ghouse has asked for the wisdom of the Perl Monks concerning the following question:

Hi, I need to read a attribute value in an XML of the As I am new to XML, I am having trouble on this ...Can you help me in this Sample XML is as following 
 

<?xml version="1.0" encoding="ISO-8859-1" ?> 
- <RESEARCH>
  <xslobject name="research.xsl" /> 
  <PGFROM>12/30/04</PGFROM>
  <PGOFFSET>0</PGOFFSET> 
  <PGCOUNT>1</PGCOUNT> 
- <LINK PUBID="584557" DOCID="100386419" CLIENT="0" HIDDEN="1" DELETED
+="0">
  <SCORE>1.00</SCORE> 
  <PUBDATE>01/06/05</PUBDATE> 
  <DOCDATE>01/06/05</DOCDATE> 
  <EXPDATE>NEVER</EXPDATE> 
  <TITLE>SC 4 DOC AND IMAGE</TITLE> 
  <URI>/DDL/servlets/dv.search?docID=100386419&searchType=INTRANET</UR
+I> 
   <FILETYPE>doc</FILETYPE> 
  </LINK>
  </RESEARCH>

[download]
I am able to read an Element but to read an attrtibute say the following line in XML " <LINK PUBID="584557" DOCID="100386419" CLIENT="0" HIDDEN="1" DELETED="0"> " How do I do it ... My sample code is 
sub    ParseCSrsp {
    my ($CSrspXMLMsg,$StatusRef) = @_;
    my $CSxmlParser;

    $CSxmlParser = new XML::Parser(ErrorContext => 2);
    $CSxmlParser->setHandlers(Init  => \&InitHandler,
                              Final => \&FinalHandler,
                              Start => \&StartHandler,
                              End   => \&EndHandler,
                              Char  => \&CharHandler);

    $CSxmlParser->parse($CSrspXMLMsg);
    return 1;
}

# XML parser handler functions
sub InitHandler {
    $CSElement = {};
    undef $XMLTagContent;
}

sub FinalHandler {
}

sub StartHandler {
    my $Element = $_[1];

    undef $XMLTagContent;
    &{$LBSubIndex{$Element}} if $LBSubIndex{$Element};
    if ($Element =~ /URI/i) {
        $TagElement = 'E';
    }            
}

sub EndHandler {
    my $Element = $_[1];
#    &{$LBSubIndex{$Element . '_'}} if $LBSubIndex{$Element . '_'};
    &{$LBSubIndex{$Element}} if $LBSubIndex{$Element};
}

sub CharHandler {
    $XMLTagContent .= $_[1];
}

# Custom handlers
sub ParseELEMENT_ {
    $CSElement = {};
}

sub LBGeneric {
    my $Element = $_[1];
       print "$Element is generic element and $XMLTagContent is tagcon
+tent\n";
    if ($TagElement eq 'E') {                 
            if ($XMLTagContent eq '') {           
            $PrevContact = ' ';               
        } else {                              
            $PrevContact = $XMLTagContent ;   
        }                                     
        print "GOT IT NOW $PrevContact\n";
        $TagElement = '';                     
        $URL = $PrevContact;
    } elsif ($XMLTagContent =~ /search/i) {  
        print "GOT IT NOW $PrevContact\n";
    }
#    $CSrsp->{$Element} = $XMLTagContent if $XMLTagContent;

}

# Custom handlers    
sub ParseELEMENT_ {  
    $CSElement = {}; 
    print $CSElement;
    }

[download]
Thanks, Ghouse

Replies are listed 'Best First'.
Re: Parsing XML - SAX events
by gellyfish (Monsignor) on Jan 06, 2005 at 13:48 UTC

    If all you are interested in is the value of a single attribute then you might be better off using XPath: 

    use XML::XPath
                                                                      
+          
$/ = undef;
                                                                      
+          
my $foo = <DATA>;
my $xp = XML::XPath->new(xml => $foo);
                                                                      
+          
my $nodes = $xp->find('/RESEARCH/LINK/@PUBID');
                                                                      
+          
foreach my $node ($nodes->get_nodelist())
{
   print $node->getData(),"\n";
}
__END__
<RESEARCH>
  <xslobject name="research.xsl" />
  <PGFROM>12/30/04</PGFROM>
  <PGOFFSET>0</PGOFFSET>
  <PGCOUNT>1</PGCOUNT>
<LINK PUBID="584557" DOCID="100386419" CLIENT="0" HIDDEN="1" DELETED="
+0">
  <SCORE>1.00</SCORE>
  <PUBDATE>01/06/05</PUBDATE>
  <DOCDATE>01/06/05</DOCDATE>
  <EXPDATE>NEVER</EXPDATE>
  <TITLE>SC 4 DOC AND IMAGE</TITLE>
  <URI>/DDL/servlets/dv.search?docID=100386419&amp;searchType=INTRANET
+</URI>
   <FILETYPE>doc</FILETYPE>
  </LINK>
</RESEARCH>

    [download]
    (BTW, Your XML is not well formed - you need to render the ampersand in the URI as the entity &amp;)

    /J\

[reply]
[d/l]
Re: Parsing XML - SAX events
by mirod (Canon) on Jan 06, 2005 at 14:40 UTC

    You are using XML::Parser, which brings 2 comments: first it is NOT a SAX parser, it has its own onterface, which actually predates the SAX spec. Secondly: don't use XML::Parser! It is a very useful low-level module, but not the easiest to use.

    Lots of nicer modules have been built on top of XML::Parser (or on top of other parsers) that will make your life much easier. In your case it looks like you could use XML::Simple. If you want SAX, then XML::SAX will do. Other options include XML::XPath, as mentioned by gellyfish, XML::LibXML, based on libxml2, and (of course ;--) XML::Twig.

    Finally, if you are new to XML in Perl, you should probably read the Perl XML FAQ.

[reply]
Re: Parsing XML - SAX events
by holli (Abbot) on Jan 06, 2005 at 12:49 UTC 
    sub StartHandler 
{
    my $expat   = shift;
    my $element = shift;
    my %attr    = @_;

    print $attr{PUBID};
    #...
}

    [download]
[reply]
[d/l]
      Thank you . But I am not sure as to how do i capture this value and take it store it in a global variable which can be accessed later
[reply]

Back to Seekers of Perl Wisdom