Adodb library updated to 5.20

libraries/adodb/README.md

+ADOdb Library for PHP5
+======================
+
+[![Join chat on Gitter](https://img.shields.io/gitter/room/form-data/form-data.svg)](https://gitter.im/adodb/adodb?utm_source=badge&utm_medium=badge&utm_campaign=pr-badge&utm_content=badge)
+[![Download ADOdb](https://img.shields.io/sourceforge/dm/adodb.svg)](https://sourceforge.net/projects/adodb/files/latest/download)
+
+(c) 2000-2013 John Lim (jlim@natsoft.com)  
+(c) 2014      Damien Regad, Mark Newnham and the ADOdb community
+
+Released under both [BSD 3-Clause](https://github.com/ADOdb/ADOdb/blob/master/LICENSE.md#bsd-3-clause-license) 
+and [GNU Lesser GPL library 2.1](https://github.com/ADOdb/ADOdb/blob/master/LICENSE.md#gnu-lesser-general-public-license) 
+licenses.  
+This means you can use it in proprietary products; 
+see [License](https://github.com/ADOdb/ADOdb/blob/master/LICENSE.md) for details.
+
+Home page: http://adodb.org/
+
+> **WARNING: known issue with Associative Fetch Mode in ADOdb v5.19
+-- PLEASE UPGRADE TO v5.20 !**  
+> When fetching data in Associative mode (i.e. when `$ADODB_FETCH_MODE` is
+> set to *ADODB_FETCH_ASSOC*), recordsets do not return any data (empty strings)
+> when using some database drivers. The problem has been reported on MSSQL,
+> Interbase and Foxpro, but possibly affects other drivers as well; all drivers
+> derived from the above are also impacted.
+> For further details, please refer to [Issue #20](https://github.com/ADOdb/ADOdb/issues/20).
+
+
+Introduction
+============
+
+PHP's database access functions are not standardized. This creates a
+need for a database class library to hide the differences between the
+different databases (encapsulate the differences) so we can easily
+switch databases.
+
+The library currently supports MySQL, Interbase, Sybase, PostgreSQL, Oracle,
+Microsoft SQL server,  Foxpro ODBC, Access ODBC, Informix, DB2,
+Sybase SQL Anywhere, generic ODBC and Microsoft's ADO.
+
+We hope more people will contribute drivers to support other databases.
+
+
+Installation
+============
+
+Unpack all the files into a directory accessible by your web server.
+
+To test, try modifying some of the tutorial examples.
+Make sure you customize the connection settings correctly.
+
+You can debug using:
+
+``` php
+<?php
+include('adodb/adodb.inc.php');
+
+$db = ADONewConnection($driver); # eg. 'mysql' or 'oci8'
+$db->debug = true;
+$db->Connect($server, $user, $password, $database);
+$rs = $db->Execute('select * from some_small_table');
+print "<pre>";
+print_r($rs->GetRows());
+print "</pre>";
+```
+
+
+Documentation and Examples
+==========================
+
+Refer to the [ADOdb website](http://adodb.org/) for library documentation and examples. The documentation can also be [downloaded for offline viewing](https://sourceforge.net/projects/adodb/files/Documentation/).
+
+- [Main documentation](http://adodb.org/dokuwiki/doku.php?id=v5:userguide:userguide_index): Query, update and insert records using a portable API.
+- [Data dictionary](http://adodb.org/dokuwiki/doku.php?id=v5:dictionary:dictionary_index) describes how to create database tables and indexes in a portable manner.
+- [Database performance monitoring](http://adodb.org/dokuwiki/doku.php?id=v5:performance:performance_index) allows you to perform health checks, tune and monitor your database.
+- [Database-backed sessions](http://adodb.org/dokuwiki/doku.php?id=v5:session:session_index).
+
+There is also a [tutorial](http://adodb.org/dokuwiki/doku.php?id=v5:userguide:mysql_tutorial) that contrasts ADOdb code with PHP native MySQL code.
+
+
+Files
+=====
+
+- `adodb.inc.php` is the library's main file. You only need to include this file.
+- `adodb-*.inc.php` are the database specific driver code.
+- `adodb-session.php` is the PHP4 session handling code.
+- `test.php` contains a list of test commands to exercise the class library.
+- `testdatabases.inc.php` contains the list of databases to apply the tests on.
+- `Benchmark.php` is a simple benchmark to test the throughput of a SELECT
+statement for databases described in testdatabases.inc.php. The benchmark
+tables are created in test.php.
+
+
+Support
+=======
+
+To discuss with the ADOdb development team and users, connect to our
+[Gitter chatroom](https://gitter.im/adodb/adodb) using your Github credentials.
+
+Please report bugs, issues and feature requests on Github:
+
+https://github.com/ADOdb/ADOdb/issues
+
+You may also find legacy issues in
+
+- the old [ADOdb forums](http://phplens.com/lens/lensforum/topics.php?id=4) on phplens.com
+- the [SourceForge tickets section](http://sourceforge.net/p/adodb/_list/tickets)
+
+However, please note that they are not actively monitored and should
+only be used as reference.

libraries/adodb/adodb-csvlib.inc.php

@@ -6,19 +6,21 @@ if (!defined('ADODB_DIR')) die();
 global $ADODB_INCLUDED_CSV;
 $ADODB_INCLUDED_CSV = 1;
 
-/* 
+/*
 
-  V4.90 8 June 2006  (c) 2000-2006 John Lim (jlim#natsoft.com.my). All rights reserved.
-  Released under both BSD license and Lesser GPL library license. 
-  Whenever there is any discrepancy between the two licenses, 
-  the BSD license will take precedence. See License.txt. 
+  @version   v5.20.9  21-Dec-2016
+  @copyright (c) 2000-2013 John Lim (jlim#natsoft.com). All rights reserved.
+  @copyright (c) 2014      Damien Regad, Mark Newnham and the ADOdb community
+  Released under both BSD license and Lesser GPL library license.
+  Whenever there is any discrepancy between the two licenses,
+  the BSD license will take precedence. See License.txt.
   Set tabs to 4 for best viewing.
-  
+
   Latest version is available at http://adodb.sourceforge.net
-  
-  Library for CSV serialization. This is used by the csv/proxy driver and is the 
-  CacheExecute() serialization format. 
-  
+
+  Library for CSV serialization. This is used by the csv/proxy driver and is the
+  CacheExecute() serialization format.
+
   ==== NOTE ====
   Format documented at http://php.weblogs.com/ADODB_CSV
   ==============
@@ -34,63 +36,64 @@ $ADODB_INCLUDED_CSV = 1;
 	function _rs2serialize(&$rs,$conn=false,$sql='')
 	{
 		$max = ($rs) ? $rs->FieldCount() : 0;
-		
+
 		if ($sql) $sql = urlencode($sql);
 		// metadata setup
-		
+
 		if ($max <= 0 || $rs->dataProvider == 'empty') { // is insert/update/delete
 			if (is_object($conn)) {
 				$sql .= ','.$conn->Affected_Rows();
 				$sql .= ','.$conn->Insert_ID();
 			} else
 				$sql .= ',,';
-			
+
 			$text = "====-1,0,$sql\n";
 			return $text;
 		}
 		$tt = ($rs->timeCreated) ? $rs->timeCreated : time();
-		
+
 		## changed format from ====0 to ====1
 		$line = "====1,$tt,$sql\n";
-		
+
 		if ($rs->databaseType == 'array') {
-			$rows =& $rs->_array;
+			$rows = $rs->_array;
 		} else {
 			$rows = array();
-			while (!$rs->EOF) {	
+			while (!$rs->EOF) {
 				$rows[] = $rs->fields;
 				$rs->MoveNext();
-			} 
+			}
 		}
-		
+
 		for($i=0; $i < $max; $i++) {
-			$o =& $rs->FetchField($i);
+			$o = $rs->FetchField($i);
 			$flds[] = $o;
 		}
-	
+
 		$savefetch = isset($rs->adodbFetchMode) ? $rs->adodbFetchMode : $rs->fetchMode;
 		$class = $rs->connection->arrayClass;
 		$rs2 = new $class();
+		$rs2->timeCreated = $rs->timeCreated; # memcache fix
 		$rs2->sql = $rs->sql;
-		$rs2->oldProvider = $rs->dataProvider; 
+		$rs2->oldProvider = $rs->dataProvider;
 		$rs2->InitArrayFields($rows,$flds);
 		$rs2->fetchMode = $savefetch;
 		return $line.serialize($rs2);
 	}
 
-	
+
 /**
-* Open CSV file and convert it into Data. 
+* Open CSV file and convert it into Data.
 *
 * @param url  		file/ftp/http url
 * @param err		returns the error message
 * @param timeout	dispose if recordset has been alive for $timeout secs
 *
 * @return		recordset, or false if error occured. If no
-*			error occurred in sql INSERT/UPDATE/DELETE, 
+*			error occurred in sql INSERT/UPDATE/DELETE,
 *			empty recordset is returned
 */
-	function &csv2rs($url,&$err,$timeout=0, $rsclass='ADORecordSet_array')
+	function csv2rs($url,&$err,$timeout=0, $rsclass='ADORecordSet_array')
 	{
 		$false = false;
 		$err = false;
@@ -102,7 +105,7 @@ $ADODB_INCLUDED_CSV = 1;
 		@flock($fp, LOCK_SH);
 		$arr = array();
 		$ttl = 0;
-		
+
 		if ($meta = fgetcsv($fp, 32000, ",")) {
 			// check if error message
 			if (strncmp($meta[0],'****',4) === 0) {
@@ -112,10 +115,10 @@ $ADODB_INCLUDED_CSV = 1;
 			}
 			// check for meta data
 			// $meta[0] is -1 means return an empty recordset
-			// $meta[1] contains a time 
-	
+			// $meta[1] contains a time
+
 			if (strncmp($meta[0], '====',4) === 0) {
-			
+
 				if ($meta[0] == "====-1") {
 					if (sizeof($meta) < 5) {
 						$err = "Corrupt first line for format -1";
@@ -123,12 +126,12 @@ $ADODB_INCLUDED_CSV = 1;
 						return $false;
 					}
 					fclose($fp);
-					
+
 					if ($timeout > 0) {
 						$err = " Illegal Timeout $timeout ";
 						return $false;
 					}
-					
+
 					$rs = new $rsclass($val=true);
 					$rs->fields = array();
 					$rs->timeCreated = $meta[1];
@@ -136,9 +139,9 @@ $ADODB_INCLUDED_CSV = 1;
 					$rs->_numOfFields = 0;
 					$rs->sql = urldecode($meta[2]);
 					$rs->affectedrows = (integer)$meta[3];
-					$rs->insertid = $meta[4];	
+					$rs->insertid = $meta[4];
 					return $rs;
-				} 
+				}
 			# Under high volume loads, we want only 1 thread/process to _write_file
 			# so that we don't have 50 processes queueing to write the same data.
 			# We use probabilistic timeout, ahead of time.
@@ -148,7 +151,7 @@ $ADODB_INCLUDED_CSV = 1;
 			# -1 sec after timeout give processes 1/4 chance of timing out
 			# +0 sec after timeout, give processes 100% chance of timing out
 				if (sizeof($meta) > 1) {
-					if($timeout >0){ 
+					if($timeout >0){
 						$tdiff = (integer)( $meta[1]+$timeout - time());
 						if ($tdiff <= 2) {
 							switch($tdiff) {
@@ -160,7 +163,7 @@ $ADODB_INCLUDED_CSV = 1;
 									return $false;
 								}
 								break;
-							case 2: 
+							case 2:
 								if ((rand() & 15) == 0) {
 									fclose($fp);
 									$err = "Timeout 2";
@@ -174,12 +177,12 @@ $ADODB_INCLUDED_CSV = 1;
 									return $false;
 								}
 								break;
-							default: 
+							default:
 								fclose($fp);
 								$err = "Timeout 0";
 								return $false;
 							} // switch
-							
+
 						} // if check flush cache
 					}// (timeout>0)
 					$ttl = $meta[1];
@@ -189,7 +192,7 @@ $ADODB_INCLUDED_CSV = 1;
 				if ($meta[0] === '====1') {
 					// slurp in the data
 					$MAXSIZE = 128000;
-					
+
 					$text = fread($fp,$MAXSIZE);
 					if (strlen($text)) {
 						while ($txt = fread($fp,$MAXSIZE)) {
@@ -205,7 +208,7 @@ $ADODB_INCLUDED_CSV = 1;
 					}
 					return $rs;
 				}
-				
+
 				$meta = false;
 				$meta = fgetcsv($fp, 32000, ",");
 				if (!$meta) {
@@ -235,15 +238,15 @@ $ADODB_INCLUDED_CSV = 1;
 			$err = "Recordset had unexpected EOF 2";
 			return $false;
 		}
-		
+
 		// slurp in the data
 		$MAXSIZE = 128000;
-		
+
 		$text = '';
 		while ($txt = fread($fp,$MAXSIZE)) {
 			$text .= $txt;
 		}
-			
+
 		fclose($fp);
 		@$arr = unserialize($text);
 		//var_dump($arr);
@@ -257,17 +260,18 @@ $ADODB_INCLUDED_CSV = 1;
 		$rs->InitArrayFields($arr,$flds);
 		return $rs;
 	}
-	
+
 
 	/**
 	* Save a file $filename and its $contents (normally for caching) with file locking
+	* Returns true if ok, false if fopen/fwrite error, 0 if rename error (eg. file is locked)
 	*/
 	function adodb_write_file($filename, $contents,$debug=false)
-	{ 
+	{
 	# http://www.php.net/bugs.php?id=9203 Bug that flock fails on Windows
 	# So to simulate locking, we assume that rename is an atomic operation.
-	# First we delete $filename, then we create a $tempfile write to it and 
-	# rename to the desired $filename. If the rename works, then we successfully 
+	# First we delete $filename, then we create a $tempfile write to it and
+	# rename to the desired $filename. If the rename works, then we successfully
 	# modified the file exclusively.
 	# What a stupid need - having to simulate locking.
 	# Risks:
@@ -277,36 +281,39 @@ $ADODB_INCLUDED_CSV = 1;
 	# 4. another process creates $filename between unlink() and rename() -- ok, rename() fails and  cache updated
 		if (strncmp(PHP_OS,'WIN',3) === 0) {
 			// skip the decimal place
-			$mtime = substr(str_replace(' ','_',microtime()),2); 
+			$mtime = substr(str_replace(' ','_',microtime()),2);
 			// getmypid() actually returns 0 on Win98 - never mind!
 			$tmpname = $filename.uniqid($mtime).getmypid();
-			if (!($fd = @fopen($tmpname,'a'))) return false;
-			$ok = ftruncate($fd,0);			
-			if (!fwrite($fd,$contents)) $ok = false;
+			if (!($fd = @fopen($tmpname,'w'))) return false;
+			if (fwrite($fd,$contents)) $ok = true;
+			else $ok = false;
 			fclose($fd);
-			chmod($tmpname,0644);
-			// the tricky moment
-			@unlink($filename);
-			if (!@rename($tmpname,$filename)) {
-				unlink($tmpname);
-				$ok = false;
-			}
-			if (!$ok) {
-				if ($debug) ADOConnection::outp( " Rename $tmpname ".($ok? 'ok' : 'failed'));
+
+			if ($ok) {
+				@chmod($tmpname,0644);
+				// the tricky moment
+				@unlink($filename);
+				if (!@rename($tmpname,$filename)) {
+					@unlink($tmpname);
+					$ok = 0;
+				}
+				if (!$ok) {
+					if ($debug) ADOConnection::outp( " Rename $tmpname ".($ok? 'ok' : 'failed'));
+				}
 			}
 			return $ok;
 		}
 		if (!($fd = @fopen($filename, 'a'))) return false;
 		if (flock($fd, LOCK_EX) && ftruncate($fd, 0)) {
-			$ok = fwrite( $fd, $contents );
+			if (fwrite( $fd, $contents )) $ok = true;
+			else $ok = false;
 			fclose($fd);
-			chmod($filename,0644);
+			@chmod($filename,0644);
 		}else {
 			fclose($fd);
 			if ($debug)ADOConnection::outp( " Failed acquiring lock for $filename<br>\n");
 			$ok = false;
 		}
-	
+
 		return $ok;
 	}
-?>
\ No newline at end of file

libraries/adodb/adodb-error.inc.php

 <?php
-/** 
- * @version V4.90 8 June 2006 (c) 2000-2006 John Lim (jlim#natsoft.com.my). All rights reserved.
- * Released under both BSD license and Lesser GPL library license. 
- * Whenever there is any discrepancy between the two licenses, 
- * the BSD license will take precedence. 
+/**
+ * @version   v5.20.9  21-Dec-2016
+ * @copyright (c) 2000-2013 John Lim (jlim#natsoft.com). All rights reserved.
+ * @copyright (c) 2014      Damien Regad, Mark Newnham and the ADOdb community
+ * Released under both BSD license and Lesser GPL library license.
+ * Whenever there is any discrepancy between the two licenses,
+ * the BSD license will take precedence.
  *
  * Set tabs to 4 for best viewing.
- * 
+ *
  * The following code is adapted from the PEAR DB error handling code.
  * Portions (c)1997-2002 The PHP Group.
  */
@@ -41,6 +43,9 @@ if (!defined("DB_ERROR_SYNTAX")) {
 	define("DB_ERROR_EXTENSION_NOT_FOUND",-25);
 	define("DB_ERROR_NOSUCHDB",           -25);
 	define("DB_ERROR_ACCESS_VIOLATION",   -26);
+	define("DB_ERROR_DEADLOCK",           -27);
+	define("DB_ERROR_STATEMENT_TIMEOUT",  -28);
+	define("DB_ERROR_SERIALIZATION_FAILURE", -29);
 }
 
 function adodb_errormsg($value)
@@ -59,27 +64,27 @@ function adodb_error($provider,$dbType,$errno)
 {
 	//var_dump($errno);
 	if (is_numeric($errno) && $errno == 0) return 0;
-	switch($provider) { 
+	switch($provider) {
 	case 'mysql': $map = adodb_error_mysql(); break;
-	
+
 	case 'oracle':
 	case 'oci8': $map = adodb_error_oci8(); break;
-	
+
 	case 'ibase': $map = adodb_error_ibase(); break;
-	
+
 	case 'odbc': $map = adodb_error_odbc(); break;
-	
+
 	case 'mssql':
 	case 'sybase': $map = adodb_error_mssql(); break;
-	
+
 	case 'informix': $map = adodb_error_ifx(); break;
-	
+
 	case 'postgres': return adodb_error_pg($errno); break;
-	
+
 	case 'sqlite': return $map = adodb_error_sqlite(); break;
 	default:
 		return DB_ERROR;
-	}	
+	}
 	//print_r($map);
 	//var_dump($errno);
 	if (isset($map[$errno])) return $map[$errno];
@@ -91,27 +96,30 @@ function adodb_error($provider,$dbType,$errno)
 function adodb_error_pg($errormsg)
 {
 	if (is_numeric($errormsg)) return (integer) $errormsg;
-    static $error_regexps = array(
-            '/(Table does not exist\.|Relation [\"\'].*[\"\'] does not exist|sequence does not exist|class ".+" not found)$/' => DB_ERROR_NOSUCHTABLE,
-            '/Relation [\"\'].*[\"\'] already exists|Cannot insert a duplicate key into (a )?unique index.*/'      => DB_ERROR_ALREADY_EXISTS,
-            '/divide by zero$/'                     => DB_ERROR_DIVZERO,
-            '/pg_atoi: error in .*: can\'t parse /' => DB_ERROR_INVALID_NUMBER,
-            '/ttribute [\"\'].*[\"\'] not found|Relation [\"\'].*[\"\'] does not have attribute [\"\'].*[\"\']/' => DB_ERROR_NOSUCHFIELD,
-            '/parser: parse error at or near \"/'   => DB_ERROR_SYNTAX,
-            '/referential integrity violation/'     => DB_ERROR_CONSTRAINT,
-			'/Relation [\"\'].*[\"\'] already exists|Cannot insert a duplicate key into (a )?unique index.*|duplicate key violates unique constraint/'     
-			 	 => DB_ERROR_ALREADY_EXISTS
-        );
+	// Postgres has no lock-wait timeout.  The best we could do would be to set a statement timeout.
+	static $error_regexps = array(
+			'(Table does not exist\.|Relation [\"\'].*[\"\'] does not exist|sequence does not exist|class ".+" not found)$' => DB_ERROR_NOSUCHTABLE,
+			'Relation [\"\'].*[\"\'] already exists|Cannot insert a duplicate key into (a )?unique index.*|duplicate key.*violates unique constraint'     => DB_ERROR_ALREADY_EXISTS,
+			'database ".+" does not exist$'       => DB_ERROR_NOSUCHDB,
+			'(divide|division) by zero$'          => DB_ERROR_DIVZERO,
+			'pg_atoi: error in .*: can\'t parse ' => DB_ERROR_INVALID_NUMBER,
+			'ttribute [\"\'].*[\"\'] not found|Relation [\"\'].*[\"\'] does not have attribute [\"\'].*[\"\']' => DB_ERROR_NOSUCHFIELD,
+			'(parser: parse|syntax) error at or near \"'   => DB_ERROR_SYNTAX,
+			'referential integrity violation'     => DB_ERROR_CONSTRAINT,
+			'deadlock detected$'                  => DB_ERROR_DEADLOCK,
+			'canceling statement due to statement timeout$' => DB_ERROR_STATEMENT_TIMEOUT,
+			'could not serialize access due to'   => DB_ERROR_SERIALIZATION_FAILURE
+		);
 	reset($error_regexps);
-    while (list($regexp,$code) = each($error_regexps)) {
-        if (preg_match($regexp, $errormsg)) {
-            return $code;
-        }
-    }
-    // Fall back to DB_ERROR if there was no mapping.
-    return DB_ERROR;
+	while (list($regexp,$code) = each($error_regexps)) {
+		if (preg_match("/$regexp/mi", $errormsg)) {
+			return $code;
+		}
+	}
+	// Fall back to DB_ERROR if there was no mapping.
+	return DB_ERROR;
 }
-	
+
 function adodb_error_odbc()
 {
 static $MAP = array(
@@ -169,7 +177,7 @@ static $MAP = array(
             -923 => DB_ERROR_CONNECT_FAILED,
             -924 => DB_ERROR_CONNECT_FAILED
         );
-		
+
 		return $MAP;
 }
 
@@ -187,7 +195,7 @@ static $MAP = array(
             '-1210'   => DB_ERROR_INVALID_DATE,
             '-1212'   => DB_ERROR_INVALID_DATE
        );
-	   
+
 	   return $MAP;
 }
 
@@ -206,7 +214,7 @@ static $MAP = array(
             2291 => DB_ERROR_CONSTRAINT,
             2449 => DB_ERROR_CONSTRAINT
         );
-	   
+
 	return $MAP;
 }
 
@@ -216,7 +224,7 @@ static $MAP = array(
 		  208 => DB_ERROR_NOSUCHTABLE,
           2601 => DB_ERROR_ALREADY_EXISTS
        );
-	   
+
 	return $MAP;
 }
 
@@ -225,7 +233,7 @@ function adodb_error_sqlite()
 static $MAP = array(
 		  1 => DB_ERROR_SYNTAX
        );
-	   
+
 	return $MAP;
 }
 
@@ -252,7 +260,6 @@ static $MAP = array(
 		    2002 => DB_ERROR_CONNECT_FAILED,
 			2005 => DB_ERROR_CONNECT_FAILED
        );
-	   
+
 	return $MAP;
 }
-?>
\ No newline at end of file

libraries/adodb/adodb-errorhandler.inc.php

 <?php
 /**
- * @version V4.90 8 June 2006  (c) 2000-2006 John Lim (jlim#natsoft.com.my). All rights reserved.
+ * @version   v5.20.9  21-Dec-2016
+ * @copyright (c) 2000-2013 John Lim (jlim#natsoft.com). All rights reserved.
+ * @copyright (c) 2014      Damien Regad, Mark Newnham and the ADOdb community
  * Released under both BSD license and Lesser GPL library license.
  * Whenever there is any discrepancy between the two licenses,
  * the BSD license will take precedence.
@@ -13,7 +15,7 @@
 
 
 // added Claudio Bustos  clbustos#entelchile.net
-if (!defined('ADODB_ERROR_HANDLER_TYPE')) define('ADODB_ERROR_HANDLER_TYPE',E_USER_ERROR); 
+if (!defined('ADODB_ERROR_HANDLER_TYPE')) define('ADODB_ERROR_HANDLER_TYPE',E_USER_ERROR);
 
 if (!defined('ADODB_ERROR_HANDLER')) define('ADODB_ERROR_HANDLER','ADODB_Error_Handler');
 
@@ -74,6 +76,5 @@ function ADODB_Error_Handler($dbms, $fn, $errno, $errmsg, $p1, $p2, &$thisConnec
 
 
 	//print "<p>$s</p>";
-	trigger_error($s,ADODB_ERROR_HANDLER_TYPE); 
+	trigger_error($s,ADODB_ERROR_HANDLER_TYPE);
 }
-?>

libraries/adodb/adodb-php4.inc.php

 <?php
 
 /*
-  V4.90 8 June 2006  (c) 2000-2006 John Lim (jlim#natsoft.com.my). All rights reserved.
-  Released under both BSD license and Lesser GPL library license. 
-  Whenever there is any discrepancy between the two licenses, 
+  @version   v5.20.9  21-Dec-2016
+  @copyright (c) 2000-2013 John Lim (jlim#natsoft.com). All rights reserved.
+  @copyright (c) 2014      Damien Regad, Mark Newnham and the ADOdb community
+  Released under both BSD license and Lesser GPL library license.
+  Whenever there is any discrepancy between the two licenses,
   the BSD license will take precedence.
-  
+
   Set tabs to 4.
 */
 
 
 class ADODB_BASE_RS {
 }
-
-?>
\ No newline at end of file