Take 3

Nathan Strum, on Tue Mar 6, 2007 8:50 PM, said:

Nathan Strum, on Tue Mar 6, 2007 7:50 PM, said:

batari, on Tue Mar 6, 2007 8:24 PM, said:

SpiceWare, on Tue Mar 6, 2007 10:06 PM, said:

Nathan Strum, on Tue Mar 6, 2007 7:50 PM, said:

The previous final build worked fine on my Krok Cart, so I don't know what I'd be looking for here. The problem I had, I would assume, was specific to that particular cart.

Possibly specific to the EEPROM - I tested 9 other Medieval Mayhem carts from Al in my 2600, and they all worked OK(well, 4 of them had the "jitter bug", but no weird crashes). I just wanted some people to test it to make sure my fix didn't break something - I've had that happen before

batari, on Tue Mar 6, 2007 8:24 PM, said:

Too bad about the bugs, but regardless it's good to hear that my utility did its job...

Great utility, just a month too late for my release What's your utility written it? I'd like to recompile for OS X if possible. I transferred the LST file from my iMac G5 to my MacBook to check under XP(via Parallels).

#include <stdlib.h>
#include <stdio.h>
#include <string.h>

int hextoi(char a, char b)
{
  int val1,val2;
  if (!((a>='0' && a<='9') || (a>='a' && a<='f') || (a>='A' && a<='F'))) return 255;
  if (!((b>='0' && b<='9') || (b>='a' && b<='f') || (b>='A' && b<='F'))) return 255;
  val1=(int)a-48;
  if (val1>30) val1=val1-32;
  if (val1>9) val1=val1-7;
  val2=(int)b-48;
  if (val2>30) val2=val2-32;
  if (val2>9) b=val2-7;
  return val1*16+val2;
}

main(){
char opcode[2];
char operand[2];
int hexopcode;
int hexoperand;
char label[10];
char mnemonic[10];
char data[100];
int i;
// opcodes:
// 24,a4,c4,e4,05,25,45,65,a5,c5,e5,a6,a7
int list[13]={0x24,0xa4,0xc4,0xe4,0x05,0x25,0x45,0x65,0xa5,0xc5,0xe5,0xa6,0xa7};
	  while (1) {
		if (!fgets(data,100,stdin)){
		  fprintf(stderr,"Done.\n");
		  exit(0);
		}
	   if (data[28]>='0') continue;
	   opcode[0]=data[22];
	   opcode[1]=data[23];
	   operand[0]=data[25];
	   operand[1]=data[26];
	   hexopcode=hextoi(opcode[0],opcode[1]);
	   hexoperand=hextoi(operand[0],operand[1]);
	   if (!((hexoperand>=0) && (hexoperand < 128))) continue;
	   for (i=0;i<10;++i) {label[i]='\0';mnemonic[i]='\0';}
	   for (i=39;i<48;++i) label[i-39]=data[i];
	   for (i=36;i<40;++i) 
	   {
		  if (data[36]=='.') mnemonic[i-36]=data[i+1];
		  else if (data[35]=='b') mnemonic[i-36]=data[i-1];
		  else if (data[35]=='w') mnemonic[i-36]=data[i-1];
		  else mnemonic[i-36]=data[i];
	   }

  for (i=0;i<13;++i)
  {
	if (hexopcode == list[i])
	{
	   if (strncmp(label,"CX\0",2) && strncmp(label,"INPT\0",4) && strncmp(mnemonic,"wor\0",3) && strncmp(mnemonic,"byt\0",3))
		 printf("%s\n",data);
	}
  }

  }
}

Cybergoth, on Wed Mar 7, 2007 10:14 AM, said:

Fred, have you considered forwarding the tool to Andrew Davie, so he can add it to the DASM homepage? I'm sure it'd be interesting for every DASM developer to have

Cybergoth, on Wed Mar 7, 2007 4:14 AM, said:

Fred, have you considered forwarding the tool to Andrew Davie, so he can add it to the DASM homepage? I'm sure it'd be interesting for every DASM developer to have

batari, on Wed Mar 7, 2007 10:59 PM, said:

Cybergoth, on Wed Mar 7, 2007 5:17 PM, said:

batari, on Wed Mar 7, 2007 10:59 PM, said:

Also, I think I found a way to remove the false positives from the results.

It seems as if

1. Checking the values wether they're *really* a hexnumber
=> (a>='0' && a<='9') || (a>='a' && a<='z') || (a>='A' && a<='Z')

2. Excluding "word" and "byte" additionally to "CX" and "INPT"

should get rid of most false positives I've seen so far

batari, on Thu Mar 8, 2007 12:43 AM, said:

Cybergoth, on Thu Mar 8, 2007 4:01 AM, said: