget is very slow and the events are lagging behind. > I think part of this is the constant rewriting of the html. Anyway to > improve this? > > -- > // Robert Rainwater > > On 1/8/2001, 12:13:02 AM EST, Robert wrote about "[Dynapi-Dev] Latest Issues": > > > I want to release a new version of the DynAPI soon, but there are a > > few things preventing it. > > > 1) In IE 5 the events inside of text in the label widget are broke > > again. > > > 2) The list widget does not work with the updated dynlayer creation > > system. > > > 3) Events in NS 6 are firing twice. There is a patch for this at > > sourceforge. Has this patch been tested by anyone? > > > If we can resolve these issues, then I think we can release a new > > version. > > _______________________________________________ > Dynapi-Dev mailing list > Dyn...@li... > http://lists.sourceforge.net/mailman/listinfo/dynapi-dev |
|
From: <no...@so...> - 2001-01-09 10:53:03
|
Bug #128150, was updated on 2001-Jan-09 02:53 Here is a current snapshot of the bug. Project: DynAPI 2 Category: None Status: Open Resolution: None Bug Group: None Priority: 5 Submitted by: digitalstrider Assigned to : nobody Summary: scrollpane Details: In IE5.5 on Win ME the scroll bars and knob don't show up. The buttons form up fine. It all works in ns on my OS. For detailed info, follow this link: http://sourceforge.net/bugs/?func=detailbug&bug_id=128150&group_id=5757 |
|
From: Pascal <pb...@oi...> - 2001-01-09 09:27:29
|
small request, but if you submit a bug could you atleast specify your name somewhere, if your not loged on to sourceforge, please type it in the description or something. thanks, Pascal Bestebroer (pb...@oi...) Software ontwikkelaar Oberon Informatiesystemen b.v. http://www.oibv.com > -----Oorspronkelijk bericht----- > Van: dyn...@li... > [mailto:dyn...@li...]Namens > no...@so... > Verzonden: dinsdag 9 januari 2001 9:50 > Aan: no...@so...; no...@so...; > dyn...@so... > Onderwerp: [Dynapi-Dev] [Bug #128146] netscape addChild() bug > > > Bug #128146, was updated on 2001-Jan-09 00:50 > Here is a current snapshot of the bug. > > Project: DynAPI 2 > Category: Core API > Status: Open > Resolution: None > Bug Group: None > Priority: 5 > Submitted by: nobody > Assigned to : nobody > Summary: netscape addChild() bug > > Details: this basic code: > > DynAPI.onLoad=function() { > lay = new DynLayer(null,0,0,280,12,'red'); > DynAPI.document.addChild(lay); > }; > > doesn't work under netscape 4.76. > ("clip is not defined. ") > > this is because the clip-check on line 75 (im using > the latest files from cvs): > > if (dlyr.clip!=null) { > > isn't correct, dlyr.clip is undefined when i check the value. > > line 272 has the same problems, dlyr.clip is undefined > but it only checks for null > > maybe a possible fix would be: > > if (dlyr.clip!=null && typeof dlyr.clip!='undefined') { > > > > > For detailed info, follow this link: > http://sourceforge.net/bugs/?func=detailbug&bug_id=128146&grou p_id=5757 _______________________________________________ Dynapi-Dev mailing list Dyn...@li... http://lists.sourceforge.net/mailman/listinfo/dynapi-dev |
|
From: <no...@so...> - 2001-01-09 08:50:04
|
Bug #128146, was updated on 2001-Jan-09 00:50
Here is a current snapshot of the bug.
Project: DynAPI 2
Category: Core API
Status: Open
Resolution: None
Bug Group: None
Priority: 5
Submitted by: nobody
Assigned to : nobody
Summary: netscape addChild() bug
Details: this basic code:
DynAPI.onLoad=function() {
lay = new DynLayer(null,0,0,280,12,'red');
DynAPI.document.addChild(lay);
};
doesn't work under netscape 4.76.
("clip is not defined. ")
this is because the clip-check on line 75 (im using
the latest files from cvs):
if (dlyr.clip!=null) {
isn't correct, dlyr.clip is undefined when i check the value.
line 272 has the same problems, dlyr.clip is undefined
but it only checks for null
maybe a possible fix would be:
if (dlyr.clip!=null && typeof dlyr.clip!='undefined') {
For detailed info, follow this link:
http://sourceforge.net/bugs/?func=detailbug&bug_id=128146&group_id=5757
|
|
From: Pascal <pb...@oi...> - 2001-01-09 07:37:56
|
the IE event problem is because of the child-layers not getting the bgImage fix that is done in the parent layers.. I'll try to fix this (haven't found a fix yet) Pascal Bestebroer (pb...@oi...) Software ontwikkelaar Oberon Informatiesystemen b.v. http://www.oibv.com > -----Oorspronkelijk bericht----- > Van: dyn...@li... > [mailto:dyn...@li...]Namens Robert Rainwater > Verzonden: maandag 8 januari 2001 23:54 > Aan: DynAPI Development List > Onderwerp: Re: [Dynapi-Dev] Latest Issues > > > > I updated CVS and the snapshot with some changes to the label and list > widget that should get the list widget working. However, the event > problems in IE are still causing problems. Also, the onmouseover event > is never called for the ListItems. This could be related to the Label > problems. > > Also, the list widget is very slow and the events are lagging behind. > I think part of this is the constant rewriting of the html. Anyway to > improve this? > > -- > // Robert Rainwater > > On 1/8/2001, 12:13:02 AM EST, Robert wrote about > "[Dynapi-Dev] Latest Issues": > > > > I want to release a new version of the DynAPI soon, but there are a > > few things preventing it. > > > 1) In IE 5 the events inside of text in the label widget are broke > > again. > > > 2) The list widget does not work with the updated dynlayer creation > > system. > > > 3) Events in NS 6 are firing twice. There is a patch for this at > > sourceforge. Has this patch been tested by anyone? > > > If we can resolve these issues, then I think we can release a new > > version. > > > > _______________________________________________ > Dynapi-Dev mailing list > Dyn...@li... > http://lists.sourceforge.net/mailman/listinfo/dynapi-dev > |
|
From: Robert R. <rra...@ya...> - 2001-01-09 05:26:26
|
I've never really looked at ButtonImage before ... but at first glance, the events are not firing and the image in the example never changes onmouseover because it never first that event. -- // Robert Rainwater On 1/8/2001, 11:15:59 PM EST, Raymond wrote about "[Dynapi-Dev] buttonimage.js and example and artwork for SVC": > I posted this to patches also, but here I added an example (for SVC) with 2 > pieces of art you can throw into the 'images/common' fileso it becomes a > standard part of debugging. Changing 'w' and 'h' to 'width' and 'height' > fixed half the problems. Mouseover still doesn't invoke an image change. > Laters |
|
From: Raymond S. <dst...@or...> - 2001-01-09 04:17:53
|
I posted this to patches also, but here I added an example (for SVC) with 2 pieces of art you can throw into the 'images/common' fileso it becomes a standard part of debugging. Changing 'w' and 'h' to 'width' and 'height' fixed half the problems. Mouseover still doesn't invoke an image change. Laters |
175 messages has been excluded from this view by a project administrator.
