Menu
▾
▴
axiomengine-developers — High volume technical list for Axiom Engine developers
You can subscribe to this list here.
| 2006 |
Jan
|
Feb
(1) |
Mar
(50) |
Apr
(6) |
May
(3) |
Jun
(13) |
Jul
(13) |
Aug
(31) |
Sep
(1) |
Oct
|
Nov
(1) |
Dec
(8) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2011 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
(1) |
Oct
|
Nov
|
Dec
|
| 2014 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
(6) |
| 2015 |
Jan
(3) |
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
|
From: Sergey Lobko-L. <ser...@gm...> - 2015-01-06 22:28:35
|
Hi guys, Haven't seen any discussions on this list for quite a while :-) On Jan 7, 2015 1:26 AM, "Andreas Bednarek" <an...@se...> wrote: > Once the LGPL clauses apply, I'm afraid that my AFAICR assumptions > regarding deployment to end-user mobile/Xbox devices are not much relevant > anyway ;) I'm sorry. > > Out of curious, what does Axiom currently prevent from going MIT; so far I > understand it, Axiom bes a derivative work of OGRE that needs to keep the > license of OGRE at its specific version? > > > ---------- Původní zpráva ---------- > Od: Andreas Bednarek <an...@se...> > Komu: A list for discussion of Axiom Engine development < > axi...@li...> > Datum: 4. 1. 2015 19:13:10 > Předmět: Re: [axiomengine-developers] can Axiom move to MIT like OGRE 1.7+? > > Earlier, a game created by an Axiom developer was accepted by Microsoft's > Dream, Build, Play competition and ran on Xbox 360. > http://axiom3d.net/forums/viewtopic.php?f=6&t=1134&p=6299 . I'm not aware > of how did he get over the LGPL limitations and whether they were a subject > for the DPB team, but that's what I've been talking about when > "remembering" a possible Axiom deployment to mobile, it was just not > mobile, however, on Xbox there is the same problem, where the LGPL > requirement to allow end-user relinking cannot be satisfied. May the DBP be > different from Xbox Live, there Axiom had run in an Xbox environment. (I > think I'm just advocating my previous AFAICR) > > ---------- Původní zpráva ---------- > Od: Andreas Bednarek <an...@se...> > Komu: A list for discussion of Axiom Engine development < > axi...@li...> > Datum: 24. 12. 2014 14:14:40 > Předmět: Re: [axiomengine-developers] can Axiom move to MIT like OGRE 1.7+? > > As far as I can remember, there was a way to deploy to mobile without > breaking the current license... > > > > > ------------------------------------------------------------------------------ > Dive into the World of Parallel Programming! The Go Parallel Website, > sponsored by Intel and developed in partnership with Slashdot Media, is > your > hub for all things parallel software development, from weekly thought > leadership blogs to news, videos, case studies, tutorials and more. Take a > look and join the conversation now. > http://goparallel.sourceforge.net_______________________________________________ > axiomengine-developers mailing list > axi...@li... > https://lists.sourceforge.net/lists/listinfo/axiomengine-developers > > > > ------------------------------------------------------------------------------ > Dive into the World of Parallel Programming! The Go Parallel Website, > sponsored by Intel and developed in partnership with Slashdot Media, is > your > hub for all things parallel software development, from weekly thought > leadership blogs to news, videos, case studies, tutorials and more. Take a > look and join the conversation now. http://goparallel.sourceforge.net > _______________________________________________ > axiomengine-developers mailing list > axi...@li... > https://lists.sourceforge.net/lists/listinfo/axiomengine-developers > > |
|
From: Andreas B. <an...@se...> - 2015-01-06 22:26:50
|
Once the LGPL clauses apply, I'm afraid that my AFAICR assumptions regarding deployment to end-user mobile/Xbox devices are not much relevant anyway ;) I'm sorry. Out of curious, what does Axiom currently prevent from going MIT; so far I understand it, Axiom bes a derivative work of OGRE that needs to keep the license of OGRE at its specific version? ---------- Původní zpráva ---------- Od: Andreas Bednarek <an...@se...> Komu: A list for discussion of Axiom Engine development <axiomengine- dev...@li...> Datum: 4. 1. 2015 19:13:10 Předmět: Re: [axiomengine-developers] can Axiom move to MIT like OGRE 1.7+? " Earlier, a game created by an Axiom developer was accepted by Microsoft's Dream, Build, Play competition and ran on Xbox 360. http://axiom3d.net/ forums/viewtopic.php?f=6&t=1134&p=6299 . I'm not aware of how did he get over the LGPL limitations and whether they were a subject for the DPB team, but that's what I've been talking about when "remembering" a possible Axiom deployment to mobile, it was just not mobile, however, on Xbox there is the same problem, where the LGPL requirement to allow end-user relinking cannot be satisfied. May the DBP be different from Xbox Live, there Axiom had run in an Xbox environment. (I think I'm just advocating my previous AFAICR) ---------- Původní zpráva ---------- Od: Andreas Bednarek <an...@se...> Komu: A list for discussion of Axiom Engine development <axiomengine- dev...@li...> Datum: 24. 12. 2014 14:14:40 Předmět: Re: [axiomengine-developers] can Axiom move to MIT like OGRE 1.7+? " As far as I can remember, there was a way to deploy to mobile without breaking the current license... " " " ---------------------------------------------------------------------------- -- Dive into the World of Parallel Programming! The Go Parallel Website, sponsored by Intel and developed in partnership with Slashdot Media, is your hub for all things parallel software development, from weekly thought leadership blogs to news, videos, case studies, tutorials and more. Take a look and join the conversation now. http://goparallel.sourceforge.net_______ ________________________________________ axiomengine-developers mailing list axi...@li... https://lists.sourceforge.net/lists/listinfo/axiomengine-developers" |
|
From: Andreas B. <an...@se...> - 2015-01-04 18:13:05
|
Earlier, a game created by an Axiom developer was accepted by Microsoft's Dream, Build, Play competition and ran on Xbox 360. http://axiom3d.net/ forums/viewtopic.php?f=6&t=1134&p=6299 . I'm not aware of how did he get over the LGPL limitations and whether they were a subject for the DPB team, but that's what I've been talking about when "remembering" a possible Axiom deployment to mobile, it was just not mobile, however, on Xbox there is the same problem, where the LGPL requirement to allow end-user relinking cannot be satisfied. May the DBP be different from Xbox Live, there Axiom had run in an Xbox environment. (I think I'm just advocating my previous AFAICR) ---------- Původní zpráva ---------- Od: Andreas Bednarek <an...@se...> Komu: A list for discussion of Axiom Engine development <axiomengine- dev...@li...> Datum: 24. 12. 2014 14:14:40 Předmět: Re: [axiomengine-developers] can Axiom move to MIT like OGRE 1.7+? " As far as I can remember, there was a way to deploy to mobile without breaking the current license... " " " |
|
From: Dan M. <da...@gm...> - 2014-12-28 06:27:56
|
As a former Axiom developer, I likewise support moving from LGPL to MIT licensing, similar to OGRE. On Wed, Dec 24, 2014 at 5:14 AM, Andreas Bednarek <an...@se...> wrote: > I add my voice in order to move Axiom from LGPL to MIT, if that is > possible. Although not having the very reasonable reasons by hand at the > moment, I think that David pretty hits that. As far as I can remember, > there was a way to deploy to mobile without breaking the current license, > if true, still that needs extra consideration that might not have been > necessary at all. > > ---------- Původní zpráva ---------- > Od: David Jeske <da...@gm...> > Komu: axi...@li... > Datum: 23. 12. 2014 2:59:34 > Předmět: [axiomengine-developers] can Axiom move to MIT like OGRE 1.7+? > > As of v1.7, ogre thankfully has moved to MIT licensing. The LGPL's > requirement that applications allow end-user relinking creates problems for > 3d games, because DLL relinking is not possible on mobile, and because a 3d > scene graph is a perfect place to install cheat hooks. > > Any chance Axiom could follow OGRE's footsetups and change the licensing > to MIT? > > http://www.ogre3d.org/licensing/licensing-faq > > ------------------------------------------------------------------------------ > Dive into the World of Parallel Programming! The Go Parallel Website, > sponsored by Intel and developed in partnership with Slashdot Media, is > your > hub for all things parallel software development, from weekly thought > leadership blogs to news, videos, case studies, tutorials and more. Take a > look and join the conversation now. > http://goparallel.sourceforge.net_______________________________________________ > axiomengine-developers mailing list > axi...@li... > https://lists.sourceforge.net/lists/listinfo/axiomengine-developers > > > > ------------------------------------------------------------------------------ > Dive into the World of Parallel Programming! The Go Parallel Website, > sponsored by Intel and developed in partnership with Slashdot Media, is > your > hub for all things parallel software development, from weekly thought > leadership blogs to news, videos, case studies, tutorials and more. Take a > look and join the conversation now. http://goparallel.sourceforge.net > _______________________________________________ > axiomengine-developers mailing list > axi...@li... > https://lists.sourceforge.net/lists/listinfo/axiomengine-developers > > -- Dan Moorehead CEO Realmware Corporation 93 S. Jackson St. #14025 Seattle, WA 98104 Work: (206) 792-9862 Fax: (630) 454-3591 Visual3D.NET Game Engine www.visual3d.net |
|
From: Andreas B. <an...@se...> - 2014-12-24 13:14:36
|
I add my voice in order to move Axiom from LGPL to MIT, if that is possible. Although not having the very reasonable reasons by hand at the moment, I think that David pretty hits that. As far as I can remember, there was a way to deploy to mobile without breaking the current license, if true, still that needs extra consideration that might not have been necessary at all. ---------- Původní zpráva ---------- Od: David Jeske <da...@gm...> Komu: axi...@li... Datum: 23. 12. 2014 2:59:34 Předmět: [axiomengine-developers] can Axiom move to MIT like OGRE 1.7+? " As of v1.7, ogre thankfully has moved to MIT licensing. The LGPL's requirement that applications allow end-user relinking creates problems for 3d games, because DLL relinking is not possible on mobile, and because a 3d scene graph is a perfect place to install cheat hooks. Any chance Axiom could follow OGRE's footsetups and change the licensing to MIT? http://www.ogre3d.org/licensing/licensing-faq (http://www.ogre3d.org/licensing/licensing-faq) ---------------------------------------------------------------------------- -- Dive into the World of Parallel Programming! The Go Parallel Website, sponsored by Intel and developed in partnership with Slashdot Media, is your hub for all things parallel software development, from weekly thought leadership blogs to news, videos, case studies, tutorials and more. Take a look and join the conversation now. http://goparallel.sourceforge.net_______ ________________________________________ axiomengine-developers mailing list axi...@li... https://lists.sourceforge.net/lists/listinfo/axiomengine-developers" |
|
From: Michael C. <cum...@li...> - 2014-12-23 05:31:25
|
------------------------------------------------------------------------------ Dive into the World of Parallel Programming! The Go Parallel Website, sponsored by Intel and developed in partnership with Slashdot Media, is your hub for all things parallel software development, from weekly thought leadership blogs to news, videos, case studies, tutorials and more. Take a look and join the conversation now. http://goparallel.sourceforge.net |
|
From: David J. <da...@gm...> - 2014-12-23 01:59:22
|
As of v1.7, ogre thankfully has moved to MIT licensing. The LGPL's requirement that applications allow end-user relinking creates problems for 3d games, because DLL relinking is not possible on mobile, and because a 3d scene graph is a perfect place to install cheat hooks. Any chance Axiom could follow OGRE's footsetups and change the licensing to MIT? http://www.ogre3d.org/licensing/licensing-faq |
|
From: Serge Lobko-L. a. a. <ser...@gm...> - 2006-12-12 13:21:02
|
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html><head><title></title>
<META http-equiv=3DContent-Type content=3D"text/html; charset=3Diso-8859-1">
<meta http-equiv=3D"Content-Style-Type" content=3D"text/css">
<style type=3D"text/css"><!--
body {
margin: 5px 5px 5px 5px;
background-color: #ffffff;
}
/* ---------- Text Styles ---------- */
hr { color: #000000}
body, table /* Normal text */
{
font-size: 9pt;
font-family: 'Courier New';
font-style: normal;
font-weight: normal;
color: #000000;
text-decoration: none;
}
span.rvts1 /* Heading */
{
font-size: 10pt;
font-family: 'Arial';
font-weight: bold;
color: #0000ff;
}
span.rvts2 /* Subheading */
{
font-size: 10pt;
font-family: 'Arial';
font-weight: bold;
color: #000080;
}
span.rvts3 /* Keywords */
{
font-size: 10pt;
font-family: 'Arial';
font-style: italic;
color: #800000;
}
a.rvts4, span.rvts4 /* Jump 1 */
{
font-size: 10pt;
font-family: 'Arial';
color: #008000;
text-decoration: underline;
}
a.rvts5, span.rvts5 /* Jump 2 */
{
font-size: 10pt;
font-family: 'Arial';
color: #008000;
text-decoration: underline;
}
span.rvts6
{
font-size: 11pt;
font-family: 'tahoma';
font-weight: bold;
color: #ffffff;
background-color: #0000ff;
}
span.rvts7
{
font-size: 11pt;
font-family: 'tahoma';
}
span.rvts8
{
font-size: 11pt;
font-family: 'tahoma';
font-weight: bold;
}
a.rvts9, span.rvts9
{
font-size: 11pt;
font-family: 'tahoma';
color: #0000ff;
text-decoration: underline;
}
span.rvts10
{
font-size: 8pt;
font-family: 'arial';
font-style: italic;
color: #c0c0c0;
}
a.rvts11, span.rvts11
{
font-size: 8pt;
font-family: 'arial';
color: #0000ff;
text-decoration: underline;
}
/* ---------- Para Styles ---------- */
p,ul,ol /* Paragraph Style */
{
text-align: left;
text-indent: 0px;
padding: 0px 0px 0px 0px;
margin: 0px 0px 0px 0px;
}
.rvps1 /* Centered */
{
text-align: center;
}
--></style>
</head>
<body>
<p>Hello Michael,</p>
<p><br></p>
<p>Axiom3d/Engine/Extenders/HierarchicalRegistry.cs</p>
<p><br></p>
<p>public IEnumerable<K> Subtree<K>(string prefix)</p>
<p> {</p>
<p> // check for consistent input =
parameters</p>
<p> if (!prefix.EndsWith("/"))</p>
<p> throw new Argume=
ntException("Namespace prefix must end with a '/' symbol");</p>
<p><br></p>
<p> // check if the passed namespa=
ce is actually extender-provided</p>
<p> if (extenders.ContainsKey(pref=
ix))</p>
<p> {</p>
<p> IEnumerable<K=
> extenderNamespace =3D extenders[prefix].Subtree<K>();</p>
<p> foreach (K obj i=
n extenderNamespace)</p>
<p> yi=
eld return obj;</p>
<p><br></p>
<p> yield break;</p>
<p> }</p>
<p><br></p>
<p> IEnumerator enu =3D ((IEnumera=
ble)objectList).GetEnumerator();</p>
<p><br></p>
<p> while (enu.MoveNext())</p>
<p> {</p>
<p> KeyValuePair<=
string, K> de =3D (KeyValuePair<string, K>)enu.Current;</p>
<p> string key =3D d=
e.Key;</p>
<p> K val =3D de.Val=
ue;</p>
<p><br></p>
<p> if (key.StartsWi=
th(prefix))</p>
<p> yi=
eld return val;</p>
<p> }</p>
<p> }</p>
<p><br></p>
<p>Tuesday, December 12, 2006, 2:45:48 PM, you wrote:</p>
<p><br></p>
<div><table border=3D0 cellpadding=3D1 cellspacing=3D2 style=3D"border-colo=
r: #000000; border-style: solid;">
<tr valign=3Dtop>
<td width=3D12 style=3D"background-color: #0000ff;">
<p><span class=3Drvts6>></span></p>
</td>
<td width=3D665 style=3D"background-color: #ffffff;">
<p><span class=3Drvts7>What does the class that implements the IEnumerator =
look like?</span></p>
<p><br></p>
<p><br></p>
<p><span class=3Drvts7>On 12/11/06, </span><span class=3Drvts8>Serge L=
obko-Lobanovsky aka arilou</span><span class=3Drvts7> <</span><a cl=
ass=3Drvts9 href=3D"mailto:ser...@gm...">serge.lobanovsky@gma=
il.com</a><span class=3Drvts7>> wrote: </span></p>
<p><span class=3Drvts7>because I don't know the indexer return object type =
;-)</span></p>
<p><br></p>
<p><span class=3Drvts7>Tuesday, December 12, 2006, 12:05:24 AM, you wrote:&=
nbsp;</span></p>
<p><br></p>
<p><span class=3Drvts7>> Well, why are you using 'object o;' in the firs=
t place? Wouldn't</span></p>
<p><span class=3Drvts7>> you want a strongly typed reference?</span></p>
<p><br></p>
<p><span class=3Drvts7>> On 12/10/06, Serge Lobko-Lobanovsky aka a=
rilou</span></p>
<p><span class=3Drvts7>> <</span><a class=3Drvts9 href=3D"mailto:serg=
e.l...@gm..."> ser...@gm...</a><span class=3Dr=
vts7>> wrote:</span></p>
<p><span class=3Drvts7>></span></p>
<p><span class=3Drvts7>></span></p>
<p><span class=3Drvts7>> Hi ppl,</span></p>
<p><span class=3Drvts7>></span></p>
<p><br></p>
<p><span class=3Drvts7>></span></p>
<p><span class=3Drvts7>> I've just stumbled across a funny bug/issue in =
.NET 2.0 iterators. Say</span></p>
<p><span class=3Drvts7>></span></p>
<p><span class=3Drvts7>> we have the following code:</span></p>
<p><span class=3Drvts7>> </span></p>
<p><br></p>
<p><span class=3Drvts7>></span></p>
<p><span class=3Drvts7>> object o ;</span></p>
<p><span class=3Drvts7>></span></p>
<p><span class=3Drvts7>> o =3D coll["key"];</span></p>
<p><span class=3Drvts7>></span></p>
<p><br></p>
<p><span class=3Drvts7>></span></p>
<p><span class=3Drvts7>> where coll.this[string] is IEnumerable, i.e. an=
iterator.</span></p>
<p><span class=3Drvts7>></span></p>
<p><br></p>
<p><span class=3Drvts7>> What the value of 'o' is going to be, you think=
? :) </span></p>
<p><span class=3Drvts7>> A reference to the method get_Item :) Spent alm=
ost an hour chasing</span></p>
<p><span class=3Drvts7>> it down (breakpoint set into the body of the in=
dexer never worked).</span></p>
<p><span class=3Drvts7>></span></p>
<p><br></p>
<p><span class=3Drvts7>> Uh-oh. </span></p>
<p><span class=3Drvts7>></span></p>
<p><br></p>
<p><span class=3Drvts7>></span></p>
<p><br></p>
<p><br></p>
<p><br></p>
<p><span class=3Drvts7>--</span></p>
<p><span class=3Drvts7>Best regards,</span></p>
<p><span class=3Drvts7>Serge &nbs=
p; mailto:</span><a class=
=3Drvts9 href=3D"mailto:ser...@gm...">serge.lobanovsky@gmail.=
com</a></p>
<p><br></p>
</td>
</tr>
</table>
</div>
<p><br></p>
<p><br></p>
<p><br></p>
<p><br></p>
<p><span class=3Drvts10>-- </span></p>
<p><span class=3Drvts10>Best regards,</span></p>
<p><span class=3Drvts10> Serge &nbs=
p; </span><a class=
=3Drvts11 href=3D"mailto:ser...@gm...">mailto:serge.lobanovsk=
y...@gm...</a></p>
</body></html>
|
|
From: Serge Lobko-L. a. a. <ser...@gm...> - 2006-12-11 22:21:43
|
because I don't know the indexer return object type ;-) Tuesday, December 12, 2006, 12:05:24 AM, you wrote: > Well, why are you using 'object o;'=A0 in the first place? Wouldn't > you want a strongly typed reference? > On 12/10/06, Serge Lobko-Lobanovsky aka arilou > <ser...@gm...> wrote:=20 > =20 > =20 > Hi ppl, > =20 > =20 > I've just stumbled across a funny bug/issue in .NET 2.0 iterators. Say > =20 > we have the following code: > =20 > =20 > object o ; > =20 > o =3D coll["key"]; > =20 > =20 > where coll.this[string] is IEnumerable, i.e. an iterator. > =20 > What the value of 'o' is going to be, you think? :)=20 > A reference to the method get_Item :) Spent almost an hour chasing > it down (breakpoint set into the body of the indexer never worked). > =20 > Uh-oh. =A0 > =20 > =20 --=20 Best regards, Serge mailto:ser...@gm... |
|
From: Michael C. <cum...@gm...> - 2006-12-11 22:05:37
|
Well, why are you using 'object o;' in the first place? Wouldn't you want a strongly typed reference? On 12/10/06, Serge Lobko-Lobanovsky aka arilou <ser...@gm...> wrote: > > Hi ppl, > > > I've just stumbled across a funny bug/issue in .NET 2.0 iterators. Say > > we have the following code: > > > object o ; > > o = coll["key"]; > > > where coll.this[string] is IEnumerable, i.e. an iterator. > > > What the value of 'o' is going to be, you think? :) > A reference to the method get_Item :) Spent almost an hour chasing it down > (breakpoint set into the body of the indexer never worked). > > > Uh-oh. > > > -- > > Best regards, > > Serge mailto:ser...@gm...<ser...@gm...> > > ------------------------------------------------------------------------- > Take Surveys. Earn Cash. Influence the Future of IT > Join SourceForge.net's Techsay panel and you'll get the chance to share > your > opinions on IT & business topics through brief surveys - and earn cash > http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV > > _______________________________________________ > axiomengine-developers mailing list > axi...@li... > https://lists.sourceforge.net/lists/listinfo/axiomengine-developers > > > |
|
From: Serge Lobko-L. a. a. <ser...@gm...> - 2006-12-10 13:39:45
|
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html><head><title></title>
<META http-equiv=3DContent-Type content=3D"text/html; charset=3Dwindows-125=
1">
<meta http-equiv=3D"Content-Style-Type" content=3D"text/css">
<style type=3D"text/css"><!--
body {
margin: 5px 5px 5px 5px;
background-color: #ffffff;
}
/* ---------- Text Styles ---------- */
hr { color: #000000}
body, table /* Normal text */
{
font-size: 9pt;
font-family: 'Courier New';
font-style: normal;
font-weight: normal;
color: #000000;
text-decoration: none;
}
span.rvts1 /* Heading */
{
font-size: 10pt;
font-family: 'Arial';
font-weight: bold;
color: #0000ff;
}
span.rvts2 /* Subheading */
{
font-size: 10pt;
font-family: 'Arial';
font-weight: bold;
color: #000080;
}
span.rvts3 /* Keywords */
{
font-size: 10pt;
font-family: 'Arial';
font-style: italic;
color: #800000;
}
a.rvts4, span.rvts4 /* Jump 1 */
{
font-size: 10pt;
font-family: 'Arial';
color: #008000;
text-decoration: underline;
}
a.rvts5, span.rvts5 /* Jump 2 */
{
font-size: 10pt;
font-family: 'Arial';
color: #008000;
text-decoration: underline;
}
span.rvts6
{
font-size: 8pt;
font-family: 'arial';
font-style: italic;
color: #c0c0c0;
}
a.rvts7, span.rvts7
{
font-size: 8pt;
font-family: 'arial';
color: #0000ff;
text-decoration: underline;
}
/* ---------- Para Styles ---------- */
p,ul,ol /* Paragraph Style */
{
text-align: left;
text-indent: 0px;
padding: 0px 0px 0px 0px;
margin: 0px 0px 0px 0px;
}
.rvps1 /* Centered */
{
text-align: center;
}
--></style>
</head>
<body>
<p>Hi ppl,</p>
<p><br></p>
<p>I've just stumbled across a funny bug/issue in .NET 2.0 iterators. Say</=
p>
<p>we have the following code:</p>
<p><br></p>
<p>object o ;</p>
<p>o =3D coll["key"];</p>
<p><br></p>
<p>where coll.this[string] is IEnumerable, i.e. an iterator.</p>
<p>
<br>What the value of 'o' is going to be, you think? :)
<br>A reference to the method get_Item :) Spent almost an hour chasing it d=
own (breakpoint set into the body of the indexer never worked).</p>
<p>
<br>Uh-oh. </p>
<p><br></p>
<p><span class=3Drvts6>-- </span></p>
<p><span class=3Drvts6>Best regards,</span></p>
<p><span class=3Drvts6> Serge  =
; </span><a class=3Drvts7 h=
ref=3D"mailto:ser...@gm...">mailto:ser...@gm...=
</a></p>
</body></html>
|
|
From: Serge Lobko-L. a. a. <ser...@gm...> - 2006-12-08 23:18:45
|
Hello Kyle, Hobitton RC2? Saturday, December 9, 2006, 1:10:06 AM, you wrote: > Yeah, I wonder if he just forgot about the mailing list. I downloaded > and built the new release just earlier this afternoon. > Serge Lobko-Lobanovsky aka arilou wrote: >> Hello Kyle, >> >> Funny. Is this mailing list no longer maintained? >> >> Borrillis, did you... ? -- Best regards, Serge mailto:ser...@gm... |
|
From: Serge Lobko-L. a. a. <ser...@gm...> - 2006-12-08 22:58:55
|
Hello Kyle, Funny. Is this mailing list no longer maintained? Borrillis, did you... ? Saturday, December 9, 2006, 12:55:24 AM, you wrote: > Strange... Seems like Borrillis just released a new RC a couple days ago. >> Hello Kyle, >> >> Glad to see you aboard. However, it seems that the development has >> stalled... -- Best regards, Serge mailto:ser...@gm... |
|
From: Serge Lobko-L. a. a. <ser...@gm...> - 2006-12-08 22:08:35
|
Hello Kyle, Glad to see you aboard. However, it seems that the development has stalled... Friday, December 8, 2006, 10:48:37 PM, you wrote: > Hi, > My name is Kyle. I've been reading up on, building, and playing with > Axiom for a couple weeks now. I've made a few forum posts and reworked > one of the tutorials, so I feel obligated to officially introduce myself. > Currently, I'm working at a large defense contractor doing maintenance > work on some internal test software, but I'm going back to school this > next fall to finish my degree. > I'm rather new to C# and .NET--having a much stronger background in > C--so I might not be of much use early on (I've been writing C#/.NET > full time for the past three months at work however, so I'm learning > quicky), but I'd love to help the project out in any way I can. > ------------------------------------------------------------------------- > Take Surveys. Earn Cash. Influence the Future of IT > Join SourceForge.net's Techsay panel and you'll get the chance to share your > opinions on IT & business topics through brief surveys - and earn cash > http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV > _______________________________________________ > axiomengine-developers mailing list > axi...@li... > https://lists.sourceforge.net/lists/listinfo/axiomengine-developers -- Best regards, Serge mailto:ser...@gm... |
|
From: Kyle K. <ky...@vi...> - 2006-12-08 20:48:48
|
Hi, My name is Kyle. I've been reading up on, building, and playing with Axiom for a couple weeks now. I've made a few forum posts and reworked one of the tutorials, so I feel obligated to officially introduce myself. Currently, I'm working at a large defense contractor doing maintenance work on some internal test software, but I'm going back to school this next fall to finish my degree. I'm rather new to C# and .NET--having a much stronger background in C--so I might not be of much use early on (I've been writing C#/.NET full time for the past three months at work however, so I'm learning quicky), but I'd love to help the project out in any way I can. |
|
From: Serge Lobko-L. a. a. <ser...@gm...> - 2006-11-23 10:43:02
|
Hello friends, Since there have been no activity here for quite a while can we say the development is stalled? -- Best regards, Serge mailto:ser...@gm... |
|
From: Michael C. <cum...@gm...> - 2006-09-18 09:12:29
|
The Axiom team is proud to announce that we have finally been able to provide the community with a download. Being the first in a series of releases, this is your chance to make sure this will be a very stable and useable release. Please take the time to download and work with it in your projects. Of course, report any issues on the forums or the SourceForge bugtracker you may have so they can be corrected before the final release. We are planning to have an RC2 in two weeks time and a final release 2 weeks after that. There are four different packages available in this release; full, binary, source and media. The full package includes all source and binaries. The source package includes all source files, dependencies and build system. The binary package has just the binary assemblies and dependencies in a ready to run form. There is only one binary package for Windows/DX initially, a cross platform SDL binary package will be added soon. The media package contains the media folder and is a requirement for both the binary and source packages. The media folder is much small than in the past as we have trimmed a lot of the fluff from it. The files are available from the SourceForge Release System. This release has been codenamed 'hobbiton'. Now, those of you that have been paying attention are going to ask 'I thought the next release was codenamed 'bree'? To quote one of my favorite non-LOTR characters, 'Let me axplain...noh, there is too much, let me summup'. The bree branch was started with some very large goals, updating to Ogre 1.06, .Net 2.0Collections, new Plugin System, and upgrade to MDX 2.0. After a year of development, we were nowhere near being able to release, let alone have a compilable system. Even though we were progressing on the goal of updating to OGRE compatibility, it was hard to gauge where we were since the source couldn't compile. This produced several side effects, one it was demoralizing to everyone who was contributing to the project as there was no visible results, second the community tends to think the project is dead because forward momentum is so slow. I decided that with the beta of XNA Game Studio and the XNA Framework that the community at large needed to know that Axiom was not a dead project and that it is a very viable solution for XNA development (yes there is an XNA renderer under way, I'll get to that ). It is still a good solution for OpenGL and DirectX and will continue to be. However there is an obvious opportunity to gain a lot of forward momentum by piggy-backing on the newness of the XNA Framework. So about a month ago, we backtracked and started development on hobbiton. We came to the realization that it was going to be very difficult to get bree to a relatively stable state. However we couldn't ignore some of the advancements that had been made in bree. So we decided that the best place to start was the last known revision to work reliably. This just so happened to be the last release that leed had put out. After several weeks of search out patches and fixing up bugs we now have a fairly stable code base to release. There are several things to note about this release, the biggest is that is is essentially the same as 0.7.0.0 with some bugfixes and minor improvements. That means that if you have been working with the SVN Head, your code will break against hobbiton. We do apologize for this, but it will be a needed step in order for us to make major leaps forward in a very short time. By taking this step we have been able to get ALL the demos working on .Net and Mono! So all of the namespace changes, most of the OGRE updates, the plugin architecture, and other changes were not moved back into hobbiton to ensure a nice fast stable release. So what happens now? hobbiton will replace SVN HEAD after the final release and bug fixes and minor updates will occur there. As for bree, it will be rebased on the hobbiton source and all OGRE 1.06 updates reapplied. The plugin system, updates to the math library and .Net 2.0 collections will be researched for viability and performance. The OGRE updating will move on a slightly different track however. Feature sets will be identified and updated, this should allow us to make progress and keep the system building and the demos running. So what's all this noise about XNA? I'll attempt to answer your questions about using Axiom and XNA here. First lets discuss the XNA framework. There are two platforms for XNA, Windows and Xbox. The Xbox platform is not yet available, which is good, because there are some heavy requirements in order to run on it. For the Windows platform we are already working on a PlatformManager and renderer for the XNA Framework. There is no completion date for this yet, but it is a priority. There are some technical hurdles to overcome, the biggest being XNA does not have a fixed-function pipeline, so Axiom's fixed-function pipeline needs to be converted to the equivalent shader model. Doable, but tedious. For the Xbox, we have the same issues as for Windows, plus there is no support for P/Invoke, which basically means we can't use DevIL. So in short, Windows XNA will be available in the short term, with minor changes to the base code. However, there will be a longer and more intensive development effort to support the Xbox platform. So enjoy this release and let us know if there are any issues, we'll get them corrected as soon as possible! The Axiom Team http://axiomengine.sf.net |
|
From: Terry L. T. <c0d...@gm...> - 2006-08-29 14:59:37
|
Hi folks, Sorry for jumping in late - the bulk of this thread happened when I was away and it seemed too awkward to try and respond inline. So I'll just read everything and attempt a general response. > borrillis said: > Is there a real value to AxiomCollection at this point? > Does it solve any issues that aren't already solved by > the types in System.Collections.Generic? > Is deriving from AxiomCollection better than > System.Collections.Generic.Dictionary<T>? This is a key question, IMHO. I've yet to see a good rationale for the Add() method that justifies the design consequences - in this case the ease with which bugs can be introduced that go unchecked until run-time, making the codebase less robust as a result. What's the use case? If the answer is "to accommodate objects that won't be tracked by key", I would have to dispute the validity of the Add() method as a design choice. My OO design sensibilities include a strong preference for objects with clear and focused responsibilities. An object with too many responsibilities is unfocused and likely needs to be refactored into separate objects. This is fundamentally what bothers me about the Add() method - it has the effect of blurring the responsibilities of the collection. Is it a dictionary or a list? If a dictionary has to be modified to easily accommodate objects that aren't going to be tracked using unique keys (name or otherwise), the situation probably really calls for another collection type, like an array or list. If there is a strong case for this kind of dual responsibility, I would definitely not bake it in at the root of the object heirarchy, but rather add it further up the tree where the effects of introducing ambiguity are isolated from the rest of the system. I'm still digesting the rest of the discussion about key generation. Makes sense so far. Cheers, T. |
|
From: Michael C. <cum...@gm...> - 2006-08-29 13:04:45
|
Already on it Serge, I'll know more when the beta is released tommorow. We will be able to support XNA on windows fairly easily, however XNA onthe XBox means removing any and all external dependencies that are not Managed. I am pretty sure that the DirectX9 Renderer using MDX 2.0 will convert fairly easily, We will also need to create a new Platform Manager for XNA as well. Windows will have support for keyboard\mouse\controller whereas the XBox will only support keyboard\controller. Borrillis On 8/29/06, Serge Lobko-Lobanovsky aka arilou <ser...@gm...> wrote: > > Hi guys, > > It seems that it may be possible to introduce support for XNA in > Axiom? > > Anyone looked into the subject? > > -- > Best regards, > Serge mailto:ser...@gm... > > > ------------------------------------------------------------------------- > Using Tomcat but need to do more? Need to support web services, security? > Get stuff done quickly with pre-integrated technology to make your job > easier > Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo > http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 > _______________________________________________ > axiomengine-developers mailing list > axi...@li... > https://lists.sourceforge.net/lists/listinfo/axiomengine-developers > |
|
From: Serge Lobko-L. a. a. <ser...@gm...> - 2006-08-29 10:58:11
|
Hi guys, It seems that it may be possible to introduce support for XNA in Axiom? Anyone looked into the subject? -- Best regards, Serge mailto:ser...@gm... |
|
From: Andreas B. <an...@se...> - 2006-08-28 12:39:41
|
Hello guys,
Serge: Guig.NewGuid() works fine.
However I did some test and believe now that GUIDs are not the best solution
either because, they seem to be damn slow in our scenario and actually that
applies to the string.GetHashCode() thing.
As to the GUIDS: I modified the AxiomCollection so that it doesn't use the
while (base.ContainsKey(key = GetNextKey()))
;
base.Add(key,item);
approach and used direct adding without previous ContainsKey() checking:
base.Add(GetNextKey(item), item);
the string key was constructed in the GetNextKey() method as follows:
return (K)Convert.ChangeType(Guid.NewGuid() + typeof(T).Name, typeof(K));
i.e. simply get a Guid and append the current type name of the collection
items.
I expected a speed up as the while(base.ContainsKey()) check was ommited but
was quite shocked, see below.
I also tried to construct the key prepending a hash value of the key to it
so let's mention it as well:
{
string s = typeof( T ). Name +
checked(nextUniqueKeyCounter++).ToString();
key = ((K) Convert. ChangeType( s. GetHashCode() + s,
typeof( K ) ) );
}
Results:
a) The original while-contains approach measure: 100000 items added in
12.0411 seconds
b) The direct adding using a GUID measure: 100000 items added in 59.1695
seconds
c) A while-contains approach additionaly prepending the result of
GetHashCode() applied with the pre-final string key: 100000 items added in
60.6140 seconds
(I think the necessary heap allocations do that...)
Come on, just use the simplest
key = (K)Convert.ChangeType(checked(++nextUniqueKeyCounter) +
typeof(T).Name, typeof(K));
approach. I.e. keys looking like this:
"1234Entity"
What do you say?
In the final still correct thing I got 100000 items added in 11.1862
seconds.
Without a while-contains check which I could omit for my simple scenario it
was about 9.8 secs. So the ContainsKey() method seems to be really fast.
(And, probably, the less chars of the key it has to compare to existing
items before equality is determined, the faster. So it would be probably
better to have a "4321Entity" key for a 1234 counter :-) however.)
Andris
----- Original Message -----
From: "Serge Lobko-Lobanovsky aka arilou" <ser...@gm...>
To: "Andreas Bednarek" <an...@se...>
Cc: "A list for discussion of Axiom Engine development"
<axi...@li...>
Sent: Monday, August 28, 2006 8:56 AM
Subject: Re: [axiomengine-developers] AxiomCollection.cs design issue
> Hello Andreas,
>
> Actually, it should've been
>
> Guid g = Guid.NewGuid(); // factory method
> Console.WriteLine(g);
> Console.ReadLine();
>
> Sorry, couldn't test it before posting (no VS.NET at home yet after a
> major upgrade :( )
>
> Sunday, August 27, 2006, 10:24:22 PM, you wrote:
>
>> I'm afraid that
>
>> (new Guid()).ToString();
>
>> produces a
>> "00000000-0000-0000-0000-000000000000"
>
>> :(
>> (did a test right now )
>
|
|
From: Andreas B. <an...@se...> - 2006-08-27 19:01:45
|
Hi,
If we cannot omit the check whether the key exists before adding a new =
one, I'd prefer some short prefix constructed from the uint counter or =
as Borrilis suggested, using GetHashCode() of the final key value to =
prepend to it, i.e. not using a GUID.
Then, in order not to replicate a commonly used naming scheme which was =
the request, what about just swap the ("TypeName" + counter) order and =
use (counter + "TypeName") ? :-)
Simpliest, fastest.
Yet to the AxiomCollection.Remove(T item) method, it could by improved =
for speed.=20
Instead of this:
public void Remove( T item )
{
if ( this.ContainsValue( item ) )
{
int index =3D this.IndexOfValue( item );
this.RemoveAt( index );
}
}
we should use this (the value needs not to be searched two times + some =
checks):
public void Remove( T item )
{
if (item =3D=3D null)
throw new ArgumentNullException("Can't remove a null =
item from a collection.");
int index =3D this.IndexOfValue( item );
if ( index >=3D 0 )
{
this. RemoveAt( index );
}
else
throw new AxiomException( "The specified item was not =
found in the collection." );
}
If you want I can make a suggestion of how the new AxiomCollection might =
look like as whole, the question still is how to create the unique =
string key...
|
|
From: Andreas B. <an...@se...> - 2006-08-25 00:13:58
|
As to the GUIDs, by using that with a string key we could omit using the = static uint counter and possibly also the check whether the key exist in = the collection already. Is that right? That could really speed up = adding. Just whether the 32-36 chars for only a GUID string won't slow = down again...(don't know how SortedList<> is implemented.. but this will = be a minor issue, won't it? :-)) For the non-string type keys you'll have to do the whether-key-exist = check though. As for generating a hash, key.GetHashCode() should suffice. Agree. (Maybe the hex/dec representation of the uint static counter could be = obtained faster with a similar effect... ;-) Why the restriction on having the has start with a letter? We are not = creating a c# variable, also special characters should be ok as well. = This is a value that is ( read should be ) only used internally in the = engine ( since it's autogenerated. ) Just that string Id's tend to start with a letter. That needs not to be = so important, however I don't feel like special characters should appear = in the name really. Although currently not intended, suddenly it can = appear in a xml file or a stream somewhere... (you might wish to list = the keys in SceneManager.SceneNodes to console for some debugging...) Else I agree this seems to remain for internal use only, the fact is, = that the autogenerated collection key won't match with an autoassigned = name for a Node for example. However a number or GUID string have no = special chars already. Anyone needing the key value for persistance would want to have their = own naming scheme and not rely on the engine to create a key for them. ( = For things like the DotScene format, or any other custom scene format ). Possibly. As to the inconsistence of Node.Name and collection key for = example, the Axiom user needs to find his own solution either. |
|
From: Michael C. <cum...@gm...> - 2006-08-24 10:56:26
|
Why the restriction on having the has start with a letter? We are not creating a c# variable, also special characters should be ok as well. This is a value that is ( read should be ) only used internally in the engine ( since it's autogenerated. ) Anyone needing the key value for persistance would want to have their own naming scheme and not rely on the engine to create a key for them. ( For things like the DotScene format, or any other custom scene format ). As for generating a hash, key.GetHashCode() should suffice. On 8/23/06, Andreas Bednarek <an...@se...> wrote: > > > > > I guess my main issues at this point are: > > > 1) Some one can inadvertantly use the same naming scheme as is proposed, > > > in fact I can see many an exporter using this exact naming scheme. > > So you mean that the while loop could currently make many iterations until > a key is found? Yeah, right. > > > How about we use a hash of the final key value and prepend that to > the type name? > Something like that sounds like a good idea, just: > - How will you be creating the hash value? When using some .NET Framework > class, if Mono supports it... > - Will you be able to create a string representation that doesn't contain > any special chars like ' ? / =' etc. and possibly not start with a digit? > - May the collection change rapidly at runtime, so it would eventually > break performance? (Except cases the hash adds to performance) > > But I'm no expert on hashing. > > Maybe just make some fast thing like converting the uint counter to a four > or maybe six chars Hex string (less signif. bytes shown) ? Ensure starting > with letter by prepending a hardcoded '_' or whatever. I.e. not using the > whole string key for hashing... > > > 2) rename NextKey to GetNextKey, NextKey sounds like a property, > GetNextKey is a definite action. > > True. > > > ------------------------------------------------------------------------- > Using Tomcat but need to do more? Need to support web services, security? > Get stuff done quickly with pre-integrated technology to make your job > easier > Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo > http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 > > _______________________________________________ > axiomengine-developers mailing list > axi...@li... > https://lists.sourceforge.net/lists/listinfo/axiomengine-developers > > > |
|
From: Serge Lobko-L. a. a. <ser...@gm...> - 2006-08-24 07:51:01
|
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html><head><title></title>
<META http-equiv=Content-Type content="text/html; charset=iso-8859-1">
<meta http-equiv="Content-Style-Type" content="text/css">
<style type="text/css"><!--
body {
margin: 5px 5px 5px 5px;
background-color: #ffffff;
}
/* ========== Text Styles ========== */
hr { color: #000000}
body, table /* Normal text */
{
font-size: 9pt;
font-family: 'Courier New';
font-style: normal;
font-weight: normal;
color: #000000;
text-decoration: none;
}
span.rvts1 /* Heading */
{
font-size: 10pt;
font-family: 'Arial';
font-weight: bold;
color: #0000ff;
}
span.rvts2 /* Subheading */
{
font-size: 10pt;
font-family: 'Arial';
font-weight: bold;
color: #000080;
}
span.rvts3 /* Keywords */
{
font-size: 10pt;
font-family: 'Arial';
font-style: italic;
color: #800000;
}
a.rvts4, span.rvts4 /* Jump 1 */
{
font-size: 10pt;
font-family: 'Arial';
color: #008000;
text-decoration: underline;
}
a.rvts5, span.rvts5 /* Jump 2 */
{
font-size: 10pt;
font-family: 'Arial';
color: #008000;
text-decoration: underline;
}
span.rvts6
{
font-size: 11pt;
font-family: 'tahoma';
font-weight: bold;
color: #ffffff;
}
span.rvts7
{
font-size: 11pt;
font-family: 'tahoma';
}
span.rvts8
{
font-size: 11pt;
font-family: 'tahoma';
font-weight: bold;
}
a.rvts9, span.rvts9
{
font-size: 11pt;
font-family: 'tahoma';
color: #0000ff;
text-decoration: underline;
}
span.rvts10
{
font-family: 'arial';
}
span.rvts11
{
}
span.rvts12
{
font-size: 8pt;
font-family: 'arial';
font-style: italic;
color: #c0c0c0;
}
a.rvts13, span.rvts13
{
font-size: 8pt;
font-family: 'arial';
color: #0000ff;
text-decoration: underline;
}
/* ========== Para Styles ========== */
p,ul,ol /* Paragraph Style */
{
text-align: left;
text-indent: 0px;
padding: 0px 0px 0px 0px;
margin: 0px 0px 0px 0px;
}
.rvps1 /* Centered */
{
text-align: center;
}
--></style>
</head>
<body>
<p>Guys,</p>
<p><br></p>
<p>Why don't we use GUIDs?</p>
<p><br></p>
<p>Wednesday, August 23, 2006, 10:03:36 PM, you wrote:</p>
<p><br></p>
<div><table border=0 cellpadding=1 cellspacing=2 style="border-color: #000000; border-style: solid;">
<tr valign=top>
<td width=14 style="background-color: #0000ff;">
<p><span class=rvts6>></span></p>
</td>
<td width=681 style="background-color: #ffffff;">
<p><span class=rvts7>I guess my main issues at this point are:</span></p>
<p><br></p>
<p><span class=rvts7>1) Some one can inadvertantly use the same naming scheme as is proposed, in fact I can see many an exporter using this exact naming scheme.</span></p>
<p><span class=rvts7> How about we use a hash of the final key value and prepend that to the type name?</span></p>
<p><br></p>
<p><span class=rvts7>2) rename NextKey to GetNextKey, NextKey sounds like a property, GetNextKey is a definite action.</span></p>
<p><br></p>
<p><br></p>
<p><br></p>
<p><span class=rvts7>On 8/23/06, </span><span class=rvts8>Andreas Bednarek</span><span class=rvts7> <</span><a class=rvts9 href="mailto:an...@se...">an...@se...</a><span class=rvts7>> wrote:</span></p>
<p><span class=rvts7> </span></p>
<p><br></p>
<p><span class=rvts7>And where does this get called from? So far I don't see much difference from the current implementation.</span></p>
<p><span class=rvts7> </span></p>
<p><span class=rvts10>The GenerateKey() should have been called before Add(key, item) when there is no actual value for the key (simply outside AxiomCollection). Not very cool either so thought of a better idea, see below. </span></p>
<p><span class=rvts10> </span></p>
<p><span class=rvts10>Also the main difference is, that the Add(T item) is removed. I remember finding some bugs in SceneManager where CreateCamera, CreateSceneNode and maybe some more methods were using the simple Add(T item) thus not using the appropriate name the objects were created with. Just like the Foo example in Terry's post.</span></p>
<p><span class=rvts7> </span></p>
<p><span class=rvts10>(Another diff is that the key is now unique.)</span></p>
<p><span class=rvts7> </span></p>
<p><span class=rvts10>How about this?</span></p>
<p><span class=rvts7> </span></p>
<p><span class=rvts7> </span></p>
<p><span class=rvts10>- RENAME the Add(T item) method as maybe it is a bit confusing, name it AddAutoKey(T item) for example. It simply does more than just adding.</span></p>
<p><span class=rvts10>- Let it generate the autokey (fix current uniquity issue) and let it return the key to the caller if maybe needed</span></p>
<p><span class=rvts10>- (nextUniqueKeyCounter could be 'uint' not 'int' and checked for overflow - just little thing)</span></p>
<p><span class=rvts10> </span></p>
<p><span class=rvts10>This might be a simple although indirect solution to the issues that initiated this thread. The 'bandaid' Terry mentioned does not go away unfortunately, but whether that hurts really... also by revisiting all the existing code that shall use the new method name, it can be easily checked whether the previous Add(item) was used correctly.</span></p>
<p><span class=rvts10>Just maybe.</span></p>
<p><span class=rvts10> </span></p>
<p><br></p>
<p><span class=rvts11> public class AxiomCollection<K, T> : SortedList<K, T> where K : IConvertible</span></p>
<p><span class=rvts11> {</span></p>
<p><span class=rvts10> </span></p>
<p><span class=rvts11> ...</span></p>
<p><span class=rvts10> </span></p>
<p><span class=rvts11> static protected uint nextUniqueKeyCounter;</span></p>
<p><span class=rvts11> </span></p>
<p><span class=rvts11> ...</span></p>
<p><span class=rvts10> </span></p>
<p><span class=rvts11> /// <summary></span></p>
<p><span class=rvts11> /// Accepts an unnamed object and names it manually. </span></p>
<p><span class=rvts11> /// </summary></span></p>
<p><span class=rvts11> /// <param name="item">Item to be added to the collection.</param></span></p>
<p><span class=rvts11> /// <returns>The autogenerated key that was used when adding the item to the collection.</returns></span></p>
<p><span class=rvts11> public virtual K AddAutoKey( T item )</span></p>
<p><span class=rvts11> {</span></p>
<p><span class=rvts11> K key; // = default( K ); ??</span></p>
<p><span class=rvts10> </span></p>
<p><span class=rvts11> while ( base. ContainsKey( key = NextKey( item ) ) )</span></p>
<p><span class=rvts11> ;</span></p>
<p><span class=rvts10> </span></p>
<p><span class=rvts11> base. Add( key, item );</span></p>
<p><span class=rvts10> </span></p>
<p><span class=rvts11> return key;</span></p>
<p><span class=rvts11> }</span></p>
<p><span class=rvts10> </span></p>
<p><br></p>
<p><span class=rvts11> private K NextKey( T item )</span></p>
<p><span class=rvts11> {</span></p>
<p><span class=rvts11> K key;</span></p>
<p><span class=rvts10> </span></p>
<p><span class=rvts11> if ( typeof( K ) == typeof( string ) )</span></p>
<p><span class=rvts11> {</span></p>
<p><span class=rvts11> key = ( K ) Convert. ChangeType( typeof( T ). Name + nextUniqueKeyCounter, typeof( K ) );</span></p>
<p><span class=rvts11> }</span></p>
<p><span class=rvts11> else</span></p>
<p><span class=rvts11> {</span></p>
<p><span class=rvts11> key = ( K ) Convert. ChangeType( nextUniqueKeyCounter, typeof( K ) );</span></p>
<p><span class=rvts11> }</span></p>
<p><span class=rvts10> </span></p>
<p><span class=rvts11> checked( nextUniqueKeyCounter++ );</span></p>
<p><span class=rvts10> </span></p>
<p><span class=rvts11> return key;</span></p>
<p><span class=rvts11> }</span></p>
<p><span class=rvts11> }</span></p>
<p><span class=rvts10> </span></p>
<p><span class=rvts10> </span></p>
<p><span class=rvts7> </span></p>
<p><br></p>
<p><span class=rvts7>-------------------------------------------------------------------------</span></p>
<p><span class=rvts7>Using Tomcat but need to do more? Need to support web services, security?</span></p>
<p><span class=rvts7>Get stuff done quickly with pre-integrated technology to make your job easier </span></p>
<p><span class=rvts7>Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo</span></p>
<p><a class=rvts9 href="http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642">http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642</a></p>
<p><br></p>
<p><span class=rvts7>_______________________________________________</span></p>
<p><span class=rvts7>axiomengine-developers mailing list</span></p>
<p><a class=rvts9 href="mailto:axi...@li...">axi...@li...</a></p>
<p><a class=rvts9 href="https://lists.sourceforge.net/lists/listinfo/axiomengine-developers">https://lists.sourceforge.net/lists/listinfo/axiomengine-developers </a></p>
<p><br></p>
<p><br></p>
</td>
</tr>
</table>
</div>
<p><br></p>
<p><br></p>
<p><br></p>
<p><br></p>
<p><span class=rvts12>-- </span></p>
<p><span class=rvts12>Best regards,</span></p>
<p><span class=rvts12> Serge </span><a class=rvts13 href="mailto:ser...@gm...">mailto:ser...@gm...</a></p>
</body></html>
|
24 messages has been excluded from this view by a project administrator.