Eliminating HTML escape sequences from string with PHP

Question

I am working on a php script to pull quest data from wowhead, particularly what starts and ends the quest, whether it is an item or a npc, and what its id or name is, respectively. This is the relevant portion of the whole script, with the rest involving database insertion. This is the completed snippet of code I came up with if anyone is interested. Also, seeing as this will run about 15,000 times, is this the best method of obtaining/storing the data?

<?php

$quests = array();
//$questlimit = 14987;
$questlimit = 5;
$currentquest = 1;
$questsprocessed = 0;
while($questsprocessed != $questlimit)
{
echo "<br>";
echo "  Start of iteration: ".$questsprocessed."  ";
echo "<br>";
echo "  Attempting to process quest: ".$currentquest."  ";
echo "<br>";

$quests[$currentquest] = array();
$baseurl = 'http://wowhead.com/quest=';
$fullurl = $baseurl.$currentquest;

$data = drupal_http_request($fullurl);

$queststartloc1 = strpos($data->data, 'quest_start'); 
$queststartloc2 = strpos($data->data, 'quest_end');

if($queststartloc1==false)
{$currentquest++; echo "No data for this quest"; echo "<br>"; continue;}


$questendloc1 = strpos($data->data, 'quest_end');
$questendloc2 = strpos($data->data, 'x5DDifficulty');

$startcaptureLength = $queststartloc2 - $queststartloc1;
$endcaptureLength = $questendloc2 - $questendloc1;


$quest_start_raw = substr($data->data,$queststartloc1, $startcaptureLength);
$quest_end_raw = substr($data->data, $questendloc1, $endcaptureLength);

$startDecoded = preg_replace('~\\\\x([A-Fa-f0-9]{2})~e', 'chr("0x$1")', $quest_start_raw);
$endDecoded = preg_replace('~\\\\x([A-Fa-f0-9]{2})~e', 'chr("0x$1")', $quest_end_raw);
$quests[$currentquest]['Start'] = array();
$quests[$currentquest]['End'] = array();

if(strstr($startDecoded, 'npc'))
  {
   $quests[$currentquest]['Start']['Type'] = "npc";
  preg_match('~npc=(\d+)~', $startDecoded, $startmatch);
  }
else
{
  $quests[$currentquest]['Start']['Type'] = "item";
  preg_match('~item=(\d+)~', $startDecoded, $startmatch);
}


$quests[$currentquest]['Start']['ID'] = $startmatch[1];


if(strstr($endDecoded, 'npc'))
  {
   $quests[$currentquest]['End']['Type'] = "npc";
  preg_match('~npc=(\d+)~', $endDecoded, $endmatch);
  }
else
{
  $quests[$currentquest]['End']['Type'] = "item";
  preg_match('~item=(\d+)~', $endDecoded, $endmatch);
}


$quests[$currentquest]['End']['ID'] = $endmatch[1];

//var_dump($quests[$currentquest]);

echo "  End of iteration: ".$questsprocessed."  ";
echo "<br>";
echo "  Processed quest: ".$currentquest."  ";
echo "<br>";
$currentquest++;
$questsprocessed++;

}
?>

I added in a couple checks for null data (quest does not exist), and database insertions based off the array type id. The only problem I have with this script is that it says "Connection timed out after 30 seconds" about every hour, how would I go about handling this error and just restarting the loop? — user28187
– user28187, Commented Apr 17, 2014 at 8:15

gog · Accepted Answer · 2014-04-16 09:55:02Z

3

These are called "escape sequences". Normally, they're used to represent characters not printable otherwise, but can encode any character. In php, you can decode them like this:

$text = '
quest_start\\x5DStart\\x3A\\x20\\x5Bitem\\x3D16305\\x5D\\x5B\\x2Ficon\\x5D\\x5B\\x2Fli\\x5D\\x5Bli\\x5D\\x5Bicon\\x20name\\x3Dquest_end\\x5DEnd\\x3A\\x20\\x5Burl\\x3D\\x2Fnpc\\x3D12696\\x5DSenani\\x20Thunderheart\\x5B\\x2Furl\\x5D\\x5B\\x2Ficon\\x5D\\x5B\\x2Fli\\x5D\\x5Bli\\x5DNot\\x20sharable\\x5B\\x2Fli\\x5D\\x5Bli
';

$decoded = preg_replace('~\\\\x([A-Fa-f0-9]{2})~e', 'chr("0x$1")', $text);

Which gives you a string similar to this:

 quest_start]Start: [item=16305][/icon][/li][li][icon name=quest_end]End: [url=/npc=12696]Senani Thunderheart[/url][/icon][/li][li]Not sharable[/li][li

(obviously, some kind of BB-code). To remove all bbcodes, yet one replacement is necessary:

$clean = preg_replace('~(\[.+?\])+~', ' ', $decoded);

answered Apr 16, 2014 at 9:55

gog

11.4k2 gold badges29 silver badges42 bronze badges

Sign up to request clarification or add additional context in comments.

2 Comments

user28187 Over a year ago

Thank you very much for this, it will get me well on my way to getting this script done. The only problem I have with this is that the clean seems to clean out my item number. From the decoded message, I need Start: item=16305 End: npc=12696. I mentioned the npc name in my post, because I didn't see the npc ID was in there as well. The IDs are much more useful to me than the name at the moment. I saw a post in a google search where they turned a string into a table, splitting at something like name:john domain:example.com id:123, but I can't seem to find it anymore.

gog Over a year ago

@user28187: you can extract the number before the cleanup with preg_match('~npc=(\d+)~', $decoded, $match)

Collectives™ on Stack Overflow

Eliminating HTML escape sequences from string with PHP

1 Answer 1

2 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

2 Comments

Your Answer

Sign up or log in

Post as a guest