Improve BubbleTree colors
Improve BubbleTree colors

file:b/.gitmodules (new)
--- /dev/null
+++ b/.gitmodules
@@ -1,1 +1,4 @@
+[submodule "lib/bubbletree"]
+	path = lib/bubbletree
+	url = https://github.com/okfn/bubbletree.git
 

--- a/admin/displayUNSPSC.php
+++ b/admin/displayUNSPSC.php
@@ -1,11 +1,7 @@
 <?php

 error_reporting(E_ALL);

+  include_once("../lib/common.inc.php");

 

-$link = mysql_connect('localhost', 'team7', 'wlUsnD2xu9');

-if (!$link) {

-    die('Could not connect: ' . mysql_error());

-}

-@mysql_select_db("team7") or die("Unable to select database");

 

 $unspscresult = mysql_query ("select * from UNSPSCcategories;");

 while ($row = mysql_fetch_assoc($unspscresult)) {

@@ -17,7 +13,7 @@
 echo "<table>";

 while ($row = mysql_fetch_assoc($catsresult)) {

 	$catName = $unspsc[$row['cat']."000000"].$row['cat'];

-	if ($row['cat'] = "") $catName = "null";

+	if ($row['cat'] == "") $catName = "null";

 	

 	echo "<tr><td>$catName</td><td>".$row['value']."</td></tr>";

 }


--- /dev/null
+++ b/displayBubbletree.php
@@ -1,1 +1,109 @@
 
+<!DOCTYPE html> 
+<html xmlns="http://www.w3.org/1999/xhtml"> 
+<head> 
+	<meta charset="UTF-8"/> 
+	<title>Minimal BubbleTree Demo</title> 
+	<script type="text/javascript" src="lib/bubbletree/lib/jquery-1.5.2.min.js"></script> 
+	<script type="text/javascript" src="lib/bubbletree/lib/jquery.history.js"></script> 
+	<script type="text/javascript" src="lib/bubbletree/lib/raphael.js"></script> 
+	<script type="text/javascript" src="lib/bubbletree/lib/vis4.js"></script> 
+	<script type="text/javascript" src="lib/bubbletree/lib/Tween.js"></script> 
+	<script type="text/javascript" src="lib/bubbletree/build/bubbletree.js"></script> 
+	<link rel="stylesheet" type="text/css" href="lib/bubbletree/build/bubbletree.css" /> 
+	<script type="text/javascript" src="lib/bubbletree/styles/cofog.js"></script> 
+	
+	
+	<script type="text/javascript"> 
+       
+		$(function() {
+		<?php
+include_once ("lib/common.inc.php");
+
+include("lib/Color.php");
+$color = new Lux_Color();
+
+
+$unspscresult = mysql_query("select * from UNSPSCcategories;");
+while ($row = mysql_fetch_assoc($unspscresult)) {
+	$unspsc[$row['UNSPSC']] = $row['Title'];
+}
+$total = 0;
+$cats = 0;
+$catsresult = mysql_query("SELECT LEFT( categoryUNSPSC, 1 ) as cat ,
+SUM( value ) as value
+FROM `contractnotice`
+WHERE childCN = 0
+GROUP BY cat ;");
+$nodes = Array();
+while ($row = mysql_fetch_assoc($catsresult)) {
+$cats++;
+$catColor = $color->hsl2hex(Array($cats/10, .7, .5));
+	$catName = $unspsc[$row['cat'] . "0000000"] . $row['cat'];
+	if ($row['cat'] == "") $catName = "null";
+	$subnodes = Array();
+	$cattwosresult = mysql_query("SELECT LEFT( categoryUNSPSC, 2 ) as cat ,
+SUM( value ) as value
+FROM `contractnotice`
+WHERE childCN = 0 and LEFT( categoryUNSPSC, 1 ) = '{$row['cat']}'
+GROUP BY cat ;");
+	while ($tworow = mysql_fetch_assoc($cattwosresult)) {
+		$subcatName = $unspsc[$tworow['cat'] . "000000"] . $tworow['cat'];
+		if ($tworow['cat'] == "") $subcatName = "null";
+		$subsubnodes = Array();
+		$catthreesresult = mysql_query("SELECT LEFT( categoryUNSPSC, 3 ) as cat ,
+SUM( value ) as value
+FROM `contractnotice`
+WHERE childCN = 0 and LEFT( categoryUNSPSC, 2 ) = '{$tworow['cat']}'
+GROUP BY cat ;");
+		$subCatColor = $color->hsl2hex(Array($cats/10, rand(1,10)/10, .5));
+		while ($threerow = mysql_fetch_assoc($catthreesresult)) {
+			$subsubcatName = $unspsc[$threerow['cat'] . "00000"] . $threerow['cat'];
+			if ($threerow['cat'] == "") $subsubcatName = "null";
+			$subsubnodes[] = Array(
+				"label" => $subsubcatName,
+				"amount" => $threerow['value'],
+				"color" => "#".$subCatColor
+			);
+		}
+		$subnodes[] = Array(
+			"label" => $subcatName,
+			"amount" => $tworow['value'],
+			"color" => "#".$subCatColor,
+			"children" => $subsubnodes
+		);
+	}
+	$nodes[] = Array(
+		"label" => $catName,
+		"amount" => $row['value'],
+		"color" => "#".$catColor,
+		"children" => $subnodes
+	);
+	$total+= $row['value'];
+}
+$data = Array(
+	"label" => "Australian Federal Government Contract Spending",
+	"amount" => $total,
+	"color" => "#000000",
+	"children" => $nodes
+);
+echo "var data =eval('('+'" . json_encode($data) . "'+')');";
+?>
+
+			new BubbleTree({
+				data: data,
+				container: '.bubbletree'
+			});
+		
+			
+		});
+     
+	</script> 
+</head> 
+<body> 
+	<div class="bubbletree-wrapper"> 
+		<div class="bubbletree"></div> 
+	</div> 
+</body> 
+</html> 
+

--- a/displayContract.php
+++ b/displayContract.php
@@ -28,9 +28,17 @@
 echo "<br>";

 }

 }

-echo '<a href="https://www.tenders.gov.au/?event=public.advancedsearch.keyword&keyword=CN'.$_REQUEST['CNID'].'"> View original record @ tenders.gov.au</a>';

+echo '<br><a href="https://www.tenders.gov.au/?event=public.advancedsearch.keyword&keyword=CN'.$_REQUEST['CNID'].'"> View original record @ tenders.gov.au</a><br>';

 

 mysql_free_result($result);

+

+$query = "SELECT * FROM `heuristic_results` where CNID = ".$_REQUEST['CNID'];

+$result = mysql_query($query);

+if (!$result) echo mysql_error().$query;

+while ($r = mysql_fetch_array($result, MYSQL_ASSOC)) {

+	echo "<b>{$r['heuristic_name']}</b>: {$r['heuristic_value']} (raw value: {$r['raw_value']}, mean: {$r['mean']}, stddev: {$r['stddev']})<br>";

+}

+

 include_footer();

 ?>

 

--- a/exportData.csv.php
+++ b/exportData.csv.php
@@ -4,10 +4,11 @@
 setlocale(LC_CTYPE, 'C');
 // source: http://stackoverflow.com/questions/81934/easy-way-to-export-a-sql-table-without-access-to-the-server-or-phpmyadmin#81951
 $result = mysql_query('
-SELECT CNID,contractnotice.agencyName,agency.abn as agencyABN,publishDate,contractStart,contractEnd,value,description,procurementMethod,category,categoryUNSPSC
+SELECT CNID,contractnotice.agencyName,agency.abn as agencyABN,DATE(publishDate),contractStart,contractEnd,value,description,procurementMethod,category,categoryUNSPSC
 ,supplierABN,supplierName,supplierCity,supplierPostcode,supplierCountry,contactPostcode,
 IF(supplierABN != "",supplierABN,supplierName) as supplierID,
-concat("https://www.tenders.gov.au/?event=public.advancedsearch.keyword&keyword=CN",CNID) as sourceURL FROM `contractnotice` join agency on contractnotice.agencyName=agency.agencyName where childCN = 0');
+concat("https://www.tenders.gov.au/?event=public.advancedsearch.keyword&keyword=CN",CNID) as sourceURL 
+FROM `contractnotice` join agency on contractnotice.agencyName=agency.agencyName where childCN = 0');
 if (!$result) die('Couldn\'t fetch records');
 $num_fields = mysql_num_fields($result);
 $headers = array();

--- a/heuristics/dateHeuristics.php
+++ b/heuristics/dateHeuristics.php
@@ -1,4 +1,102 @@
-        - long contract period (number of weeks/days?)
-            - Reported late
-        - 45 days? A late contract is a dodgy contract except maybe for variations?
+<?php
+//long contract period (number of weeks/days?)
+$heuristics["DATE_LONG_CONTRACT_PERIOD"] = Array(
+	"description" => "long contract period (number of weeks/days?)"
+);
+function DATE_LONG_CONTRACT_PERIOD($cn)
+{
+	$averageContractPeriod = getAverageContractPeriod();
+	$stddevContractPeriod = getstddevContractPeriod();
+	$diff = strtotime($cn['contractEnd']) - strtotime($cn['contractStart']);
+	$days = intval($diff / (60 * 60 * 24));
+	$value = abs($days - $averageContractPeriod) / $stddevContractPeriod;
+	return Array(
+		"heuristic_value" => $value,
+		"raw_value" => $days,
+		"mean" => $averageContractPeriod,
+		"stddev" => $stddevContractPeriod
+	);
+}
+$averageContractPeriod;
+function getAverageContractPeriod()
+{
+	global $averageContractPeriod, $stddevContractPeriod;
+	if (!$averageContractPeriod) {
+		getStddevAverageContractPeriod();
+	}
+	return $averageContractPeriod;
+}
+$stddevContractPeriod;
+function getstddevContractPeriod()
+{
+	global $averageContractPeriod, $stddevContractPeriod;
+	if (!$stddevContractPeriod) {
+		getStddevAverageContractPeriod();
+	}
+	return $stddevContractPeriod;
+}
+function getStddevAverageContractPeriod()
+{
+	global $averageContractPeriod, $stddevContractPeriod;
+	$query = "select AVG(dateDiff(contractEnd,contractStart)),stddev(dateDiff(contractEnd,contractStart)) from contractnotice";
+	$result = mysql_query($query);
+	$r = mysql_fetch_array($result, MYSQL_BOTH);
+	$averageContractPeriod = $r[0];
+	$stddevContractPeriod = $r[1];
+}
 
+
+//Reported late, 45 days? A late contract is a dodgy contract except maybe for variations?
+$heuristics["DATE_REPORTED_LATE"] = Array(
+	"description" => "Reported late, 45 days?"
+);
+function DATE_REPORTED_LATE($cn)
+{
+	$averageDaysLate = getAverageDaysLate();
+	$stddevDaysLate = getStddevDaysLate();
+	$diff = strtotime($cn['publishDate']) - strtotime($cn['contractStart']);
+	$days = intval($diff / (60 * 60 * 24));
+	if ($days <= 0) {
+		$value = 0;
+	}
+	else {
+		// +1 demerit for exceeding 45 day requirement
+		$value = (abs($days - $averageDaysLate) / $stddevDaysLate) + ($days < 45 ? 0 : 1);
+	}
+	return Array(
+		"heuristic_value" => $value,
+		"raw_value" => $days,
+		"mean" => $averageDaysLate,
+		"stddev" => $stddevDaysLate
+	);
+}
+$averageDaysLate;
+function getAverageDaysLate()
+{
+	global $averageDaysLate;
+	if (!$averageDaysLate) {
+            getDaysLate();
+	}
+	return $averageDaysLate;
+}
+$stddevDaysLate;
+function getStddevDaysLate()
+{
+	global $stddevDaysLate;
+	if (!$stddevDaysLate) {
+	getDaysLate();
+	}
+	return $stddevDaysLate;
+}
+function getDaysLate() {
+    
+	global $averageDaysLate,$stddevDaysLate;
+    
+    		$query = "select AVG(dateDiff(publishDate,contractStart)), STDDEV(dateDiff(publishDate,contractStart)) from contractnotice";
+		$result = mysql_query($query);
+		$r = mysql_fetch_array($result, MYSQL_BOTH);
+		$averageDaysLate = $r[0];
+                	$stddevDaysLate = $r[1];
+
+}
+?>

--- /dev/null
+++ b/heuristics/heuristics.inc.php
@@ -1,1 +1,40 @@
+<?php
+include_once ("../lib/common.inc.php");
+$heuristics = Array();
+//each heuristic adds self to description array
+include ("dateHeuristics.php");
+include ("historyHeuristics.php");
+//include ("metadataHeuristics.php");
+//include ("valueHeuristics.php");
+function runHeuristic($heuristicName, $cn)
+{
+	// check  if already ran
+	$query = "select count(*) from heuristic_results where heuristic_name = '$heuristicName' and CNID = '{$cn['CNID']}'";
+	$result = mysql_query($query);
+	$r = mysql_fetch_array($result);
+	if ($r[0] == 0) {
+		// if not, run now
+		$hresults = call_user_func($heuristicName, $cn);
+		if (!isset($hresults["heuristic_value"]) || !isset($hresults["raw_value"]) || !isset($hresults["mean"]) || !isset($hresults["stddev"])) {
+			print_r($hresults);
+			die("Missing field in heurtistic $heuristicName result");
+		}
+		$query = "insert into heuristic_results values('$heuristicName',
+    '{$hresults["heuristic_value"]}',
+     '{$hresults["raw_value"]}',
+      '{$hresults["mean"]}',
+       '{$hresults["stddev"]}',
+           '{$cn["CNID"]}',
+           NOW(),
+           '{$cn["publishDate"]}',
+           '{$cn["agencyABN"]}',
+           '{$cn["supplierID"]}'
+    )";
+		// save value and cn data via sql
+		$result = mysql_query($query);
+		if ($result) echo "Saved $heuristicName for {$cn["CNID"]} <br>\n";
+		elseif (strpos(mysql_error() , "Duplicate entry") === false) echo $hresults . " failed insert.<br>" . mysql_error() . " <br>  $query <br><br>\n";
+	}
+}
+?>
 

--- a/heuristics/historyHeuristics.php
+++ b/heuristics/historyHeuristics.php
@@ -1,6 +1,126 @@
-      - unusual for agency/supplier
-        - previous low number of transactions 
-          - zero ie. new agency/supplier is huge score
-    - unusual value for time of year
-        - compare to all other records in last 2 weeks
-        - ie. many large contracts in june so takes more to standout
+<?php
+$heuristics["HISTORY_LOW_TRANSACTIONS_AGENCY"] = Array(
+	"description" => "unusual for agency due to previous low number of transactions "
+);
+function HISTORY_LOW_TRANSACTIONS_AGENCY($cn)
+{
+	$thisAgencyTransactions = getAgencyTransactions($cn['agencyName']);
+	$averageAgencyTransactions = getAverageAgencyTransactions();
+	$stddevAgencyTransactions = getstddevAgencyTransactions();
+	$diff = strtotime($cn['contractEnd']) - strtotime($cn['contractStart']);
+	$days = intval($diff / (60 * 60 * 24));
+	$value = abs($days - $averageAgencyTransactions) / $stddevAgencyTransactions;
+	return Array(
+		"heuristic_value" => $value,
+		"raw_value" => $days,
+		"mean" => $averageAgencyTransactions,
+		"stddev" => $stddevAgencyTransactions
+	);
+}
+$agencyTransactions = Array();
+function getAgencyTransactions($agencyName)
+{
+	global $agencyTransactions;
+	if (!$agencyTransactions[$agencyName]) {
+		$query = 'select count(*) from contractnotice where agencyName = "' . $agencyName . '"';
+		$result = mysql_query($query);
+		$r = mysql_fetch_array($result, MYSQL_BOTH);
+		$agencyTransactions[$agencyName] = $r[0];
+	}
+	return $agencyTransactions[$agencyName];
+}
+$averageAgencyTransactions;
+function getAverageAgencyTransactions()
+{
+	global $averageAgencyTransactions;
+	if (!$averageAgencyTransactions) {
+		getStatsAgencyTransactions();
+	}
+	return $averageAgencyTransactions;
+}
+$stddevAgencyTransactions;
+function getstddevAgencyTransactions()
+{
+	global $stddevAgencyTransactions;
+	if (!$stddevAgencyTransactions) {
+		getStatsAgencyTransactions();
+	}
+	return $stddevAgencyTransactions;
+}
+function getStatsAgencyTransactions()
+{
+	global $averageAgencyTransactions, $stddevAgencyTransactions;
+	$query = "select avg(count), STDDEV(count) from (select count(*) as count
+                from contractnotice group by agencyName) as a;";
+	$result = mysql_query($query);
+	$r = mysql_fetch_array($result, MYSQL_BOTH);
+	$averageAgencyTransactions = $r[0];
+	$stddevAgencyTransactions = $r[1];
+}
+$heuristics["HISTORY_LOW_TRANSACTIONS_SUPPLIER"] = Array(
+	"description" => "unusual for supplier due to previous low number of transactions "
+);
+function HISTORY_LOW_TRANSACTIONS_SUPPLIER($cn)
+{
+	$thisSupplierTransactions = getSupplierTransactions($cn['supplierName'], $cn['supplierABN']);
+	$averageSupplierTransactions = getAverageSupplierTransactions();
+	$stddevSupplierTransactions = getstddevSupplierTransactions();
+	$diff = strtotime($cn['contractEnd']) - strtotime($cn['contractStart']);
+	$days = intval($diff / (60 * 60 * 24));
+	$value = abs($days - $averageSupplierTransactions) / $stddevSupplierTransactions;
+	return Array(
+		"heuristic_value" => $value,
+		"raw_value" => $days,
+		"mean" => $averageSupplierTransactions,
+		"stddev" => $stddevSupplierTransactions
+	);
+}
+$supplierTransactions = Array();
+function getSupplierTransactions($supplierName, $supplierABN)
+{
+	global $supplierTransactions;
+	if ($supplierABN != 0 && $supplierABN != "") {
+		if (!$supplierTransactions[$supplierABN]) {
+			$query = 'select count(*) from contractnotice where supplierABN = "' . $supplierABN . '"';
+			$result = mysql_query($query);
+			$r = mysql_fetch_array($result, MYSQL_BOTH);
+			$supplierTransactions[$supplierABN] = $r[0];
+		}
+		return $supplierTransactions[$supplierABN];
+	}
+	if (!$supplierTransactions[$supplierName]) {
+		$query = 'select count(*) from contractnotice where supplierName = "' . $supplierName . '"';
+		$result = mysql_query($query);
+		$r = mysql_fetch_array($result, MYSQL_BOTH);
+		$supplierTransactions[$supplierName] = $r[0];
+	}
+	return $supplierTransactions[$supplierName];
+}
+$averageSupplierTransactions;
+function getAverageSupplierTransactions()
+{
+	global $averageSupplierTransactions;
+	if (!$averageSupplierTransactions) {
+		getStatsSupplierTransactions();
+	}
+	return $averageSupplierTransactions;
+}
+$stddevSupplierTransactions;
+function getstddevSupplierTransactions()
+{
+	global $stddevSupplierTransactions;
+	if (!$stddevSupplierTransactions) {
+		getStatsSupplierTransactions();
+	}
+	return $stddevSupplierTransactions;
+}
+function getStatsSupplierTransactions()
+{
+	global $averageSupplierTransactions, $stddevSupplierTransactions;
+	$query = 'select avg(count), stddev(count) from (select IF(supplierABN != "",supplierABN,supplierName) as supplierID, count(*) as count from contractnotice group by supplierID) as a;';
+	$result = mysql_query($query);
+	$r = mysql_fetch_array($result, MYSQL_BOTH);
+	$averageSupplierTransactions = $r[0];
+	$stddevSupplierTransactions = $r[1];
+}
+

--- a/heuristics/metadataHeuristics.php
+++ b/heuristics/metadataHeuristics.php
@@ -1,3 +1,59 @@
-      - duplicated description
-        - most duplicated overall, most duplicated per agency/category/supplier etc.
-  
+ <?php
+/* all
+   SELECT description, count(*) as count
+FROM `contractnotice` 
+group by description having count > 1 order by count
+*/
+/*- duplicated description
+ - most duplicated overall, most duplicated per agency/category/supplier etc. */
+$heuristics["METADATA_DUPLICATED_DESCRIPTION"] = Array(
+	"description" => ""
+);
+function METADATA_DUPLICATED_DESCRIPTION($cn)
+{
+	$averageDuplicatedDescriptions = getAverageDuplicatedDescriptions();
+	$stddevDuplicatedDescriptions = getstddevDuplicatedDescriptions();
+	$query = 'select count(*) from contractnotice where description = "' . $agencyName . '"';
+	$result = mysql_query($query);
+	$r = mysql_fetch_array($result, MYSQL_BOTH);
+	$dupeDesc = $r[0];
+        if ($dupeDesc == 1) $value = 0;
+	else $value = abs($dupeDesc - $averageDuplicatedDescriptions) / $stddevDuplicatedDescriptions;
+	return Array(
+		"heuristic_value" => $value,
+		"raw_value" => $dupeDesc,
+		"mean" => $averageDuplicatedDescriptions,
+		"stddev" => $stddevDuplicatedDescriptions
+	);
+}
+$averageDuplicatedDescriptions;
+function getAverageDuplicatedDescriptions()
+{
+	global $averageDuplicatedDescriptions;
+	if (!$averageDuplicatedDescriptions) {
+		getStatsDuplicatedDescriptions();
+	}
+	return $averageDuplicatedDescriptions;
+}
+$stddevDuplicatedDescriptions;
+function getstddevDuplicatedDescriptions()
+{
+	global $stddevDuplicatedDescriptions;
+	if (!$stddevDuplicatedDescriptions) {
+		getStatsDuplicatedDescriptions();
+	}
+	return $stddevDuplicatedDescriptions;
+}
+function getStatsDuplicatedDescriptions()
+{
+	$query = "select avg(count),STDDEV(count) from (
+        SELECT description, count(*) as count
+FROM `contractnotice` 
+group by description having count > 1 
+        ) as a;";
+	$result = mysql_query($query);
+	$r = mysql_fetch_array($result, MYSQL_BOTH);
+	$averageDuplicatedDescriptions = $r[0];
+	$stddevDuplicatedDescriptions = $r[1];
+}
+

file:a/heuristics/readme.txt (deleted)
--- a/heuristics/readme.txt
+++ /dev/null
@@ -1,11 +1,1 @@
-heuristicResults
-heuristic = BY_DATE...
-value = 0.0
-cnPubDate = 
-lastUpdated = 1/1/1970
-contractNotice =
-agency = 
-supplier =
 
-work out total value by summing value
-

--- a/heuristics/runHeuristics.php
+++ b/heuristics/runHeuristics.php
@@ -1,6 +1,32 @@
-<?
-if agency
-if supplier
-if CN
+<?php
+include_once("heuristics.inc.php");
+$query = "SELECT *, agency.abn as agencyABN, IF(supplierABN != '',supplierABN,supplierName) as supplierID
+FROM contractnotice JOIN agency ON contractnotice.agencyName=agency.agencyName";
+
+$query = "SELECT *, agency.abn as agencyABN, IF(supplierABN != '',supplierABN,supplierName) as supplierID
+FROM contractnotice JOIN agency ON contractnotice.agencyName=agency.agencyName
+WHERE  DATE(importDate) = (select * from (SELECT DATE(importDate) 
+FROM contractnotice ORDER BY importDate DESC limit 1) alias)";
+$result = mysql_query($query);
+if (!$result) echo mysql_error().$query;
+while ($cn = mysql_fetch_array($result, MYSQL_BOTH)) {
+	//get each new CN from latest update
+	foreach ($heuristics as $heuristic => $description) {
+		// run all heuristics
+		runHeuristic($heuristic, $cn);
+	}
+        flush();
+}
+/*foreach agency
+
+aggregate agency metrics
+
+foreach supplier
+
+aggreate supplier metrics
+
+foreach CN
+
+aggregate CN metrics */
 ?>
 

--- a/heuristics/valueHeuristics.php
+++ b/heuristics/valueHeuristics.php
@@ -1,10 +1,58 @@
+<?php 
+      /*  - large contract value
+          - standard dev from mean/median
+          - percent of total contracts for supplier/agency*/
+   $heuristics["VALUE_LARGE_CONTRACT_OVERALL"] = Array(
+	"description" => "unusual value for time of year");
+function METADATA_DUPLICATED_DESCRIPTION($cn)
+{
+    	$averageContractPeriod = getAverageContractPeriod();
+	$diff = strtotime($cn['contractStart']) - strtotime($cn['publishDate']);
+	$days = intval($diff / 24);
+	return ($days > 45 ? 1 : 0);
+}
 
-        - large contract value
-          - chi-square test for outliers / standard dev from mean/median
-          - percent of total contracts for supplier/agency
+      /*  - peculiar value
+        - Just under 80k, amplified if other contracts with same supplier are just under
+      */
+         $heuristics["VALUE_NEAR_THRESHOLD"] = Array(
+	"description" => "unusual value for time of year");
+      /*
+    - unusual variation amount - absolute value; large reductions as well as large increases
+      */
+           $heuristics["VALUE_LARGE_VARIATION"] = Array(
+	"description" => "unusual value for time of year");
+function METADATA_DUPLICATED_DESCRIPTION($cn)
+{
+    	$averageContractPeriod = getAverageContractPeriod();
+	$diff = strtotime($cn['contractStart']) - strtotime($cn['publishDate']);
+	$days = intval($diff / 24);
+	return ($days > 45 ? 1 : 0);
+}
 
-
-        - peculiar value
-        - Just under 80k, amplified if other contracts with same supplier are just under
-    - unusual variation amount
-        - absolute value; large reductions as well as large increases
+/*   - unusual value for time of year
+        - compare to all other records in last 2 weeks
+        - ie. many large contracts in june so takes more to standout*/
+$heuristics["VALUE_HIGH_FOR_MONTH"] = Array(
+	"description" => "unusual value for time of year"
+);
+function VALUE_HIGH_FOR_MONTH($cn, $monthAsInt)
+{
+	$averageContractPeriod = getAverageContractPeriod();
+	$diff = strtotime($cn['contractStart']) - strtotime($cn['publishDate']);
+	$days = intval($diff / 24);
+	return ($days > 45 ? 1 : 0);
+}
+$monthlyValueAverage = Array();
+function getAgencyTransactions($agencyName)
+{
+	global $agencyTransactions;
+	if (!$agencyTransactions[$agencyName]) {
+		$query = 'select count(*) from contractnotice where agencyName = "' . $agencyName . '"';
+		$result = mysql_query($query);
+		$r = mysql_fetch_array($result, MYSQL_BOTH);
+		$agencyTransactions[$agencyName] = $r[0];
+	}
+	return $agencyTransactions[$agencyName];
+}
+?>

--- /dev/null
+++ b/heuristics/viewHeuristicsColormap.php
@@ -1,1 +1,60 @@
+<?php    
+  include_once("../lib/common.inc.php");
+  echo '<style>
+  div {
+  padding: 5px;
+  display: inline-block;
+  }
+  </style>';
+// http://www.herethere.net/~samson/php/color_gradient/color_gradient_generator.php.txt
+// return the interpolated value between pBegin and pEnd
+function interpolate($pBegin, $pEnd, $pStep, $pMax)
+{
+	if ($pBegin < $pEnd) {
+		return (($pEnd - $pBegin) * ($pStep / $pMax)) + $pBegin;
+	}
+	else {
+		return (($pBegin - $pEnd) * (1 - ($pStep / $pMax))) + $pEnd;
+	}
+}
+function Gradient($HexFrom, $HexTo, $ColorSteps)
+{
+	$theColorBegin = hexdec($HexFrom);
+	$theColorEnd = hexdec($HexTo);
+	$theNumSteps = intval($ColorSteps);
+	$theR0 = ($theColorBegin & 0xff0000) >> 16;
+	$theG0 = ($theColorBegin & 0x00ff00) >> 8;
+	$theB0 = ($theColorBegin & 0x0000ff) >> 0;
+	$theR1 = ($theColorEnd & 0xff0000) >> 16;
+	$theG1 = ($theColorEnd & 0x00ff00) >> 8;
+	$theB1 = ($theColorEnd & 0x0000ff) >> 0;
+	$GradientColors = array();
+	// generate gradient swathe now
+	for ($i = 0; $i <= $theNumSteps; $i++) {
+		$theR = interpolate($theR0, $theR1, $i, $theNumSteps);
+		$theG = interpolate($theG0, $theG1, $i, $theNumSteps);
+		$theB = interpolate($theB0, $theB1, $i, $theNumSteps);
+		$theVal = ((($theR << 8) | $theG) << 8) | $theB;
+		$GradientColors[] = sprintf("%06X", $theVal);
+	}
+	return $GradientColors;
+}
+$Gradients = Gradient("66FF00" , "FF0000" , 10); 
+  
+$query = "select max(sum) from (SELECT sum(heuristic_value) 
+as sum FROM heuristic_results group by CNID) as a";
+$result = mysql_query($query);
+$r = mysql_fetch_array($result, MYSQL_BOTH);
+$maxVal = $r[0];
+  
+$query = "SELECT sum(heuristic_value) as sum, CNID
+FROM `heuristic_results` group by CNID order by sum DESC LIMIT 300";
+$result = mysql_query($query);
+if (!$result) echo mysql_error().$query;
+while ($r = mysql_fetch_array($result, MYSQL_BOTH)) {
+    echo '<div style="background: #'.$Gradients[floor(($r['sum']/$maxVal) * 10)].';">';
+    echo '<a title="'.$r['sum'].'" href="../displayContract.php?CNID='.$r['CNID'].'">X</a>';
+    echo "</div>";
+}
 
+?>

--- /dev/null
+++ b/heuristics/viewHeuristicsDistribution.php
@@ -1,1 +1,82 @@
+<?php    
 
+/*// most interesting
+SELECT sum(heuristic_value) as sum, CNID
+FROM `heuristic_results` group by CNID order by sum DESC limit 30
+
+// spread of values
+select floor(sum) as val,count(*) from (SELECT sum(heuristic_value) 
+as sum FROM heuristic_results group by CNID) as a group by val*/
+
+ /* CAT:Spline chart */ 
+
+ /* pChart library inclusions */ 
+ include("../lib/pChart2.1.0/class/pData.class.php"); 
+ include("../lib/pChart2.1.0/class/pDraw.class.php"); 
+ include("../lib/pChart2.1.0/class/pImage.class.php"); 
+
+$series = Array();
+
+  include_once("../lib/common.inc.php");
+$query = "select heuristic_name, floor(heuristic_value) as val,count(*) from heuristic_results group by heuristic_name, val";
+$result = mysql_query($query);
+if (!$result) echo mysql_error().$query;
+while ($r = mysql_fetch_array($result, MYSQL_BOTH)) {
+    $series[$r["heuristic_name"]][$r["val"]] = $r[2];
+}
+
+ /* Create and populate the pData object */ 
+ $MyData = new pData();
+ $labels = Array();