lib/rotjs-library/src/text.js

/**
 * @namespace
 * Contains text tokenization and breaking routines
 */
ROT.Text = {
	RE_COLORS: /%([bc]){([^}]*)}/g,

	/* token types */
	TYPE_TEXT:		0,
	TYPE_NEWLINE:	1,
	TYPE_FG:		2,
	TYPE_BG:		3,

	/**
	 * Measure size of a resulting text block
	 */
	measure: function(str, maxWidth) {
		var result = {width:0, height:1};
		var tokens = this.tokenize(str, maxWidth);
		var lineWidth = 0;

		for (var i=0;i<tokens.length;i++) {
			var token = tokens[i];
			switch (token.type) {
				case this.TYPE_TEXT:
					lineWidth += token.value.length;
				break;

				case this.TYPE_NEWLINE:
					result.height++;
					result.width = Math.max(result.width, lineWidth);
					lineWidth = 0;
				break;
			}
		}
		result.width = Math.max(result.width, lineWidth);

		return result;
	},

	/**
	 * Convert string to a series of a formatting commands
	 */
	tokenize: function(str, maxWidth) {
		var result = [];

		/* first tokenization pass - split texts and color formatting commands */
		var offset = 0;
		str.replace(this.RE_COLORS, function(match, type, name, index) {
			/* string before */
			var part = str.substring(offset, index);
			if (part.length) {
				result.push({
					type: ROT.Text.TYPE_TEXT,
					value: part
				});
			}

			/* color command */
			result.push({
				type: (type == "c" ? ROT.Text.TYPE_FG : ROT.Text.TYPE_BG),
				value: name.trim()
			});

			offset = index + match.length;
			return "";
		});

		/* last remaining part */
		var part = str.substring(offset);
		if (part.length) {
			result.push({
				type: ROT.Text.TYPE_TEXT,
				value: part
			});
		}

		return this._breakLines(result, maxWidth);
	},

	/* insert line breaks into first-pass tokenized data */
	_breakLines: function(tokens, maxWidth) {
		if (!maxWidth) { maxWidth = Infinity; };

		var i = 0;
		var lineLength = 0;
		var lastTokenWithSpace = -1;

		while (i < tokens.length) { /* take all text tokens, remove space, apply linebreaks */
			var token = tokens[i];
			if (token.type == ROT.Text.TYPE_NEWLINE) { /* reset */
				lineLength = 0; 
				lastTokenWithSpace = -1;
			}
			if (token.type != ROT.Text.TYPE_TEXT) { /* skip non-text tokens */
				i++;
				continue; 
			}

			/* remove spaces at the beginning of line */
			while (lineLength == 0 && token.value.charAt(0) == " ") { token.value = token.value.substring(1); }

			/* forced newline? insert two new tokens after this one */
			var index = token.value.indexOf("\n");
			if (index != -1) { 
				token.value = this._breakInsideToken(tokens, i, index, true); 

				/* if there are spaces at the end, we must remove them (we do not want the line too long) */
				var arr = token.value.split("");
				while (arr[arr.length-1] == " ") { arr.pop(); }
				token.value = arr.join("");
			}

			/* token degenerated? */
			if (!token.value.length) {
				tokens.splice(i, 1);
				continue;
			}

			if (lineLength + token.value.length > maxWidth) { /* line too long, find a suitable breaking spot */

				/* is it possible to break within this token? */
				var index = -1;
				while (1) {
					var nextIndex = token.value.indexOf(" ", index+1);
					if (nextIndex == -1) { break; }
					if (lineLength + nextIndex > maxWidth) { break; }
					index = nextIndex;
				}

				if (index != -1) { /* break at space within this one */
					token.value = this._breakInsideToken(tokens, i, index, true);
				} else if (lastTokenWithSpace != -1) { /* is there a previous token where a break can occur? */
					var token = tokens[lastTokenWithSpace];
					var breakIndex = token.value.lastIndexOf(" ");
					token.value = this._breakInsideToken(tokens, lastTokenWithSpace, breakIndex, true);
					i = lastTokenWithSpace;
				} else { /* force break in this token */
					token.value = this._breakInsideToken(tokens, i, maxWidth-lineLength, false);
				}

			} else { /* line not long, continue */
				lineLength += token.value.length;
				if (token.value.indexOf(" ") != -1) { lastTokenWithSpace = i; }
			}
			
			i++; /* advance to next token */
		}


		tokens.push({type: ROT.Text.TYPE_NEWLINE}); /* insert fake newline to fix the last text line */

		/* remove trailing space from text tokens before newlines */
		var lastTextToken = null;
		for (var i=0;i<tokens.length;i++) {
			var token = tokens[i];
			switch (token.type) {
				case ROT.Text.TYPE_TEXT: lastTextToken = token; break;
				case ROT.Text.TYPE_NEWLINE: 
					if (lastTextToken) { /* remove trailing space */
						var arr = lastTextToken.value.split("");
						while (arr[arr.length-1] == " ") { arr.pop(); }
						lastTextToken.value = arr.join("");
					}
					lastTextToken = null;
				break;
			}
		}

		tokens.pop(); /* remove fake token */

		return tokens;
	},

	/**
	 * Create new tokens and insert them into the stream
	 * @param {object[]} tokens
	 * @param {int} tokenIndex Token being processed
	 * @param {int} breakIndex Index within current token's value
	 * @param {bool} removeBreakChar Do we want to remove the breaking character?
	 * @returns {string} remaining unbroken token value
	 */
	_breakInsideToken: function(tokens, tokenIndex, breakIndex, removeBreakChar) {
		var newBreakToken = {
			type: ROT.Text.TYPE_NEWLINE
		}
		var newTextToken = {
			type: ROT.Text.TYPE_TEXT,
			value: tokens[tokenIndex].value.substring(breakIndex + (removeBreakChar ? 1 : 0))
		}
		tokens.splice(tokenIndex+1, 0, newBreakToken, newTextToken);
		return tokens[tokenIndex].value.substring(0, breakIndex);
	}
}