Talk:Trinity in Trouble - Revision history

Alb42: /* Not up to date items */

2016-01-25T17:05:42Z

Not up to date items

Alb42: /* Not up to date items */ not your fault

2016-01-25T17:04:57Z

Not up to date items: not your fault

Molly: add section not up to date items

2016-01-24T23:48:30Z

add section not up to date items

Molly: /* What is "Isrvstr"? */ explain what structure Issrvstr is.

2016-01-24T23:32:13Z

What is "Isrvstr"?: explain what structure Issrvstr is.

Alb42: /* What is "Isrvstr"? */ new section

2016-01-24T14:42:20Z

What is "Isrvstr"?: new section

Molly: /* untyped Pointer in structures */ question whether to add to wiki or not.

2015-09-30T17:20:23Z

untyped Pointer in structures: question whether to add to wiki or not.

Molly: /* untyped Pointer in structures */ thx + further intentions

2015-09-27T17:37:18Z

untyped Pointer in structures: thx + further intentions

Alb42: untyped pointer, not a critics

2015-09-27T15:28:20Z

untyped pointer, not a critics

Molly: /* untyped Pointer in structures */ Answer

2015-09-27T12:55:05Z

untyped Pointer in structures: Answer

Alb42: untyped Pointer

2015-09-27T11:03:52Z

untyped Pointer

New page

== untyped Pointer in structures ==

Notice to: "structure TWindow field WScreen" and all the others of this kind:
You will find dozens of this kind in any kind of structures. The reason are recursive type definitions which does work little bit strange in Pascal.
Forwarding of types only within the same Block.
So this works:
<source lang="pascal">
type
PScreen = ^TScreen;
PWindow = ^TWindow;

TWindow = record
WScreen: PScreen
end;

TScreen = record
FirstWindow: PWindow;
end;
const
IDCMP_NEWSIZE = 1 shl 1;

</source>

and this does not work:

<source lang="pascal">
type
PScreen = ^TScreen;
PWindow = ^TWindow;

TWindow = record
WScreen: PScreen
end;

const
IDCMP_NEWSIZE = 1 shl 1;

type
TScreen = record
FirstWindow: PWindow;
end;
</source>

So a complete Rewrite/Resort of the complete unit is needed, as I did for AROS. I will not do that for MorphOS/Amiga again.
--[[User:Alb42|ALB42]] ([[User talk:Alb42|talk]]) 13:03, 27 September 2015 (CEST)

@@ Line 72: / Line 72: @@
 : against the wiki list again.
 : too soon? for me more already too later, sadly not in 3.0, same for Amiga, I planed to tell you but you was not online
-: in IRC last WE, so I thought you will see it in the commit message and/or wiki changes.
+: in IRC last WE, so I thought you will see it in the commit message and/or wiki changes. [[User:Alb42|ALB42]] ([[User talk:Alb42|talk]]) 18:05, 25 January 2016 (CET)

← Older revision		Revision as of 17:04, 25 January 2016
Line 68:		Line 68:

	I noticed "ASLRequestTags already in AROS a while" which is true, so apologies. It is a bit 'difficult' to keep 3.0 and 3.1 separated so am happy with any correction. fwiw: i try to keep trinity compatible for 3.0, and intended to maintain compatibility with 3.1 trunk in future -> recent changes might render that impossible (unless duplicating a lot of existing code within trinity unit(s)), so i'm currently leaning towards keeping trinity in comparison to 3.0, and work on 3.1 to get on par. In that regards your array of const changes came a bit too soon ;-p BTW: unit amigalib is as clean as possible (not having current changes for array of const though), and could be copied 1:1 to aros repository when done -> no weird types used as far as possible.		I noticed "ASLRequestTags already in AROS a while" which is true, so apologies. It is a bit 'difficult' to keep 3.0 and 3.1 separated so am happy with any correction. fwiw: i try to keep trinity compatible for 3.0, and intended to maintain compatibility with 3.1 trunk in future -> recent changes might render that impossible (unless duplicating a lot of existing code within trinity unit(s)), so i'm currently leaning towards keeping trinity in comparison to 3.0, and work on 3.1 to get on par. In that regards your array of const changes came a bit too soon ;-p BTW: unit amigalib is as clean as possible (not having current changes for array of const though), and could be copied 1:1 to aros repository when done -> no weird types used as far as possible.
		+
		+	: eh, no I just forget to mark it as done, I did it long time before, when rewriting the ASL unit, I didn't check
		+	: against the wiki list again.
		+	: too soon? for me more already too later, sadly not in 3.0, same for Amiga, I planed to tell you but you was not online
		+	: in IRC last WE, so I thought you will see it in the commit message and/or wiki changes.

← Older revision		Revision as of 23:48, 24 January 2016
Line 64:		Line 64:
	: Currently (only) used in [http://amigadev.elowar.com/read/ADCD_2.1/Includes_and_Autodocs_3._guide/node001C.html AddTOF()] and [http://amigadev.elowar.com/read/ADCD_2.1/Includes_and_Autodocs_3._guide/node004A.html RemTOF()]		: Currently (only) used in [http://amigadev.elowar.com/read/ADCD_2.1/Includes_and_Autodocs_3._guide/node001C.html AddTOF()] and [http://amigadev.elowar.com/read/ADCD_2.1/Includes_and_Autodocs_3._guide/node004A.html RemTOF()]
	: Choice of existence: whether we implement (or can be implemented) AddTOF() and RemTOF() functionality.		: Choice of existence: whether we implement (or can be implemented) AddTOF() and RemTOF() functionality.
		+
		+	== Not up to date items ==
		+
		+	I noticed "ASLRequestTags already in AROS a while" which is true, so apologies. It is a bit 'difficult' to keep 3.0 and 3.1 separated so am happy with any correction. fwiw: i try to keep trinity compatible for 3.0, and intended to maintain compatibility with 3.1 trunk in future -> recent changes might render that impossible (unless duplicating a lot of existing code within trinity unit(s)), so i'm currently leaning towards keeping trinity in comparison to 3.0, and work on 3.1 to get on par. In that regards your array of const changes came a bit too soon ;-p BTW: unit amigalib is as clean as possible (not having current changes for array of const though), and could be copied 1:1 to aros repository when done -> no weird types used as far as possible.

← Older revision		Revision as of 23:32, 24 January 2016
Line 59:		Line 59:

	Isrvstr seems not present in the AROS SDK		Isrvstr seems not present in the AROS SDK
		+
		+	: You are correct. It indeed is not, but should be there (at least for AROS 68k).
		+	: it is a structure representing an interrupt server chain, e.g. used to 'link' in your code to the interrupt list.
		+	: Currently (only) used in [http://amigadev.elowar.com/read/ADCD_2.1/Includes_and_Autodocs_3._guide/node001C.html AddTOF()] and [http://amigadev.elowar.com/read/ADCD_2.1/Includes_and_Autodocs_3._guide/node004A.html RemTOF()]
		+	: Choice of existence: whether we implement (or can be implemented) AddTOF() and RemTOF() functionality.

← Older revision		Revision as of 14:42, 24 January 2016
Line 55:		Line 55:
	What to do with things like TPoint, TRectangle etc. ? Should we list then, do we ignore those ? ATM i do not have been able to find a satisfactory answer for myself. Maybe one of you guys have ?		What to do with things like TPoint, TRectangle etc. ? Should we list then, do we ignore those ? ATM i do not have been able to find a satisfactory answer for myself. Maybe one of you guys have ?
	--[[User:Molly\|Molly]] ([[User talk:Molly\|talk]]) 19:20, 30 September 2015 (CEST)		--[[User:Molly\|Molly]] ([[User talk:Molly\|talk]]) 19:20, 30 September 2015 (CEST)
		+
		+	== What is "Isrvstr"? ==
		+
		+	Isrvstr seems not present in the AROS SDK

← Older revision		Revision as of 17:20, 30 September 2015
Line 50:		Line 50:
	:: oh do not missunderstand this as critics (thats the reason its only as comment on Discussion) just a side note, and because there are some cases, maybe would be better to make an own part for this cases. (P.S. I reformated your entry to a std. Wiki discussion entry) --[[User:Alb42\|ALB42]] ([[User talk:Alb42\|talk]]) 17:28, 27 September 2015 (CEST)		:: oh do not missunderstand this as critics (thats the reason its only as comment on Discussion) just a side note, and because there are some cases, maybe would be better to make an own part for this cases. (P.S. I reformated your entry to a std. Wiki discussion entry) --[[User:Alb42\|ALB42]] ([[User talk:Alb42\|talk]]) 17:28, 27 September 2015 (CEST)
	::: okidoki, no problem :-) thanks for the formatting, as i had no idea how todo that. BTW: i'm offline working on a complete unit table, so that it would be easier to see in one go what units are missing for what platform (or which units still needs attention). IBesides that i wanted to use the table to link to individual list entries (which would then be categorized per unit). It's taking me a bit longer than anticipated. --[[User:Molly\|Molly]] ([[User talk:Molly\|talk]]) 19:37, 27 September 2015 (CEST)		::: okidoki, no problem :-) thanks for the formatting, as i had no idea how todo that. BTW: i'm offline working on a complete unit table, so that it would be easier to see in one go what units are missing for what platform (or which units still needs attention). IBesides that i wanted to use the table to link to individual list entries (which would then be categorized per unit). It's taking me a bit longer than anticipated. --[[User:Molly\|Molly]] ([[User talk:Molly\|talk]]) 19:37, 27 September 2015 (CEST)
		+
		+	== other (obvious) conflicting types ==
		+
		+	What to do with things like TPoint, TRectangle etc. ? Should we list then, do we ignore those ? ATM i do not have been able to find a satisfactory answer for myself. Maybe one of you guys have ?
		+	--[[User:Molly\|Molly]] ([[User talk:Molly\|talk]]) 19:20, 30 September 2015 (CEST)

← Older revision		Revision as of 17:37, 27 September 2015
Line 49:		Line 49:
	:Yups, i am aware this is indeed the case.Please forgive me for just adding things as i go along and run into them, as it should perhaps be easier to just check the individual units one at a time. My goal is just to list everything i encounter, and at a later point in time start doing something about it. Hopefully it will also point out that writing source for multi-platform is a pita. ATM and imho there are bigger fish to fry to make things a bit more multi-platform compatible, such as the auto opening of libraries in general and intuition for MorphOS in particular. the different varargs implementations are also a pita to work with.I would be happy if the procedures/functions are getting a bit more compatible and on same par as well as remove/fix some of the most obvious and most encountered incompatibilities. Most things i can 'solve' temporary with using unit trinity, but some things can't be fixed that way (or very difficult). Thanks for paying attention, which i presume means it's not all in vain :-) --[[User:Molly\|Molly]] ([[User talk:Molly\|talk]]) 14:55, 27 September 2015 (CEST)		:Yups, i am aware this is indeed the case.Please forgive me for just adding things as i go along and run into them, as it should perhaps be easier to just check the individual units one at a time. My goal is just to list everything i encounter, and at a later point in time start doing something about it. Hopefully it will also point out that writing source for multi-platform is a pita. ATM and imho there are bigger fish to fry to make things a bit more multi-platform compatible, such as the auto opening of libraries in general and intuition for MorphOS in particular. the different varargs implementations are also a pita to work with.I would be happy if the procedures/functions are getting a bit more compatible and on same par as well as remove/fix some of the most obvious and most encountered incompatibilities. Most things i can 'solve' temporary with using unit trinity, but some things can't be fixed that way (or very difficult). Thanks for paying attention, which i presume means it's not all in vain :-) --[[User:Molly\|Molly]] ([[User talk:Molly\|talk]]) 14:55, 27 September 2015 (CEST)
	:: oh do not missunderstand this as critics (thats the reason its only as comment on Discussion) just a side note, and because there are some cases, maybe would be better to make an own part for this cases. (P.S. I reformated your entry to a std. Wiki discussion entry) --[[User:Alb42\|ALB42]] ([[User talk:Alb42\|talk]]) 17:28, 27 September 2015 (CEST)		:: oh do not missunderstand this as critics (thats the reason its only as comment on Discussion) just a side note, and because there are some cases, maybe would be better to make an own part for this cases. (P.S. I reformated your entry to a std. Wiki discussion entry) --[[User:Alb42\|ALB42]] ([[User talk:Alb42\|talk]]) 17:28, 27 September 2015 (CEST)
		+	::: okidoki, no problem :-) thanks for the formatting, as i had no idea how todo that. BTW: i'm offline working on a complete unit table, so that it would be easier to see in one go what units are missing for what platform (or which units still needs attention). IBesides that i wanted to use the table to link to individual list entries (which would then be categorized per unit). It's taking me a bit longer than anticipated. --[[User:Molly\|Molly]] ([[User talk:Molly\|talk]]) 19:37, 27 September 2015 (CEST)

@@ Line 46: / Line 46: @@
 --[[User:Alb42|ALB42]] ([[User talk:Alb42|talk]]) 13:03, 27 September 2015 (CEST)
-== re: untyped Pointer in structures ==
+:Hi ALB42,
-Hi ALB42,
+:Yups, i am aware this is indeed the case.Please forgive me for just adding things as i go along and run into them, as it should perhaps be easier to just check the individual units one at a time. My goal is just to list everything i encounter, and at a later point in time start doing something about it. Hopefully it will also point out that writing source for multi-platform is a pita. ATM and imho there are bigger fish to fry to make things a bit more multi-platform compatible, such as the auto opening of libraries in general and intuition for MorphOS in particular. the different varargs implementations are also a pita to work with.I would be happy if the procedures/functions are getting a bit more compatible and on same par as well as remove/fix some of the most obvious and most encountered incompatibilities. Most things i can 'solve' temporary with using unit trinity, but some things can't be fixed that way (or very difficult). Thanks for paying attention, which i presume means it's not all in vain :-) --[[User:Molly|Molly]] ([[User talk:Molly|talk]]) 14:55, 27 September 2015 (CEST)
+:: oh do not missunderstand this as critics (thats the reason its only as comment on Discussion) just a side note, and because there are some cases, maybe would be better to make an own part for this cases. (P.S. I reformated your entry to a std. Wiki discussion entry) --[[User:Alb42|ALB42]] ([[User talk:Alb42|talk]]) 17:28, 27 September 2015 (CEST)

← Older revision		Revision as of 12:55, 27 September 2015
Line 45:		Line 45:
	So a complete Rewrite/Resort of the complete unit is needed, as I did for AROS. I will not do that for MorphOS/Amiga again.		So a complete Rewrite/Resort of the complete unit is needed, as I did for AROS. I will not do that for MorphOS/Amiga again.
	--[[User:Alb42\|ALB42]] ([[User talk:Alb42\|talk]]) 13:03, 27 September 2015 (CEST)		--[[User:Alb42\|ALB42]] ([[User talk:Alb42\|talk]]) 13:03, 27 September 2015 (CEST)
		+
		+	== re: untyped Pointer in structures ==
		+	Hi ALB42,
		+
		+	Yups, i am aware this is indeed the case.
		+
		+	Please forgive me for just adding things as i go along and run into them, as it should perhaps be easier to just check the individual units one at a time.
		+
		+	My goal is just to list everything i encounter, and at a later point in time start doing something about it. Hopefully it will also point out that writing source for multi-platform is a pita.
		+
		+	ATM and imho there are bigger fish to fry to make things a bit more multi-platform compatible, such as the auto opening of libraries in general and intuition for MorphOS in particular. the different varargs implementations are also a pita to work with.
		+
		+	I would be happy if the procedures/functions are getting a bit more compatible and on same par as well as remove/fix some of the most obvious and most encountered incompatibilities.
		+
		+	Most things i can 'solve' temporary with using unit trinity, but some things can't be fixed that way (or very difficult).
		+
		+
		+	Thanks for paying attention, which i presume means it's not all in vain :-)
		+
		+	--[[User:Molly\|Molly]] ([[User talk:Molly\|talk]]) 14:55, 27 September 2015 (CEST)