This is a great move from you Dietmar, thank you very much for sticking with us :)

Regards,
Joachim Birging

_________________
Common sense - So rare it's almost like a super power

Thanx for this gift dietmar!!

_________________
--

Amiga x5000 ı o2o ı 4GB RAM ı RadeonRX580 | SBlaster Audigy Fx - AmigaOS4.1 FInal Edition

A1200 sandwich :)

Croatian Amiga portal

Thanks Dietmar. People are sure happy for this.

So when next update comes the earlier version
of Cubic IDE will be for free (hehe)

@Kicko

Maybe that depends on the sale number of Cubic IDE ?

It's better to considder this as a special gift and to keep
supporting new software while it is developed and updated.

Not directed directly at you ofcourse, but to everybody.

_________________
A1G3-SE + OS4.1 u1 iso (x2)

Thanks!
I look forward to giving this a workout.
Maybe I'll even give up SE one day... maybe.

LyleHaze

_________________
question=(2b||!(2b))

I was going to download it, then I noticed this on the requirements page:

"The software is NOT compatible with OS4 (because the memory management functions of OS4 are not exactly identical with the same functions of AmigaOS3). The result is memory ocrruption, leading to generally unstable behaviour."

What a pity

Varthall

_________________
AmigaOne XE - AmigaOS 4.1 - Freescale 7457 1GHz - 1GB ram

Quote:

"The software is NOT compatible with OS4 (because the memory management functions of OS4 are not exactly identical with the same functions of AmigaOS3). The result is memory ocrruption, leading to generally unstable behaviour."

The statement is not entirely fair. Golded uses an undocumented feature of OS 3's memory management, namely that the OS used to store the size of a memory chunk allocated with AllocVec in front of the allocation itself. This is no longer true for AmigaOS4.0, so any program that relies on that will fail.

AmigaOS4.0 memory functions still perform the same task that was specified for OS 3, however programs (ab)using undocumented side effects of older revisions will cease to work since the memory handling changed.

Later versions of GoldEd fix this problem.

_________________
Seriously, if you want to contact me do not bother sending me a PM here. Write me a mail

@Rogue & Dietmar

Well, let's hope it's the case ! I have a copy of CubicIDE, and there's just too much problem between CubicIDE and OS4 Upd#4 :

- flashing screen
- scroll wheel no more working
- edited document's directory locked !!! (reboot to free them)

Will they be adressed ? Anyway, it's still a great editor !!!

Quote:

- flashing screen

This has started with update #4 and is apparently related to the blank borders option (a video property). The flicker will disappear if you disable the "blank borders" checkbox in Cubic's display configuration dialog.

Quote:

- scroll wheel no more working

OS4 #4 does not use the newmouse protocol (how AmigaOS software has addressed the mouse wheel in recent years). I'm not going to change my software, last but not least because I have no SDK nor documentation nor test opportunity. This has to be addressed in other ways, ie. a general solution for newmouse-oriented software.

Quote:

- edited document's directory locked !!! (reboot to free them)

Have you configured the editor to "stay resident", ie. run without window? See checkmark in the first menu. In that case the current directory - obviously - will remain locked. Otherwise, it would be an update #4 problem I haven't heard of before; please provide useful information in that case (snoopdos).

Would that help:

ISig = Wait( 1 window->UserPort->mp_SigBit );
while ( (IMsg = (struct IntuiMessage *)GetMsg( vi->window->UserPort )) && !out )
{
IClass = IMsg->Class;
ICode = IMsg->Code;
MouseX = IMsg->MouseX;
MouseY = IMsg->MouseY;
IAddress = IMsg->IAddress;
ReplyMsg((struct Message *)IMsg);

switch (IClass)
{
...

#ifdef __amigaos4__
case IDCMP_EXTENDEDMOUSE:
if (ICode & IMSGCODE_INTUIWHEELDATA)
{
struct IntuiWheelData *wd = (struct IntuiWheelData*)IAddress;

px = MouseX;
py = MouseY;

angle += 10 * wd->WheelY;
refresh = TRUE;
}
break;
#else
...
#endif
...

and

#include "intuition/intuition.h"

struct IntuiWheelData {
UWORD Version; /* version of this structure (see below) */
UWORD Reserved; /* always 0, reserved for future use */
WORD WheelX; /* horizontal wheel movement delta */
WORD WheelY; /* vertical wheel movement delta */
};

(kind of)

I am willing to test this...

And no, the new mouse hack eats my new style events. (that's possible with that tool from os4depot).

(someone tells me how to format that text ?)

--
Actually, I think that code is broken. Cannot believe that the IAdress is still valid after the ReplyMsg(). But, at least it works...you'd propably want to check the wheel data before the ReplyMsg (?).

Last edited by Jorge on 01-Apr-2006 at 07:06 AM.

_________________
AmigaOne XE G3/933/VIA/FM801/R200 (fixed),
G3SE/600/Voodoo3/Sil680/RTL8139/SBLive! (noiseless!)
µA1-MK2/G3/933/R200/CMI8738
XE/G4 (broken 7450/800)

@Dietmar

Ok for the first one, nuchecking the option solve the problem.

For the locking problem, yes, CubicIDE is resident, but i'm talking about the edited document's directory, not CubicIDE's one. I.e. when you close a document (let's say SDK:local/common/MyApp/foo.c), you should be able to delete the MyApp directory. It's no more possible with Upd4 when a document contained in MyApp has been edited once.

I would love to give you more information, but... erm... i don't know how to interpret Snoopdos informations. I see "lock" operations, but no "unlock".

@Hombre
Sorry, no idea why you have this difficulty. Report it as problem for Upd4 if you are sure it is not user error, maybe it's a known issue. File a reasonable bug report. What you do, how you do it (detailed and step by step), snoopdos log, version numbers etc.

This version of golded seems to be working fine on OS4 here, aint had a problem yet.

_________________
A1XE G4 800mhz,512mb,Radeon9200,SonyDRU-510A and 800a,ESI Juli@, Sil680, WiFi.

@Curty
Quote:

This version of golded seems to be working fine on OS4 here, aint had a problem yet.

The reason is that many allocations are still (I don't know how long) executed by OS4 as the program expects, ie. allocation size before allocation, just as OS3 + MorphOS + AROS implement AllocVec(). Not consistently though: If certain conditions are met - if the allocation is big and if the MEMF_PUBLIC flag is not used - the size is not stored by the OS and my program sees a random number. That's when you will get the corruption. During normal use, this is a very rare event but ultimately unavoidable.

This issue between OS4 and GoldED was addressed with service pack 20.

Excellent news! Keep up the good work!

_________________
Dark Lord Design Wicked Solutions For Damned Problems