X-Git-Url: https://git.stderr.nl/gitweb?p=matthijs%2Fupstream%2Fblosxom.git;a=blobdiff_plain;f=blosxom.cgi;h=f6473743edd2d5e229b24526b00583902e3c3fa4;hp=a49237827cedb6cb6f9efd80753aae7abe269210;hb=HEAD;hpb=15dfb8340ae2dcc2daf7f426a60fe5a8edeb950c diff --git a/blosxom.cgi b/blosxom.cgi index a492378..f647374 100755 --- a/blosxom.cgi +++ b/blosxom.cgi @@ -1,8 +1,8 @@ #!/usr/bin/perl # Blosxom -# Author: Rael Dornfest (2002-2003), The Blosxom Development Team (2005-2008) -# Version: 2.1.2 ($Id: blosxom.cgi,v 1.88 2008/11/13 17:19:51 alfie Exp $) +# Author: Rael Dornfest (2002-2003), The Blosxom Development Team (2005-2009) +# Version: 2.1.2 ($Id: blosxom.cgi,v 1.98 2009/07/19 17:18:37 xtaran Exp $) # Home/Docs/Licensing: http://blosxom.sourceforge.net/ # Development/Downloads: http://sourceforge.net/projects/blosxom @@ -75,12 +75,16 @@ $blog_encoding = "UTF-8"; # Where are this blog's entries kept? $datadir = "/Library/WebServer/Documents/blosxom"; -# What's my preferred base URL for this blog (leave blank for automatic)? +# What's my preferred base URL for this blog (leave blank for +# automatic)? $url = ""; # Should I stick only to the datadir for items or travel down the # directory hierarchy looking for items? If so, to what depth? -# 0 = infinite depth (aka grab everything), 1 = datadir only, n = n levels down +# +# 0 = infinite depth (aka grab everything), 1 = datadir only, +# n = n levels down + $depth = 0; # How many entries should I show on the home page? @@ -97,8 +101,8 @@ $show_future_entries = 0; # --- Plugins (Optional) ----- -# File listing plugins blosxom should load -# (if empty blosxom will load all plugins in $plugin_dir and $plugin_path directories) +# File listing plugins blosxom should load (if empty blosxom will load +# all plugins in $plugin_dir and $plugin_path directories) $plugin_list = ""; # Where are my plugins kept? @@ -107,8 +111,8 @@ $plugin_dir = ""; # Where should my plugins keep their state information? $plugin_state_dir = "$plugin_dir/state"; -# Additional plugins location -# List of directories, separated by ';' on windows, ':' everywhere else +# Additional plugins location. A list of directories, separated by ';' +# on windows, ':' everywhere else. $plugin_path = ""; # --- Static Rendering ----- @@ -116,7 +120,8 @@ $plugin_path = ""; # Where are this blog's static files to be created? $static_dir = "/Library/WebServer/Documents/blog"; -# What's my administrative password (you must set this for static rendering)? +# What's my administrative password (you must set this for static +# rendering)? $static_password = ""; # What flavours should I generate statically? @@ -126,9 +131,22 @@ $static_password = ""; # 0 = no, 1 = yes $static_entries = 0; -# Should I encode entities for xml content-types? (plugins can turn this off if they do it themselves) +# --- Advanced Encoding Options ----- + +# Should I encode entities for xml content-types? (plugins can turn +# this off if they do it themselves) $encode_xml_entities = 1; +# Should I encode 8 bit special characters, e.g. umlauts in URLs, e.g. +# convert an ISO-Latin-1 \"o to %F6? (off by default for now; plugins +# can change this, too) +$encode_8bit_chars = 0; + +# RegExp matching all characters which should be URL encoded in links. +# Defaults to anything but numbers, letters, slash, colon, dash, +# underscore and dot. +$url_escape_re = qr([^-/a-zA-Z0-9:._]); + # -------------------------------- =head1 ENVIRONMENT @@ -177,9 +195,54 @@ development was picked up by a team of dedicated users of blosxom since =cut - -use vars - qw! $version $blog_title $blog_description $blog_language $blog_encoding $datadir $url %template $template $depth $num_entries $file_extension $default_flavour $static_or_dynamic $config_dir $plugin_list $plugin_path $plugin_dir $plugin_state_dir @plugins %plugins $static_dir $static_password @static_flavours $static_entries $path_info_full $path_info $path_info_yr $path_info_mo $path_info_da $path_info_mo_num $flavour $static_or_dynamic %month2num @num2month $interpolate $entries $output $header $show_future_entries %files %indexes %others $encode_xml_entities $content_type !; +use vars qw! + $version + $blog_title + $blog_description + $blog_language + $blog_encoding + $datadir + $url + %template + $template + $depth + $num_entries + $file_extension + $default_flavour + $static_or_dynamic + $config_dir + $plugin_list + $plugin_path + $plugin_dir + $plugin_state_dir + @plugins + %plugins + $static_dir + $static_password + @static_flavours + $static_entries + $path_info_full + $path_info + $path_info_yr + $path_info_mo + $path_info_da + $path_info_mo_num + $flavour + %month2num + @num2month + $interpolate + $entries + $output + $header + $show_future_entries + %files + %indexes + %others + $encode_xml_entities + $encode_8bit_chars + $url_escape_re + $content_type + !; use strict; use FileHandle; @@ -188,7 +251,7 @@ use File::stat; use Time::Local; use CGI qw/:standard :netscape/; -$version = "2.1.2"; +$version = "2.1.2+dev"; # Load configuration from $ENV{BLOSXOM_CONFIG_DIR}/blosxom.conf, if it exists my $blosxom_config; @@ -247,7 +310,7 @@ unless ($url) { $url = url(); # Unescape %XX hex codes (from URI::Escape::uri_unescape) - $url =~ s/%([0-9A-Fa-f]{2})/chr(hex($1))/eg; + $url =~ s/%([0-9A-Fa-f]{2})/chr(hex($1))/eg; # Support being called from inside a SSI document $url =~ s/^included:/http:/ if $ENV{SERVER_PROTOCOL} eq 'INCLUDED'; @@ -298,14 +361,14 @@ else { # Path Info Magic # Take a gander at HTTP's PATH_INFO for optional blog name, archive yr/mo/day my @path_info = split m{/}, path_info() || param('path'); -$path_info_full = join '/', @path_info; # Equivalent to $ENV{PATH_INFO} +$path_info_full = join '/', @path_info; # Equivalent to $ENV{PATH_INFO} shift @path_info; # Flavour specified by ?flav={flav} or index.{flav} $flavour = ''; -if (! ($flavour = param('flav'))) { +if ( !( $flavour = param('flav') ) ) { if ( $path_info[$#path_info] =~ /(.+)\.(.+)$/ ) { - $flavour = $2; + $flavour = $2; pop @path_info if $1 eq 'index'; } } @@ -315,45 +378,50 @@ $flavour ||= $default_flavour; $flavour = blosxom_html_escape($flavour); sub blosxom_html_escape { - my $string = shift; - my %escape = ( - '<' => '<', - '>' => '>', - '&' => '&', - '"' => '"', - "'" => ''' - ); - my $escape_re = join '|' => keys %escape; - $string =~ s/($escape_re)/$escape{$1}/g; - $string; + my $string = shift; + my %escape = ( + '<' => '<', + '>' => '>', + '&' => '&', + '"' => '"', + "'" => ''' + ); + my $escape_re = join '|' => keys %escape; + $string =~ s/($escape_re)/$escape{$1}/g; + $string; } # Global variable to be used in head/foot.{flavour} templates $path_info = ''; + # Add all @path_info elements to $path_info till we come to one that could be a year -while ( $path_info[0] && $path_info[0] !~ /^(19|20)\d{2}$/) { +while ( $path_info[0] && $path_info[0] !~ /^(19|20)\d{2}$/ ) { $path_info .= '/' . shift @path_info; } # Pull date elements out of path -if ($path_info[0] && $path_info[0] =~ /^(19|20)\d{2}$/) { - $path_info_yr = shift @path_info; - if ($path_info[0] && - ($path_info[0] =~ /^(0\d|1[012])$/ || - exists $month2num{ ucfirst lc $path_info_mo })) { - $path_info_mo = shift @path_info; - # Map path_info_mo to numeric $path_info_mo_num - $path_info_mo_num = $path_info_mo =~ /^\d{2}$/ - ? $path_info_mo - : $month2num{ ucfirst lc $path_info_mo }; - if ($path_info[0] && $path_info[0] =~ /^[0123]\d$/) { - $path_info_da = shift @path_info; +if ( $path_info[0] && $path_info[0] =~ /^(19|20)\d{2}$/ ) { + $path_info_yr = shift @path_info; + if ($path_info[0] + && ( $path_info[0] =~ /^(0\d|1[012])$/ + || exists $month2num{ ucfirst lc $path_info_mo } ) + ) + { + $path_info_mo = shift @path_info; + + # Map path_info_mo to numeric $path_info_mo_num + $path_info_mo_num + = $path_info_mo =~ /^\d{2}$/ + ? $path_info_mo + : $month2num{ ucfirst lc $path_info_mo }; + if ( $path_info[0] && $path_info[0] =~ /^[0123]\d$/ ) { + $path_info_da = shift @path_info; + } } - } } # Add remaining path elements to $path_info -$path_info .= '/' . join('/', @path_info); +$path_info .= '/' . join( '/', @path_info ); # Strip spurious slashes $path_info =~ s!(^/*)|(/*$)!!g; @@ -396,7 +464,7 @@ my @plugin_list = (); my %plugin_hash = (); # If $plugin_list is set, read plugins to use from that file -if ( $plugin_list ) { +if ($plugin_list) { if ( -r $plugin_list and $fh->open("< $plugin_list") ) { @plugin_list = map { chomp $_; $_ } grep { /\S/ && !/^#/ } <$fh>; $fh->close; @@ -408,7 +476,7 @@ if ( $plugin_list ) { } # Otherwise walk @plugin_dirs to get list of plugins to use -if ( ! @plugin_list && @plugin_dirs ) { +if ( !@plugin_list && @plugin_dirs ) { for my $plugin_dir (@plugin_dirs) { next unless -d $plugin_dir; if ( opendir PLUGINS, $plugin_dir ) { @@ -438,7 +506,7 @@ foreach my $plugin (@plugin_list) { my $on_off = $off eq '_' ? -1 : 1; # Allow perl module plugins - # The -z test is a hack to allow a zero-length placeholder file in a + # The -z test is a hack to allow a zero-length placeholder file in a # $plugin_path directory to indicate an @INC module should be loaded if ( $plugin =~ m/::/ && ( $plugin_list || -z $plugin_hash{$plugin} ) ) { @@ -482,22 +550,18 @@ sub load_template { # Define default entries subroutine $entries = sub { my ( %files, %indexes, %others ); + my $param_all = param('-all'); find( sub { - my $d; my $curr_depth = $File::Find::dir =~ tr[/][]; return if $depth and $curr_depth > $depth; + return if !-r $File::Find::name; - if ( - - # a match - $File::Find::name - =~ m!^$datadir/(?:(.*)/)?(.+)\.$file_extension$! - - # not an index, .file, and is readable - and $2 ne 'index' and $2 !~ /^\./ and ( -r $File::Find::name ) - ) + # if a $file_extension file and not a .file or an index + if ( m/^([^.].*)\.$file_extension$/ + and $1 ne 'index' ) { + my $basename_noext = $1; # read modification time my $mtime = stat($File::Find::name)->mtime or return; @@ -509,22 +573,23 @@ $entries = sub { $files{$File::Find::name} = $mtime; # static rendering bits + (my $dirname = $File::Find::dir) =~ s!^$datadir/?!!; my $static_file - = "$static_dir/$1/index." . $static_flavours[0]; - if ( param('-all') + = "$static_dir/${dirname}index.$static_flavours[0]"; + if ( $param_all or !-f $static_file or stat($static_file)->mtime < $mtime ) { - $indexes{$1} = 1; - $d = join( '/', ( nice_date($mtime) )[ 5, 2, 3 ] ); + $indexes{$dirname} = 1; + my $d = join( '/', ( nice_date($mtime) )[ 5, 2, 3 ] ); $indexes{$d} = $d; - $indexes{ ( $1 ? "$1/" : '' ) . "$2.$file_extension" } = 1 + $indexes{"$dirname$basename_noext.$file_extension"} = 1 if $static_entries; } } # not an entries match - elsif ( !-d $File::Find::name and -r $File::Find::name ) { + elsif ( !-d $File::Find::name ) { $others{$File::Find::name} = stat($File::Find::name)->mtime; } }, @@ -550,11 +615,7 @@ my ( $files, $indexes, $others ) = &$entries(); %indexes = %$indexes; # Static -if ( !$ENV{GATEWAY_INTERFACE} - and param('-password') - and $static_password - and param('-password') eq $static_password ) -{ +if ( $static_or_dynamic eq 'static' ) { param('-quiet') or print "Blosxom is generating static index pages...\n"; @@ -655,10 +716,13 @@ sub generate { # Define default interpolation subroutine $interpolate = sub { + package blosxom; my $template = shift; + # Interpolate scalars, namespaced scalars, and hash/hashref scalars - $template =~ s/(\$\w+(?:::\w+)*(?:(?:->)?{(['"]?)[-\w]+\2})?)/"defined $1 ? $1 : ''"/gee; + $template + =~ s/(\$\w+(?:::\w+)*(?:(?:->)?{([\'\"]?)[-\w]+\2})?)/"defined $1 ? $1 : ''"/gee; return $template; }; @@ -706,8 +770,7 @@ sub generate { # Define a default sort subroutine my $sort = sub { my ($files_ref) = @_; - return - sort { $files_ref->{$b} <=> $files_ref->{$a} } + return sort { $files_ref->{$b} <=> $files_ref->{$a} } keys %$files_ref; }; @@ -793,18 +856,22 @@ sub generate { } } - if ( $encode_xml_entities && - $content_type =~ m{\bxml\b} && - $content_type !~ m{\bxhtml\b} ) { - # Escape special characters inside the container + # Save unescaped versions and allow them to be used in + # flavour templates. + use vars qw/$url_unesc $path_unesc $fn_unesc/; + $url_unesc = $url; + $path_unesc = $path; + $fn_unesc = $fn; + + # Fix special characters in links inside XML content + if ( $encode_xml_entities + && $content_type =~ m{\bxml\b} + && $content_type !~ m{\bxhtml\b} ) + { - # The following line should be moved more towards to top for - # performance reasons -- Axel Beckert, 2008-07-22 - my $url_escape_re = qr([^-/a-zA-Z0-9:._]); + # Escape special characters inside the container - $url =~ s($url_escape_re)(sprintf('%%%02X', ord($&)))eg; - $path =~ s($url_escape_re)(sprintf('%%%02X', ord($&)))eg; - $fn =~ s($url_escape_re)(sprintf('%%%02X', ord($&)))eg; + &url_escape_url_path_and_fn(); # Escape <, >, and &, and to produce valid RSS $title = blosxom_html_escape($title); @@ -814,6 +881,11 @@ sub generate { $fn = blosxom_html_escape($fn); } + # Fix special characters in links inside XML content + if ($encode_8bit_chars) { + &url_escape_url_path_and_fn(); + } + $story = &$interpolate($story); $output .= $story; @@ -871,6 +943,12 @@ sub nice_date { return ( $dw, $mo, $mo_num, $da, $ti, $yr, $utc_offset ); } +sub url_escape_url_path_and_fn { + $url =~ s($url_escape_re)(sprintf('%%%02X', ord($&)))eg; + $path =~ s($url_escape_re)(sprintf('%%%02X', ord($&)))eg; + $fn =~ s($url_escape_re)(sprintf('%%%02X', ord($&)))eg; +} + # Default HTML and RSS template bits __DATA__ html content_type text/html; charset=$blog_encoding