out/js/search.js

/**
 search.js searchs an XML index to HTML files.

 A part of the jsfind project (http://projects.elucidsoft.net/jsfind)
 Copyright (C) 2003 Shawn Garbett

 This program is free software; you can redistribute it and/or
 modify it under the terms of the GNU General Public License
 as published by the Free Software Foundation; either version 2
 of the License, or (at your option) any later version.
 
 This program is distributed in the hope that it will be useful,
 but WITHOUT ANY WARRANTY; without even the implied warranty of
 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 GNU General Public License for more details.
 
 You should have received a copy of the GNU General Public License
 along with this program; if not, write to the Free Software
 Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA  02111-1307, USA.
 
 Contact Info:
 Shawn Garbett <Shawn@eLucidSoft.net>
 http://www.elucidsoft.net
 4037 General Bate Drive
 Nashville, TN 37204
*/

// Constants
var conversion = new String
        ("0123456789abcdefghijklmnopqrstuvwxyz");

// State variables
var query_left = "";
var search_err = "";
var results    = null;
var index_path = "";

// wildcard data about found keys URLs and position
var wildcard_url_pos = new Array();

var watchdog_id = 0;
var watchdog_callback = null;

// Object to hold search results
function Result(title, link, freq) {
        this.title=title;
        this.link=link;
        this.frequency=Number(freq);
}

// Function to merge (intersect) two result sets
function intersect_results(data) {
        // If there are no stored results, then these are the results
        if (! results) {
                results = data;
                return;
        }

        var output=new Array();

        // There are existing results, to do an intersect...
        for (var i=0; i<results.length; i++) {
                for (var j=0; j<data.length; j++) {
                        if (data[j].title == results[i].title) {
                                results[i].frequency += data[j].frequency;
                                output.push(results[i]);  
                                break;
                        }
                }
        }

        results = output;
}

var debug_div = null;

function debug(msg) {

//      return; // Disable debugging

        if (! debug_div) debug_div = document.getElementById('debug');

        // this will create debug div if it doesn't exist.
        if (! debug_div) {
                debug_div = document.createElement('div');
                if (document.body) document.body.appendChild(debug_div);
                else debug_div = null;
        }
        if (debug_div) {
                debug_div.appendChild(document.createTextNode(msg));
                debug_div.appendChild(document.createElement("br"));
        }
}

// Convert a number into a base x alphanumeric number string
function convert(num) {
        var base = conversion.length;
        var pow = 1;
        var pos = 0;
        var out = "";  

        if (num == 0) return "0";

        while (num > 0) {
                pos = num % base;
                out = conversion.charAt(pos) + out;
                num = Math.floor(num/base);
                pow *= base;
        }

        return out;
}

function watchdog() {
        debug ("TIMEOUT!");
        watchdog_callback(new Array());
}

var xmldoc;

// This function loads the XML document from the specified URL, and when
// it is fully loaded, passes that document and the url to the specified
// handler function.  This function works with any XML document

function loadXML(url, handler, data, result_handler) {
        debug("loadXML("+url+","+data+")");

        // Timeout operation in 10 seconds
        watchdog_callback = result_handler;
        watchdog_id=setTimeout("watchdog()", 20000);

        //debug("setTimeout = "+watchdog_id);

        try {
                // Use the standard DOM Level 2 technique, if it is supported
                if (document.implementation && document.implementation.createDocument) {
                        // Create a new Document object
                        xmldoc = document.implementation.createDocument("", "", null);

                        // Specify what should happen when it finishes loading
                        xmldoc.onload = function() { handler(xmldoc, url, data, result_handler); }

                        //xmldoc.onerror = docError;
                        //xmldoc.addEventListener("load",docError,false);

                        // And tell it what URL to load
                        xmldoc.load(url);
                        return true;
                }
                // Otherwise use Microsoft's proprietary API for Internet Explorer
                // Something about not following standards once again
                else if (window.ActiveXObject) {  
                        xmldoc = new ActiveXObject("Microsoft.XMLDOM"); // Create doc.
                        if (! xmldoc) xmldoc = new ActiveXObject("MSXML2.DOMDocument"); // Create doc.
                        // Specify onload
                        xmldoc.onreadystatechange = function() {              
                                if (xmldoc.readyState == 4) handler(xmldoc, url, data, result_handler);
                        }
                        xmldoc.load(url);                                     // Start loading!
                        return true;
                }
                // else fallback on usage of iframes to load xml (Opera 7.53 without Java and maybe old Mac browsers)
                else {
                        debug("using iframe xml loader - experimental and slow");
                        if (! window.xml_iframe) {
                                debug("creating iframe");
                                window.xml_iframe = document.createElement('div');
                                window.xml_iframe.innerHTML = '<iframe src="'+url+'" name="xml_iframe" height="0" width="0" style="display: none;"></iframe>';
                                document.body.appendChild(window.xml_iframe);
                        } else {
                                debug("loading xml in existing iframe");
                                window.frames.xml_iframe.window.document.location.href = url;
                        }

                        // set timeout to re-check if iframe is loaded
                        window.iframe_timeout = window.setInterval('iframe_xml_loaded();',100);

                        // save some data for iframe_xml_loaded()
                        window.xml_handler = handler;
                        window.xml_url = url;
                        window.xml_data = data;
                        window.xml_result_handler = result_handler;
                        return true;
                }

                clearTimeout(watchdog_id);
                debug("Browser incompatilibity: can't request XML document by one of supported methods");
                return false;
        }

        catch(ex) {
                clearTimeout(watchdog_id);
                //debug("clearTimeout = "+watchdog_id);
                debug ("CAUGHT EXCEPTION!");
                result_handler(new Array());
                return false;
        }

        return true;
}

function iframe_xml_loaded() {
        debug("iframe_xmldoc_loaded");
        if (! window.frames['xml_iframe']) return;
        var xml = eval('window.frames.xml_iframe.window.document');
        if (xml) {
        clearTimeout(window.iframe_timeout);
                debug("calling handler with ("+window.xml_url+","+window.xml_data+",...)");
                window.xml_handler(window.frames.xml_iframe.window.document, window.xml_url, window.xml_data, window.xml_result_handler);
        } else {
                debug("can't eval iframe with xml");
        }
}

var data = new Array();

function loadData_intersect(xmldoc, url, pos, result_handler) {
        data = new Array();
        if (loadData(xmldoc, url, pos)) {
                intersect_results(data); 
                search_query_left(result_handler);
        } else {
                debug("INTERNAL ERROR, Inconsistent index");
                search_err="INTERNAL ERROR, Inconsistent index";
        }
}

function loadData(xmldoc, url, pos) {

        clearTimeout(watchdog_id);
        //debug("clearTimeout = "+watchdog_id);

        debug ("loadData("+url+","+pos+")");

        // Get all entries
        var entries = xmldoc.getElementsByTagName("e");

        if (entries.length > pos) {
                // Get the links associated with this query
                var links = entries[pos].getElementsByTagName("l");

                debug("loaded "+links.length+" links");

                // Dynamically append results to output
                var ret = false;
                for(i=0; i<links.length; i++) {
                        data.push(new Result(
                                links[i].getAttribute("t"),
                                links[i].firstChild.data,
                                links[i].getAttribute("f"))
                        );
                        ret = true;
                }
                return ret;
        } else {
                debug("ERROR: seek to "+pos+" with only "+entries.length+" elements");
        }
}


function search_query_left(result_handler) {
        if (query_left.length > 0) {
                doSearch(index_path, query_left, result_handler);  
        } else {
                results.sort(sortResults);
                result_handler(results);
        }
}

// you may override this function to sort by something else
function sortResults(a, b) {
        return a.frequency - b.frequency;
}

function end_traverseTree(wildcard,query,result_handler) {
        if (! wildcard) {
                debug("Unable to locate key "+query);
                result_handler(new Array());
        } else {
                debug("wildcard "+query+" produced "+(wildcard_url_pos.length / 2)+" results: "+wildcard_url_pos.join(" "));
        }

}

function traverseTree(xmldoc, url, query, result_handler) {
        clearTimeout(watchdog_id);
        //debug("clearTimeout = "+watchdog_id);
 
        debug("traverseTree("+xmldoc+","+url+","+query+")");

        var keys = xmldoc.getElementsByTagName("k");
        var i;

        // support for wildcard
        var qlen = query.length;
        var wildcard = false;
        var query_full = query;
        if (query.charAt(qlen-1) == '*') {
                wildcard = true;
                query = query.substr(0,--qlen);
                debug("using wildcard "+query+"*");
        }

        for(i = 0; i < keys.length; i++) {
                var key = keys[i].firstChild.data;

                if (wildcard) key = key.substr(0,qlen);

                debug("? "+key+" -- "+query);

                if (key != '' && key != null) {
                        // Case where current key is greater than query, descend
                        if (key > query) {
                                if (! loadXML(url.replace(".xml","/"+convert(i)+".xml"), traverseTree, query_full, result_handler)) {
                                        end_traverseTree(wildcard, query_full, result_handler);
                                }
                                // make sure of garbage collection
                                xmldoc=null;
                                return;
                        }
                        // Found it!
                        else if (key==query) {
                                if (wildcard) {
                                        wildcard_url_pos.push(url.replace(/(\w+\.xml)/, "_$1"));
                                        wildcard_url_pos.push(i);
                                        debug("+"+i+": "+keys[i].firstChild.data);
                                } else {
                                        // exact match
                                        if (! loadXML(url.replace(/(\w+\.xml)/, "_$1"), loadData_intersect, i, result_handler)) {
                                                debug("ERROR: Unable to locate data "+query_full);
                                                result_handler(new Array());
                                        }
                                        // make sure of garbage collection
                                        xmldoc=null;
                                        return;
                                }
                        } // key < query
                } // if key
        } // for

        // Look past the end...
        if (keys.length == 0 || !loadXML(url.replace(".xml","/"+convert(i)+".xml"), traverseTree, query_full, result_handler)) {
                end_traverseTree(wildcard, query_full, result_handler);
        }

        // make sure of garbage collection
        xmldoc=null;
        return;
}

function doSearch(index_name,query, result_func) {

        //alert("doSearch("+index_name+","+query+")");
        var pos=query.search(/[\s\+]/);
        if (index_name) index_path = index_name+'/';

        if (pos < 0) {
                query_left = "";
        } else {
                query_left = query.slice(pos+1);
                query = query.slice(0,pos);
        } 

        if (! loadXML(index_path+"0.xml", traverseTree, query.toLowerCase(), result_func)) {
                debug("ERROR: Couldn't find main index 0.xml");
                search_err = "INTERNAL ERROR: Unable to load main index 0.xml";
        }
}
1	dpavlin	391	/**
2			search.js searchs an XML index to HTML files.
3
4			A part of the jsfind project (http://projects.elucidsoft.net/jsfind)
5			Copyright (C) 2003 Shawn Garbett
6
7			This program is free software; you can redistribute it and/or
8			modify it under the terms of the GNU General Public License
9			as published by the Free Software Foundation; either version 2
10			of the License, or (at your option) any later version.
11
12			This program is distributed in the hope that it will be useful,
13			but WITHOUT ANY WARRANTY; without even the implied warranty of
14			MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
15			GNU General Public License for more details.
16
17			You should have received a copy of the GNU General Public License
18			along with this program; if not, write to the Free Software
19			Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
20
21			Contact Info:
22			Shawn Garbett <Shawn@eLucidSoft.net>
23			http://www.elucidsoft.net
24			4037 General Bate Drive
25			Nashville, TN 37204
26			*/
27
28			// Constants
29			var conversion = new String
30	dpavlin	560	("0123456789abcdefghijklmnopqrstuvwxyz");
31	dpavlin	391
32			// State variables
33			var query_left = "";
34			var search_err = "";
35			var results = null;
36			var index_path = "";
37
38	dpavlin	535	// wildcard data about found keys URLs and position
39			var wildcard_url_pos = new Array();
40
41	dpavlin	391	var watchdog_id = 0;
42			var watchdog_callback = null;
43
44			// Object to hold search results
45	dpavlin	534	function Result(title, link, freq) {
46			this.title=title;
47			this.link=link;
48			this.frequency=Number(freq);
49	dpavlin	391	}
50
51			// Function to merge (intersect) two result sets
52	dpavlin	534	function intersect_results(data) {
53			// If there are no stored results, then these are the results
54			if (! results) {
55			results = data;
56			return;
57			}
58	dpavlin	391
59	dpavlin	534	var output=new Array();
60	dpavlin	391
61	dpavlin	534	// There are existing results, to do an intersect...
62			for (var i=0; i<results.length; i++) {
63			for (var j=0; j<data.length; j++) {
64			if (data[j].title == results[i].title) {
65			results[i].frequency += data[j].frequency;
66			output.push(results[i]);
67			break;
68			}
69			}
70			}
71	dpavlin	391
72	dpavlin	534	results = output;
73	dpavlin	391	}
74
75	dpavlin	496	var debug_div = null;
76
77	dpavlin	534	function debug(msg) {
78	dpavlin	391
79	dpavlin	534	// return; // Disable debugging
80	dpavlin	391
81	dpavlin	534	if (! debug_div) debug_div = document.getElementById('debug');
82
83			// this will create debug div if it doesn't exist.
84			if (! debug_div) {
85			debug_div = document.createElement('div');
86			if (document.body) document.body.appendChild(debug_div);
87			else debug_div = null;
88			}
89			if (debug_div) {
90			debug_div.appendChild(document.createTextNode(msg));
91			debug_div.appendChild(document.createElement("br"));
92			}
93	dpavlin	391	}
94
95	dpavlin	560	// Convert a number into a base x alphanumeric number string
96	dpavlin	534	function convert(num) {
97			var base = conversion.length;
98			var pow = 1;
99			var pos = 0;
100			var out = "";
101	dpavlin	391
102	dpavlin	534	if (num == 0) return "0";
103	dpavlin	391
104	dpavlin	534	while (num > 0) {
105			pos = num % base;
106			out = conversion.charAt(pos) + out;
107			num = Math.floor(num/base);
108			pow *= base;
109			}
110	dpavlin	391
111	dpavlin	534	return out;
112	dpavlin	391	}
113
114	dpavlin	534	function watchdog() {
115			debug ("TIMEOUT!");
116			watchdog_callback(new Array());
117	dpavlin	391	}
118
119	dpavlin	469	var xmldoc;
120
121	dpavlin	391	// This function loads the XML document from the specified URL, and when
122			// it is fully loaded, passes that document and the url to the specified
123			// handler function. This function works with any XML document
124
125	dpavlin	534	function loadXML(url, handler, data, result_handler) {
126			debug("loadXML("+url+","+data+")");
127	dpavlin	391
128	dpavlin	534	// Timeout operation in 10 seconds
129			watchdog_callback = result_handler;
130			watchdog_id=setTimeout("watchdog()", 20000);
131	dpavlin	391
132	dpavlin	535	//debug("setTimeout = "+watchdog_id);
133	dpavlin	391
134	dpavlin	534	try {
135			// Use the standard DOM Level 2 technique, if it is supported
136			if (document.implementation && document.implementation.createDocument) {
137			// Create a new Document object
138			xmldoc = document.implementation.createDocument("", "", null);
139	dpavlin	391
140	dpavlin	534	// Specify what should happen when it finishes loading
141			xmldoc.onload = function() { handler(xmldoc, url, data, result_handler); }
142	dpavlin	391
143	dpavlin	534	//xmldoc.onerror = docError;
144			//xmldoc.addEventListener("load",docError,false);
145	dpavlin	496
146	dpavlin	534	// And tell it what URL to load
147			xmldoc.load(url);
148			return true;
149			}
150			// Otherwise use Microsoft's proprietary API for Internet Explorer
151			// Something about not following standards once again
152			else if (window.ActiveXObject) {
153			xmldoc = new ActiveXObject("Microsoft.XMLDOM"); // Create doc.
154			if (! xmldoc) xmldoc = new ActiveXObject("MSXML2.DOMDocument"); // Create doc.
155			// Specify onload
156			xmldoc.onreadystatechange = function() {
157			if (xmldoc.readyState == 4) handler(xmldoc, url, data, result_handler);
158			}
159			xmldoc.load(url); // Start loading!
160			return true;
161			}
162			// else fallback on usage of iframes to load xml (Opera 7.53 without Java and maybe old Mac browsers)
163			else {
164			debug("using iframe xml loader - experimental and slow");
165			if (! window.xml_iframe) {
166			debug("creating iframe");
167			window.xml_iframe = document.createElement('div');
168			window.xml_iframe.innerHTML = '<iframe src="'+url+'" name="xml_iframe" height="0" width="0" style="display: none;"></iframe>';
169			document.body.appendChild(window.xml_iframe);
170			} else {
171			debug("loading xml in existing iframe");
172			window.frames.xml_iframe.window.document.location.href = url;
173			}
174	dpavlin	496
175	dpavlin	534	// set timeout to re-check if iframe is loaded
176			window.iframe_timeout = window.setInterval('iframe_xml_loaded();',100);
177	dpavlin	391
178	dpavlin	534	// save some data for iframe_xml_loaded()
179			window.xml_handler = handler;
180			window.xml_url = url;
181			window.xml_data = data;
182			window.xml_result_handler = result_handler;
183			return true;
184			}
185
186			clearTimeout(watchdog_id);
187			debug("Browser incompatilibity: can't request XML document by one of supported methods");
188			return false;
189			}
190
191			catch(ex) {
192			clearTimeout(watchdog_id);
193	dpavlin	535	//debug("clearTimeout = "+watchdog_id);
194	dpavlin	534	debug ("CAUGHT EXCEPTION!");
195			result_handler(new Array());
196			return false;
197			}
198
199			return true;
200	dpavlin	391	}
201
202	dpavlin	496	function iframe_xml_loaded() {
203	dpavlin	534	debug("iframe_xmldoc_loaded");
204			if (! window.frames['xml_iframe']) return;
205			var xml = eval('window.frames.xml_iframe.window.document');
206			if (xml) {
207	dpavlin	496	clearTimeout(window.iframe_timeout);
208	dpavlin	534	debug("calling handler with ("+window.xml_url+","+window.xml_data+",...)");
209			window.xml_handler(window.frames.xml_iframe.window.document, window.xml_url, window.xml_data, window.xml_result_handler);
210			} else {
211			debug("can't eval iframe with xml");
212			}
213	dpavlin	496	}
214
215	dpavlin	534	var data = new Array();
216	dpavlin	391
217	dpavlin	534	function loadData_intersect(xmldoc, url, pos, result_handler) {
218			data = new Array();
219	dpavlin	535	if (loadData(xmldoc, url, pos)) {
220	dpavlin	534	intersect_results(data);
221			search_query_left(result_handler);
222			} else {
223			debug("INTERNAL ERROR, Inconsistent index");
224			search_err="INTERNAL ERROR, Inconsistent index";
225			}
226			}
227	dpavlin	391
228	dpavlin	535	function loadData(xmldoc, url, pos) {
229	dpavlin	391
230	dpavlin	534	clearTimeout(watchdog_id);
231	dpavlin	535	//debug("clearTimeout = "+watchdog_id);
232	dpavlin	391
233	dpavlin	534	debug ("loadData("+url+","+pos+")");
234	dpavlin	391
235	dpavlin	534	// Get all entries
236			var entries = xmldoc.getElementsByTagName("e");
237	dpavlin	391
238	dpavlin	534	if (entries.length > pos) {
239			// Get the links associated with this query
240			var links = entries[pos].getElementsByTagName("l");
241	dpavlin	391
242	dpavlin	534	debug("loaded "+links.length+" links");
243
244			// Dynamically append results to output
245			var ret = false;
246			for(i=0; i<links.length; i++) {
247			data.push(new Result(
248			links[i].getAttribute("t"),
249			links[i].firstChild.data,
250			links[i].getAttribute("f"))
251			);
252			ret = true;
253			}
254			return ret;
255			} else {
256			debug("ERROR: seek to "+pos+" with only "+entries.length+" elements");
257			}
258	dpavlin	391	}
259
260	dpavlin	534
261			function search_query_left(result_handler) {
262			if (query_left.length > 0) {
263			doSearch(index_path, query_left, result_handler);
264			} else {
265			results.sort(sortResults);
266			result_handler(results);
267			}
268	dpavlin	391	}
269
270	dpavlin	534	// you may override this function to sort by something else
271			function sortResults(a, b) {
272			return a.frequency - b.frequency;
273			}
274
275	dpavlin	535	function end_traverseTree(wildcard,query,result_handler) {
276			if (! wildcard) {
277			debug("Unable to locate key "+query);
278			result_handler(new Array());
279			} else {
280			debug("wildcard "+query+" produced "+(wildcard_url_pos.length / 2)+" results: "+wildcard_url_pos.join(" "));
281			}
282
283			}
284
285	dpavlin	534	function traverseTree(xmldoc, url, query, result_handler) {
286			clearTimeout(watchdog_id);
287	dpavlin	535	//debug("clearTimeout = "+watchdog_id);
288	dpavlin	391
289	dpavlin	534	debug("traverseTree("+xmldoc+","+url+","+query+")");
290	dpavlin	391
291	dpavlin	534	var keys = xmldoc.getElementsByTagName("k");
292			var i;
293	dpavlin	391
294	dpavlin	534	// support for wildcard
295			var qlen = query.length;
296			var wildcard = false;
297			var query_full = query;
298			if (query.charAt(qlen-1) == '*') {
299			wildcard = true;
300			query = query.substr(0,--qlen);
301			debug("using wildcard "+query+"*");
302			}
303
304			for(i = 0; i < keys.length; i++) {
305			var key = keys[i].firstChild.data;
306
307	dpavlin	535	if (wildcard) key = key.substr(0,qlen);
308	dpavlin	534
309	dpavlin	535	debug("? "+key+" -- "+query);
310	dpavlin	534
311			if (key != '' && key != null) {
312			// Case where current key is greater than query, descend
313			if (key > query) {
314	dpavlin	535	if (! loadXML(url.replace(".xml","/"+convert(i)+".xml"), traverseTree, query_full, result_handler)) {
315			end_traverseTree(wildcard, query_full, result_handler);
316	dpavlin	534	}
317	dpavlin	535	// make sure of garbage collection
318			xmldoc=null;
319			return;
320	dpavlin	534	}
321			// Found it!
322			else if (key==query) {
323			if (wildcard) {
324	dpavlin	535	wildcard_url_pos.push(url.replace(/(\w+\.xml)/, "_$1"));
325			wildcard_url_pos.push(i);
326			debug("+"+i+": "+keys[i].firstChild.data);
327	dpavlin	534	} else {
328			// exact match
329			if (! loadXML(url.replace(/(\w+\.xml)/, "_$1"), loadData_intersect, i, result_handler)) {
330	dpavlin	535	debug("ERROR: Unable to locate data "+query_full);
331	dpavlin	534	result_handler(new Array());
332			}
333			// make sure of garbage collection
334			xmldoc=null;
335			return;
336			}
337			} // key < query
338			} // if key
339			} // for
340
341			// Look past the end...
342	dpavlin	535	if (keys.length == 0 \|\| !loadXML(url.replace(".xml","/"+convert(i)+".xml"), traverseTree, query_full, result_handler)) {
343			end_traverseTree(wildcard, query_full, result_handler);
344	dpavlin	534	}
345
346			// make sure of garbage collection
347			xmldoc=null;
348			return;
349	dpavlin	391	}
350
351	dpavlin	534	function doSearch(index_name,query, result_func) {
352	dpavlin	391
353	dpavlin	534	//alert("doSearch("+index_name+","+query+")");
354			var pos=query.search(/[\s\+]/);
355			if (index_name) index_path = index_name+'/';
356	dpavlin	391
357	dpavlin	534	if (pos < 0) {
358			query_left = "";
359			} else {
360			query_left = query.slice(pos+1);
361			query = query.slice(0,pos);
362			}
363
364			if (! loadXML(index_path+"0.xml", traverseTree, query.toLowerCase(), result_func)) {
365			debug("ERROR: Couldn't find main index 0.xml");
366			search_err = "INTERNAL ERROR: Unable to load main index 0.xml";
367			}
368	dpavlin	391	}