Author Topic: Oxygen Basic alpha  (Read 58642 times)

jcfuller

  • Guest
Re: Oxygen Basic alpha
« Reply #15 on: September 11, 2010, 12:25:54 PM »
Charles is there a limit with this construct?
James
Code: [Select]
dim as  byte b(600)=>(1,0,255,255,0,0,0,0,0,0,0,0,64,8,202,16,9,0,10,0,10,0,104,1,132,0,0,0,0,0,82,0,67,0,68,0,76,0,71,0,
     50,0,83,0,76,0,76,0,0,0,10,0,144,1,0,0,84,0,97,0,104,0,111,0,109,0,97,0,0,0,0,0,0,0,0,0,0,0,0,0,
     0,0,1,80,240,0,99,0,57,0,15,0,233,3,0,0,255,255,128,0,68,0,105,0,115,0,109,0,105,0,115,0,115,0,0,0,0,0,0,0,
     0,0,0,0,0,0,0,0,7,0,0,80,6,0,3,0,83,1,27,0,234,3,0,0,255,255,128,0,82,0,67,0,32,0,70,0,105,0,108,0,
     101,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,18,0,80,15,0,12,0,44,1,12,0,235,3,0,0,255,255,130,0,0,0,0,0,
     0,0,0,0,0,0,0,0,0,0,1,80,68,1,12,0,15,0,12,0,236,3,0,0,255,255,128,0,46,0,46,0,46,0,46,0,46,0,0,0,
     0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,80,63,0,99,0,57,0,15,0,237,3,0,0,255,255,128,0,67,0,114,0,101,0,97,0,
     116,0,101,0,32,0,67,0,111,0,100,0,101,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,7,0,0,80,6,0,39,0,83,1,39,0,
     238,3,0,0,255,255,128,0,83,0,101,0,108,0,101,0,99,0,116,0,32,0,67,0,111,0,100,0,101,0,32,0,67,0,114,0,101,0,97,0,
     116,0,105,0,111,0,110,0,32,0,84,0,121,0,112,0,101,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,9,0,3,80,27,0,54,0,
     69,0,12,0,239,3,0,0,255,255,128,0,70,0,117,0,108,0,108,0,32,0,83,0,76,0,76,0,32,0,77,0,111,0,100,0,117,0,108,0,
     101,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,9,0,1,80,120,0,54,0,87,0,15,0,240,3,0,0,255,255,128,0,83,0,76,0,
     76,0,32,0,65,0,83,0,77,0,68,0,65,0,84,0,65,0,32,0,84,0,111,0,32,0,70,0,105,0,108,0,101,0,0,0,0,0,0,0,
     0,0,0,0,0,0,0,0,9,0,1,80,231,0,54,0,105,0,15,0,241,3,0,0,255,255,128,0,83,0,76,0,76,0,32,0,65,0,83,0,
     77,0,68,0,65,0,84,0,65,0,32,0,84,0,111,0,32,0,67,0,108,0,105,0,112,0,66,0,111,0,97,0,114,0,100,0,0,0,0,0)


cevpegge

  • Guest
Re: Oxygen Basic alpha
« Reply #16 on: September 11, 2010, 11:09:42 PM »

Many megabytes I would think James. The compiler just builds up a large string of Assembler. But boundary checks are not made on arrays. They are primitive beasts.

Charles

ahadev

  • Guest
Re: Oxygen Basic alpha
« Reply #17 on: September 12, 2010, 05:06:16 AM »
Charles, I have unpacked the archive on a Win2K system. If I want to start Oxygen Basic only a requester opens and tells me that this is no Windows executable or something like that. Is this a known problem on Win2K?

Andreas

cevpegge

  • Guest
Re: Oxygen Basic alpha
« Reply #18 on: September 12, 2010, 06:27:22 AM »

Hi Andreas,

I had one problematic posting. could you try downloading again to get the latest version and see if this works.

thanks

Charles

jcfuller

  • Guest
Re: Oxygen Basic alpha
« Reply #19 on: September 12, 2010, 06:59:33 AM »
Charles,
  I know this is only an alpha, but because you update so often, would you add a build or ver or something to the upload so we know if we have the latest.

James

jcfuller

  • Guest
Re: Oxygen Basic alpha
« Reply #20 on: September 12, 2010, 07:00:43 AM »
Charles,
Listobject4 fails to run after compiling.

James

cevpegge

  • Guest
Re: Oxygen Basic alpha
« Reply #21 on: September 12, 2010, 08:20:12 AM »
I've included a log recording the updates as we go.

OxygenUdates.txt
Code: [Select]
OXYGEN UPDATES LOG


03:39 12/09/2010  Resources may now be attached to compiled files. (rsrc is las section in file)
03:39 12/09/2010  File paths containing spaces supported.
04:45 12/09/2010  tools updated: cco2 co2 eo2

Thanks James for letting me know about ListObject4 - the compiled version does not work unlike ListObject3 so I hope to trace the problem fairly quickly.

Charles
« Last Edit: September 12, 2010, 08:31:29 AM by cevpegge »

jcfuller

  • Guest
Re: Oxygen Basic alpha
« Reply #22 on: September 12, 2010, 09:15:11 AM »
Actually I meant on the sourceforge site.

James

jcfuller

  • Guest
Re: Oxygen Basic alpha
« Reply #23 on: September 12, 2010, 10:21:42 AM »
Charles,
This will not compile.

James

Code: [Select]
#basic
dim as  byte b(600)=>(1,0,255,255,0,0,0,0,0,0,0,0,64,8,202,16,9,0,10,0,10,0,104,1,132,0,0,0,0,0,82,0,67,0,68,0,76,0,71,0,
     50,0,83,0,76,0,76,0,0,0,10,0,144,1,0,0,84,0,97,0,104,0,111,0,109,0,97,0,0,0,0,0,0,0,0,0,0,0,0,0,
     0,0,1,80,240,0,99,0,57,0,15,0,233,3,0,0,255,255,128,0,68,0,105,0,115,0,109,0,105,0,115,0,115,0,0,0,0,0,0,0,
     0,0,0,0,0,0,0,0,7,0,0,80,6,0,3,0,83,1,27,0,234,3,0,0,255,255,128,0,82,0,67,0,32,0,70,0,105,0,108,0,
     101,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,18,0,80,15,0,12,0,44,1,12,0,235,3,0,0,255,255,130,0,0,0,0,0,
     0,0,0,0,0,0,0,0,0,0,1,80,68,1,12,0,15,0,12,0,236,3,0,0,255,255,128,0,46,0,46,0,46,0,46,0,46,0,0,0,
     0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,80,63,0,99,0,57,0,15,0,237,3,0,0,255,255,128,0,67,0,114,0,101,0,97,0,
     116,0,101,0,32,0,67,0,111,0,100,0,101,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,7,0,0,80,6,0,39,0,83,1,39,0,
     238,3,0,0,255,255,128,0,83,0,101,0,108,0,101,0,99,0,116,0,32,0,67,0,111,0,100,0,101,0,32,0,67,0,114,0,101,0,97,0,
     116,0,105,0,111,0,110,0,32,0,84,0,121,0,112,0,101,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,9,0,3,80,27,0,54,0,
     69,0,12,0,239,3,0,0,255,255,128,0,70,0,117,0,108,0,108,0,32,0,83,0,76,0,76,0,32,0,77,0,111,0,100,0,117,0,108,0,
     101,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,9,0,1,80,120,0,54,0,87,0,15,0,240,3,0,0,255,255,128,0,83,0,76,0,
     76,0,32,0,65,0,83,0,77,0,68,0,65,0,84,0,65,0,32,0,84,0,111,0,32,0,70,0,105,0,108,0,101,0,0,0,0,0,0,0,
     0,0,0,0,0,0,0,0,9,0,1,80,231,0,54,0,105,0,15,0,241,3,0,0,255,255,128,0,83,0,76,0,76,0,32,0,65,0,83,0,
     77,0,68,0,65,0,84,0,65,0,32,0,84,0,111,0,32,0,67,0,108,0,105,0,112,0,66,0,111,0,97,0,114,0,100,0,0,0,0,0)

print b(3)
terminate


cevpegge

  • Guest
Re: Oxygen Basic alpha
« Reply #24 on: September 12, 2010, 02:11:24 PM »

Yes that was an interesting one James. I use a counter to prevent macro recursion explosions (!), and this was not cleared between multi-assignments.

I have also fixed ListObject4 but I am not certain what causes it - It seems to be associated with string concatenation.

To distinguish Oxygen alpha versions I will serialise the update log and put the same version number on the SourceForge Download button.

Thanks.

Charles


ahadev

  • Guest
Re: Oxygen Basic alpha
« Reply #25 on: September 13, 2010, 12:51:11 AM »
Hi Charles,

I had one problematic posting. could you try downloading again to get the latest version and see if this works.

The problem on Win2K is still there. If I doubleclick the file OxideSc.exe I get the requester that this file is not a valid Win32 application.

Andreas

cevpegge

  • Guest
Re: Oxygen Basic alpha
« Reply #26 on: September 13, 2010, 01:16:58 AM »

Ah yes. I think I know what the problem might be. This version of Scit (IDE derived from Scintilla) has compressed sections in its EXE which your OS may not be able to recognise. I will try out the uncompacted version and let you know.

Thanks!

Charles

ahadev

  • Guest
Re: Oxygen Basic alpha
« Reply #27 on: September 13, 2010, 02:58:41 AM »
Thank you, Charles! Then I will test the next update.

cevpegge

  • Guest
Re: Oxygen Basic alpha
« Reply #28 on: September 13, 2010, 04:28:55 AM »

Andreas,

I have just posted the other Scit IDE which I hope will work on your OS. This posting also includes fixes for the problems reported here by James.

#Alpha003
http://oxygenbasic.sourceforge.net


ahadev

  • Guest
Re: Oxygen Basic alpha
« Reply #29 on: September 13, 2010, 04:57:24 AM »
Charles,

unfortunately the binary files OxideSc.exe (old) AND SciTE.exe are both from 10-08-28 and still don't work. I think the file or a dll is still compressed?

On which Windows versions is it tested?

Andreas