XML newbee - Extract tag value

[zzman]

Hi,
I have a 3GB XML file thatI need to parse. since the file is too big to read all at once in to an array, I am reading it one line at a time. I need help on how to extract the contents of some tags, say name, address etc. Here is the code:

$file = '../test.xml';
$handle = @fopen($file, "r");

if (!$handle) die('Error opening file');
else{
	$go = false;
	$lines = '';
	$recCt = 0;
	$res = array();
	while (!feof($handle)) {
		$line = fgets($handle);
		if (strpos($line, '</lstgSect>')) {
			$p = xml_parser_create();
			xml_parse_into_struct($p, $lines, $vals, $index);
			xml_parser_free($p);
			/////////////////////////////////////////////////
			//extract the tag value here
                        //$name = getTagCont($vals, 'name');
                        //or something similar
                        ///////////////////////////////////////////////// 		
			$go = false;
			$lines = '';
			$res = array();

	}//end if (strpos...
	if ($go) { $lines .= $line; }				
	if (strpos($line, '<lstgSect>')) { $go = true; $recCt++; }

	//for testing only
	if ($recCt > 3) break;
}//end while
}//end else of if(!handle

The vals array looks like this:
Array ( [0] => Array ( [tag] => LSTGREQ [type] => open [level] => 1 [value] => ) [1] => Array ( [tag] => ALICD [type] => complete [level] => 2 [value] => C ) [2] => Array ( [tag] => LSTGREQ [value] => [type] => cdata [level] => 1 ) [3] => Array ( [tag] => LSTGREQ [value] => [type] => cdata [level] => 1 ) [4] => Array ( [tag] => LACT [type] => open [level] => 2 [value] => ) [5] => Array ( [tag] => LACTCD [type] => complete [level] => 3 [value] => N ) [6] => Array ( [tag] => LACT [value] => [type] => cdata [level] => 2 ) [7] => Array ( [tag] => LACT [value] => [type] => cdata [level] => 2 ) [8] => Array ( [tag] => LACT [type] => close [level] => 2 ) [9] => Array ( [tag] => LSTGREQ [value] => [type] => cdata [level] => 1 ) [10] => Array ( [tag] => LSTGREQ [value] => [type] => cdata [level] => 1 ) [11] => Array ( [tag] => ACTSRCCD [type] => complete [level] => 2 [value] => O ) [12] => Array ( [tag] => LSTGREQ [value] => [type] => cdata [level] => 1 ) [13] => Array ( [tag] => LSTGREQ [value] => [type] => cdata [level] => 1 ) [14] => Array ( [tag] => LCTL [type] => open [level] => 2 [value] => ) [15] => Array ( [tag] => RTY [type] => open [level] => 3 [value] => ) [16] => Array ( [tag] => LSTGSCOPECD [type] => complete [level] => 4 [value] => F ) [17] => Array ( [tag] => RTY [value] => [type] => cdata [level] => 3 ) [18] => Array ( [tag] => RTY [value] => [type] => cdata [level] => 3 ) [19] => Array ( [tag] => RTYCD [type] => complete [level] => 4 [value] => ) [20] => Array ( [tag] => RTY [value] => [type] => cdata [level] => 3 ) [21] => Array ( [tag] => RTY [value] => [type .......

[BobLennon]

zzman wrote:

Hi,
I have a 3GB XML file thatI need to parse. since the file is too big to read all at once in to an array, I am reading it one line at a time. I need help on how to extract the contents of some tags, say name, address etc. Here is the code:

 .......[/QUOTE]

Maybe you might consider reading and parsing 4k blocks at a time.  Something like:

$xml_parser = xml_parser_create();  //Create a reference to the xml parser
xml_set_element_handler($xml_parser, "startElement", "endElement");  // Set up start and end element callback functions
xml_set_character_data_handler($xml_parser, "characterData");  //Set up Character Data callback function

$fp = fopen($feed, "r") or die("could not open XML input");

while ($data = fread($fp, 4096)) {  //Read Parse and load $dataArray array until  end  

	if (!xml_parse($xml_parser, $data, feof($fp))){
	 	die(sprintf("XML error: %s at line %d",	xml_error_string(xml_get_error_code($xml_parser)),
		xml_get_current_line_number($xml_parser)));
	}
}
xml_parser_free($xml_parser);  //  Release the parser.

[zzman]

Reading the file is not the problem. I just want to know how do I get the contents of the tag from the $vals array.

[BobLennon]

zzman wrote:

Reading the file is not the problem. I just want to know how do I get the contents of the tag from the $vals array.

When using the xml_parse_into_struct function, the $vals array is made up of a series of arrays containing information about each tag, its level, its type and its contents. It is generated in the same order that the tags occur in the input xml.

The $index array in this call can be addressed associatively by the tag name. It returns an array of the index values to the $vals array that are associated with that tag. You can sequence through the $index array to get the indices in the $vals array that occur when a tag is opened, closed, or completed. When the [type] (state) associative index in the $vals array has the value of complete, the value enclosed between that tags opening and closing can be accessed using the value of the [value] associative index.

I'm not sure I know how to handle the CDATA that may be contained in the XML input using the xml_parse_into_struct function. I've always found it easier to use callback routines registered with the xml_set_element_handler, and the xml_set_character_data_handler for this kind of thing.

[zzman]

Thanks for you help. But i am getting nowhere with this, I thought XML was supposed to make lefe easier ... any ways if someone can help that'll be great. here is the schema of the <lstgSect> tag ...

<lstgSect>
    <lstgReq>
      <aliCd>C     </aliCd>
      <lact>
        <lactCd>N</lactCd>
      </lact>
      <actSrcCd>O</actSrcCd>
      <lctl>
        <rty>
          <lstgScopeCd>F</lstgScopeCd>
          <rtyCd>
          </rtyCd>
        </rty>
        <lstgTypCd>L</lstgTypCd>
        <broCd>D</broCd>
        <revDaccInd>N</revDaccInd>
        <dspScopeCd>563366</dspScopeCd>
        <langCd>E</langCd>
        <pndgCd>N</pndgCd>
      </lctl>
      <lappCtl>
        <stylCd>CI</stylCd>
        <doiNbr>2</doiNbr>
        <tyaCd>B</tyaCd>
      </lappCtl>
      <lstgMgmt>
        <lnPlcmtCd>W</lnPlcmtCd>
        <noslInd>N</noslInd>
      </lstgMgmt>
      <inProd>
        <prodScopeCd>P</prodScopeCd>
        <inProdIdTxt>
          <inProdTxt>CX</inProdTxt>
          <inSectTxt>CX01</inSectTxt>
        </inProdIdTxt>
      </inProd>
      <ltnFlds>
        <tn>
          <countryCd>001</countryCd>
          <areaCd>800</areaCd>
          <coCd>469</coCd>
          <lineNbr>4663</lineNbr>
        </tn>
        <omtnInd>N</omtnInd>
      </ltnFlds>
      <ln>
        <dlnmInd>N</dlnmInd>
        <lnAbbInd>Y</lnAbbInd>
        <lnlnNm>Sears</lnlnNm>
        <lnFirstNm>
          <lnfnNm>Roebuck And Co</lnfnNm>
        </lnFirstNm>
      </ln>
      <la>
        <omaddrInd>Y</omaddrInd>
        <omlocInd>N</omlocInd>
        <alocAbbInd>Y</alocAbbInd>
        <athoroAbbInd>Y</athoroAbbInd>
      </la>
      <ltxt>
        <indTxt>Service/Parts Information</indTxt>
      </ltxt>
      <dspData>
        <dsp1Nbr>218</dsp1Nbr>
        <dsp2Nbr>2002</dsp2Nbr>
        <dsp1Txt>0</dsp1Txt>
        <dsp2Txt>0</dsp2Txt>
      </dspData>
    </lstgReq>
    <setPlcmtReq>
      <hdrLev>
        <hdrStatCd>N</hdrStatCd>
      </hdrLev>
      <priorLev>
        <levDoiNbr>1</levDoiNbr>
        <levStatCd>N</levStatCd>
        <levSeqOvr>
          <seqOverCd>F</seqOverCd>
        </levSeqOvr>
        <levTxt>Product Repair Services</levTxt>
      </priorLev>
    </setPlcmtReq>
  </lstgSect>

[BobLennon]

Do you have a complete schema? Or would you validate and publish a cut down version?

[BobLennon]

Sorry I didn't notice it was complete. I'll parse it out and post the code, but it won't read a line at a time and it will not use the xml_parse_into_struct function if that's OK.

[BobLennon]

<?php
/************************************************************************  

 *						This is a Example script to demonstrate XML input parsing.
 *	
 *  The script fills an array from an abbreviated input file Example.xml.  It parses the input into
 *	the following array structure to  allow easy output formatting.
 *  

 * 		[Example] = Array{
 *  		[lstgSect] = Array {	
 *  			[lstgReq] = Array{											 
 *  				[aliCd]
 *  				[lact] = Array{
 *						[lactCd]
 *					}
 *					
 *					[actSrcCd] 
 *					[lctl]=Array{
 *						[rty]=Array{
 *							[lstgScopeCd]
 *							[rtyCd]
 *						}
 *						[lstgTypCd]
 *						[broCd]
 *						[revDaccInd]
 *						[dspScopeCd]
 *						[langCd]
 *						[pndgCd]
 *					}
 *					[lappCtl]=Array{
 *						[stylCd]
 *						[doiNbr]
 *					    [tyaCd]
 *					}
 *				}  //Truncate input for example
 * 			}
 *  	}
 **********************************************************************/

	// Global Variables
$xmlInput= "Example.xml";
$Example = array();  //This is the parsed array data container

$currentData = "";	// Index names for Example array access
$section = "";

		// Parser Initialization
$xml_parser = xml_parser_create();  //Create a reference to the xml parser
xml_set_element_handler($xml_parser, "startElement", "endElement");  // Set up start and end element callback functions
xml_set_character_data_handler($xml_parser, "characterData");  //Set up Character Data callback function

$fp = fopen($xmlInput, "r") or die("could not open XML input");

while ($data = fread($fp, 4096)) {  //Read Parse and load $Example array until  end  

	if (!xml_parse($xml_parser, $data, feof($fp))){
	 	die(sprintf("XML error: %s at line %d",	xml_error_string(xml_get_error_code($xml_parser)),
		xml_get_current_line_number($xml_parser)));
	}
}
xml_parser_free($xml_parser);  //  Release the parser.

	// Dump Array
print_r($Example);

	// PHP XML Parser Call Back Functions
function startElement($parser, $name, $attrs) {
   	global $Example, $currentData,$section;
   	switch($name) {  // Set up Index References for arrays in Example container
   		case "LSTGSECT": // Parser returns case folded tags
   			$currentData = ""; 
			break;
   		case "LSTGREQ":
   			$section = "LSTGREQ";
   			break;
   		case "LACT":
   			$section = "LACT";
   			break;
  		case "LCTL":
   			$section = "LCTL";
   			break;
  		case "RTY":
   			$section = "RTY";
   			break;
  		case "LAPPCTL":
   			$section = "LAPPCTL";
   			break;
   		default:
   			$currentData = $name;  //Remember element name if not an array
   			break;
   	}
}
function endElement($parser, $name) {
   	global $currentData;
   	$currentData = "";  //Clear index name at end of element parse.
}
function characterData($parser, $data) {
	global $Example, $currentData,$section;
	if ($currentData != "") {
		switch($section) {
		case "LSTGSECT": // Parser returns case folded tags
   			$Example[$section][$currentData] .= $data;
			break;
   		case "LSTGREQ":
   			$Example['LSTGSECT'][$section][$currentData] .= $data;
   			break;
   		case "LACT":
   			$Example['LSTGSECT']['LSTGREQ'][$section][$currentData].=$data;
   			break;
  		case "LCTL":
   			$Example['LSTGSECT']['LSTGREQ'][$section][$currentData].=$data;
   			break;
  		case "RTY":
   			$Example['LSTGSECT']['LSTGREQ'][$section][$currentData].=$data;
   			break;
  		case "LAPPCTL":
   			$Example['LSTGSECT']['LSTGREQ'][$section][$currentData].=$data;
   			break;
		}
	}
}
?>

[zzman]

Thanks man. Appreciate your help...