Another printf question...

[Dale Botkin]

OK, let's say I have an array of messages:

const char messages[5][10] = {
"Message 1",
"Message 2",
"Message 3",
"Message 4",
"Message 5" };

and I want to select and print one of the messages. Can I do this:

printf(lcd_putc,"\%C\n",messages[x][0]);

or do I need to:

char buffer[10];
memcpy(buffer, messages[x][0], 10);
printf(lcd_putc,"\%C\n",buffer);

Either way it compiles without error messages. I doubt the first way will work, since it's too easy... am I right?
___________________________
This message was ported from CCS's old forum
Original Post ID: 1412

[Tomi]

The first way will work fine. I have just a small note about your code: it would be better to use strings with length of 2^N to help the compiler to calculate the address of the string.
For example, the address calculation and byte access in your case looks like this:
const char messages[5][10] // length: 10 bytes
.................... printf("\%C\n",messages[x][0]);
*
003A: CLRF 77 // "implies a multiplication by 10"
003B: CLRF 78
003C: MOVF 22,W
003D: BCF 03,0
003E: BTFSC 23,0
003F: ADDWF 77,F
0040: RRF 77,F
0041: RRF 78,F
0042: BTFSC 23,1
0043: ADDWF 77,F
0044: RRF 77,F
0045: RRF 78,F
0046: BTFSC 23,2
0047: ADDWF 77,F
0048: RRF 77,F
0049: RRF 78,F
004A: BTFSC 23,3
004B: ADDWF 77,F
004C: RRF 77,F
004D: RRF 78,F
004E: BTFSC 23,4
004F: ADDWF 77,F
0050: RRF 77,F
0051: RRF 78,F
0052: BTFSC 23,5
0053: ADDWF 77,F
0054: RRF 77,F
0055: RRF 78,F
0056: BTFSC 23,6
0057: ADDWF 77,F
0058: RRF 77,F
0059: RRF 78,F
005A: BTFSC 23,7
005B: ADDWF 77,F
005C: RRF 77,F
005D: RRF 78,F
005E: BCF 0A,3
005F: BCF 0A,4
0060: GOTO 0A7
........
00A7: MOVF 78,W
00A8: CALL 004 // access the string

And if you use 16-bytes long strings:
const char messages[5][16] // 16-byte strings

.................... printf("\%C\n",messages[x][0]);
*
0099: SWAPF 21,W
009A: MOVWF 77
009B: MOVLW F0
009C: ANDWF 77,F
009D: MOVF 77,W
009E: CALL 004 // access the string
___________________________
This message was ported from CCS's old forum
Original Post ID: 1413

[Dale B.]

Cool, thanks for the response! On the one hand the compiler didn't complain so I thought it might be OK, but wasn't sure what I would end up with and I haven't breadboarded this yet.

I'm debating what to do about the message length. I'm storing messages for a 2x16 LCD, so I really want 16-character strings, but that makes the array 17 bytes (16 + \0). I'd cut it to 15 characters, but a couple of the messages really need to be 16 characters long... I'll have to play with that some if memory gets tight.

Thanks!

Dale
___________________________
This message was ported from CCS's old forum
Original Post ID: 1414

[Tomi]

<font face="Courier New" size=-1>Here is my code snippet to use 16chr LCDs without that terminating \0.
The definition is not too elegant:
(The "Message 4" is terminated after the "4" character)
const char messages[5][16] = {
{'M','e','s','s','a','g','e',' ','1','a','b','c','d','e','f','g'},
{'M','e','s','s','a','g','e',' ','2','a','b','c','d','e','f','g'},
{'M','e','s','s','a','g','e',' ','3','a','b','c','d','e','f','g'},
{'M','e','s','s','a','g','e',' ','4',0},
{'M','e','s','s','a','g','e',' ','5','a','b','c','d','e','f','g'}
};
These are NOT strings but character arrays (no terminating \0 added automatically). Printing is terminated when either a terminating \0 found or the 16th character is displayed:

void DisplayString(char stringID)
{
char i,what;
for (i=0;i<16;i++) { // max. 16 chars are displayed
what = messages[stringID][i]; // copy the char into RAM to avoid the double access of the ROM
if (!what) break; // Terminating NULL found
lcd_putc(what);
}
} ___________________________
This message was ported from CCS's old forum
Original Post ID: 1421