Merge remote-tracking branch 'upstream/master' into TGUIs_Nexties

code/__HELPERS/files.dm

@@ -20,7 +20,7 @@
 				continue
 		path += choice
 
-		if(copytext(path,-1,0) != "/")		//didn't choose a directory, no need to iterate again
+		if(copytext_char(path, -1) != "/")		//didn't choose a directory, no need to iterate again
 			break
 	var/extensions
 	for(var/i in valid_extensions)

code/__HELPERS/icons.dm

@@ -985,7 +985,7 @@ world
 	var/icon/atom_icon = new(A.icon, A.icon_state)
 
 	if(!letter)
-		letter = copytext(A.name, 1, 2)
+		letter = A.name[1]
 		if(uppercase == 1)
 			letter = uppertext(letter)
 		else if(uppercase == -1)
@@ -1113,7 +1113,7 @@ GLOBAL_LIST_INIT(freon_color_matrix, list("#2E5E69", "#60A2A8", "#A1AFB1", rgb(0
 	WRITE_FILE(GLOB.iconCache[iconKey], icon)
 	var/iconData = GLOB.iconCache.ExportText(iconKey)
 	var/list/partial = splittext(iconData, "{")
-	return replacetext(copytext(partial[2], 3, -5), "\n", "")
+	return replacetext(copytext_char(partial[2], 3, -5), "\n", "")
 
 /proc/icon2html(thing, target, icon_state, dir, frame = 1, moving = FALSE)
 	if (!thing)

code/__HELPERS/pronouns.dm

@@ -28,10 +28,10 @@
 	. = "does"
 
 /datum/proc/p_theyve(capitalized, temp_gender)
-	. = p_they(capitalized, temp_gender) + "'" + copytext(p_have(temp_gender), 3)
+	. = p_they(capitalized, temp_gender) + "'" + copytext_char(p_have(temp_gender), 3)
 
 /datum/proc/p_theyre(capitalized, temp_gender)
-	. = p_they(capitalized, temp_gender) + "'" + copytext(p_are(temp_gender), 2)
+	. = p_they(capitalized, temp_gender) + "'" + copytext_char(p_are(temp_gender), 2)
 
 /datum/proc/p_s(temp_gender) //is this a descriptive proc name, or what?
 	. = "s"

code/__HELPERS/sanitize_values.dm

@@ -43,24 +43,24 @@
 	if(!istext(color))
 		color = ""
 
-	var/start = 1 + (text2ascii(color,1)==35)
+	var/start = 1 + (text2ascii(color, 1) == 35)
 	var/len = length(color)
-	var/step_size = 1 + ((len+1)-start != desired_format)
+	var/char = ""
 
 	. = ""
-	for(var/i=start, i<=len, i+=step_size)
-		var/ascii = text2ascii(color,i)
-		switch(ascii)
-			if(48 to 57)
-				. += ascii2text(ascii)		//numbers 0 to 9
-			if(97 to 102)
-				. += ascii2text(ascii)		//letters a to f
-			if(65 to 70)
-				. += ascii2text(ascii+32)	//letters A to F - translates to lowercase
+	for(var/i = start, i <= len, i += length(char))
+		char = color[i]
+		switch(text2ascii(char))
+			if(48 to 57)	//numbers 0 to 9
+				. += char
+			if(97 to 102)	//letters a to f
+				. += char
+			if(65 to 70)	//letters A to F - translates to lowercase
+				. += lowertext(char)
 			else
 				break
 
-	if(length(.) != desired_format)
+	if(length_char(.) != desired_format)
 		if(default)
 			return default
 		return crunch + repeat_string(desired_format, "0")
@@ -68,7 +68,9 @@
 	return crunch + .
 
 /proc/sanitize_ooccolor(color)
-	var/list/HSL = rgb2hsl(hex2num(copytext(color,2,4)),hex2num(copytext(color,4,6)),hex2num(copytext(color,6,8)))
+	if(length(color) != length_char(color))
+		CRASH("Invalid characters in color '[color]'")
+	var/list/HSL = rgb2hsl(hex2num(copytext(color, 2, 4)), hex2num(copytext(color, 4, 6)), hex2num(copytext(color, 6, 8)))
 	HSL[3] = min(HSL[3],0.4)
 	var/list/RGB = hsl2rgb(arglist(HSL))
 	return "#[num2hex(RGB[1],2)][num2hex(RGB[2],2)][num2hex(RGB[3],2)]"

code/__HELPERS/shell.dm

@@ -52,6 +52,6 @@
 		if(bad_chars)
 			bad_match = url_encode(bad_chars_regex.match)
 			scrubbed_url += bad_match
-			last_good = bad_chars + length(bad_match)
+			last_good = bad_chars + length(bad_chars_regex.match)
 	while(bad_chars)
 	. = scrubbed_url

code/__HELPERS/text.dm

@@ -40,8 +40,8 @@
 	for(var/char in repl_chars)
 		var/index = findtext(t, char)
 		while(index)
-			t = copytext(t, 1, index) + repl_chars[char] + copytext(t, index+1)
-			index = findtext(t, char, index+1)
+			t = copytext(t, 1, index) + repl_chars[char] + copytext(t, index + length(char))
+			index = findtext(t, char, index + length(char))
 	return t
 
 /proc/sanitize_filename(t)
@@ -63,22 +63,28 @@
 
 
 //Returns null if there is any bad text in the string
-/proc/reject_bad_text(text, max_length=512)
-	if(length(text) > max_length)
-		return			//message too long
-	var/non_whitespace = 0
-	for(var/i=1, i<=length(text), i++)
-		switch(text2ascii(text,i))
-			if(62,60,92,47)
-				return			//rejects the text if it contains these bad characters: <, >, \ or /
-			if(127 to 255)
-				return			//rejects weird letters like �
+/proc/reject_bad_text(text, max_length = 512, ascii_only = TRUE)
+	var/char_count = 0
+	var/non_whitespace = FALSE
+	var/lenbytes = length(text)
+	var/char = ""
+	for(var/i = 1, i <= lenbytes, i += length(char))
+		char = text[i]
+		char_count++
+		if(char_count > max_length)
+			return
+		switch(text2ascii(char))
+			if(62,60,92,47) // <, >, \, /
+				return
 			if(0 to 31)
-				return			//more weird stuff
+				return
 			if(32)
 				continue		//whitespace
+			if(127 to INFINITY)
+				if(ascii_only)
+					return
 			else
-				non_whitespace = 1
+				non_whitespace = TRUE
 	if(non_whitespace)
 		return text		//only accepts the text if it has some non-spaces
 
@@ -101,73 +107,84 @@
 	else
 		return trim(html_encode(name), max_length)
 
+#define NO_CHARS_DETECTED 0
+#define SPACES_DETECTED 1
+#define SYMBOLS_DETECTED 2
+#define NUMBERS_DETECTED 3
+#define LETTERS_DETECTED 4
+
 //Filters out undesirable characters from names
-/proc/reject_bad_name(t_in, allow_numbers=0, max_length=MAX_NAME_LEN)
-	if(!t_in || length(t_in) > max_length)
-		return //Rejects the input if it is null or if it is longer then the max length allowed
+/proc/reject_bad_name(t_in, allow_numbers = FALSE, max_length = MAX_NAME_LEN, ascii_only = TRUE)
+	if(!t_in)
+		return //Rejects the input if it is null
 
-	var/number_of_alphanumeric	= 0
-	var/last_char_group			= 0
+	var/number_of_alphanumeric = 0
+	var/last_char_group = NO_CHARS_DETECTED
 	var/t_out = ""
+	var/t_len = length(t_in)
+	var/charcount = 0
+	var/char = ""
 
-	for(var/i=1, i<=length(t_in), i++)
-		var/ascii_char = text2ascii(t_in,i)
-		switch(ascii_char)
+
+	for(var/i = 1, i <= t_len, i += length(char))
+		char = t_in[i]
+
+		switch(text2ascii(char))
 			// A  .. Z
 			if(65 to 90)			//Uppercase Letters
-				t_out += ascii2text(ascii_char)
 				number_of_alphanumeric++
-				last_char_group = 4
+				last_char_group = LETTERS_DETECTED
 
 			// a  .. z
 			if(97 to 122)			//Lowercase Letters
-				if(last_char_group<2)
-					t_out += ascii2text(ascii_char-32)	//Force uppercase first character
-				else
-					t_out += ascii2text(ascii_char)
+				if(last_char_group == NO_CHARS_DETECTED || last_char_group == SPACES_DETECTED || last_char_group == SYMBOLS_DETECTED) //start of a word
+					char = uppertext(char)
 				number_of_alphanumeric++
-				last_char_group = 4
+				last_char_group = LETTERS_DETECTED
 
 			// 0  .. 9
 			if(48 to 57)			//Numbers
-				if(!last_char_group)
-					continue	//suppress at start of string
-				if(!allow_numbers)
+				if(last_char_group == NO_CHARS_DETECTED || !allow_numbers) //suppress at start of string
 					continue
-				t_out += ascii2text(ascii_char)
 				number_of_alphanumeric++
-				last_char_group = 3
+				last_char_group = NUMBERS_DETECTED
 
 			// '  -  .
 			if(39,45,46)			//Common name punctuation
-				if(!last_char_group)
+				if(last_char_group == NO_CHARS_DETECTED)
 					continue
-				t_out += ascii2text(ascii_char)
-				last_char_group = 2
+				last_char_group = SYMBOLS_DETECTED
 
 			// ~   |   @  :  #  $  %  &  *  +
 			if(126,124,64,58,35,36,37,38,42,43)			//Other symbols that we'll allow (mainly for AI)
-				if(!last_char_group)
-					continue	//suppress at start of string
-				if(!allow_numbers)
+				if(last_char_group == NO_CHARS_DETECTED || !allow_numbers) //suppress at start of string
 					continue
-				t_out += ascii2text(ascii_char)
-				last_char_group = 2
+				last_char_group = SYMBOLS_DETECTED
 
 			//Space
 			if(32)
-				if(last_char_group <= 1)
-					continue	//suppress double-spaces and spaces at start of string
-				t_out += ascii2text(ascii_char)
-				last_char_group = 1
+				if(last_char_group == NO_CHARS_DETECTED || last_char_group == SPACES_DETECTED) //suppress double-spaces and spaces at start of string
+					continue
+				last_char_group = SPACES_DETECTED
+
+			if(127 to INFINITY)
+				if(ascii_only)
+					continue
+				last_char_group = SYMBOLS_DETECTED //for now, we'll treat all non-ascii characters like symbols even though most are letters
+
 			else
-				return
+				continue
+
+		t_out += char
+		charcount++
+		if(charcount >= max_length)
+			break
 
 	if(number_of_alphanumeric < 2)
 		return		//protects against tiny names like "A" and also names like "' ' ' ' ' ' ' '"
 
-	if(last_char_group == 1)
-		t_out = copytext(t_out,1,length(t_out))	//removes the last character (in this case a space)
+	if(last_char_group == SPACES_DETECTED)
+		t_out = copytext_char(t_out, 1, -1) //removes the last character (in this case a space)
 
 	for(var/bad_name in list("space","floor","wall","r-wall","monkey","unknown","inactive ai"))	//prevents these common metagamey names
 		if(cmptext(t_out,bad_name))
@@ -175,6 +192,11 @@
 
 	return t_out
 
+#undef NO_CHARS_DETECTED
+#undef SPACES_DETECTED
+#undef NUMBERS_DETECTED
+#undef LETTERS_DETECTED
+
 //html_encode helper proc that returns the smallest non null of two numbers
 //or 0 if they're both null (needed because of findtext returning 0 when a value is not present)
 /proc/non_zero_min(a, b)
@@ -184,39 +206,6 @@
 		return a
 	return (a < b ? a : b)
 
-/*
- * Text searches
- */
-
-//Checks the beginning of a string for a specified sub-string
-//Returns the position of the substring or 0 if it was not found
-/proc/dd_hasprefix(text, prefix)
-	var/start = 1
-	var/end = length(prefix) + 1
-	return findtext(text, prefix, start, end)
-
-//Checks the beginning of a string for a specified sub-string. This proc is case sensitive
-//Returns the position of the substring or 0 if it was not found
-/proc/dd_hasprefix_case(text, prefix)
-	var/start = 1
-	var/end = length(prefix) + 1
-	return findtextEx(text, prefix, start, end)
-
-//Checks the end of a string for a specified substring.
-//Returns the position of the substring or 0 if it was not found
-/proc/dd_hassuffix(text, suffix)
-	var/start = length(text) - length(suffix)
-	if(start)
-		return findtext(text, suffix, start, null)
-	return
-
-//Checks the end of a string for a specified substring. This proc is case sensitive
-//Returns the position of the substring or 0 if it was not found
-/proc/dd_hassuffix_case(text, suffix)
-	var/start = length(text) - length(suffix)
-	if(start)
-		return findtextEx(text, suffix, start, null)
-
 //Checks if any of a given list of needles is in the haystack
 /proc/text_in_list(haystack, list/needle_list, start=1, end=0)
 	for(var/needle in needle_list)
@@ -231,23 +220,19 @@
 			return 1
 	return 0
 
-//Adds 'u' number of zeros ahead of the text 't'
-/proc/add_zero(t, u)
-	while (length(t) < u)
-		t = "0[t]"
-	return t
+//Adds 'char' ahead of 'text' until there are 'count' characters total
+/proc/add_leading(text, count, char = " ")
+	text = "[text]"
+	var/charcount = count - length_char(text)
+	var/list/chars_to_add[max(charcount + 1, 0)]
+	return jointext(chars_to_add, char) + text
 
-//Adds 'u' number of spaces ahead of the text 't'
-/proc/add_lspace(t, u)
-	while(length(t) < u)
-		t = " [t]"
-	return t
-
-//Adds 'u' number of spaces behind the text 't'
-/proc/add_tspace(t, u)
-	while(length(t) < u)
-		t = "[t] "
-	return t
+//Adds 'char' behind 'text' until there are 'count' characters total
+/proc/add_trailing(text, count, char = " ")
+	text = "[text]"
+	var/charcount = count - length_char(text)
+	var/list/chars_to_add[max(charcount + 1, 0)]
+	return text + jointext(chars_to_add, char)
 
 //Returns a string with reserved characters and spaces before the first letter removed
 /proc/trim_left(text)
@@ -267,57 +252,41 @@
 //Returns a string with reserved characters and spaces before the first word and after the last word removed.
 /proc/trim(text, max_length)
 	if(max_length)
-		text = copytext(text, 1, max_length)
+		text = copytext_char(text, 1, max_length)
 	return trim_left(trim_right(text))
 
 //Returns a string with the first element of the string capitalized.
 /proc/capitalize(t as text)
-	return uppertext(copytext(t, 1, 2)) + copytext(t, 2)
-
-//Centers text by adding spaces to either side of the string.
-/proc/dd_centertext(message, length)
-	var/new_message = message
-	var/size = length(message)
-	var/delta = length - size
-	if(size == length)
-		return new_message
-	if(size > length)
-		return copytext(new_message, 1, length + 1)
-	if(delta == 1)
-		return new_message + " "
-	if(delta % 2)
-		new_message = " " + new_message
-		delta--
-	var/spaces = add_lspace("",delta/2-1)
-	return spaces + new_message + spaces
-
-//Limits the length of the text. Note: MAX_MESSAGE_LEN and MAX_NAME_LEN are widely used for this purpose
-/proc/dd_limittext(message, length)
-	var/size = length(message)
-	if(size <= length)
-		return message
-	return copytext(message, 1, length + 1)
-
+	. = t
+	if(t)
+		. = t[1]
+		return uppertext(.) + copytext(t, 1 + length(.))
 
 /proc/stringmerge(text,compare,replace = "*")
 //This proc fills in all spaces with the "replace" var (* by default) with whatever
 //is in the other string at the same spot (assuming it is not a replace char).
 //This is used for fingerprints
 	var/newtext = text
-	if(length(text) != length(compare))
-		return 0
-	for(var/i = 1, i < length(text), i++)
-		var/a = copytext(text,i,i+1)
-		var/b = copytext(compare,i,i+1)
+	var/text_it = 1 //iterators
+	var/comp_it = 1
+	var/newtext_it = 1
+	var/text_length = length(text)
+	var/comp_length = length(compare)
+	while(comp_it <= comp_length && text_it <= text_length)
+		var/a = text[text_it]
+		var/b = compare[comp_it]
 //if it isn't both the same letter, or if they are both the replacement character
 //(no way to know what it was supposed to be)
 		if(a != b)
 			if(a == replace) //if A is the replacement char
-				newtext = copytext(newtext,1,i) + b + copytext(newtext, i+1)
+				newtext = copytext(newtext, 1, newtext_it) + b + copytext(newtext, newtext_it + length(newtext[newtext_it]))
 			else if(b == replace) //if B is the replacement char
-				newtext = copytext(newtext,1,i) + a + copytext(newtext, i+1)
+				newtext = copytext(newtext, 1, newtext_it) + a + copytext(newtext, newtext_it + length(newtext[newtext_it]))
 			else //The lists disagree, Uh-oh!
 				return 0
+		text_it += length(a)
+		comp_it += length(b)
+		newtext_it += length(newtext[newtext_it])
 	return newtext
 
 /proc/stringpercent(text,character = "*")
@@ -326,16 +295,21 @@
 	if(!text || !character)
 		return 0
 	var/count = 0
-	for(var/i = 1, i <= length(text), i++)
-		var/a = copytext(text,i,i+1)
+	var/lentext = length(text)
+	var/a = ""
+	for(var/i = 1, i <= lentext, i += length(a))
+		a = text[i]
 		if(a == character)
 			count++
 	return count
 
 /proc/reverse_text(text = "")
 	var/new_text = ""
-	for(var/i = length(text); i > 0; i--)
-		new_text += copytext(text, i, i+1)
+	var/lentext = length(text)
+	var/letter = ""
+	for(var/i = 1, i <= lentext, i += length(letter))
+		letter = text[i]
+		new_text = letter + new_text
 	return new_text
 
 GLOBAL_LIST_INIT(zero_character_only, list("0"))
@@ -358,15 +332,6 @@ GLOBAL_LIST_INIT(binary, list("0","1"))
 /proc/random_color()
 	return random_string(6, GLOB.hex_characters)
 
-/proc/add_zero2(t, u)
-	var/temp1
-	while (length(t) < u)
-		t = "0[t]"
-	temp1 = t
-	if (length(t) > u)
-		temp1 = copytext(t,2,u+1)
-	return temp1
-
 //merges non-null characters (3rd argument) from "from" into "into". Returns result
 //e.g. into = "Hello World"
 //     from = "Seeya______"
@@ -379,41 +344,48 @@ GLOBAL_LIST_INIT(binary, list("0","1"))
 		into = ""
 	if(!istext(from))
 		from = ""
-	var/null_ascii = istext(null_char) ? text2ascii(null_char,1) : null_char
-
-	var/previous = 0
+	var/null_ascii = istext(null_char) ? text2ascii(null_char, 1) : null_char
+	var/copying_into = FALSE
+	var/char = ""
 	var/start = 1
-	var/end = length(into) + 1
-
-	for(var/i=1, i<end, i++)
-		var/ascii = text2ascii(from, i)
-		if(ascii == null_ascii)
-			if(previous != 1)
-				. += copytext(from, start, i)
-				start = i
-				previous = 1
+	var/end_from = length(from)
+	var/end_into = length(into)
+	var/into_it = 1
+	var/from_it = 1
+	while(from_it <= end_from && into_it <= end_into)
+		char = from[from_it]
+		if(text2ascii(char) == null_ascii)
+			if(!copying_into)
+				. += copytext(from, start, from_it)
+				start = into_it
+				copying_into = TRUE
 		else
-			if(previous != 0)
-				. += copytext(into, start, i)
-				start = i
-				previous = 0
+			if(copying_into)
+				. += copytext(into, start, into_it)
+				start = from_it
+				copying_into = FALSE
+		into_it += length(into[into_it])
+		from_it += length(char)
 
-	if(previous == 0)
-		. += copytext(from, start, end)
+	if(copying_into)
+		. += copytext(into, start)
 	else
-		. += copytext(into, start, end)
+		. += copytext(from, start, from_it)
+		if(into_it <= end_into)
+			. += copytext(into, into_it)
 
 //finds the first occurrence of one of the characters from needles argument inside haystack
 //it may appear this can be optimised, but it really can't. findtext() is so much faster than anything you can do in byondcode.
 //stupid byond :(
 /proc/findchar(haystack, needles, start=1, end=0)
-	var/temp
+	var/char = ""
 	var/len = length(needles)
-	for(var/i=1, i<=len, i++)
-		temp = findtextEx(haystack, ascii2text(text2ascii(needles,i)), start, end)	//Note: ascii2text(text2ascii) is faster than copytext()
-		if(temp)
-			end = temp
-	return end
+	for(var/i = 1, i <= len, i += length(char))
+		char = needles[i]
+		. = findtextEx(haystack, char, start, end)
+		if(.)
+			return
+	return 0
 
 /proc/parsemarkdown_basic_step1(t, limited=FALSE)
 	if(length(t) <= 0)
@@ -571,23 +543,28 @@ GLOBAL_LIST_INIT(binary, list("0","1"))
 
 	return t
 
-#define string2charlist(string) (splittext(string, regex("(.)")) - splittext(string, ""))
+/proc/text2charlist(text)
+	var/char = ""
+	var/lentext = length(text)
+	. = list()
+	for(var/i = 1, i <= lentext, i += length(char))
+		char = text[i]
+		. += char
 
 /proc/rot13(text = "")
-	var/list/textlist = string2charlist(text)
-	var/list/result = list()
-	for(var/c in textlist)
-		var/ca = text2ascii(c)
-		if(ca >= text2ascii("a") && ca <= text2ascii("m"))
-			ca += 13
-		else if(ca >= text2ascii("n") && ca <= text2ascii("z"))
-			ca -= 13
-		else if(ca >= text2ascii("A") && ca <= text2ascii("M"))
-			ca += 13
-		else if(ca >= text2ascii("N") && ca <= text2ascii("Z"))
-			ca -= 13
-		result += ascii2text(ca)
-	return jointext(result, "")
+	var/lentext = length(text)
+	var/char = ""
+	var/ascii = 0
+	. = ""
+	for(var/i = 1, i <= lentext, i += length(char))
+		char = text[i]
+		ascii = text2ascii(char)
+		switch(ascii)
+			if(65 to 77, 97 to 109) //A to M, a to m
+				ascii += 13
+			if(78 to 90, 110 to 122) //N to Z, n to z
+				ascii -= 13
+		. += ascii2text(ascii)
 
 //Takes a list of values, sanitizes it down for readability and character count,
 //then exports it as a json file at data/npc_saves/[filename].json.
@@ -599,7 +576,8 @@ GLOBAL_LIST_INIT(binary, list("0","1"))
 		return
 
 	//Regular expressions are, as usual, absolute magic
-	var/regex/all_invalid_symbols = new("\[^ -~]+")
+	//Any characters outside of 32 (space) to 126 (~) because treating things you don't understand as "magic" is really stupid
+	var/regex/all_invalid_symbols = new(@"[^ -~]{1}")
 
 	var/list/accepted = list()
 	for(var/string in proposed)
@@ -607,34 +585,44 @@ GLOBAL_LIST_INIT(binary, list("0","1"))
 			continue
 		var/buffer = ""
 		var/early_culling = TRUE
-		for(var/pos = 1, pos <= length(string), pos++)
-			var/let = copytext(string, pos, (pos + 1) % length(string))
-			if(early_culling && !findtext(let,GLOB.is_alphanumeric))
+		var/lentext = length(string)
+		var/let = ""
+
+		for(var/pos = 1, pos <= lentext, pos += length(let))
+			let = string[pos]
+			if(!findtext(let, GLOB.is_alphanumeric))
 				continue
 			early_culling = FALSE
-			buffer += let
-		if(!findtext(buffer,GLOB.is_alphanumeric))
+			buffer = copytext(string, pos)
+			break
+		if(early_culling) //Never found any letters! Bail!
 			continue
+
 		var/punctbuffer = ""
-		var/cutoff = length(buffer)
-		for(var/pos = length(buffer), pos >= 0, pos--)
-			var/let = copytext(buffer, pos, (pos + 1) % length(buffer))
-			if(findtext(let,GLOB.is_alphanumeric))
+		var/cutoff = 0
+		lentext = length_char(buffer)
+		for(var/pos = 1, pos <= lentext, pos++)
+			let = copytext_char(buffer, -pos, -pos + 1)
+			if(!findtext(let, GLOB.is_punctuation)) //This won't handle things like Nyaaaa!~ but that's fine
 				break
-			if(findtext(let,GLOB.is_punctuation))
-				punctbuffer = let + punctbuffer //Note this isn't the same thing as using +=
-				cutoff = pos
+			punctbuffer += let
+			cutoff += length(let)
 		if(punctbuffer) //We clip down excessive punctuation to get the letter count lower and reduce repeats. It's not perfect but it helps.
 			var/exclaim = FALSE
 			var/question = FALSE
 			var/periods = 0
-			for(var/pos = length(punctbuffer), pos >= 0, pos--)
-				var/punct = copytext(punctbuffer, pos, (pos + 1) % length(punctbuffer))
-				if(!exclaim && findtext(punct,"!"))
+			lentext = length(punctbuffer)
+			for(var/pos = 1, pos <= lentext, pos += length(let))
+				let = punctbuffer[pos]
+				if(!exclaim && findtext(let, "!"))
 					exclaim = TRUE
-				if(!question && findtext(punct,"?"))
+					if(question)
+						break
+				if(!question && findtext(let, "?"))
 					question = TRUE
-				if(!exclaim && !question && findtext(punct,"."))
+					if(exclaim)
+						break
+				if(!exclaim && !question && findtext(let, ".")) //? and ! take priority over periods
 					periods += 1
 			if(exclaim)
 				if(question)
@@ -643,15 +631,13 @@ GLOBAL_LIST_INIT(binary, list("0","1"))
 					punctbuffer = "!"
 			else if(question)
 				punctbuffer = "?"
-			else if(periods)
-				if(periods > 1)
-					punctbuffer = "..."
-				else
-					punctbuffer = "" //Grammer nazis be damned
-			buffer = copytext(buffer, 1, cutoff) + punctbuffer
-		if(!findtext(buffer,GLOB.is_alphanumeric))
-			continue
-		if(!buffer || length(buffer) > 280 || length(buffer) <= cullshort || buffer in accepted)
+			else if(periods > 1)
+				punctbuffer = "..."
+			else
+				punctbuffer = "" //Grammer nazis be damned
+			buffer = copytext(buffer, 1, -cutoff) + punctbuffer
+		lentext = length_char(buffer)
+		if(!buffer || lentext > 280 || lentext <= cullshort || (buffer in accepted))
 			continue
 
 		accepted += buffer
@@ -688,7 +674,7 @@ GLOBAL_LIST_INIT(binary, list("0","1"))
 
 	var/leng = length(string)
 
-	var/next_space = findtext(string, " ", next_backslash + 1)
+	var/next_space = findtext(string, " ", next_backslash + length(string[next_backslash]))
 	if(!next_space)
 		next_space = leng - next_backslash
 
@@ -696,8 +682,8 @@ GLOBAL_LIST_INIT(binary, list("0","1"))
 		return string
 
 	var/base = next_backslash == 1 ? "" : copytext(string, 1, next_backslash)
-	var/macro = lowertext(copytext(string, next_backslash + 1, next_space))
-	var/rest = next_backslash > leng ? "" : copytext(string, next_space + 1)
+	var/macro = lowertext(copytext(string, next_backslash + length(string[next_space]), next_space))
+	var/rest = next_backslash > leng ? "" : copytext(string, next_space + length(string[next_space]))
 
 	//See https://secure.byond.com/docs/ref/info.html#/DM/text/macros
 	switch(macro)
@@ -769,28 +755,24 @@ GLOBAL_LIST_INIT(binary, list("0","1"))
 			return "[number]\th"
 
 /proc/unintelligize(message)
-	var/prefix=copytext(message,1,2)
+	var/regex/word_boundaries = regex(@"\b[\S]+\b", "g")
+	var/prefix = message[1]
 	if(prefix == ";")
-		message = copytext(message,2)
-	else if(prefix in list(":","#"))
-		prefix += copytext(message,2,3)
-		message = copytext(message,3)
+		message = copytext(message, 1 + length(prefix))
+	else if(prefix in list(":", "#"))
+		prefix += message[1 + length(prefix)]
+		message = copytext(message, length(prefix))
 	else
-		prefix=""
+		prefix = ""
 
-	var/list/words = splittext(message," ")
 	var/list/rearranged = list()
-	for(var/i=1;i<=words.len;i++)
-		var/cword = pick(words)
-		words.Remove(cword)
-		var/suffix = copytext(cword,length(cword)-1,length(cword))
-		while(length(cword)>0 && suffix in list(".",",",";","!",":","?"))
-			cword  = copytext(cword,1              ,length(cword)-1)
-			suffix = copytext(cword,length(cword)-1,length(cword)  )
+	while(word_boundaries.Find(message))
+		var/cword = word_boundaries.match
 		if(length(cword))
 			rearranged += cword
-	message = "[prefix][jointext(rearranged," ")]"
-	. = message
+	shuffle_inplace(rearranged)
+	return "[prefix][jointext(rearranged, " ")]"
+
 
 #define is_alpha(X) ((text2ascii(X) <= 122) && (text2ascii(X) >= 97))
 #define is_digit(X) ((length(X) == 1) && (length(text2num(X)) == 1))

code/__HELPERS/text_vr.dm

@@ -8,14 +8,15 @@
 			index = findtext(t, char)
 	return t
 
-proc/TextPreview(var/string,var/len=40)
-	if(length(string) <= len)
-		if(!length(string))
+/proc/TextPreview(string, len = 40)
+	var/char_len = length_char(string)
+	if(char_len <= len)
+		if(char_len)
 			return "\[...\]"
 		else
 			return string
 	else
-		return "[copytext(string, 1, 37)]..."
+		return "[copytext_char(string, 1, 37)]..."
 
 GLOBAL_LIST_EMPTY(mentorlog)
 GLOBAL_PROTECT(mentorlog)

code/__HELPERS/type2type.dm

@@ -542,17 +542,17 @@
 
 //assumes format #RRGGBB #rrggbb
 /proc/color_hex2num(A)
-	if(!A)
+	if(!A || length(A) != length_char(A))
 		return 0
-	var/R = hex2num(copytext(A,2,4))
-	var/G = hex2num(copytext(A,4,6))
-	var/B = hex2num(copytext(A,6,0))
+	var/R = hex2num(copytext(A, 2, 4))
+	var/G = hex2num(copytext(A, 4, 6))
+	var/B = hex2num(copytext(A, 6, 0))
 	return R+G+B
 
 //word of warning: using a matrix like this as a color value will simplify it back to a string after being set
 /proc/color_hex2color_matrix(string)
 	var/length = length(string)
-	if(length != 7 && length != 9)
+	if((length != 7 && length != 9) || length != length_char(string))
 		return color_matrix_identity()
 	var/r = hex2num(copytext(string, 2, 4))/255
 	var/g = hex2num(copytext(string, 4, 6))/255

code/__HELPERS/unsorted.dm

@@ -6,25 +6,18 @@
 
 //Inverts the colour of an HTML string
 /proc/invertHTML(HTMLstring)
-
-	if (!( istext(HTMLstring) ))
+	if(!istext(HTMLstring))
 		CRASH("Given non-text argument!")
 		return
-	else
-		if (length(HTMLstring) != 7)
-			CRASH("Given non-HTML argument!")
-			return
+	else if(length(HTMLstring) != 7)
+		CRASH("Given non-HTML argument!")
+		return
+	else if(length_char(HTMLstring) != 7)
+		CRASH("Given non-hex symbols in argument!")
 	var/textr = copytext(HTMLstring, 2, 4)
 	var/textg = copytext(HTMLstring, 4, 6)
 	var/textb = copytext(HTMLstring, 6, 8)
-	var/r = hex2num(textr)
-	var/g = hex2num(textg)
-	var/b = hex2num(textb)
-	textr = num2hex(255 - r, 2)
-	textg = num2hex(255 - g, 2)
-	textb = num2hex(255 - b, 2)
-	return text("#[][][]", textr, textg, textb)
-	return
+	return rgb(255 - hex2num(textr), 255 - hex2num(textg), 255 - hex2num(textb))
 
 /proc/Get_Angle(atom/movable/start,atom/movable/end)//For beams.
 	if(!start || !end)
@@ -184,15 +177,15 @@ Turf and target are separate in case you want to teleport some distance from a t
 //Returns whether or not a player is a guest using their ckey as an input
 /proc/IsGuestKey(key)
 	if (findtext(key, "Guest-", 1, 7) != 1) //was findtextEx
-		return 0
+		return FALSE
 
 	var/i, ch, len = length(key)
 
-	for (i = 7, i <= len, ++i)
+	for (i = 7, i <= len, ++i) //we know the first 6 chars are Guest-
 		ch = text2ascii(key, i)
-		if (ch < 48 || ch > 57)
-			return 0
-	return 1
+		if (ch < 48 || ch > 57) //0-9
+			return FALSE
+	return TRUE
 
 //Generalised helper proc for letting mobs rename themselves. Used to be clname() and ainame()
 /mob/proc/apply_pref_name(role, client/C)
@@ -1314,7 +1307,7 @@ GLOBAL_DATUM_INIT(dview_mob, /mob/dview, new)
 /proc/GUID()
 	var/const/GUID_VERSION = "b"
 	var/const/GUID_VARIANT = "d"
-	var/node_id = copytext(md5("[rand()*rand(1,9999999)][world.name][world.hub][world.hub_password][world.internet_address][world.address][world.contents.len][world.status][world.port][rand()*rand(1,9999999)]"), 1, 13)
+	var/node_id = copytext_char(md5("[rand()*rand(1,9999999)][world.name][world.hub][world.hub_password][world.internet_address][world.address][world.contents.len][world.status][world.port][rand()*rand(1,9999999)]"), 1, 13)
 
 	var/time_high = "[num2hex(text2num(time2text(world.realtime,"YYYY")), 2)][num2hex(world.realtime, 6)]"