Script went crazy this morning

[TDotBoy2011]

<?php
// Declare variables //
$debug = false;
//$url = "http://www.mto.gov.on.ca/english/traveller/conditions/rdclosure.htm";
$url = "http://www.mto.gov.on.ca/english/traveller/trip/road_closures.shtml";
//$url = "./road_closures.html";

$email_recip = "";
#$email_recip = "t";
$email_from = "s";
$email_from_addr = "t";
$message_header = "";
$message_footer = "";
$arfield1 = array("summary","highway","direction","fromat","to","lanesaffected","trafficimpact","reason","others","eventstart","eventend");
$cntarfield1 = count($arfield1);
$timenow = date("Y-m-d H:i:s");
// Connect to database
$dbname = "mevents";
$dbuser = "";
$dbpass = "";

$link = mysql_connect('localhost',$dbuser,$dbpass);
if (!$link) { 
    die('Could not connect: ' . mysql_error());
}
mysql_select_db($dbname);


// DON'T EDIT PAST HERE UNLESS YOU KNOW WHAT YOU'RE DOING //
// MAIN PROGRAM ///
$webpage = read_web($url);
$content = find_data($webpage, "<!-- MIDDLE COLUMN", "</html>"); 
$content = find_data($content, "</div><a ", "</html>");
$curr_events = read_html_table($content,true);       // Parse the events HTML table

if ($curr_events["totalevents"]>0)
{
  $sql = "update mEvents set eventid = 'xRoadInfo' where eventid like 'RoadInfo%'";
  $result = mysql_query($sql);
}

for ($curr_event_num = 1; $curr_event_num <= $curr_events["totalevents"]; $curr_event_num++) {
  $send_email = false;     // Flag used to see if an email should be sent for new closure events
  $event_message = $message_header;
  $curr_event_found = false;

  if ($curr_events[$curr_event_num]["unscheduledevents"]==0)
  {
  	$sql = "select * from mEvents where eventid = '".($curr_events[$curr_event_num]["eventid"])."' limit 1";
  }
  else
  {
  	$unscd = explode("\n",$curr_events[$curr_event_num]["others"]["data"]);
  	$sql = "select * from mEvents where eventid='xRoadInfo' and highway='".$curr_events[$curr_event_num]["highway"]["data"]."' and direction='".$curr_events[$curr_event_num]["direction"]["data"]."' and fromat='".$curr_events[$curr_event_num]["fromat"]["data"]."'";
  	if ($unscd[0]!="") $sql .= " and others like '".$unscd[0]."%'";
  }
  $result = mysql_query($sql);
  if (mysql_num_rows($result)>0) $curr_event_found = true;

  if (!$curr_event_found) {    // Add the not found current event to the message
$send_email = true;

if ($curr_events[$curr_event_num]["eventend"]["data"]=="")
    {
        $event_message .= "*-NEW-*\n";
    }
else
{
	$event_message .= "-CLEARED-\n";
}
if ($debug) $event_message .= $curr_events[$curr_event_num]["eventid"];
$sqlquery = "insert into mEvents (eventid,date_insert";
$sqlval = "('".$curr_events[$curr_event_num]["eventid"]."','".$timenow."'";
for ($a=0;$a<$cntarfield1;$a++)
{
   $sqlquery .= ",`".$arfield1[$a]."`";
   
   $sqlval .= ",'".addslashes($curr_events[$curr_event_num][$arfield1[$a]]["data"])."'";
   if ($curr_events[$curr_event_num][$arfield1[$a]]["data"]!="")
   {
     $event_message .= $curr_events[$curr_event_num][$arfield1[$a]]["field"]."".$curr_events[$curr_event_num][$arfield1[$a]]["data"]."\n";
   }
   if ($debug) $event_message .= "<br/>";
}
$sqlquery .= ") values".$sqlval.")";
mysql_query($sqlquery);
$event_message .= "====END====\n";
$email_subject = "*-NEW-* ";
$email_subject .= $curr_events[$curr_event_num]["summary"]["data"];
  } else {
$row = mysql_fetch_array($result);
if ($row["eventend"]=="")
{
	if ($curr_events[$curr_event_num]["unscheduledevents"]==0)
	{
	if ($curr_events[$curr_event_num]["eventend"]["data"]!="")
	{
		$send_email = true;
		$event_message .= "-CLEARED-\n";
		if ($debug) $event_message .= $curr_events[$curr_event_num]["eventid"];
		$sqlquery = "update mEvents set date_update='$timenow'";
		for ($a=0;$a<$cntarfield1;$a++)
		{
		   $sqlquery .= ",`".$arfield1[$a]."`='".addslashes($curr_events[$curr_event_num][$arfield1[$a]]["data"])."'";
		   if ($curr_events[$curr_event_num][$arfield1[$a]]["data"]!="")
   		   {
		     $event_message .= $curr_events[$curr_event_num][$arfield1[$a]]["field"]."".$curr_events[$curr_event_num][$arfield1[$a]]["data"]."\n";
		   }
		   if ($debug) $event_message .= "<br/>";
		}
		$sqlquery .= " where eventid='".$curr_events[$curr_event_num]["eventid"]."'";
		mysql_query($sqlquery);
		$event_message .= "====END====\n";
		$email_subject = "*-CLR-* ";
		$email_subject .= $curr_events[$curr_event_num]["summary"]["data"];

	}
	}
	else
	{
		$sqlquery = "update mEvents set eventid='".$curr_events[$curr_event_num]["eventid"]."' where id=".$row["id"];
		mysql_query($sqlquery);
	}
}
}
$event_message .= $message_footer;

//Send an email if there is a new/resolved event 
if ($send_email) { 
if ($debug) print ("Sending the message below to " . $email_recip . ", from \"" . $email_from . "\" <" .$email_from_addr . ">.\n" . $event_message . "\n");
else mail ($email_recip, $email_subject, $event_message, "From: \"" . $email_from . "\" <" .$email_from_addr . ">");
}

}
// FOUR LINES BELOW COMMENTED OUT  TO STOP REPEATING MESSAGE
//if ($curr_events["totalevents"]>0)
//{
// $sql = "delete from mEvents where eventid = 'xRoadInfo'";
  //$result = mysql_query($sql);
//}

mysql_close();

// Functions //
//read_web - Read the web page
//           $strURL ==> URL of the webpage
function read_web($strURL)
{
  global $debug;
  $buffer = "";
  if($debug){ print("Reading \"$strURL\".\n"); }
  $fh = fopen($strURL, "rb");
  if ($fh === false) {
    return "";
  }
  while (!feof($fh)) {
    $buffer .= fread($fh, 1024);
  }
  fclose($fh);
  return $buffer;
} // end function read_web

// find)data - Gets a substring of the webpage by scraping the data
//             $strFile ==> text of the webpage
//             $strStart_Pattern ==> start of the substring
//             $strEnd_Pattern ==> end of the substring
function find_data($strFile,$strStart_Pattern,$strEnd_Pattern, $intStart_Position = 0)
{
  global $debug;
  if($debug){ print("Searching for \"$strStart_Pattern\"...\"$strEnd_Pattern\".\n<!-- //-->"); }
  $first_match = strpos($strFile,$strStart_Pattern, $intStart_Position);
  if ($first_match) {
    # find the begining of the tag
    for ($i = $first_match; $i>0;$i--) {
      $char = $strFile[$i];
      if ($char == "<" ) {
        $first_match = $i; //record the location of the tag
        break;
      }
    }
    $partialbuffer = substr($strFile,$first_match,strlen($strFile) - $first_match);
    # find the end of the sub string
    $second_match = strpos($partialbuffer,$strEnd_Pattern);
    return substr($partialbuffer,0,$second_match + strlen($strEnd_Pattern));
  } else {
    return(false);
  }
} //end function find_data

// read_html_table - Read the contents of an HTML table and return the array(s)
//                   strHTMLTable ==> HTML table text
//                   boolSkipFirstRow ==> Skip the first row if it contains column titles (true/false)
function read_html_table($strHTMLTable, $boolSkipFirstRow)
{
  global $debug,$arfield1,$cntarfield1;
  $arrevents = array();

  preg_match_all('/<a name="Event.+<\/table>/Us',$strHTMLTable,$dataevents);
  $dataevents = $dataevents[0];

  $countdataevents = count($dataevents);

  $j = 0;
  for($i=0; $i<$countdataevents; $i++)
  {
preg_match('/<a name="Event(.+)" id/', $dataevents[$i], $matches); 
$eventid = $matches[1];

//if ($eventid>0)
{
	$j++;
	$arrevents[$j]["eventid"] = $eventid;
	preg_match('/summary="(.+)" style/', $dataevents[$i], $matches);
	$arrevents[$j]["summary"]["field"] = "";
	$arrevents[$j]["summary"]["data"] = $matches[1];

	preg_match_all('/<tr>.+<\/tr>/Us',$dataevents[$i],$tr);
	$tr = $tr[0];
	$counttr = count($tr);

	for ($k=1;$k<$cntarfield1;$k++)
	{
		$arrevents[$j][$arfield1[$k]]["field"] = "";
		$arrevents[$j][$arfield1[$k]]["data"] = "";

	}
	if ($eventid>0) $arrevents[$j]["unscheduledevents"] = 0;
	else $arrevents[$j]["unscheduledevents"] = 1;

	for ($k=1;$k<$counttr;$k++)
	{
		preg_match_all('/<td.+<\/td>/Us',$tr[$k],$td);
		$td = $td[0];
		$td1 = cleandatafield(cleandata($td[0]));

		if (checkdatafield($td1)==1)
		{
			$arrevents[$j][$td1]["field"] = cleandata($td[0])." ";
			$arrevents[$j][$td1]["data"] = cleandata($td[1]);
		}
		else
		{
			if (cleandata($td[1])!="")
			{
				$arrevents[$j]["others"]["field"] = "";
				$arrevents[$j]["others"]["data"] .= cleandata($td[0])." ".cleandata($td[1])."\n";
			}
		}
	}
}
  }
  echo "<br/>\ntotal ".$j;
  $arrevents["totalevents"] = $j;
  return $arrevents;
} //end function readhtmltable

function cleandata($x)
{
  $y = $x;
  $y = preg_replace('/<td.+>/U',"",$y);
  $y = str_replace("</td>","",$y);
  $y = str_replace("<strong>","",$y);
  $y = str_replace("</strong>","",$y);
  $y = str_replace("/r/n","",$y);
  $y = trim($y);

  return $y;
}
function cleandatafield($x)
{
  $y = $x;
  $y = str_replace("/","",$y);
  $y = str_replace(":","",$y);
  $y = str_replace(" ","",$y);
  $y = strtolower($y);
  return $y;
}
function checkdatafield($x)
{
  global $debug,$arfield1,$cntarfield1;
  $found = 0;
  for ($i=0;$i<$cntarfield1;$i++)
  {
if ($x==$arfield1[$i]) $found = 1;
  }
  return $found;
}

print ("ClosedRoadScript");
print date("     G:i:s");
?>

 

I have the above script that reads this website  http://www.mto.gov.on.ca/english/traveller/trip/road_closures.shtml  and emails out the changes.  Up until this morning it has been working fine but then it started repeating the same one over and over.

 

 

It didn't put the normal line in the database it would for any other event which would explain why the same email got sent over and over but I am trying to figure out why the database was empty for this event. The only idea I have is something to do with this line and the script not liking it  "VASEY ROAD (WAVERLEY)/SIMCOE ROAD 27/SIMCOE ROAD 23"  or that it's too long for database.

 

Someone on this site has been helping me with this for the last couple of years but he has since passed away. I've been making small adjustments myself with my limited php know how but this one has me stumped.

 

 

 

[Adam]

Do you have access to the database server?

[TDotBoy2011]

Do you have access to the database server?

 

Yes.  It's all controlled by me.

[Adam]

Check your database for the entries. If the column's data type isn't long enough, then the MySQL server will truncate it to the right size in order to insert it. When you run the query to check if the event already exists:

 

and fromat='".$curr_events[$curr_event_num]["fromat"]["data"]."'

 

This condition will be evaluating the truncated version to the full length version -- obviously never going to find it.

[TDotBoy2011]

Check your database for the entries. If the column's data type isn't long enough, then the MySQL server will truncate it to the right size in order to insert it. When you run the query to check if the event already exists:

 

and fromat='".$curr_events[$curr_event_num]["fromat"]["data"]."'

 

This condition will be evaluating the truncated version to the full length version -- obviously never going to find it.

 

It was set to 250 so I am going to guess that was the issue. I've increased it but my luck I will never see one that long again.