From e3462e56068a991aa8ea715883d41c0ad783fed3 Mon Sep 17 00:00:00 2001
From: Jeremy Ruston <jeremy@osmosoft.com>
Date: Sat, 15 Dec 2012 11:39:58 +0000
Subject: [PATCH] Added more parse rules

---
 .../wikiparser/rules/block/classblock.js      |  49 ++++++
 .../wikiparser/rules/block/codeblock.js       |  58 +++++++
 .../parsers/wikiparser/rules/block/table.js   | 161 ++++++++++++++++++
 .../parsers/wikiparser/rules/run/classrun.js  |  51 ++++++
 .../parsers/wikiparser/rules/run/coderun.js   |  58 +++++++
 .../parsers/wikiparser/rules/run/comment.js   |  34 ++++
 .../parsers/wikiparser/rules/run/dash.js      |  41 +++++
 .../parsers/wikiparser/rules/run/emphasis.js  |  78 +++++++++
 8 files changed, 530 insertions(+)
 create mode 100644 core/modules/parsers/wikiparser/rules/block/classblock.js
 create mode 100644 core/modules/parsers/wikiparser/rules/block/codeblock.js
 create mode 100644 core/modules/parsers/wikiparser/rules/block/table.js
 create mode 100644 core/modules/parsers/wikiparser/rules/run/classrun.js
 create mode 100644 core/modules/parsers/wikiparser/rules/run/coderun.js
 create mode 100644 core/modules/parsers/wikiparser/rules/run/comment.js
 create mode 100644 core/modules/parsers/wikiparser/rules/run/dash.js
 create mode 100644 core/modules/parsers/wikiparser/rules/run/emphasis.js

diff --git a/core/modules/parsers/wikiparser/rules/block/classblock.js b/core/modules/parsers/wikiparser/rules/block/classblock.js
new file mode 100644
index 000000000..4b0df5c4b
--- /dev/null
+++ b/core/modules/parsers/wikiparser/rules/block/classblock.js
@@ -0,0 +1,49 @@
+/*\
+title: $:/core/modules/parsers/wikiparser/rules/block/classblock.js
+type: application/javascript
+module-type: wikiblockrule
+
+Wiki text block rule for assigning classes to paragraphs and other blocks. For example:
+
+{{{
+{{myClass{
+This paragraph will have the CSS class `myClass`.
+
+* The `<ul>` around this list will also have the class `myClass`
+* List item 2
+
+ }}}
+}}}
+
+Note that the opening and closing braces both must be immediately followed by a newline.
+
+\*/
+(function(){
+
+/*jslint node: true, browser: true */
+/*global $tw: false */
+"use strict";
+
+exports.name = "classblock";
+
+exports.init = function(parser) {
+	this.parser = parser;
+	// Regexp to match
+	this.matchRegExp = /\{\{([^\{\r\n]+)\{\r?\n/mg;
+};
+
+exports.parse = function() {
+	var reEndString = "(\\}\\}\\}$(?:\\r?\\n)?)";
+	// Get the class
+	var classString = this.match[1];
+	// Move past the match
+	this.parser.pos = this.matchRegExp.lastIndex;
+	// Parse the body
+	var tree = this.parser.parseBlocks(reEndString);
+	for(var t=0; t<tree.length; t++) {
+		$tw.utils.addClassToParseTreeNode(tree[t],classString);
+	}
+	return tree;
+};
+
+})();
diff --git a/core/modules/parsers/wikiparser/rules/block/codeblock.js b/core/modules/parsers/wikiparser/rules/block/codeblock.js
new file mode 100644
index 000000000..074884159
--- /dev/null
+++ b/core/modules/parsers/wikiparser/rules/block/codeblock.js
@@ -0,0 +1,58 @@
+/*\
+title: $:/core/modules/parsers/wikiparser/rules/block/codeblock.js
+type: application/javascript
+module-type: wikiblockrule
+
+Wiki text rule for code blocks. For example:
+
+{{{
+	{{{
+	This text will not be //wikified//
+	}}}
+}}}
+
+Note that the opening curly braces and the closing curly braces must each be on a line of their own, and not be preceded or followed by white space.
+
+\*/
+(function(){
+
+/*jslint node: true, browser: true */
+/*global $tw: false */
+"use strict";
+
+exports.name = "codeblock";
+
+exports.init = function(parser) {
+	this.parser = parser;
+	// Regexp to match
+	this.matchRegExp = /\{\{\{\r?\n/mg;
+};
+
+exports.parse = function(match,isBlock) {
+	var reEnd = /(\r?\n\}\}\}$)/mg;
+	// Move past the match
+	this.parser.pos = this.matchRegExp.lastIndex;
+	// Look for the end of the block
+	reEnd.lastIndex = this.parser.pos;
+	var match = reEnd.exec(this.parser.source),
+		text;
+	// Process the block
+	if(match) {
+		text = this.parser.source.substring(this.parser.pos,match.index);
+		this.parser.pos = match.index + match[0].length;
+	} else {
+		text = this.parser.source.substr(this.parser.pos);
+		this.parser.pos = this.parser.sourceLength;
+	}
+	// Return the pre element
+	return [{
+		type: "element",
+		tag: "pre",
+		children: [{
+			type: "text",
+			text: text
+		}]
+	}];
+};
+
+})();
diff --git a/core/modules/parsers/wikiparser/rules/block/table.js b/core/modules/parsers/wikiparser/rules/block/table.js
new file mode 100644
index 000000000..f96aceda5
--- /dev/null
+++ b/core/modules/parsers/wikiparser/rules/block/table.js
@@ -0,0 +1,161 @@
+/*\
+title: $:/core/modules/parsers/wikiparser/rules/block/table.js
+type: application/javascript
+module-type: wikiblockrule
+
+Wiki text block rule for tables.
+
+\*/
+(function(){
+
+/*jslint node: true, browser: true */
+/*global $tw: false */
+"use strict";
+
+exports.name = "table";
+
+exports.init = function(parser) {
+	this.parser = parser;
+	// Regexp to match
+	this.matchRegExp = /^\|(?:[^\n]*)\|(?:[fhck]?)\r?\n/mg;
+};
+
+var processRow = function(prevColumns) {
+	var cellRegExp = /(?:\|([^\n\|]*)\|)|(\|[fhck]?\r?\n)/mg,
+		cellTermRegExp = /((?:\x20*)\|)/mg,
+		tree = [],
+		col = 0,
+		colSpanCount = 1,
+		prevCell;
+	// Match a single cell
+	cellRegExp.lastIndex = this.parser.pos;
+	var cellMatch = cellRegExp.exec(this.parser.source);
+	while(cellMatch && cellMatch.index === this.parser.pos) {
+		if(cellMatch[1] === "~") {
+			// Rowspan
+			var last = prevColumns[col];
+			if(last) {
+				last.rowSpanCount++;
+				$tw.utils.addAttributeToParseTreeNode(last.element,"rowspan",last.rowSpanCount);
+				$tw.utils.addAttributeToParseTreeNode(last.element,"valign","center");
+				if(colSpanCount > 1) {
+					$tw.utils.addAttributeToParseTreeNode(last.element,"colspan",colSpanCount);
+					colSpanCount = 1;
+				}
+			}
+			// Move to just before the `|` terminating the cell
+			this.parser.pos = cellRegExp.lastIndex - 1;
+		} else if(cellMatch[1] === ">") {
+			// Colspan
+			colSpanCount++;
+			// Move to just before the `|` terminating the cell
+			this.parser.pos = cellRegExp.lastIndex - 1;
+		} else if(cellMatch[2]) {
+			// End of row
+			if(prevCell && colSpanCount > 1) {
+				$tw.utils.addAttributeToParseTreeNode(prevCell,"colspan",colSpanCount);
+			}
+			this.parser.pos = cellRegExp.lastIndex - 1;
+			break;
+		} else {
+			// For ordinary cells, step beyond the opening `|`
+			this.parser.pos++;
+			// Look for a space at the start of the cell
+			var spaceLeft = false,
+				chr = this.parser.source.substr(this.parser.pos,1);
+			while(chr === " ") {
+				spaceLeft = true;
+				this.parser.pos++;
+				chr = this.parser.source.substr(this.parser.pos,1);
+			}
+			// Check whether this is a heading cell
+			var cell;
+			if(chr === "!") {
+				this.parser.pos++;
+				cell = {type: "element", tag: "th", children: []};
+			} else {
+				cell = {type: "element", tag: "td", children: []};
+			}
+			tree.push(cell);
+			// Record information about this cell
+			prevCell = cell;
+			prevColumns[col] = {rowSpanCount:1,element:cell};
+			// Check for a colspan
+			if(colSpanCount > 1) {
+				$tw.utils.addAttributeToParseTreeNode(cell,"colspan",colSpanCount);
+				colSpanCount = 1;
+			}
+			// Parse the cell
+			cell.children = this.parser.parseRun(cellTermRegExp,{eatTerminator: true});
+			// Set the alignment for the cell
+			if(cellMatch[1].substr(cellMatch[1].length-1,1) === " ") { // spaceRight
+				$tw.utils.addAttributeToParseTreeNode(cell,"align",spaceLeft ? "center" : "left");
+			} else if(spaceLeft) {
+				$tw.utils.addAttributeToParseTreeNode(cell,"align","right");
+			}
+			// Move back to the closing `|`
+			this.parser.pos--;
+		}
+		col++;
+		cellRegExp.lastIndex = this.parser.pos;
+		cellMatch = cellRegExp.exec(this.parser.source);
+	}
+	return tree;
+};
+
+exports.parse = function() {
+	var rowContainerTypes = {"c":"caption", "h":"thead", "":"tbody", "f":"tfoot"},
+		table = {type: "element", tag: "table", children: []},
+		rowRegExp = /^\|([^\n]*)\|([fhck]?)\r?\n/mg,
+		rowTermRegExp = /(\|(?:[fhck]?)\r?\n)/mg,
+		prevColumns = [],
+		currRowType,
+		rowContainer,
+		rowCount = 0;
+	// Match the row
+	rowRegExp.lastIndex = this.parser.pos;
+	var rowMatch = rowRegExp.exec(this.parser.source);
+	while(rowMatch && rowMatch.index === this.parser.pos) {
+		var rowType = rowMatch[2];
+		// Check if it is a class assignment
+		if(rowType === "k") {
+			$tw.utils.addClassToParseTreeNode(table,rowMatch[1]);
+			this.parser.pos = rowMatch.index + rowMatch[0].length;
+		} else {
+			// Otherwise, create a new row if this one is of a different type
+			if(rowType != currRowType) {
+				rowContainer = {type: "element", tag: rowContainerTypes[rowType], children: []};
+				table.children.push(rowContainer);
+				currRowType = rowType;
+			}
+			// Is this a caption row?
+			if(currRowType === "c") {
+				// If so, move past the opening `|` of the row
+				this.parser.pos++;
+				// Move the caption to the first row if it isn't already
+				if(table.children.length !== 1) {
+					table.children.pop(); // Take rowContainer out of the children array
+					table.children.splice(0,0,rowContainer); // Insert it at the bottom						
+				}
+				// Set the alignment - TODO: figure out why TW did this
+//				rowContainer.attributes.align = rowCount === 0 ? "top" : "bottom";
+				// Parse the caption
+				rowContainer.children = this.parser.parseRun(rowTermRegExp,{eatTerminator: true});
+			} else {
+				// Create the row
+				var theRow = {type: "element", tag: "tr", children: []};
+				$tw.utils.addClassToParseTreeNode(theRow,rowCount%2 ? "oddRow" : "evenRow");
+				rowContainer.children.push(theRow);
+				// Process the row
+				theRow.children = processRow.call(this,prevColumns);
+				this.parser.pos = rowMatch.index + rowMatch[0].length;
+				// Increment the row count
+				rowCount++;
+			}
+		}
+		rowMatch = rowRegExp.exec(this.parser.source);
+	}
+	return [table];
+};
+
+})();
diff --git a/core/modules/parsers/wikiparser/rules/run/classrun.js b/core/modules/parsers/wikiparser/rules/run/classrun.js
new file mode 100644
index 000000000..0151a407f
--- /dev/null
+++ b/core/modules/parsers/wikiparser/rules/run/classrun.js
@@ -0,0 +1,51 @@
+/*\
+title: $:/core/modules/parsers/wikiparser/rules/run/classrun.js
+type: application/javascript
+module-type: wikirunrule
+
+Wiki text run rule for assigning classes to runs of text. For example:
+
+{{{
+{{myClass{This text will have the CSS class `myClass`.
+
+* This will not be recognised as a list
+
+List item 2}}}
+}}}
+
+
+\*/
+(function(){
+
+/*jslint node: true, browser: true */
+/*global $tw: false */
+"use strict";
+
+exports.name = "classrun";
+
+exports.init = function(parser) {
+	this.parser = parser;
+	// Regexp to match
+	this.matchRegExp = /\{\{([^\{\r\n]+)\{/mg;
+};
+
+exports.parse = function() {
+	var reEnd = /(\}\}\})/g;
+	// Get the class
+	var classString = this.match[1];
+	// Move past the match
+	this.parser.pos = this.matchRegExp.lastIndex;
+	// Parse the run up to the terminator
+	var tree = this.parser.parseRun(reEnd,{eatTerminator: true});
+	// Return the classed span
+	return [{
+		type: "element",
+		tag: "span",
+		attributes: {
+			"class": {type: "string", value: classString}
+		},
+		children: tree
+	}];
+};
+
+})();
diff --git a/core/modules/parsers/wikiparser/rules/run/coderun.js b/core/modules/parsers/wikiparser/rules/run/coderun.js
new file mode 100644
index 000000000..513178ee4
--- /dev/null
+++ b/core/modules/parsers/wikiparser/rules/run/coderun.js
@@ -0,0 +1,58 @@
+/*\
+title: $:/core/modules/parsers/wikiparser/rules/run/coderun.js
+type: application/javascript
+module-type: wikirunrule
+
+Wiki text run rule for code runs. For example:
+
+{{{
+	This is a {{{code run}}} and `so is this`.
+}}}
+
+\*/
+(function(){
+
+/*jslint node: true, browser: true */
+/*global $tw: false */
+"use strict";
+
+exports.name = "coderun";
+
+exports.init = function(parser) {
+	this.parser = parser;
+	// Regexp to match
+	this.matchRegExp = /(\{\{\{)|(`)/mg;
+};
+
+exports.parse = function() {
+	// Move past the match
+	this.parser.pos = this.matchRegExp.lastIndex;
+	var reEnd;
+	if(this.match[0] === "{{{") {
+		reEnd = /(\}\}\})/mg;
+	} else {
+		reEnd = /(`)/mg;
+	}
+	// Look for the end marker
+	reEnd.lastIndex = this.parser.pos;
+	var match = reEnd.exec(this.parser.source),
+		text;
+	// Process the text
+	if(match) {
+		text = this.parser.source.substring(this.parser.pos,match.index);
+		this.parser.pos = match.index + match[0].length;
+	} else {
+		text = this.parser.source.substr(this.parser.pos);
+		this.parser.pos = this.parser.sourceLength;
+	}
+	return [{
+		type: "element",
+		tag: "code",
+		children: [{
+			type: "text",
+			text: text
+		}]
+	}];
+};
+
+})();
diff --git a/core/modules/parsers/wikiparser/rules/run/comment.js b/core/modules/parsers/wikiparser/rules/run/comment.js
new file mode 100644
index 000000000..5eae8a3cc
--- /dev/null
+++ b/core/modules/parsers/wikiparser/rules/run/comment.js
@@ -0,0 +1,34 @@
+/*\
+title: $:/core/modules/parsers/wikiparser/rules/run/comment.js
+type: application/javascript
+module-type: wikirunrule
+
+Wiki text run rule for HTML comments. For example:
+
+{{{
+<!-- This is a comment -->
+}}}
+
+\*/
+(function(){
+
+/*jslint node: true, browser: true */
+/*global $tw: false */
+"use strict";
+
+exports.name = "comment";
+
+exports.init = function(parser) {
+	this.parser = parser;
+	// Regexp to match - HTML comment regexp by Stephen Ostermiller, http://ostermiller.org/findhtmlcomment.html
+	this.matchRegExp = /\<![ \r\n\t]*(?:--(?:[^\-]|[\r\n]|-[^\-])*--[ \r\n\t]*)\>/mg;
+};
+
+exports.parse = function(match,isBlock) {
+	// Move past the match
+	this.parser.pos = this.matchRegExp.lastIndex;
+	// Don't return any elements
+	return [];
+};
+
+})();
diff --git a/core/modules/parsers/wikiparser/rules/run/dash.js b/core/modules/parsers/wikiparser/rules/run/dash.js
new file mode 100644
index 000000000..60ddb3e9e
--- /dev/null
+++ b/core/modules/parsers/wikiparser/rules/run/dash.js
@@ -0,0 +1,41 @@
+/*\
+title: $:/core/modules/parsers/wikiparser/rules/run/dash.js
+type: application/javascript
+module-type: wikirunrule
+
+Wiki text run rule for dashes. For example:
+
+{{{
+This is an en-dash: --
+
+This is an em-dash: ---
+}}}
+
+Dashes must be followed by whitespace in order to be distinguished from strikethrough notation (`--strikethrough--`).
+
+\*/
+(function(){
+
+/*jslint node: true, browser: true */
+/*global $tw: false */
+"use strict";
+
+exports.name = "dash";
+
+exports.init = function(parser) {
+	this.parser = parser;
+	// Regexp to match
+	this.matchRegExp = /-{2,3}(?=\s)/mg;
+};
+
+exports.parse = function() {
+	// Move past the match
+	this.parser.pos = this.matchRegExp.lastIndex;
+	var dash = this.match[0].length === 2 ? "&ndash;" : "&mdash;";
+	return [{
+		type: "entity",
+		entity: dash
+	}];
+};
+
+})();
diff --git a/core/modules/parsers/wikiparser/rules/run/emphasis.js b/core/modules/parsers/wikiparser/rules/run/emphasis.js
new file mode 100644
index 000000000..939473a7f
--- /dev/null
+++ b/core/modules/parsers/wikiparser/rules/run/emphasis.js
@@ -0,0 +1,78 @@
+/*\
+title: $:/core/modules/parsers/wikiparser/rules/run/emphasis.js
+type: application/javascript
+module-type: wikirunrule
+
+Wiki text run rule for emphasis. For example:
+
+{{{
+	This is ''bold'' text
+
+	This is //italic// text
+
+	This is __underlined__ text
+
+	This is ^^superscript^^ text
+
+	This is ,,subscript,, text
+
+	This is ~~strikethrough~~ text
+}}}
+
+\*/
+(function(){
+
+/*jslint node: true, browser: true */
+/*global $tw: false */
+"use strict";
+
+exports.name = "emphasis";
+
+exports.init = function(parser) {
+	this.parser = parser;
+	// Regexp to match
+	this.matchRegExp = /''|\/\/|__|\^\^|,,|~~/mg;
+};
+
+exports.parse = function() {
+	// Move past the match
+	this.parser.pos = this.matchRegExp.lastIndex;
+	// Figure out which element and closing regexp to use
+	var tag,reEnd;
+	switch(this.match[0]) {
+		case "''": // Bold
+			tag = "strong";
+			reEnd = /''/mg;
+			break;
+		case "//": // Italics
+			tag = "em";
+			reEnd = /\/\//mg;
+			break;
+		case "__": // Underline
+			tag = "u";
+			reEnd = /__/mg;
+			break;
+		case "^^": // Superscript
+			tag = "sup";
+			reEnd = /\^\^/mg;
+			break;
+		case ",,": // Subscript
+			tag = "sub";
+			reEnd = /,,/mg;
+			break;
+		case "~~": // Strikethrough
+			tag = "strike";
+			reEnd = /~~/mg;
+			break;
+	}
+	// Parse the run including the terminator
+	var tree = this.parser.parseRun(reEnd,{eatTerminator: true});
+	// Return the classed span
+	return [{
+		type: "element",
+		tag: tag,
+		children: tree
+	}];
+};
+
+})();