[Bugsy]

Not sure I understand, but if you're asking if you can replace individual characters, yes you can. You just find the position of the chars you want to replace and increment or decrement as needed.

You also have the replace string natives.

PHP Code:

new szSource[ 64 ] = "May I please have a warm beer?";

new iFindPos = strfind( szSource , "warm" );

szSource[ iFindPos + 0 ] = 'c';
szSource[ iFindPos + 1 ] = 'o';
szSource[ iFindPos + 2 ] = 'l';
szSource[ iFindPos + 3 ] = 'd';

server_print( szSource ); 


[abdobiskra]

Not exactly. There are letters, for example, whose position in the word differs and takes a different encoding of the bytes. Therefore, I want to specify the shape of the byte before the letter that I specified
What i try :

PHP Code:

                for (new i = 0; i < sizeof(said_to_utf8); i += 3)// Each character contains 3 bytes
                {
                    
                    if ((said_to_utf8[i] == 0xEF) && (said_to_utf8[i+1] == 0xBB ))
                    {
                        //Check the third byte of the character "0x9F"
                        if (i + 5 < sizeof(said_to_utf8) && said_to_utf8[i+5] == 0x9F) 
                        {
                            // The character before the current character 
                            if (i - 3 >= 0)
                            {
                                new byteBefore1 = said_to_utf8[i - 3];
                                new byteBefore2 = said_to_utf8[i - 2];
                                new byteBefore3 = said_to_utf8[i - 1];
                                
                                client_print(0, print_chat, "(debug1) 1) %d | 2) %d | 3) %d", byteBefore1, byteBefore2, byteBefore3)
                            
                            }
                
                            // The character after the current character
                            if (i + 6 < sizeof(said_to_utf8))
                            {
                                new byteAfter1 = said_to_utf8[i + 6];
                                new byteAfter2 = said_to_utf8[i + 7];
                                new byteAfter3 = said_to_utf8[i + 8];
                                client_print(0, print_chat, "(debug2) 1) %d | 2) %d | 3) %d", byteAfter1, byteAfter2, byteAfter3)
            
                            }
                        }
                    }
                } 


I didn't get results (maybe the method is wrong I'm not sure..)

[Bugsy]

Can you provide an example?

Source = "Hello Test"
Outcome = "Goodbyte Test"

or

Source = 0xFF 0x80 0x55 0x22
Outcome = 0x80 0x55 0x22 0x00

Once I fully understand I can likely make it happen.

[abdobiskra]

Do you want it in Arabic?
Source = "ﺎﺒﺣﺮﻣ"
to be = "ﺎﻋﺍﺩﻭ"
should save"Encode in UTF-8-BOM"
So:
Source be like : "ï»؟ï؛ژï؛’ï؛£ï؛®ï»£" ===> مرحبا ===> Hello
to be : "ï»؟ï؛ژﻋï؛چï؛©ï»Â*" ==> وداعا ==> Goodbye

Source = "ﺎﺒﺣﺮﻣ" to be = "ﺎﻋﺍﺩﻭ"

For me here, when I read thes words up, it gives the writing upside down from left to right, but it must be really like this in order to get a good result inside the screen (as if it were a trick or a hack or something like that)
Because when I want to write it in programs that support the Arabic language, for example, or websites, or in this forum or others, it is much easier and more correct than that

The Arabic language is not supported properly within the game screen
Sorry, This is why it is difficult to give an example
What I do is reverse the letters and words from left to right and then replace the letters with what suits their shape according to their position within the word (Isolated - End - Middle - Beginning)
I think fixing it by bytes makes more sense.
Some sites that I use to decode the letters:
https://www.compart.com/en/unicode/bidiclass/AL
https://en.wikipedia.org/wiki/Arabic_script_in_Unicode
https://mothereff.in/utf-8#%EF%BB%A3

[Bugsy]

Again, I can likely do it but I don't understand what you are trying to do. If you can somehow convey that using an English example, I can likely help.

Let's take the language out of the conversation if possible. Maybe something like "find byte position 2, replace with X. Replace the next byte with Y. Skip a byte and replace the following byte with Z."

[abdobiskra]

If you mean the English language, it is somewhat different because the number of bytes in it is 1 for each letter

I will try to explain more:

for (new i = 0; i < sizeof(said_to_utf; i++)// {

To calculate the bytes

if (said_to_utf8[i] == 0xEF && (said_to_utf8[i+1] == 0xBB || said_to_utf8[i+1] == 0xBA))) {

Here the first and second bytes of each character in words are specified .. that Arabic has 3 bytes per letter ..Only the third byte changes
So I want to check the third byte of one of the characters I'm looking for and replace it with another byte.

if (said_to_utf8[i+2] == 0x9F) { said_to_utf8[i+2] = 0x9E; } {

But it is missing a condition and it must not be another character before it being set for me to replace it

So what I thought was to subtract 3 bytes from "i" to get it

if (i - 3 >= 0) { new byteBefore = said_to_utf8[i - 3]; if (byteBefore == 0x8E || byteBefore == 0x8B) { said_to_utf8[i+2] = 0x9E; client_print(0, print_chat, "(debug) before the letter ") } }

But I didn't succeed in that maybe wrong way..

P.s:
I am not trying to replace a letter or word with another letter or word , I just want to replace the byte of the letter according to its shape and position in the word.

If we were to apply the example to your data it would become:

// Source = 0xFF 0x80 0x55 0x22

PHP Code:

        for (new i = 0; i < sizeof(said_to_utf8); i++)// Source = 0xFF 0x80 0x55 0x22
        {
            if (said_to_utf8[i] == 0x55)
            {
                if (i - 1 >= 0)
                {
                    new byteBefore = said_to_utf8[i - 1];
                        
                    if (byteBefore == 0x80) {
                            
                        
                        said_to_utf8[i] = 0xXX; //EXAMPLE
                    }
                }
            }
        } 


Sorry for not explaining well

[Bugsy]

Ok so here's what I understand.

- Each character is 3 bytes
- You want to check the first two bytes, and if they match what you're looking for, check if the 3rd is X before changing to Y

Reviewing your code here and now that I understand a bit better, maybe your problem was your for-loop is incrementing by 1 byte per iteration instead of incrementing by 3.

PHP Code:

#include <amxmodx>

public plugin_init() 
{
    new i;
    
    //5 letters based on your explanation
    new said_to_utf8[] = 
    {  
        0xEF , 0xBB , 0x9F ,    //239 187 159
        0xEF , 0xBB , 0x9F , 
        0xEF , 0xBB , 0x9F , 
        0xEF , 0xBB , 0x9F , 
        0xEF , 0xBB , 0x9F
    };
    
    //Output the original data
    for ( new i = 0 ; i <= charsmax( said_to_utf8 ) ; i+= 3 )
        server_print( "Original %d %d %d" , said_to_utf8[ i ] ,  said_to_utf8[ i + 1] , said_to_utf8[ i + 2 ] );
        
    do 
    {
        //If byte 1 is 0xEF and byte 2 is 0xBB then..
        if  ( ( said_to_utf8[ i ] == 0xEF ) && ( said_to_utf8[ i + 1 ] == 0xBB ) )
        {
            //if the third byte is 0x9F, change it to 0x9E
            if ( said_to_utf8[ i + 2 ] == 0x9F )
                said_to_utf8[ i + 2 ] = 0x9E;
        }
    } 
    while ( ( i += 3 ) < charsmax( said_to_utf8 ) )
    
    //Output the changed data
    for ( new i = 0 ; i <= charsmax( said_to_utf8 ) ; i+= 3 )
        server_print( "Changed %d %d %d" , said_to_utf8[ i ] ,  said_to_utf8[ i + 1] , said_to_utf8[ i + 2 ] );
} 


Output:

Original 239 187 159
Original 239 187 159
Original 239 187 159
Original 239 187 159
Original 239 187 159

Changed 239 187 158
Changed 239 187 158
Changed 239 187 158
Changed 239 187 158
Changed 239 187 158

[abdobiskra]

Yes, now I want to get the last byte of the character before the character I specified earlier

[Bugsy]

You're losing me... So then just subtract 3 from the index, and increment by 2?

[abdobiskra]

Thank youu Bugsy
I had some confusion in my code I think this method is enough

PHP Code:

                        if (said_to_utf8[i+2] == 0x9F) {
                            
                            if (i + 3 < sizeof(said_to_utf8)) {
                                
                                if ( said_to_utf8[i+5] == 0x8E) {