in reply to looping over xml

Without the XML data you are using this code on it is quite difficult to figure out what happens.

Just a couple of remarks though: if you replace your main loop by check_tag( $xso) then you will get the document root, as it is now you miss it:

my $parser = XML::Parser->new(ErrorContext => 2, Style => "Tree");
my $xso = XML::SimpleObject->new( $parser->parse($content) ) or die "c
+ould not parse!";
check_tag($xso);
...

[download]

A second remark is that I hope you are dealing with data-oriented XML, as opposed to document-oriented XML: XML::SimpleObjects does not deal properly with mixed content (<p>this is <b>mixed</b> content</p> (just as XML::Simple BTW, must be the name): in this case the content of the p element will be "this is content". This is a weakness of all modules that stick to a model of the XML document where the data is directly within an element. You have to have an extra layer, with PCDATA or CDATA pseudo-elements (or text nodes) within all elements, or you cannot deal with mixed content.

If you post a minimum example of your XML data we might be able to help you more.

And now the obligatory XML::Twig version. Note that I will probably add the is_field method in the next release, as it seems to make sense for people who do data-oriented XML.

#!/bin/perl -w
use strict;

use XML::Twig;

my $twig = XML::Twig->new();
$twig->parse( \*DATA)
            or die "could not parse!";
# the #ELT argument means that we will get only the
# "real" elements, as opposed to the ones containing the text
foreach my $tag($twig->descendants( '#ELT')) {
    print $tag->gi;
    print " ", $tag->text if( is_field( $tag));
    print "\n";
  }

sub is_field
  { my $tag= shift;
    return 1 if( ($tag->children == 1) && $tag->first_child->is_text);
    return 0;
  }

__DATA__
<doc id="id1">
  <elt id="id2">elt 1</elt>
  <elt id="id3">
    <subelt id="id4">subelt 1</subelt>
    <subelt id="id5">subelt 2</subelt>
    <subelt id="id6">subelt 3</subelt>
  </elt>
  <elt id="id7">elt 3</elt>
  <elt id="id8"><subelt id="id9">subelt 4</subelt></elt>
</doc>

[download]

Replies are listed 'Best First'.
Re: Re: looping over xml
by Micz (Beadle) on Aug 06, 2001 at 12:14 UTC
    hello,
    thank you for your remarks. a sample of the code I am trying to parse: 
    <?xml version="1.0" encoding="UTF-8" ?>
<vxml version="1.0" application="news_root.xml">
<form id="Kategorie">
    <field name="kategorie">
        <grammar src="gram/kategorie.grammar"/>
        <prompt bargein="true">
            <audio src="http://xxx.61.xxx.xxx/audio/portal/news/10.wav
+">xaxaxaxa</audio>
        <audio src="http://xxx.xxx.xxx.xxx/audio/portal/news/input_sta
+rt.wav">xaxaxaxa</audio>
        </prompt>
        <filled>
            <audio src="http://xxx.xxx.xxx.xxx/audio/portal/common/inp
+ut_end.wav">input end</audio>
            <if cond="kategorie == 'menue'"><goto next="../index.jsp"/
+><elseif cond="kategorie == 'schlagzeilen'"/><goto next="#Schlagzeile
+n"/><elseif cond="kategorie == 'wirtschaft'"/><goto next="#Wirtschaft
+"/><elseif cond="kategorie == 'boerse'"/><goto next="#Boerse"/><elsei
+f cond="kategorie == 'sport'"/><goto next="#Sport"/><elseif cond="kat
+egorie == 'unterhaltung'"/><goto next="#Unterhaltung"/><else/><clear 
+namelist="kategorie"/>123.123.123.123/nee<reprompt/></if>
        </filled> 
    </field>    
    <noinput count="1">
        <audio src="http://xxx.xxx.xxx.xxx/audio/portal/common/noinput
+.wav">noinput</audio>        
        <audio src="http://xxx.xxx.xxx.xxx/audio/portal/news/40.wav">I
+ch habe sie wirklich nicht gehoert.</audio>
        <audio src="http://xxx.xxx.xxx.xxx/audio/portal/common/pauses/
+500.wav">kleine Pause</audio>
        <audio src="http://xxx.xxx.xxx.xxx/audio/portal/news/input_sta
+rt.wav">input start</audio>
    </noinput>
    <noinput count="2">
        <audio src="http://xxx.xxx.xxx.xxx/audio/portal/common/noinput
+.wav">noinput</audio>        
        <audio src="http://xxx.xxx.xxx.xxx/audio/portal/news/50.wav">I
+ch habe sie wirklich nicht gehoert.</audio>
        <audio src="http://xxx.xxx.xxx.xxx/audio/portal/common/pauses/
+500.wav">kleine Pause</audio>
        <audio src="http://xxx.xxx.xxx.xxx/audio/portal/news/input_sta
+rt.wav">input start</audio>
    </noinput>
</form>
</vxml>

    [download]
    I will implement the first point, and post further feedback. If there is a "well-formed-ness" error in the code above, it is because I triaged a bit, the document itself is longer (and validates).
    A further note: The point I want to reach is to have all the audio and goto tags in a hash, if my approach is totally wrong please warn me.

    thanks again, jan
[reply]
[d/l]
      okay, here is more of my code. a simple link checker: 
      $tasks{$URL} = $URL;

while (keys(%tasks) != 0) {
  foreach $key (keys (%tasks)) {
   if (!$opt_v) {
     print "#";
   } else {
     print "checking $key\n";
   }
    check_page($key);
    delete $tasks{key};
  }
}


sub check_page{
  unless (defined ($content = get $_[0])) {
    print "\n\n NASTY ERROR! \n\n bad link: $_[0]\n\n\n";
    return;
  }

  my $parser = XML::Parser->new(ErrorContext => 2, Style => "Tree");
  my $xso = XML::SimpleObject->new( $parser->parse($content) ) or die 
+"could not parse!";

  check_tag($xso);

# output the information

  foreach $file (keys %soundfiles) {
    print "\n BAD ".$file." called in $_[0]\n";
  }

  foreach $link (keys %links) {
    next if make_url($link) =~ /$_[0]/ ;
    next if $link =~ /^#/ ;
    $tasks{make_url($link)} = make_url($link);
        print "checking ".make_url($link)."\n";
  }
}

sub check_tag{
  foreach my $tag($_[0]->children()) {
   if ($tag->name =~ /audio/) {
    build_audio($tag);
   }
   if ($tag->name =~ /goto/) {
    $links{$tag->attribute('next')} = "goto";
   }
    check_tag($tag);
  }
}

sub make_url {
  if ($_[0] =~ /http:/) {
    return $_[0];
  } elsif ($_[0] =~ /^#/) {
    return $_[0];
  } else {
    return $base_url.$_[0];
  }
}

sub build_audio {
  if (!exists $soundfiles{$_[0]->attribute('src')}) {
   unless (head(make_url($_[0]->attribute('src')))) {
     $soundfiles{make_url($_[0]->attribute('src'))} = "BAD";
   }
  }
}

      [download]
[reply]
[d/l]

        Using XML::Twig you create and parse the twig with:

        my $t= XML::Twig->new();
$t->parse( $content);

        [download]

        You can get all the audio elements by doing: my @audio= $twig->descendants( 'audio') (I'll let you figure out how to get all the goto elements ;--)

        You can then access the src attribute using $elt->att( 'src')

[reply]
[d/l]
[select]

In Section Seekers of Perl Wisdom