shell bypass 403
package Test::XML::XPath;
# @(#) $Id$
use strict;
use warnings;
use Carp;
use Test::More;
use Test::Builder;
our $VERSION = '0.03';
# Call this early so that lack of a suitable class will be picked up
# when we're imported, not on first use.
_find_xpath_class();
#---------------------------------------------------------------------
# Import shenanigans. Copied from Test::Pod...
#---------------------------------------------------------------------
sub import {
my $self = shift;
my $caller = caller;
no strict 'refs';
*{ $caller . '::is_xpath' } = \&is_xpath;
*{ $caller . '::like_xpath' } = \&like_xpath;
*{ $caller . '::set_xpath_processor' } = \&set_xpath_processor;
*{ $caller . '::unlike_xpath' } = \&unlike_xpath;
my $Test = Test::Builder->new;
$Test->exported_to( $caller );
$Test->plan( @_ );
}
#---------------------------------------------------------------------
# Tool.
#---------------------------------------------------------------------
sub like_xpath {
my ($input, $statement, $test_name) = @_;
croak "usage: like_xpath(xml,xpath[,name])"
unless $input && $statement;
my $Test = Test::Builder->new;
my $ok = eval {
my $xp = _make_xpath( $input );
return $xp->exists( $statement );
};
if ($@) {
$Test->ok( 0, $test_name );
$Test->diag( " Parse Failure: $@" );
return 0;
} else {
ok( $ok, $test_name );
unless ( $ok ) {
diag ( " input: $input" );
diag ( " does not match: $statement" );
}
return $ok;
}
}
sub unlike_xpath {
my ($input, $statement, $test_name) = @_;
croak "usage: unlike_xpath(xml,xpath[,name])"
unless $input && $statement;
my $Test = Test::Builder->new;
my $ok = eval {
my $xp = _make_xpath( $input );
return ! $xp->exists( $statement );
};
if ($@) {
$Test->ok( 0, $test_name );
$Test->diag( " Parse Failure: $@" );
return 0;
} else {
ok( $ok, $test_name );
unless ( $ok ) {
diag ( " input: $input" );
diag ( " does match: $statement" );
}
return $ok;
}
}
sub is_xpath {
my ($input, $statement, $expected, $test_name) = @_;
croak "usage: is_xpath(xml,xpath,expected[,name])"
unless $input && $statement && $expected;
my $Test = Test::Builder->new;
my $got = eval {
my $xp = _make_xpath( $input );
$xp->findvalue( $statement );
};
if ($@) {
$Test->ok( 0, $test_name );
$Test->diag( " Parse Failure: $@" );
return 0;
} else {
my $retval = $Test->is_eq( $got, $expected, $test_name );
unless ( $retval ) {
diag( " evaluating: $statement" );
diag( " against: $input" );
}
return $retval;
}
}
#---------------------------------------------------------------------
# Abstract interface to XPath processing.
#---------------------------------------------------------------------
{
my $xpath_class;
sub set_xpath_processor {
$xpath_class = join('::', __PACKAGE__, @_ );
}
sub _make_xpath {
$xpath_class ||= _find_xpath_class();
return $xpath_class->new( @_ );
}
}
sub _find_xpath_class {
foreach (qw( XML::LibXML XML::XPath )) {
eval "use $_";
return __PACKAGE__ . "::$_" unless $@;
}
# Ooops, we're unusable.
die $@;
}
{
package Test::XML::XPath::XML::XPath;
sub new {
my $class = shift;
bless { xpath => XML::XPath->new( xml => @_ ) }, $class;
}
sub exists {
my $self = shift;
return $self->{xpath}->exists( @_ );
}
sub findvalue {
my $self = shift;
return $self->{xpath}->findvalue( @_ );
}
}
{
package Test::XML::XPath::XML::LibXML;
sub new {
my $class = shift;
my $p = XML::LibXML->new;
bless { xpath => $p->parse_string( @_ ) }, $class;
}
sub exists {
my $self = shift;
my @nodes = $self->{xpath}->findnodes( @_ );
return @nodes ? 1 : 0;
}
sub findvalue {
my $self = shift;
return $self->{xpath}->findvalue( @_ );
}
}
1;
__END__
=head1 NAME
Test::XML::XPath - Test XPath assertions
=head1 SYNOPSIS
use Test::XML::XPath tests => 3;
like_xpath( '<foo />', '/foo' ); # PASS
like_xpath( '<foo />', '/bar' ); # FAIL
unlike_xpath( '<foo />', '/bar' ); # PASS
is_xpath( '<foo>bar</foo>', '/foo', 'bar' ); # PASS
is_xpath( '<foo>bar</foo>', '/bar', 'foo' ); # FAIL
# More interesting examples of xpath assertions.
my $xml = '<foo attrib="1"><bish><bosh args="42">pub</bosh></bish></foo>';
# Do testing for attributes.
like_xpath( $xml, '/foo[@attrib="1"]' ); # PASS
# Find an element anywhere in the document.
like_xpath( $xml, '//bosh' ); # PASS
# Both.
like_xpath( $xml, '//bosh[@args="42"]' ); # PASS
=head1 DESCRIPTION
This module allows you to assert statements about your XML in the form
of XPath statements. You can say that a piece of XML must contain
certain tags, with so-and-so attributes, etc. It will try to use any
installed XPath module that it knows about. Currently, this means
XML::LibXML and XML::XPath, in that order.
B<NB>: Normally in XPath processing, the statement occurs from a
I<context> node. In the case of like_xpath(), the context node will
always be the root node. In practice, this means that these two
statements are identical:
# Absolute path.
like_xpath( '<foo/>', '/foo' );
# Path relative to root.
like_xpath( '<foo/>', 'foo' );
It's probably best to use absolute paths everywhere in order to keep
things simple.
B<NB>: Beware of specifying attributes. Because they use an @-sign,
perl will complain about trying to interpolate arrays if you don't
escape them or use single quotes.
=head1 FUNCTIONS
=over 4
=item like_xpath ( XML, XPATH [, NAME ] )
Assert that XML (a string containing XML) matches the statement
XPATH. NAME is the name of the test.
Returns true or false depending upon test success.
=item unlike_xpath ( XML, XPATH [, NAME ] )
This is the reverse of like_xpath(). The test will only pass if XPATH
I<does not> generates any matches in XML.
Returns true or false depending upon test success.
=item is_xpath ( XML, XPATH, EXPECTED [, NAME ] )
Evaluates XPATH against XML, and pass the test if the is EXPECTED. Uses
findvalue() internally.
Returns true or false depending upon test success.
=item set_xpath_processor ( CLASS )
Set the class name of the XPath processor used. It is up to you to
ensure that this class is loaded.
=back
In all cases, XML must be well formed, or the test will fail.
=head1 SEE ALSO
L<Test::XML>.
L<XML::XPath>, which is the basis for this module.
If you are not conversant with XPath, there are many tutorials
available on the web. Google will point you at them. The first one
that I saw was: L<http://www.zvon.org/xxl/XPathTutorial/>, which
appears to offer interactive XPath as well as the tutorials.
=head1 AUTHOR
Dominic Mitchell E<lt>cpan2 (at) semantico.comE<gt>
=head1 COPYRIGHT AND LICENSE
Copyright 2002 by semantico
This library is free software; you can redistribute it and/or modify
it under the same terms as Perl itself.
=cut
# Local Variables:
# mode: cperl
# cperl-indent-level: 4
# indent-tabs-mode: nil
# End:
# vim: set ai et sw=4 :