;
+ close(INFILE) ;
+}
-Header
+my $own_builder = new HTML::TreeBuilder;
+isa_ok( $own_builder, 'HTML::TreeBuilder' );
-This is a link to
-Schibsted Nett in Norway.
+my $obj_h = parse_html $HTML, $own_builder;
+isa_ok( $obj_h, "HTML::TreeBuilder", "existing TreeBuilder handled OK." );
-Sofie Amundsen var på vei hjem fra skolen. Det første stykket
-hadde hun gått sammen med Jorunn. De hadde snakket om roboter.
-Jorunn hadde ment at menneskets hjerne var som en komplisert
-datamaskin. Sofie var ikke helt sikker på om hun var enig. Et
-menneske måtte da være noe mer enn en maskin?
+my $h = parse_html $HTML;
+isa_ok( $h, "HTML::TreeBuilder" );
+# This ensures that the output from $h->dump goes to STDOUT
+my $html;
+ok ($html = $h->as_HTML(undef, ' '), "Get html as string." );
- <-- this one did not terminate the comment
- because "--" on the previous line
-more comment
+my $h2 = parse_htmlfile( "t/oldparse.html" );
+isa_ok( $h2, "HTML::TreeBuilder" );
--->
+ok ($html = $h2->as_HTML(undef, ' '), "Get html as string." );
-
-
-| Name
- |
|---|
| Aas | Gisle
- |
| Koster | Martijn
- |
-
-EOT
-
-
-$h = parse_html $HTML;
-
-# This ensures that the output from $h->dump goes to STDOUT
-
-$html = $h->as_HTML(undef, ' ');
-
-{
- my $h = $html;
- $h =~ s/^/\# /mg;
- print "# HTML: $h#\n";
-}
-
# This is a very simple test just to ensure that we get something
# sensible back.
-if( $html =~ //i && $html =~ /www\.sn\.no/
- && $html !~ /comment/ && $html =~ /Gisle/
-) {
- print "ok 1\n\n";
-} else {
- print "not ok 1\n\n";
-}
+like( $html, qr//i, "parse_htmlfile: found OK." );
+like( $html, qr/www\.sn\.no/, "parse_htmlfile: found www.sn.no link" );
+unlike( $html, qr/comment/, "parse_htmlfile: found comment" );
+like( $html, qr/Gisle/, "parse_htmlfile: found Gisle" );
-$h->delete;
-
-
-exit;
Index: trunk/t/oldparse.html
===================================================================
--- trunk/t/oldparse.html (revision 0)
+++ trunk/t/oldparse.html (revision 128)
@@ -0,0 +1,37 @@
+
+Test page
+
+
+Header
+
+This is a link to
+Schibsted Nett in Norway.
+
+Sofie Amundsen var på vei hjem fra skolen. Det første stykket
+hadde hun gått sammen med Jorunn. De hadde snakket om roboter.
+Jorunn hadde ment at menneskets hjerne var som en komplisert
+datamaskin. Sofie var ikke helt sikker på om hun var enig. Et
+menneske måtte da være noe mer enn en maskin?
+
+
+ <-- this one did not terminate the nested-comment
+ because "--" on the previous line
+
+more comment
+
+-->
+
+
+
+
+
+| Name
+ |
|---|
| Aas | Gisle
+ |
| Koster | Martijn
+ |
Index: trunk/t/parse.t
===================================================================
--- trunk/t/parse.t (revision 18)
+++ trunk/t/parse.t (revision 128)
@@ -1,4 +1,4 @@
-
+#!perl -Tw
# -*-Perl-*-
# Time-stamp: "2003-09-15 01:45:14 ADT"
Index: trunk/t/parsefile.t
===================================================================
--- trunk/t/parsefile.t (revision 18)
+++ trunk/t/parsefile.t (revision 128)
@@ -1,7 +1,7 @@
-
+#!perl -Tw
# -*-Perl-*-
# Time-stamp: "2003-09-15 01:44:58 ADT"
-use Test;
+use Test::More;
BEGIN { plan tests => 3 }
use HTML::TreeBuilder;
use strict;
@@ -53,8 +53,8 @@
}
# Just make a few samples to check that we got what we expected
-ok //i;
-ok //i;
-ok /this is a simple/;
+like($_, qr//i, "Matches Head");
+like($_, qr//i, "Matches isindex");
+like($_, qr/this is a simple/, "Matches simple text");
# /foo\s*a=b/ || $bad++; # too version-dependent
Index: trunk/t/split.t
===================================================================
--- trunk/t/split.t (revision 18)
+++ trunk/t/split.t (revision 128)
@@ -1,4 +1,4 @@
-
+#!perl -Tw
# -*-Perl-*-
# Time-stamp: "2003-09-15 01:48:48 ADT"
#
@@ -9,6 +9,8 @@
# Now we use a shorter document, because we don't have all day on
# this.
+my ($HTML, $notests);
+BEGIN {
$HTML = <<'EOT';
Tittel
@@ -23,94 +25,66 @@
some entities (å)
EOT
-$| = 1;
+$notests = length($HTML); # A test for each char in the test doc
+$notests *= 2; # done twice
+$notests += 3; # plus more for the the rest of the tests
+}
+use strict;
-$notests = length($HTML);
-print "1..$notests\n";
+use Test::More tests=>$notests; # Tests
-use HTML::TreeBuilder;
+BEGIN {
+ use_ok( 'HTML::TreeBuilder');
+}
-print "#Using HTML::TreeBuilder version v$HTML::TreeBuilder::VERSION\n";
-print "#Using HTML::Element version v$HTML::Element::VERSION\n";
-print "#Using HTML::Parser version v", $HTML::Parser::VERSION || "?", "\n";
-print "#Using HTML::Entities version v", $HTML::Entities::VERSION || "?", "\n";
-print "#Using HTML::Tagset version v", $HTML::Tagset::VERSION || "?", "\n";
-print "# Running under perl version $] for $^O",
- (chr(65) eq 'A') ? "\n" : " in a non-ASCII world\n";
-print "# Win32::BuildNumber ", &Win32::BuildNumber(), "\n"
- if defined(&Win32::BuildNumber) and defined &Win32::BuildNumber();
-print "# MacPerl verison $MacPerl::Version\n"
- if defined $MacPerl::Version;
-printf
- "# Current time local: %s\n# Current time GMT: %s\n",
- scalar(localtime($^T)), scalar(gmtime($^T));
-
-
-$h = new HTML::TreeBuilder;
+my $h = new HTML::TreeBuilder;
+isa_ok ( $h, "HTML::TreeBuilder");
$h->parse($HTML)->eof;
-$html = $h->as_HTML;
+my $html = $h->as_HTML;
$h->delete;
-{
- my $h = $html;
- $h =~ s/^/# /mg;
- print "# Parsing: $h#\n";
-}
-
# Each test here tries to parse the doc when we split it in two.
-for $pos (1 .. length($HTML) - 1) {
- $first = substr($HTML, 0, $pos);
- $last = substr($HTML, $pos);
- die "This is bad" unless $HTML eq ($first . $last);
+for my $pos (0 .. length($HTML) - 1) {
+ my $first = substr($HTML, 0, $pos);
+ my $last = substr($HTML, $pos);
+ is ($first.$last, $HTML, "File split okay");
+ my $h1;
eval {
- $h = new HTML::TreeBuilder;
- $h->parse($first);
- $h->parse($last);
- $h->eof;
+ $h1 = new HTML::TreeBuilder;
+ $h1->parse($first);
+ $h1->parse($last);
+ $h1->eof;
};
if ($@) {
- print "Died when splitting at position $pos:\n";
- $before = 10;
- $before = $pos if $pos < $before;
- print "«", substr($HTML, $pos - $before, $before);
- print "»\n«";
- print substr($HTML, $pos, 10);
- print "»\n";
- print "not ok $pos\n";
- $h->delete;
- next;
+ print "Died when splitting at position $pos:\n";
+ my $before = 10;
+ $before = $pos if $pos < $before;
+ print "«", substr($HTML, $pos - $before, $before);
+ print "»\n«";
+ print substr($HTML, $pos, 10);
+ print "»\n";
+ print "not ok $pos\n";
+ $h1->delete;
+ next;
}
- $new_html = $h->as_HTML;
- if ($new_html ne $html) {
- print "\n\nSomething is different when splitting at position $pos:\n";
- $before = 10;
+ my $new_html = $h1->as_HTML;
+ my $before = 10;
$before = $pos if $pos < $before;
- print "«", substr($HTML, $pos - $before, $before);
- print "»\n«";
- print substr($HTML, $pos, 10);
- print "»\n";
- print "\n$html$new_html\n";
- print "not ok $pos";
-
- } else {
- print "ok $pos\n";
- }
- $h->delete;
+ is($new_html, $html, "Still Parsing as the same after split at $pos") or
+ diag("Something is different when splitting at position $pos:\n",
+ "«", substr($HTML, $pos - $before, $before), "»\n«",
+ substr($HTML, $pos, 10), "»\n", "\n$html$new_html\n",
+ );
+ $h1->delete;
}
# Also try what happens when we feed the document one-char at a time
-print "#\n#\nNow parsing document once char at a time...\n";
-$h = new HTML::TreeBuilder;
+# print "#\n#\nNow parsing document once char at a time...\n";
+my $perChar = new HTML::TreeBuilder;
while ($HTML =~ /(.)/sg) {
- $h->parse($1);
+ $perChar->parse($1);
}
-$h->eof;
-$new_html = $h->as_HTML;
-if ($new_html ne $html) {
- print "Also different when parsed one char at a time\n";
- print "\n$html$new_html\n";
- $BAD = 1;
-}
-
-print join '', $BAD ? "not " : '', "ok $notests\n";
-
+$perChar->eof;
+my $new_html = $perChar->as_HTML;
+is ($new_html, $html, "Testing per Char parsing");
+$perChar->delete;
Index: trunk/TODO
===================================================================
--- trunk/TODO (revision 18)
+++ trunk/TODO (revision 128)
@@ -28,4 +28,5 @@
+
========================================================================
Index: trunk/lib/HTML/Tree/AboutObjects.pod
===================================================================
--- trunk/lib/HTML/Tree/AboutObjects.pod (revision 18)
+++ trunk/lib/HTML/Tree/AboutObjects.pod (revision 128)
@@ -430,6 +430,8 @@
in and of itself -- but in the appropriate context, it's understood to
I<mean> the credit card account you're accessing.
+=back
+
This is exactly the relationship between objects and object values, and
from this analogy, several facts about object values are a bit more
explicable:
Index: trunk/lib/HTML/TreeBuilder.pm
===================================================================
--- trunk/lib/HTML/TreeBuilder.pm (revision 18)
+++ trunk/lib/HTML/TreeBuilder.pm (revision 128)
@@ -101,7 +101,7 @@
return $new;
}
-# TODO: document?
+# TODO: document more fully?
sub parse_content { # from any number of scalars
my $tree = shift;
my $retval;
@@ -1568,6 +1568,11 @@
once you've finished feeding all the chunks to parse(...), and
before you actually start doing anything else with the tree in C<$root>.
+=item C<< $root->parse_content(...) >>
+
+Basically a happly alias for C<< $root->parse(...); $root->eof >>.
+Takes the exact same arguments as C<< $root->parse() >>.
+
=item $root->delete()
[An important method inherited from L<HTML::Element|HTML::Element>, which
Index: trunk/lib/HTML/Element.pm
===================================================================
--- trunk/lib/HTML/Element.pm (revision 18)
+++ trunk/lib/HTML/Element.pm (revision 128)
@@ -83,7 +83,7 @@
"I like potatoes"
Incidentally, diagramming with indenting works much better for very
-large trees, and is easier for a program to generate. The $tree->dump
+large trees, and is easier for a program to generate. The C<<$tree->dump>>
method uses indentation just that way.
However you diagram the tree, it's stored the same in memory -- it's a
@@ -126,7 +126,7 @@
child's I<third> child", you're more likely to have to scan the contents
of a tree, looking for whatever nodes, or kinds of nodes, you want to
do something with. The most straightforward way to look over a tree
-is to "traverse" it; an HTML::Element method ($h->traverse) is
+is to "traverse" it; an HTML::Element method (C<<$h->traverse>>) is
provided for this purpose; and several other HTML::Element methods are
based on it.
@@ -250,11 +250,11 @@
If setting a new value, the old value of that attribute is
returned.
-If methods are provided for accessing an attribute (like $h->tag for
-"_tag", $h->content_list, etc. below), use those instead of calling
-attr $h->attr, whether for reading or setting.
+If methods are provided for accessing an attribute (like C<< $h->tag >> for
+"_tag", C<< $h->content_list >>, etc. below), use those instead of calling
+attr C<< $h->attr >>, whether for reading or setting.
-Note that setting an attribute to undef (as opposed to "", the empty
+Note that setting an attribute to C<undef> (as opposed to "", the empty
string) actually deletes the attribute.
=cut
Index: trunk/lib/HTML/AsSubs.pm
===================================================================
--- trunk/lib/HTML/AsSubs.pm (revision 18)
+++ trunk/lib/HTML/AsSubs.pm (revision 128)
@@ -104,16 +104,29 @@
frame frameset noframe
);
-my @code;
for (@TAGS) {
- push(@code, "sub $_ { _elem('$_', \@_); }\n");
- push(@EXPORT, $_);
+ my $code;
+ $code = "sub $_ { _elem('$_', \@_); }\n" ;
+ push(@EXPORT, $_);
+
+=head1 Generated functions
+Here we use a loop and generate functions as strings. Then eval to 'define' the function. 1 subroutine per tag.
+=cut
+ eval $code;
+ if ($@) {
+ die $@;
+ }
}
-eval join('', @code);
-if ($@) {
- die $@;
-}
+=head1 Private Functions
+=cut
+
+=head2 _elem()
+
+The _elem() function is wrapped by all the html 'tag' functions. It takes a tag-name, optional hashref of attributes and a list of content as parameters.
+
+=cut
+
sub _elem
{
my $tag = shift;
Index: trunk/MANIFEST
===================================================================
--- trunk/MANIFEST (revision 18)
+++ trunk/MANIFEST (revision 128)
@@ -17,7 +17,10 @@
t/00system.t
t/building.t
t/clonei.t
+t/construct_tree.t
+t/oldparse.html
t/oldparse.t
t/parse.t
t/parsefile.t
t/split.t
+META.yml
Index: trunk/Makefile.PL
===================================================================
--- trunk/Makefile.PL (revision 18)
+++ trunk/Makefile.PL (revision 128)
@@ -18,11 +18,14 @@
package MY;
+# This could probably be tossed in favor of the ExtUtils::MakeMaker default
sub libscan
{ # Determine things that should *not* be installed
my($self, $path) = @_;
return '' if $path =~ m/~/;
+ return '' if $path =~ /(?:RCS|CVS|SCCS|\.svn)/;
$path;
+
}
__END__
Property changes on: trunk
___________________________________________________________________
Name: svn:ignore
+
.project