Parent Directory
| 
Revision Log
++ ChangeLog 27 May 2007 11:15:58 -0000 * parser.cgi: Output how long each process takes. Document element testing is moved to the |Whatpm::ContentChecker| module. 2007-05-27 Wakaba <wakaba@suika.fam.cx>
| 1 | wakaba | 1.1 | #!/usr/bin/perl |
| 2 | use strict; | ||
| 3 | |||
| 4 | wakaba | 1.2 | use lib qw[/home/httpd/html/www/markup/html/whatpm |
| 5 | /home/wakaba/public_html/-temp/wiki/lib]; | ||
| 6 | wakaba | 1.5 | use CGI::Carp qw[fatalsToBrowser]; |
| 7 | wakaba | 1.6 | use Time::HiRes qw/time/; |
| 8 | wakaba | 1.1 | |
| 9 | use SuikaWiki::Input::HTTP; ## TODO: Use some better CGI module | ||
| 10 | |||
| 11 | my $http = SuikaWiki::Input::HTTP->new; | ||
| 12 | |||
| 13 | ## TODO: _charset_ | ||
| 14 | |||
| 15 | my $mode = $http->meta_variable ('PATH_INFO'); | ||
| 16 | ## TODO: decode unreserved characters | ||
| 17 | |||
| 18 | wakaba | 1.2 | if ($mode eq '/html' or $mode eq '/test') { |
| 19 | wakaba | 1.1 | require Encode; |
| 20 | wakaba | 1.3 | require Whatpm::HTML; |
| 21 | require Whatpm::NanoDOM; | ||
| 22 | wakaba | 1.1 | |
| 23 | my $s = $http->parameter ('s'); | ||
| 24 | if (length $s > 1000_000) { | ||
| 25 | print STDOUT "Status: 400 Document Too Long\nContent-Type: text/plain; charset=us-ascii\n\nToo long"; | ||
| 26 | exit; | ||
| 27 | } | ||
| 28 | |||
| 29 | wakaba | 1.6 | my $time1 = time; |
| 30 | wakaba | 1.1 | $s = Encode::decode ('utf-8', $s); |
| 31 | wakaba | 1.6 | my $time2 = time; |
| 32 | my %time = (decode => $time2 - $time1); | ||
| 33 | |||
| 34 | wakaba | 1.1 | print STDOUT "Content-Type: text/plain; charset=utf-8\n\n"; |
| 35 | |||
| 36 | wakaba | 1.2 | print STDOUT "#errors\n"; |
| 37 | |||
| 38 | my $onerror = sub { | ||
| 39 | wakaba | 1.4 | my (%opt) = @_; |
| 40 | print STDOUT "$opt{line},$opt{column},$opt{type}\n"; | ||
| 41 | wakaba | 1.2 | }; |
| 42 | wakaba | 1.1 | |
| 43 | wakaba | 1.6 | $time1 = time; |
| 44 | wakaba | 1.3 | my $doc = Whatpm::HTML->parse_string |
| 45 | ($s => Whatpm::NanoDOM::Document->new, $onerror); | ||
| 46 | wakaba | 1.6 | $time2 = time; |
| 47 | $time{parse} = $time2 - $time1; | ||
| 48 | wakaba | 1.1 | |
| 49 | wakaba | 1.2 | print "#document\n"; |
| 50 | |||
| 51 | my $out; | ||
| 52 | wakaba | 1.6 | $time1 = time; |
| 53 | wakaba | 1.2 | if ($mode eq '/html') { |
| 54 | wakaba | 1.3 | $out = Whatpm::HTML->get_inner_html ($doc); |
| 55 | wakaba | 1.2 | } else { # test |
| 56 | $out = test_serialize ($doc); | ||
| 57 | wakaba | 1.1 | } |
| 58 | wakaba | 1.6 | $time2 = time; |
| 59 | $time{serialize} = $time2 - $time1; | ||
| 60 | wakaba | 1.2 | print STDOUT Encode::encode ('utf-8', $$out); |
| 61 | wakaba | 1.4 | print STDOUT "\n"; |
| 62 | |||
| 63 | if ($http->parameter ('dom5')) { | ||
| 64 | require Whatpm::ContentChecker; | ||
| 65 | print STDOUT "#domerrors\n"; | ||
| 66 | wakaba | 1.6 | $time1 = time; |
| 67 | Whatpm::ContentChecker->check_document ($doc, sub { | ||
| 68 | wakaba | 1.4 | my %opt = @_; |
| 69 | print STDOUT get_node_path ($opt{node}) . ';' . $opt{type} . "\n"; | ||
| 70 | }); | ||
| 71 | wakaba | 1.6 | $time2 = time; |
| 72 | $time{check} = $time2 - $time1; | ||
| 73 | wakaba | 1.4 | } |
| 74 | wakaba | 1.6 | |
| 75 | print STDOUT "#log\n"; | ||
| 76 | print STDOUT "byte->char\t", $time{decode}, "s\n"; | ||
| 77 | print STDOUT "html5->dom5\t", $time{parse}, "s\n"; | ||
| 78 | print STDOUT "dom5->serialize\t", $time{serialize}, "s\n"; | ||
| 79 | print STDOUT "dom5 check\t", $time{check}, "s\n" if defined $time{check}; | ||
| 80 | wakaba | 1.1 | } else { |
| 81 | print STDOUT "Status: 404 Not Found\nContent-Type: text/plain; charset=us-ascii\n\n404"; | ||
| 82 | } | ||
| 83 | wakaba | 1.2 | |
| 84 | exit; | ||
| 85 | |||
| 86 | sub test_serialize ($) { | ||
| 87 | my $node = shift; | ||
| 88 | my $r = ''; | ||
| 89 | |||
| 90 | my @node = map { [$_, ''] } @{$node->child_nodes}; | ||
| 91 | while (@node) { | ||
| 92 | my $child = shift @node; | ||
| 93 | my $nt = $child->[0]->node_type; | ||
| 94 | if ($nt == $child->[0]->ELEMENT_NODE) { | ||
| 95 | $r .= '| ' . $child->[1] . '<' . $child->[0]->tag_name . ">\x0A"; ## ISSUE: case? | ||
| 96 | |||
| 97 | for my $attr (sort {$a->[0] cmp $b->[0]} map { [$_->name, $_->value] } | ||
| 98 | @{$child->[0]->attributes}) { | ||
| 99 | $r .= '| ' . $child->[1] . ' ' . $attr->[0] . '="'; ## ISSUE: case? | ||
| 100 | $r .= $attr->[1] . '"' . "\x0A"; | ||
| 101 | } | ||
| 102 | |||
| 103 | unshift @node, | ||
| 104 | map { [$_, $child->[1] . ' '] } @{$child->[0]->child_nodes}; | ||
| 105 | } elsif ($nt == $child->[0]->TEXT_NODE) { | ||
| 106 | $r .= '| ' . $child->[1] . '"' . $child->[0]->data . '"' . "\x0A"; | ||
| 107 | } elsif ($nt == $child->[0]->COMMENT_NODE) { | ||
| 108 | $r .= '| ' . $child->[1] . '<!-- ' . $child->[0]->data . " -->\x0A"; | ||
| 109 | } elsif ($nt == $child->[0]->DOCUMENT_TYPE_NODE) { | ||
| 110 | $r .= '| ' . $child->[1] . '<!DOCTYPE ' . $child->[0]->name . ">\x0A"; | ||
| 111 | } else { | ||
| 112 | $r .= '| ' . $child->[1] . $child->[0]->node_type . "\x0A"; # error | ||
| 113 | } | ||
| 114 |