Hi All (and a happy holiday to those that will get a break),

I am trying to read in an XML file of addresses. I need to remove all 
the address data from the file where code =~ /^000/ (there are none 
in the example data below). I need to reproduce that data 'as is', so 
I need to honour the tag structure, although the order of the tags 
doesn't need to be honoured.

I have been trying to use XML::Simple and I had a go with XML::Smart 
but I haven't been able to get the results I want.

My best effort is below (with XML::Simple). There are a couple of 
differences I can't seem to get around. 1) How to get a name on the 
anon tags (<anon> should be <address>) . 2) how to get the number 
data as an attribute or a tag.

My script effort is below also. If anyone know of any options with 
XML::Simple that will help please let me know. Failing that, any 
other suggestions on how to tackle the problem would be appreciated.

TIA,
Dp.

======== my script ================
#!/usr/bin/perl

use strict;
use warnings;
use XML::Simple;
use Data::Dumper;

my $fh = shift;
my $out = 'Addresses-no-phgr.xml';
my $xml = eval {XMLin($fh, SuppressEmpty => 1)};

my @addresses;

my $c = @{$xml->{'address'}};
for (my $i = 0; $i < $c; ++$i) {
        my %hash;
        my %lines;
        print $xml->{'address'}->[$i]->{'code'}."\n";
        my @add_lines;
        if ( $xml->{'address'}[$i]->{'lines'} ) {
                my $d = @{$xml->{'address'}[$i]->{'lines'}-
>{'line'}};
                for (my $l = 0; $l < $d; ++$l ) {
                        push(@add_lines, $xml->{'address'}[$i]-
>{'lines'}->{'line'}[$l] );
                }


        }

        
@hash{'code','record_type','address_type','Postcode','Country','number
','lines'} =
                ($xml->{'address'}[$i]->{'code'},$xml-
>{'address'}[$i]->{'record_type'},
                 $xml->{'address'}[$i]->{'address_type'},$xml-
>{'address'}[$i]->{'Postcode'},
                 $xml->{'address'}[$i]->{'Country'},$xml-
>{'address'}[$i]->{'number'},
                 [\@add_lines]);

        push @addresses,\%hash;
}


open my $ofh, '>:encoding(iso-8859-1)', $out  or die "Can't write to 
$out: $!\n!";
XMLout(\@addresses,  XMLDecl => 1,
                RootName => 'Addresses',
                ValueAttr => {'lines' => 'line'},
                NoAttr => 1,
                Outputfile => $ofh,
        );

=======================

============= my best effort ==========
<?xml version='1.0' standalone='yes'?>
<Addresses>
  <anon>
    <Country>GBR</Country>
    <Postcode>EC1Y 4RX</Postcode>
    <address_type>shipping</address_type>
    <code>BJPPHO00</code>
    <lines>
      <anon>INCISIVE MEDIA</anon>
      <anon>HAYMARKET HOUSE</anon>
      <anon>28-29 HAYMARKET</anon>
      <anon>LONDON</anon>
      <anon>EC1Y 4RX</anon>
    </lines>
    <number>2577</number>
    <record_type>client</record_type>
  </anon>
  <anon>
    <Country>GBR</Country>
    <Postcode>SU1V 0AQ</Postcode>
    <address_type>shipping</address_type>
    <code>BJPUBL00</code>
    <lines>
      <anon>RIVER HOUSE</anon>
      <anon>12-14 BERRY STREET</anon>
      <anon>SURRY</anon>
      <anon>SU1V 0AQ</anon>
    </lines>
    <number>2578</number>
    <record_type>client</record_type>
  </anon>
</Addresses>
========================

=========== The data file ===========
<?xml version = "1.0" encoding= "utf-8"?>
<addresses>
        <address number="2577">
                <code>BJPPHO00</code>
                <record_type>client</record_type>
                <address_type>shipping</address_type>
                <Postcode>EC1Y 4RX</Postcode>
                <Country>GBR</Country>
                <lines>
                        <line>PIGEON MEDIA</line>
                        <line>HAYMARKET HOUSE</line>
                        <line>28-29 HAYMARKET</line>
                        <line>LONDON</line>
                        <lines>EC1Y 4RX</line>
                </lines>
        </address>
        <address number="2578">
                <code>BJPUBL00</code>
                <record_type>client</record_type>
                <address_type>shipping</address_type>
                <Postcode>SU1V 0AQ</Postcode>
                <Country>GBR</Country>
                <lines>
                        <line>RIVER HOUSE</line>
                        <line>12-14 BERRY STREET</line>
                        <line>SURREY</line>
                        <line>SU1V 0AQ</line>
                </lines>
        </address>
</addresses>
=============================

>>>>> ""Beginner"" == "Beginner"  <dermot@sciencephoto.com> writes:

"Beginner"> I have been trying to use XML::Simple and I had a go with XML::Smart 
"Beginner"> but I haven't been able to get the results I want.

I'd use XML::LibXML, with an XPath expression, as follows.  I'm demoing by
deleting the one that starts "BJPU", but replace that with your "0000" and
you're all set.

Note that the XPATH says "find any address node that contains a code node
whose text begins with BJPU".  And those are the ones we want to delete.

    #!/usr/bin/env perl
    use strict;

    use XML::LibXML;
    my $parser = XML::LibXML->new;
    my $doc = $parser->parse_fh(\*DATA);

    my @bad = $doc->findnodes(q{//address[.//code[starts-with(., "BJPU")]]});

    for (@bad) {
      $_->unbindNode;               # delete it!
    }

    print $doc->toString(1);

    __END__
    <?xml version = "1.0" encoding= "utf-8"?>
    <addresses>
            <address number="2577">
                    <code>BJPPHO00</code>
                    <record_type>client</record_type>
                    <address_type>shipping</address_type>
                    <Postcode>EC1Y 4RX</Postcode>
                    <Country>GBR</Country>
                    <lines>
                            <line>PIGEON MEDIA</line>
                            <line>HAYMARKET HOUSE</line>
                            <line>28-29 HAYMARKET</line>
                            <line>LONDON</line>
                            <lines>EC1Y 4RX</lines>
                    </lines>
            </address>
            <address number="2578">
                    <code>BJPUBL00</code>
                    <record_type>client</record_type>
                    <address_type>shipping</address_type>
                    <Postcode>SU1V 0AQ</Postcode>
                    <Country>GBR</Country>
                    <lines>
                            <line>RIVER HOUSE</line>
                            <line>12-14 BERRY STREET</line>
                            <line>SURREY</line>
                            <line>SU1V 0AQ</line>
                    </lines>
            </address>
    </addresses>

-- 
Randal L. Schwartz - Stonehenge Consulting Services, Inc. - +1 503 777 0095
<merlyn@stonehenge.com> <URL:http://www.stonehenge.com/merlyn/>
Perl/Unix/security consulting, Technical writing, Comedy, etc. etc.
See PerlTraining.Stonehenge.com for onsite and open-enrollment Perl training!

Randal L. Schwartz wrote:
>>>>>> ""Beginner"" == "Beginner"  <dermot@sciencephoto.com> writes:
> 
> "Beginner"> I have been trying to use XML::Simple and I had a go with XML::Smart 
> "Beginner"> but I haven't been able to get the results I want.
> 
> I'd use XML::LibXML, with an XPath expression, as follows.  I'm demoing by
> deleting the one that starts "BJPU", but replace that with your "0000" and
> you're all set.
> 
> Note that the XPATH says "find any address node that contains a code node
> whose text begins with BJPU".  And those are the ones we want to delete.
> 
>     #!/usr/bin/env perl
>     use strict;
> 
>     use XML::LibXML;
>     my $parser = XML::LibXML->new;
>     my $doc = $parser->parse_fh(\*DATA);
> 
>     my @bad = $doc->findnodes(q{//address[.//code[starts-with(., "BJPU")]]});

my @bad = $doc->findnodes(q{//address[starts-with(code, "BJPU")]});

Rob

>>>>> "Rob" == Rob Dixon <rob.dixon@350.com> writes:

>> my @bad = $doc->findnodes(q{//address[.//code[starts-with(., "BJPU")]]});

Rob> my @bad = $doc->findnodes(q{//address[starts-with(code, "BJPU")]});

It's clear to me that myy XPath-fu is still weak.  Thanks for showing
me the better way.  At least I figured out that I needed starts-with. :)

-- 
Randal L. Schwartz - Stonehenge Consulting Services, Inc. - +1 503 777 0095
<merlyn@stonehenge.com> <URL:http://www.stonehenge.com/merlyn/>
Perl/Unix/security consulting, Technical writing, Comedy, etc. etc.
See PerlTraining.Stonehenge.com for onsite and open-enrollment Perl training!

>>>>> "Rob" == Rob Dixon <rob.dixon@350.com> writes:

Rob> my @bad = $doc->findnodes(q{//address[starts-with(code, "BJPU")]});

Actualy, doesn't that require code immediately below addess?

Don't you want .//code there ?

-- 
Randal L. Schwartz - Stonehenge Consulting Services, Inc. - +1 503 777 0095
<merlyn@stonehenge.com> <URL:http://www.stonehenge.com/merlyn/>
Perl/Unix/security consulting, Technical writing, Comedy, etc. etc.
See PerlTraining.Stonehenge.com for onsite and open-enrollment Perl training!

Randal L. Schwartz wrote:
>>>>>> "Rob" == Rob Dixon <rob.dixon@350.com> writes:
> 
> Rob> my @bad = $doc->findnodes(q{//address[starts-with(code, "BJPU")]});
> 
> Actualy, doesn't that require code immediately below addess?

Yes

> Don't you want .//code there ?

Almost certainly not. In the example XML <code> appeared only as a child element
of <address>. In the general case it is unlikely to reappear at a lower level
and // would force the XSLT engine to search further than was necessary.

By the way the dot is also unnecessary as it is implicit before // anywhere
except at the beginning of a path expression. (A path starting with // covers
all descendants of the document root.)

HTH & Happy Christmas

Rob

From: "Beginner" <dermot@sciencephoto.com>
> Hi All (and a happy holiday to those that will get a break),
> 
> I am trying to read in an XML file of addresses. I need to remove all
> the address data from the file where code =~ /^000/ (there are none in
> the example data below). I need to reproduce that data 'as is', so I
> need to honour the tag structure, although the order of the tags
> doesn't need to be honoured.
> 
> I have been trying to use XML::Simple and I had a go with XML::Smart
> but I haven't been able to get the results I want.
> 
> =========== The data file ===========
> <?xml version = "1.0" encoding= "utf-8"?>
> <addresses>
>         <address number="2577">
>                 <code>BJPPHO00</code>
>                 <record_type>client</record_type>
>                 <address_type>shipping</address_type>
>                 <Postcode>EC1Y 4RX</Postcode>
>                 <Country>GBR</Country>
>                 <lines>
>                         <line>PIGEON MEDIA</line>
>                         <line>HAYMARKET HOUSE</line>
>                         <line>28-29 HAYMARKET</line>
>                         <line>LONDON</line>
>                         <lines>EC1Y 4RX</line>
>                 </lines>
>         </address>
>         <address number="2578">
>                 <code>BJPUBL00</code>
>                 <record_type>client</record_type>
>                 <address_type>shipping</address_type>
>                 <Postcode>SU1V 0AQ</Postcode>
>                 <Country>GBR</Country>
>                 <lines>
>                         <line>RIVER HOUSE</line>
>                         <line>12-14 BERRY STREET</line>
>                         <line>SURREY</line>
>                         <line>SU1V 0AQ</line>
>                 </lines>
>         </address>
> </addresses>
> =============================

Another options is:

use XML::Rules;

my $parser = XML::Rules->new(
	style => 'filter',
	rules => [
		_default => 'raw', 
			# keep as much details (order, whitespace) as possible
		code => 'raw extended',
			# keep the details, but also make the data easier 
			# to access while handling the <address> tag
		address => sub {
			my $code = $_[1]->{':code'}{_content};
			if ($code =~ /^000/) {
				return;
			} else {
				return $_[0] => $_[1];
			}
		},
	],
);

$parser->filterfile($xmlfile, \*STDOUT);

__END__

HTH, Jenda
P.S.: Please use the just uploaded 0.13, not the 0.09. There was a 
bug in escaping the data written to the XML while filtering.
===== Jenda@Krynicky.cz === http://Jenda.Krynicky.cz =====
When it comes to wine, women and song, wizards are allowed 
to get drunk and croon as much as they like.
	-- Terry Pratchett in Sourcery

>>>>> ""Beginner"" == "Beginner"  <dermot@sciencephoto.com> writes:

"Beginner"> "running xml2-config...failed
"Beginner"> The installed version of libxml2 not compatible with XML::LibXML."

It's not just a minimum.  There are some versions of libxml2 that are
sufficiently broken that the the author of XML::LibXML is protecting you from
them.  You probably have one of those.  See the source of the Makefile.PL
for details.

-- 
Randal L. Schwartz - Stonehenge Consulting Services, Inc. - +1 503 777 0095
<merlyn@stonehenge.com> <URL:http://www.stonehenge.com/merlyn/>
Perl/Unix/security consulting, Technical writing, Comedy, etc. etc.
See PerlTraining.Stonehenge.com for onsite and open-enrollment Perl training!