CSRD-NameFilter/scripting/csrd_name_filter.sp

/**
 * [CSRD] Name Filter
 * 
 * An in-house plugin that tidies up names.
 * Made in response to a few script kiddies that are (ab)using newlines and other non-printing
 * characters in their name.
 * 
 * If their existing name has no printable characters, their name will be changed to
 * "unprintable" plus their userid and their name will be locked.
 */
#pragma semicolon 1
#include <sourcemod>

#include <sdktools_functions>

#pragma newdecls required

#define PLUGIN_VERSION "1.0.5"
public Plugin myinfo = {
	name = "[CSRD] Name Filter",
	author = "nosoop",
	description = "Filters out rather unscrupulous characters from names and locks the name if "
			... "they are using a completely non-printable name.",
	version = PLUGIN_VERSION,
	url = "https://pika.nom-nom-nom.us/git/"
}

#define UNPRINTABLE_NAME_USER_FORMAT "unprintable (%d)"

int g_MultiByteNonPrinting[] = {
	0xe2a080, // braille pattern blank U+2800
	0
};

public void OnPluginStart() {
	for (int i = 1; i <= MaxClients; i++) {
		if (IsClientConnected(i)) {
			OnNameUpdated(i);
		}
	}
}

public void OnClientConnected(int client) {
	OnNameUpdated(client);
}

public void OnClientSettingsChanged(int client) {
	OnNameUpdated(client);
}

void OnNameUpdated(int client) {
	char currentName[MAX_NAME_LENGTH], newNameBuffer[MAX_NAME_LENGTH];
	
	GetClientName(client, currentName, sizeof(currentName));
	
	bool bNameChanged;
	
	/**
	 * Copies valid characters into another buffer (while also dealing with Unicode).
	 * The buffer is used if any characters are filtered.
	 */
	for (int i = 0; i < strlen(currentName); /* ... */) {
		int fullChar, nCharBytes;
		fullChar = GetMultiByteCharacterAt(currentName, i, nCharBytes);
		
		bool bCharFiltered;
		
		/**
		 * Filter out characters.
		 * 
		 * Note that while Steam may filter out whitespace characters in its interface, other
		 * external applications may be used to bypass this restriction.
		 * 
		 * (We use `sm_rename` to test the filtering process.)
		 */
		switch (nCharBytes) {
			case 3: {
				// filter unicode invisible characters
				bCharFiltered |= IsUnicodeCharInvisible(fullChar);
			}
			case 1: {
				// filter out newlines
				bCharFiltered |= (fullChar == 0x0d || fullChar == 0x0a);
			}
			default: {
				// unhandled
			}
		}
		
		if (!bCharFiltered) {
			// it's a character we're not filtering
			StrCatMultiByteChar(newNameBuffer, sizeof(newNameBuffer), fullChar, nCharBytes);
		}
		
		bNameChanged |= bCharFiltered;
		
		i += nCharBytes;
	}
	
	if (bNameChanged) {
		// Remove surrounding whitespace
		TrimString(newNameBuffer);
		
		// TODO remove duplicate whitespace
		
		// If there's no printable characters in the buffer, the player is a total dongbus.
		if (strlen(newNameBuffer) == 0) {
			Format(newNameBuffer, sizeof(newNameBuffer), UNPRINTABLE_NAME_USER_FORMAT,
					GetClientUserId(client));
			
			SetClientNameLock(client);
		}
		
		SetClientName(client, newNameBuffer);
	}
	// TODO else not name changed, but check if we can do other non-filter stuff? (trim)
}

/**
 * Determines if the specified character is non-printing (whitespace or not supported).
 * 
 * References:
 * http://www.unicode.org/faq/unsup_char.html
 * https://en.wikipedia.org/wiki/Whitespace_character
 */
bool IsUnicodeCharInvisible(int uchar) {
	// handle invisible math operators, &c. (U+205F ... U+206F)
	if (uchar >= 0xe2819f && uchar <= 0xe281af) {
		return true;
	}
	
	// handle other invisible characters
	int c;
	do {
		if (uchar == g_MultiByteNonPrinting[c]) {
			return true;
		}
	} while (g_MultiByteNonPrinting[++c] != 0);
	
	return false;
}

/**
 * Attempts to lock the specified client's name with the existing `namelockid` command.
 */
stock bool SetClientNameLock(int client, bool locked = true) {
	if (CommandExists("namelockid")) {
		ServerCommand("namelockid %d %d", GetClientUserId(client), locked ? 0 : 1);
		return true;
	}
	return false;
}

/**
 * Gets a possibly multi-byte character at the specified position in a string as an integer.
 * 
 * If the character is an ASCII character, the integer representation of the character will be
 * returned, and the number of characters read will be 1.
 * 
 * This stock is made specifically for the purpose of handling strings with multi-byte
 * characters.
 */
stock int GetMultiByteCharacterAt(const char[] text, int pos, int &read = 0) {
	int nCharBytes, charOutput;
	
	nCharBytes = GetCharBytes(text[pos]);
	
	for (int b = 0; b < nCharBytes; b++) {
		charOutput = charOutput << 8;
		
		// read in the current byte
		charOutput |= text[pos + b] & 0xFF;
		
		read++;
	}
	
	return charOutput;
}

/**
 * Appends a multi-byte character to a string.
 */
stock void StrCatMultiByteChar(char[] buffer, int maxlen, int character, int size) {
	char[] appended = new char[size];
	
	MultiByteCharacterToCharArray(appended, size, character);
	
	StrCat(buffer, maxlen, appended);
}

/**
 * Converts a multi-byte character to a char array.
 */
stock void MultiByteCharacterToCharArray(char[] buffer, int charsize, int character) {
	for (int i = 0; i < charsize; i++) {
		buffer[i] = (character >> ((charsize - i - 1) * 8 )) & 0xFF;
	}
}