milksteak 0.0.1 → 0.0.2

data/lib/public/milksteak/codemirror2/mode/xmlpure/xmlpure.js

@@ -0,0 +1,490 @@
+/**
+ * xmlpure.js
+ * 
+ * Building upon and improving the CodeMirror 2 XML parser
+ * @author: Dror BG (deebug.dev@gmail.com)
+ * @date: August, 2011
+ */
+
+CodeMirror.defineMode("xmlpure", function(config, parserConfig) {
+    // constants
+    var STYLE_ERROR = "error";
+    var STYLE_INSTRUCTION = "comment";
+    var STYLE_COMMENT = "comment";
+    var STYLE_ELEMENT_NAME = "tag";
+    var STYLE_ATTRIBUTE = "attribute";
+    var STYLE_WORD = "string";
+    var STYLE_TEXT = "atom";
+    var STYLE_ENTITIES = "string";
+
+    var TAG_INSTRUCTION = "!instruction";
+    var TAG_CDATA = "!cdata";
+    var TAG_COMMENT = "!comment";
+    var TAG_TEXT = "!text";
+    
+    var doNotIndent = {
+        "!cdata": true,
+        "!comment": true,
+        "!text": true,
+        "!instruction": true
+    };
+
+    // options
+    var indentUnit = config.indentUnit;
+
+    ///////////////////////////////////////////////////////////////////////////
+    // helper functions
+    
+    // chain a parser to another parser
+    function chain(stream, state, parser) {
+        state.tokenize = parser;
+        return parser(stream, state);
+    }
+    
+    // parse a block (comment, CDATA or text)
+    function inBlock(style, terminator, nextTokenize) {
+        return function(stream, state) {
+            while (!stream.eol()) {
+                if (stream.match(terminator)) {
+                    popContext(state);
+                    state.tokenize = nextTokenize;
+                    break;
+                }
+                stream.next();
+            }
+            return style;
+        };
+    }
+    
+    // go down a level in the document
+    // (hint: look at who calls this function to know what the contexts are)
+    function pushContext(state, tagName) {
+        var noIndent = doNotIndent.hasOwnProperty(tagName) || (state.context && state.context.doIndent);
+        var newContext = {
+            tagName: tagName,
+            prev: state.context,
+            indent: state.context ? state.context.indent + indentUnit : 0,
+            lineNumber: state.lineNumber,
+            indented: state.indented,
+            noIndent: noIndent
+        };
+        state.context = newContext;
+    }
+    
+    // go up a level in the document
+    function popContext(state) {
+        if (state.context) {
+            var oldContext = state.context;
+            state.context = oldContext.prev;
+            return oldContext;
+        }
+        
+        // we shouldn't be here - it means we didn't have a context to pop
+        return null;
+    }
+    
+    // return true if the current token is seperated from the tokens before it
+    // which means either this is the start of the line, or there is at least
+    // one space or tab character behind the token
+    // otherwise returns false
+    function isTokenSeparated(stream) {
+        return stream.sol() ||
+            stream.string.charAt(stream.start - 1) == " " ||
+            stream.string.charAt(stream.start - 1) == "\t";
+    }
+    
+    ///////////////////////////////////////////////////////////////////////////
+    // context: document
+    // 
+    // an XML document can contain:
+    // - a single declaration (if defined, it must be the very first line)
+    // - exactly one root element
+    // @todo try to actually limit the number of root elements to 1
+    // - zero or more comments
+    function parseDocument(stream, state) {
+        if(stream.eat("<")) {
+            if(stream.eat("?")) {
+                // processing instruction
+                pushContext(state, TAG_INSTRUCTION);
+                state.tokenize = parseProcessingInstructionStartTag;
+                return STYLE_INSTRUCTION;
+            } else if(stream.match("!--")) {
+                // new context: comment
+                pushContext(state, TAG_COMMENT);
+                return chain(stream, state, inBlock(STYLE_COMMENT, "-->", parseDocument));
+            } else if(stream.eatSpace() || stream.eol() ) {
+                stream.skipToEnd();
+                return STYLE_ERROR;
+            } else {
+                // element
+                state.tokenize = parseElementTagName;
+                return STYLE_ELEMENT_NAME;
+            }
+        }
+        
+        // error on line
+        stream.skipToEnd();
+        return STYLE_ERROR;
+    }
+
+    ///////////////////////////////////////////////////////////////////////////
+    // context: XML element start-tag or end-tag
+    //
+    // - element start-tag can contain attributes
+    // - element start-tag may self-close (or start an element block if it doesn't)
+    // - element end-tag can contain only the tag name
+    function parseElementTagName(stream, state) {
+        // get the name of the tag
+        var startPos = stream.pos;
+        if(stream.match(/^[a-zA-Z_:][-a-zA-Z0-9_:.]*/)) {
+            // element start-tag
+            var tagName = stream.string.substring(startPos, stream.pos);
+            pushContext(state, tagName);
+            state.tokenize = parseElement;
+            return STYLE_ELEMENT_NAME;
+        } else if(stream.match(/^\/[a-zA-Z_:][-a-zA-Z0-9_:.]*( )*>/)) {
+            // element end-tag
+            var endTagName = stream.string.substring(startPos + 1, stream.pos - 1).trim();
+            var oldContext = popContext(state);
+            state.tokenize = state.context == null ? parseDocument : parseElementBlock;
+            if(oldContext == null || endTagName != oldContext.tagName) {
+                // the start and end tag names should match - error
+                return STYLE_ERROR;
+            }
+            return STYLE_ELEMENT_NAME;
+        } else {
+            // no tag name - error
+            state.tokenize = state.context == null ? parseDocument : parseElementBlock;
+            stream.eatWhile(/[^>]/);
+            stream.eat(">");
+            return STYLE_ERROR;
+        }
+        
+        stream.skipToEnd();
+        return null;
+    }
+    
+    function parseElement(stream, state) {
+        if(stream.match(/^\/>/)) {
+            // self-closing tag
+            popContext(state);
+            state.tokenize = state.context == null ? parseDocument : parseElementBlock;
+            return STYLE_ELEMENT_NAME;
+        } else if(stream.eat(/^>/)) {
+            state.tokenize = parseElementBlock;
+            return STYLE_ELEMENT_NAME;
+        } else if(isTokenSeparated(stream) && stream.match(/^[a-zA-Z_:][-a-zA-Z0-9_:.]*( )*=/)) {
+            // attribute
+            state.tokenize = parseAttribute;
+            return STYLE_ATTRIBUTE;
+        }
+        
+        // no other options - this is an error
+        state.tokenize = state.context == null ? parseDocument : parseDocument;
+        stream.eatWhile(/[^>]/);
+        stream.eat(">");
+        return STYLE_ERROR;
+    }
+    
+    ///////////////////////////////////////////////////////////////////////////
+    // context: attribute
+    // 
+    // attribute values may contain everything, except:
+    // - the ending quote (with ' or ") - this marks the end of the value
+    // - the character "<" - should never appear
+    // - ampersand ("&") - unless it starts a reference: a string that ends with a semi-colon (";")
+    // ---> note: this parser is lax in what may be put into a reference string,
+    // ---> consult http://www.w3.org/TR/REC-xml/#NT-Reference if you want to make it tighter
+    function parseAttribute(stream, state) {
+        var quote = stream.next();
+        if(quote != "\"" && quote != "'") {
+            // attribute must be quoted
+            stream.skipToEnd();
+            state.tokenize = parseElement;
+            return STYLE_ERROR;
+        }
+        
+        state.tokParams.quote = quote;    
+        state.tokenize = parseAttributeValue;
+        return STYLE_WORD;
+    }
+
+    // @todo: find out whether this attribute value spans multiple lines,
+    //        and if so, push a context for it in order not to indent it
+    //        (or something of the sort..)
+    function parseAttributeValue(stream, state) {
+        var ch = "";
+        while(!stream.eol()) {
+            ch = stream.next();
+            if(ch == state.tokParams.quote) {
+                // end quote found
+                state.tokenize = parseElement;
+                return STYLE_WORD;
+            } else if(ch == "<") {
+                // can't have less-than signs in an attribute value, ever
+                stream.skipToEnd()
+                state.tokenize = parseElement;
+                return STYLE_ERROR;
+            } else if(ch == "&") {
+                // reference - look for a semi-colon, or return error if none found
+                ch = stream.next();
+                
+                // make sure that semi-colon isn't right after the ampersand
+                if(ch == ';') {
+                    stream.skipToEnd()
+                    state.tokenize = parseElement;
+                    return STYLE_ERROR;
+                }
+                
+                // make sure no less-than characters slipped in
+                while(!stream.eol() && ch != ";") {
+                    if(ch == "<") {
+                        // can't have less-than signs in an attribute value, ever
+                        stream.skipToEnd()
+                        state.tokenize = parseElement;
+                        return STYLE_ERROR;
+                    }
+                    ch = stream.next();
+                }
+                if(stream.eol() && ch != ";") {
+                    // no ampersand found - error
+                    stream.skipToEnd();
+                    state.tokenize = parseElement;
+                    return STYLE_ERROR;
+                }                
+            }
+        }
+        
+        // attribute value continues to next line
+        return STYLE_WORD;
+    }
+    
+    ///////////////////////////////////////////////////////////////////////////
+    // context: element block
+    //
+    // a block can contain:
+    // - elements
+    // - text
+    // - CDATA sections
+    // - comments
+    function parseElementBlock(stream, state) {
+        if(stream.eat("<")) {
+            if(stream.match("?")) {
+                pushContext(state, TAG_INSTRUCTION);
+                state.tokenize = parseProcessingInstructionStartTag;
+                return STYLE_INSTRUCTION;
+            } else if(stream.match("!--")) {
+                // new context: comment
+                pushContext(state, TAG_COMMENT);
+                return chain(stream, state, inBlock(STYLE_COMMENT, "-->",
+                    state.context == null ? parseDocument : parseElementBlock));
+            } else if(stream.match("![CDATA[")) {
+                // new context: CDATA section
+                pushContext(state, TAG_CDATA);
+                return chain(stream, state, inBlock(STYLE_TEXT, "]]>",
+                    state.context == null ? parseDocument : parseElementBlock));
+            } else if(stream.eatSpace() || stream.eol() ) {
+                stream.skipToEnd();
+                return STYLE_ERROR;
+            } else {
+                // element
+                state.tokenize = parseElementTagName;
+                return STYLE_ELEMENT_NAME;
+            }
+        } else if(stream.eat("&")) {
+            stream.eatWhile(/[^;]/);
+            stream.eat(";");
+            return STYLE_ENTITIES;
+        } else {
+            // new context: text
+            pushContext(state, TAG_TEXT);
+            state.tokenize = parseText;
+            return null;
+        }
+        
+        state.tokenize = state.context == null ? parseDocument : parseElementBlock;
+        stream.skipToEnd();
+        return null;
+    }
+    
+    function parseText(stream, state) {
+        stream.eatWhile(/[^<]/);
+        if(!stream.eol()) {
+            // we cannot possibly be in the document context,
+            // just inside an element block
+            popContext(state);
+            state.tokenize = parseElementBlock;
+        }
+        return STYLE_TEXT;
+    }
+
+    ///////////////////////////////////////////////////////////////////////////
+    // context: XML processing instructions
+    //
+    // XML processing instructions (PIs) allow documents to contain instructions for applications.
+    // PI format: <?name data?>
+    // - 'name' can be anything other than 'xml' (case-insensitive)
+    // - 'data' can be anything which doesn't contain '?>'
+    // XML declaration is a special PI (see XML declaration context below)
+    function parseProcessingInstructionStartTag(stream, state) {
+        if(stream.match("xml", true, true)) {
+            // xml declaration
+            if(state.lineNumber > 1 || stream.pos > 5) {
+                state.tokenize = parseDocument;
+                stream.skipToEnd();
+                return STYLE_ERROR;
+            } else {
+                state.tokenize = parseDeclarationVersion;
+                return STYLE_INSTRUCTION;
+            }
+        }
+
+        // regular processing instruction
+        if(isTokenSeparated(stream) || stream.match("?>")) {
+            // we have a space after the start-tag, or nothing but the end-tag
+            // either way - error!
+            state.tokenize = parseDocument;
+            stream.skipToEnd();
+            return STYLE_ERROR;
+        }
+
+        state.tokenize = parseProcessingInstructionBody;
+        return STYLE_INSTRUCTION;
+    }
+
+    function parseProcessingInstructionBody(stream, state) {
+        stream.eatWhile(/[^?]/);
+        if(stream.eat("?")) {
+            if(stream.eat(">")) {
+                popContext(state);
+                state.tokenize = state.context == null ? parseDocument : parseElementBlock;
+            }
+        }
+        return STYLE_INSTRUCTION;
+    }
+
+    
+    ///////////////////////////////////////////////////////////////////////////
+    // context: XML declaration
+    //
+    // XML declaration is of the following format:
+    // <?xml version="1.0" encoding="UTF-8" standalone="no" ?>
+    // - must start at the first character of the first line
+    // - may span multiple lines
+    // - must include 'version'
+    // - may include 'encoding' and 'standalone' (in that order after 'version')
+    // - attribute names must be lowercase
+    // - cannot contain anything else on the line
+    function parseDeclarationVersion(stream, state) {
+        state.tokenize = parseDeclarationEncoding;
+        
+        if(isTokenSeparated(stream) && stream.match(/^version( )*=( )*"([a-zA-Z0-9_.:]|\-)+"/)) {
+            return STYLE_INSTRUCTION;
+        }
+        stream.skipToEnd();
+        return STYLE_ERROR;
+    }
+
+    function parseDeclarationEncoding(stream, state) {
+        state.tokenize = parseDeclarationStandalone;
+        
+        if(isTokenSeparated(stream) && stream.match(/^encoding( )*=( )*"[A-Za-z]([A-Za-z0-9._]|\-)*"/)) {
+            return STYLE_INSTRUCTION;
+        }
+        return null;
+    }
+
+    function parseDeclarationStandalone(stream, state) {
+        state.tokenize = parseDeclarationEndTag;
+        
+        if(isTokenSeparated(stream) && stream.match(/^standalone( )*=( )*"(yes|no)"/)) {
+            return STYLE_INSTRUCTION;
+        }
+        return null;
+    }
+
+    function parseDeclarationEndTag(stream, state) {
+        state.tokenize = parseDocument;
+        
+        if(stream.match("?>") && stream.eol()) {
+            popContext(state);
+            return STYLE_INSTRUCTION;
+        }
+        stream.skipToEnd();
+        return STYLE_ERROR;
+    }
+
+    ///////////////////////////////////////////////////////////////////////////
+    // returned object
+    return {
+        electricChars: "/[",
+        
+        startState: function() {
+            return {
+                tokenize: parseDocument,
+                tokParams: {},
+                lineNumber: 0,
+                lineError: false,
+                context: null,
+                indented: 0
+            };
+        },
+
+        token: function(stream, state) {
+            if(stream.sol()) {
+                // initialize a new line
+                state.lineNumber++;
+                state.lineError = false;
+                state.indented = stream.indentation();
+            }
+
+            // eat all (the spaces) you can
+            if(stream.eatSpace()) return null;
+
+            // run the current tokenize function, according to the state
+            var style = state.tokenize(stream, state);
+            
+            // is there an error somewhere in the line?
+            state.lineError = (state.lineError || style == "error");
+
+            return style;
+        },
+        
+        blankLine: function(state) {
+            // blank lines are lines too!
+            state.lineNumber++;
+            state.lineError = false;
+        },
+        
+        indent: function(state, textAfter) {
+            if(state.context) {
+                if(state.context.noIndent == true) {
+                    // do not indent - no return value at all
+                    return;
+                }
+                if(textAfter.match(/^<\/.*/)) {
+                    // end-tag - indent back to last context
+                    return state.context.indent;
+                }
+                if(textAfter.match(/^<!\[CDATA\[/)) {
+                    // a stand-alone CDATA start-tag - indent back to column 0
+                    return 0;                
+                }
+                // indent to last context + regular indent unit
+                return state.context.indent + indentUnit;
+            }
+            return 0;
+        },
+        
+        compareStates: function(a, b) {
+            if (a.indented != b.indented) return false;
+            for (var ca = a.context, cb = b.context; ; ca = ca.prev, cb = cb.prev) {
+                if (!ca || !cb) return ca == cb;
+                if (ca.tagName != cb.tagName) return false;
+            }
+        }
+    };
+});
+
+CodeMirror.defineMIME("application/xml", "purexml");
+CodeMirror.defineMIME("text/xml", "purexml");

data/lib/public/milksteak/codemirror2/mode/yaml/index.html

@@ -0,0 +1,67 @@
+<!doctype html>
+<html>
+  <head>
+    <title>CodeMirror: YAML mode</title>
+    <link rel="stylesheet" href="../../lib/codemirror.css">
+    <script src="../../lib/codemirror.js"></script>
+    <script src="yaml.js"></script>
+    <style>.CodeMirror { border-top: 1px solid #ddd; border-bottom: 1px solid #ddd; }</style>
+    <link rel="stylesheet" href="../../doc/docs.css">
+  </head>
+  <body>
+    <h1>CodeMirror: YAML mode</h1>
+    <form><textarea id="code" name="code">
+--- # Favorite movies
+- Casablanca
+- North by Northwest
+- The Man Who Wasn't There
+--- # Shopping list
+[milk, pumpkin pie, eggs, juice]
+--- # Indented Blocks, common in YAML data files, use indentation and new lines to separate the key: value pairs
+  name: John Smith
+  age: 33
+--- # Inline Blocks, common in YAML data streams, use commas to separate the key: value pairs between braces
+{name: John Smith, age: 33}
+---
+receipt:     Oz-Ware Purchase Invoice
+date:        2007-08-06
+customer:
+    given:   Dorothy
+    family:  Gale
+
+items:
+    - part_no:   A4786
+      descrip:   Water Bucket (Filled)
+      price:     1.47
+      quantity:  4
+
+    - part_no:   E1628
+      descrip:   High Heeled "Ruby" Slippers
+      size:       8
+      price:     100.27
+      quantity:  1
+
+bill-to:  &id001
+    street: |
+            123 Tornado Alley
+            Suite 16
+    city:   East Centerville
+    state:  KS
+
+ship-to:  *id001
+
+specialDelivery:  >
+    Follow the Yellow Brick
+    Road to the Emerald City.
+    Pay no attention to the
+    man behind the curtain.
+...
+</textarea></form>
+    <script>
+      var editor = CodeMirror.fromTextArea(document.getElementById("code"), {});
+    </script>
+
+    <p><strong>MIME types defined:</strong> <code>text/x-yaml</code>.</p>
+
+  </body>
+</html>

data/lib/public/milksteak/codemirror2/mode/yaml/yaml.js

@@ -0,0 +1,95 @@
+CodeMirror.defineMode("yaml", function() {
+	
+	var cons = ['true', 'false', 'on', 'off', 'yes', 'no'];
+	var keywordRegex = new RegExp("\\b(("+cons.join(")|(")+"))$", 'i');
+	
+	return {
+		token: function(stream, state) {
+			var ch = stream.peek();
+			var esc = state.escaped;
+			state.escaped = false;
+			/* comments */
+			if (ch == "#") { stream.skipToEnd(); return "comment"; }
+			if (state.literal && stream.indentation() > state.keyCol) {
+				stream.skipToEnd(); return "string";
+			} else if (state.literal) { state.literal = false; }
+			if (stream.sol()) {
+				state.keyCol = 0;
+				state.pair = false;
+				state.pairStart = false;
+				/* document start */
+				if(stream.match(/---/)) { return "def"; }
+				/* document end */
+				if (stream.match(/\.\.\./)) { return "def"; }
+				/* array list item */
+				if (stream.match(/\s*-\s+/)) { return 'meta'; }
+			}
+			/* pairs (associative arrays) -> key */
+			if (!state.pair && stream.match(/^\s*([a-z0-9\._-])+(?=\s*:)/i)) {
+				state.pair = true;
+				state.keyCol = stream.indentation();
+				return "atom";
+			}
+			if (state.pair && stream.match(/^:\s*/)) { state.pairStart = true; return 'meta'; }
+			
+			/* inline pairs/lists */
+			if (stream.match(/^(\{|\}|\[|\])/)) {
+				if (ch == '{')
+					state.inlinePairs++;
+				else if (ch == '}')
+					state.inlinePairs--;
+				else if (ch == '[')
+					state.inlineList++;
+				else
+					state.inlineList--;
+				return 'meta';
+			}
+			
+			/* list seperator */
+			if (state.inlineList > 0 && !esc && ch == ',') {
+				stream.next();
+				return 'meta';
+			}
+			/* pairs seperator */
+			if (state.inlinePairs > 0 && !esc && ch == ',') {
+				state.keyCol = 0;
+				state.pair = false;
+				state.pairStart = false;
+				stream.next();
+				return 'meta';
+			}
+			
+			/* start of value of a pair */
+			if (state.pairStart) {
+				/* block literals */
+				if (stream.match(/^\s*(\||\>)\s*/)) { state.literal = true; return 'meta'; };
+				/* references */
+				if (stream.match(/^\s*(\&|\*)[a-z0-9\._-]+\b/i)) { return 'variable-2'; }
+				/* numbers */
+				if (state.inlinePairs == 0 && stream.match(/^\s*-?[0-9\.\,]+\s?$/)) { return 'number'; }
+				if (state.inlinePairs > 0 && stream.match(/^\s*-?[0-9\.\,]+\s?(?=(,|}))/)) { return 'number'; }
+				/* keywords */
+				if (stream.match(keywordRegex)) { return 'keyword'; }
+			}
+
+			/* nothing found, continue */
+			state.pairStart = false;
+			state.escaped = (ch == '\\');
+			stream.next();
+			return null;
+		},
+		startState: function() {
+			return {
+				pair: false,
+				pairStart: false,
+				keyCol: 0,
+				inlinePairs: 0,
+				inlineList: 0,
+				literal: false,
+				escaped: false
+			};
+		}
+	};
+});
+
+CodeMirror.defineMIME("text/x-yaml", "yaml");

data/lib/public/milksteak/codemirror2/theme/cobalt.css

@@ -0,0 +1,18 @@
+.cm-s-cobalt { background: #002240; color: white; }
+.cm-s-cobalt div.CodeMirror-selected { background: #b36539 !important; }
+.cm-s-cobalt .CodeMirror-gutter { background: #002240; border-right: 1px solid #aaa; }
+.cm-s-cobalt .CodeMirror-gutter-text { color: #d0d0d0; }
+.cm-s-cobalt .CodeMirror-cursor { border-left: 1px solid white !important; }
+
+.cm-s-cobalt span.cm-comment { color: #08f; }
+.cm-s-cobalt span.cm-atom { color: #845dc4; }
+.cm-s-cobalt span.cm-number, .cm-s-cobalt span.cm-attribute { color: #ff80e1; }
+.cm-s-cobalt span.cm-keyword { color: #ffee80; }
+.cm-s-cobalt span.cm-string { color: #3ad900; }
+.cm-s-cobalt span.cm-meta { color: #ff9d00; }
+.cm-s-cobalt span.cm-variable-2, .cm-s-cobalt span.cm-tag { color: #9effff; }
+.cm-s-cobalt span.cm-variable-3, .cm-s-cobalt span.cm-def { color: white; }
+.cm-s-cobalt span.cm-error { color: #9d1e15; }
+.cm-s-cobalt span.cm-bracket { color: #d8d8d8; }
+.cm-s-cobalt span.cm-builtin, .cm-s-cobalt span.cm-special { color: #ff9e59; }
+.cm-s-cobalt span.cm-link { color: #845dc4; }