closes #241, improved date handling

- Scholar.strToDate() accepts a string date and returns an object containing year, month, day, and part
- capture access date whenever URL is captured
- updated Zotero.dot to use new namespaces

chrome/chromeFiles/content/scholar/xpcom/cite.js

@@ -89,13 +89,13 @@ Scholar.Cite = new function() {
  * want to use the Scholar data model, but does want to use CSL in JavaScript
  */
 CSL = function(csl) {
-	this._csl = new XML(this._cleanXML(csl));
+	this._csl = new XML(CSL._cleanXML(csl));
 	
 	// initialize CSL
-	this._init();
+	CSL.init();
 	
 	// load localizations
-	this._terms = this._parseLocales(this._csl.terms);
+	this._terms = CSL._parseLocales(this._csl.terms);
 	
 	// load class defaults
 	this.class =  this._csl["@class"].toString();
@@ -272,12 +272,6 @@ CSL.prototype.createBibliography = function(format) {
 	return output;
 }
 
-
-CSL._months = ["January", "February", "March", "April", "May", "June", "July",
-               "August", "September", "October", "November", "December"];
-CSL._monthsShort = ["Jan", "Feb", "Mar", "Apr", "May", "Jun",
-                    "Jul", "Aug", "Sep", "Oct", "Nov", "Dec"];
-
 // for elements that inherit defaults from each other
 CSL._inherit = {
 	author:"contributor",
@@ -307,11 +301,10 @@ CSL._classDefaults["author-date"] = {
 
 CSL.ns = "http://purl.org/net/xbiblio/csl";
 
-CSL.prototype._cleanXML = function(xml) {
-	return xml.replace(/<\?[^>]*\?>/g, "");
-}
-
-CSL.prototype._init = function() {
+/*
+ * initializes CSL interpreter
+ */
+CSL.init = function() {
 	if(!CSL._xmlLang) {
 		// get XML lang
 		var localeService = Components.classes['@mozilla.org/intl/nslocaleservice;1'].
@@ -326,12 +319,30 @@ CSL.prototype._init = function() {
 		req.send(null);
 		
 		// get default terms
-		var locales = new XML(this._cleanXML(req.responseText));
-		CSL._defaultTerms = this._parseLocales(locales);
+		var locales = new XML(CSL._cleanXML(req.responseText));
+		CSL._defaultTerms = CSL._parseLocales(locales);
 	}
 }
 
-CSL.prototype._parseLocales = function(termXML) {
+/*
+ * returns an array of short or long month strings
+ */
+CSL.getMonthStrings = function(form) {
+	CSL.init();
+	return CSL._defaultTerms[form]["_months"];
+}
+
+/*
+ * removes parse instructions from XML
+ */
+CSL._cleanXML = function(xml) {
+	return xml.replace(/<\?[^>]*\?>/g, "");
+}
+
+/*
+ * parses locale strings into CSL._defaultTerms
+ */
+CSL._parseLocales = function(termXML) {
 	// return defaults if there are no terms
 	if(!termXML.length()) {
 		return (CSL._defaultTerms ? CSL._defaultTerms : {});
@@ -403,7 +414,17 @@ CSL.prototype._parseLocales = function(termXML) {
 				}
 			}
 		} else {
-			termArray[form][name] = term.text().toString();
+			if(name.substr(0, 6) == "month-") {
+				// place months into separate array
+				if(!termArray[form]["_months"]) {
+					termArray[form]["_months"] = new Array();
+				}				
+				var monthIndex = parseInt(name.substr(6),10)-1;
+				var term = term.text().toString();
+				termArray[form]["_months"][monthIndex] = term[0].toUpperCase()+term.substr(1).toLowerCase();
+			} else {
+				termArray[form][name] = term.text().toString();
+			}
 		}
 	}
 	
@@ -550,8 +571,8 @@ CSL.prototype._parseEtAl = function(etAl, bibCitElement) {
 			for each(var etAlElement in etAl) {
 				if(etAlElement.@position == "subsequent") {
 					bibCitElement.subsequentEtAl = new Object();
-					bibCitElement.subsequentEtAl.minCreators = parseInt(etAlElement['@min-authors']);
-					bibCitElement.subsequentEtAl.useFirst = parseInt(etAlElement['@use-first']);
+					bibCitElement.subsequentEtAl.minCreators = parseInt(etAlElement['@min-authors'], 10);
+					bibCitElement.subsequentEtAl.useFirst = parseInt(etAlElement['@use-first'], 10);
 				} else {
 					var parseElement = etAlElement;
 				}
@@ -560,8 +581,8 @@ CSL.prototype._parseEtAl = function(etAl, bibCitElement) {
 			var parseElement = etAl;
 		}
 		
-		bibCitElement.etAl.minCreators = parseInt(parseElement['@min-authors']);
-		bibCitElement.etAl.useFirst = parseInt(parseElement['@use-first']);
+		bibCitElement.etAl.minCreators = parseInt(parseElement['@min-authors'], 10);
+		bibCitElement.etAl.useFirst = parseInt(parseElement['@use-first'], 10);
 	}
 }
 
@@ -764,46 +785,6 @@ CSL.prototype._getTerm = function(term, plural, form) {
 	return this._terms[form][term];
 }
 
-/*
- * process the date "string" into a useful object
- */
-CSL.prototype._processDate = function(string) {
-	var date = new Object();
-	
-	var dateRe = /^([0-9]{4})-([0-9]{2})-([0-9]{2})$/;
-	var m = dateRe.exec(string);
-	if(m) {		// sql date
-		var jsDate = new Date(m[1], m[2]-1, m[3], false, false, false);
-	} else {	// not an sql date
-		var yearRe = /^[0-9]+$/;
-		if(yearRe.test(string)) {	// is a year
-			date.year = string;
-			return date;
-		} else {					// who knows what this is
-			var jsDate = new Date(string)
-		}
-	}
-	
-	if(isNaN(jsDate.valueOf())) { // couldn't parse
-		// get year and say other parts are month
-		var yearRe = /^(.*)([0-9]{4})(.*)$/
-		var m = yearRe.exec(string);
-		
-		if(m) {
-			date.year = m[2];
-			date.month = m[1];
-			if(m[2] && m[3]) date.month += " ";
-			date.month += m[3];
-		}
-	} else {
-		date.year = jsDate.getFullYear();
-		date.month = jsDate.getMonth();
-		date.day = jsDate.getDay();
-	}
-	
-	return date;
-}
-
 /*
  * escapes a string for a given format
  */
@@ -972,15 +953,19 @@ CSL.prototype._formatDate = function(element, date, format) {
 					string += date.disambiguation;
 				}
 			}
-		} else if(child.name == "month" && date.month) {
-			if(format == "compare") {
-				string = this._lpad(date.month+1, "0", 2);
-			} else {
-				if(element.form == "short") {
-					string = CSL._monthsShort[date.month];
+		} else if(child.name == "month") {
+			if(date.month) {
+				if(format == "compare") {
+					string = this._lpad(date.month+1, "0", 2);
 				} else {
-					string = CSL._months[date.month];
+					if(element.form == "short") {
+						string = this._terms["short"]["_months"][date.month];
+					} else {
+						string = this._terms["long"]["_months"][date.month];
+					}
 				}
+			} else if(date.part && format != "compare") {
+				string = date.part;
 			}
 		} else if(child.name == "day" && date.day) {
 			if(format == "compare") {
@@ -1518,6 +1503,9 @@ CSL.prototype._getTypeFromItem = function(item) {
 	return [CSL._optionalTypeMappings[scholarType], CSL._fallbackTypeMappings[scholarType]];
 }
 
+/*
+ * separate creators object into authors, editors, and translators
+ */
 CSL.prototype._separateItemCreators = function(item) {
 	var authors = new Array();
 	var editors = new Array();
@@ -1543,6 +1531,13 @@ CSL.prototype._separateItemCreators = function(item) {
 	
 	return [authors, editors, translators];
 }
+
+/*
+ * return an object containing year, month, and day
+ */
+CSL.prototype._processDate = function(string) {
+	return Scholar.strToDate(string);
+}
 /*
  * END SCHOLAR-SPECIFIC CODE
  */

chrome/chromeFiles/content/scholar/xpcom/scholar.js

@@ -35,10 +35,12 @@ var Scholar = new function(){
 	this.randomString = randomString;
 	this.getRandomID = getRandomID;
 	this.moveToUnique = moveToUnique;
+	this.strToDate = strToDate;
 	
 	// Public properties
 	this.version;
 	this.platform;
+	this.isMac;
 	
 	/*
 	 * Initialize the extension
@@ -424,6 +426,91 @@ var Scholar = new function(){
 		file.moveTo(newFile.parent, newName);
 		return file;
 	}
+	
+	/*
+	 * converts a string to an object containing:
+	 *    day: integer form of the day
+	 *    month: integer form of the month (indexed from 0, not 1)
+	 *    year: 4 digit year (or, year + BC/AD/etc.)
+	 *    part: anything that does not fall under any of the above categories
+	 *          (e.g., "Summer," etc.)
+	 */
+	function strToDate(string) {
+		var date = new Object();
+		
+		// skip empty things
+		if(!string) {
+			return date;
+		}
+		
+		// get short month strings from CSL interpreter
+		var months = CSL.getMonthStrings("short");
+		
+		string = string.replace(/^\s+/, "").replace(/\s+$/, "").replace(/\s+/, " ");
+		
+		var dateRe = /^([0-9]{4})[\-\/]([0-9]{2})[\-\/]([0-9]{2})$/;
+		var m = dateRe.exec(string);
+		if(m) {		// sql date
+			Scholar.debug("DATE: used form 1: SQL");
+			var jsDate = new Date(m[1], m[2]-1, m[3], false, false, false);
+		} else {	// not an sql date
+			var yearRe = /^((?:circa |around |about |c\.? ?)[0-9]{1,4}(?: ?B\.? ?C\.?(?: ?E\.?)?| ?C\.? ?E\.?| ?A\.? ?D\.?)|[0-9]{4})$/i;
+			if(yearRe.test(string)) {
+				// is just a year
+				Scholar.debug("DATE: used form 2: year-only");
+				date.year = string;
+				return date;
+			}
+			
+			// who knows what this is, but try JavaScript's date handling first
+			var jsDate = new Date(string)
+		}
+		
+		if(!isNaN(jsDate.valueOf())) {
+			Scholar.debug("DATE: retrieved from JavaScript");
+			// got a javascript date
+			date.year = jsDate.getFullYear();
+			date.month = jsDate.getMonth();
+			date.day = jsDate.getDate();
+			return date;
+		}
+		
+		// no javascript date. time for cruder things.
+		
+		// first, see if we have anything resembling a year
+		var yearRe = /^(.*)\b((?:circa |around |about |c\.? ?)[0-9]{1,4}(?: ?B\.? ?C\.?(?: ?E\.?)?| ?C\.? ?E\.?| ?A\.? ?D\.?)|[0-9]{4})\b(.*)$/i;
+		
+		var m = yearRe.exec(string);
+		if(m) {
+			date.year = m[2];
+			date.part = m[1]+m[3];
+			Scholar.debug("DATE: got year ("+date.year+", "+date.part+")");
+			
+			// then, see if have anything resembling a month anywhere
+			var monthRe = new RegExp("^(.*)\\b("+months.join("|")+")[^ ]* (.*)$", "i");
+			var m = monthRe.exec(date.part);
+			if(m) {
+				date.month = months.indexOf(m[2][0].toUpperCase()+m[2].substr(1).toLowerCase());
+				date.part = m[1]+m[3];
+				Scholar.debug("DATE: got month ("+date.month+", "+date.part+")");
+				
+				// then, see if there's a day 
+				var dayRe = /^(.*)\b([0-9]{1,2})\b(.*)$/i;
+				var m = dayRe.exec(date.part);
+				if(m) {
+					date.day = m[2];
+					date.part = m[1]+m[3];
+					Scholar.debug("DATE: got day ("+date.day+", "+date.part+")");
+				}
+			}
+		}
+		
+		if(date.part) {
+			date.part = date.part.replace(/^[^A-Za-z0-9]+/, "").replace(/[^A-Za-z0-9]+$/, "");
+		}
+		
+		return date;
+	}
 };
 
 

chrome/chromeFiles/content/scholar/xpcom/translate.js

@@ -1035,6 +1035,11 @@ Scholar.Translate.prototype._itemDone = function(item) {
 			// makes looping through easier
 			item.itemType = item.complete = undefined;
 			
+			// automatically set access date if URL is set
+			if(item.url && !item.accessDate) {
+				item.accessDate = (new Date()).toLocaleString();
+			}
+			
 			var fieldID, field;
 			for(var i in item) {
 				// loop through item fields

scrapers.sql

@@ -1,4 +1,4 @@
--- 65
+-- 66
 
 -- Set the following timestamp to the most recent scraper update date
 REPLACE INTO "version" VALUES ('repository', STRFTIME('%s', '2006-08-15 15:42:00'));
@@ -6076,7 +6076,7 @@ REPLACE INTO "csl" VALUES('http://purl.org/net/xbiblio/csl/styles/apa.csl', '200
   </bibliography>
 </style>');
 
-REPLACE INTO "csl" VALUES('http://purl.org/net/xbiblio/csl/styles/chicago-note.csl', '2006-08-29 23:05:00', 'Chicago Manual of Style (Note)',
+REPLACE INTO "csl" VALUES('http://purl.org/net/xbiblio/csl/styles/chicago-note.csl', '2006-08-30 17:40:00', 'Chicago Manual of Style (Note)',
 '<?xml version="1.0" encoding="UTF-8"?>
 <?oxygen RNGSchema="../schema/trunk/csl.rnc" type="compact"?>
 <style xmlns="http://purl.org/net/xbiblio/csl" class="note" xml:lang="en">
@@ -6173,8 +6173,8 @@ REPLACE INTO "csl" VALUES('http://purl.org/net/xbiblio/csl/styles/chicago-note.c
           <type name="article">
             <author suffix=", "/>
             <titles prefix="&#8220;" suffix=",&#8221; "/>
-            <titles relation="container" font-style="italic"/>
-            <date prefix=", ">
+            <titles relation="container" font-style="italic" suffix=", "/>
+            <date>
               <day suffix=" "/>
               <month suffix=" " text-transform="capitalize"/>
               <year/>