banned-regexps: updated copyright year

BannedRegexps needs to check for BannedContent before it does
anything. BannedContent checks the URLs, and so BannedRegexps wants to
check everything except for the URLs. The order is important: the old
code used to remove the URLs before handing the text to BannedContent,
so it never found a thing. Changing the order fixed this.

Makefile

@@ -25,7 +25,7 @@ release:
 	perl stuff/release ~/oddmuse.org/releases 2.3.3
 
 build/wiki.pl: wiki.pl
-	perl -lne "s/(\\\$$q->a\(\{-href=>'https:\/\/www.oddmuse.org\/'\}, 'Oddmuse'\))/\\\$$q->a({-href=>'https:\/\/git.savannah.gnu.org\/cgit\/oddmuse.git\/tag\/?id=$(VERSION_NO)'}, 'wiki.pl') . ' ($(VERSION_NO)), see ' . \$$1/; print" < $< > $@
+	perl -lne "s/(\\\$$q->a\(\{-href=>'https:\/\/www.oddmuse.org\/'\}, 'Oddmuse'\))/\\\$$q->a({-href=>'https:\/\/alexschroeder.ch\/cgit\/oddmuse\/tag\/?id=$(VERSION_NO)'}, 'wiki.pl') . ' ($(VERSION_NO)), see ' . \$$1/; print" < $< > $@
 
 build/%-utf8.pl: modules/translations/%-utf8.pl
 	perl -lne "s/(AddModuleDescription\('[^']+', '[^']+')\)/\$$1, 'translations\/', '$(VERSION_NO)')/; print" < $< > $@

css/alex-2018.css

@@ -2,9 +2,9 @@
 html{ text-align: center; }
 
 body, rss {
-    font-family: "Palatino Linotype", "Book Antiqua", Palatino, serif;
+    font-family: "DejaVu Serif", Palatino, serif;
     font-style: normal;
-    font-size: 14pt;
+    font-size: 16pt;
     padding: 1em 3em;
     max-width: 72ex;
     display: inline-block;
@@ -13,10 +13,31 @@ body, rss {
     background-color: #fff;
 }
 
-@media only screen and (max-device-width: 480px) {
+@media only screen and (max-device-width: 600px) {
     body {
 	padding: 1ex;
     }
+    textarea {
+	font-size: inherit;
+    }
+}
+
+/* code */
+
+textarea, pre, code, tt {
+    font-family: "DejaVu Mono", "Andale Mono", Monaco, "Courier New", Courier, monospace;
+}
+
+pre, code, tt {
+    font-size: 12pt; /* fits 80ex */
+}
+pre {
+    overflow:hidden;
+    white-space: pre-wrap;      /* CSS 3 */
+    white-space: -moz-pre-wrap; /* Mozilla, since 1999 */
+    white-space: -pre-wrap;     /* Opera 4-6 */
+    white-space: -o-pre-wrap;   /* Opera 7 */
+    word-wrap: break-word;      /* Internet Explorer 5.5+ */
 }
 
 @import url(file:///home/alex/alexschroeder.ch/css/alex-2017.css) print;
@@ -191,22 +212,6 @@ input[type="text"] {
     line-height: 125%;
 }
 
-/* code */
-
-textarea, pre, code, tt {
-    font-family: "Andale Mono", Monaco, "Courier New", Courier, monospace, "Symbola";
-    font-size: 75%; /* fits 80ex */
-}
-
-pre {
-    overflow:hidden;
-    white-space: pre-wrap;      /* CSS 3 */
-    white-space: -moz-pre-wrap; /* Mozilla, since 1999 */
-    white-space: -pre-wrap;     /* Opera 4-6 */
-    white-space: -o-pre-wrap;   /* Opera 7 */
-    word-wrap: break-word;      /* Internet Explorer 5.5+ */
-}
-
 /* styling for divs that will be invisible when printing
    when printing. */
 
@@ -223,6 +228,9 @@ div.header, div.footer, div.near, div.definition, p.comment, a.tag {
 div.footer form.search {
     display: none;
 }
+
+/* Recent Changes */
+
 div.rc {
     overflow: hidden;
 }
@@ -233,12 +241,57 @@ div.rc li strong, table.history strong, strong.description {
     font-family: inherit;
     font-weight: inherit;
 }
+.red {
+    background: red;
+    color: red;
+}
+
+.orange {
+    background: orange;
+    color: orange;
+}
+
+.yellow {
+    background: yellow;
+    color: yellow;
+}
+
+.green {
+    background: green;
+    color: green;
+}
+
+.blue {
+    background: blue;
+    color: blue;
+}
+
+.indigo {
+    background: indigo;
+    color: indigo;
+}
+
+.violet {
+    background: violet;
+    color: violet;
+}
+
+.white {
+    background: white;
+    color: white;
+}
+
+.ip-code {
+    border: 1px solid #666;
+}
+
+/* Diff */
+
 div.diff {
     padding-left: 5%;
     padding-right: 5%;
     font-size: 12pt;
     color: #000;
-
 }
 div.old {
     background-color: #ffffaf;

css/wiki.css

@@ -263,3 +263,19 @@ code {
     background: #eee;
     white-space: pre-wrap;
 }
+
+/* for https://oddmuse.org/wiki/All_Modules */
+.foo_list + .journal h1 {
+    font: inherit;
+    border: none;
+    display: list-item;
+    margin-top: 0;
+    margin-left: 1em;
+}
+
+.foo_list + .journal a {
+    font: inherit;
+    border: none;
+    text-decoration: none;
+    color: #a00;
+}

modules/ban-mixed-scripts.pl

@@ -0,0 +1,72 @@
+# Copyright (C) 2018  Alex Schroeder <alex@gnu.org>
+
+# This program is free software: you can redistribute it and/or modify it under
+# the terms of the GNU General Public License as published by the Free Software
+# Foundation, either version 3 of the License, or (at your option) any later
+# version.
+#
+# This program is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License along with
+# this program. If not, see <http://www.gnu.org/licenses/>.
+
+=encoding utf8
+
+=head1 Mixed Scripts
+
+This module disallows ordinary users from posting words that consist of multiple
+scripts. Stuff like this: "It's diffіcult to find knowledgeable people on this
+topic, but youu sound like you know wgat you're taⅼkіng аboսt!" Did you notice
+the confusable characters? The sentence contains the following:
+ARMENIAN SMALL LETTER SEH
+CYRILLIC SMALL LETTER A
+CYRILLIC SMALL LETTER BYELORUSSIAN-UKRAINIAN I
+SMALL ROMAN NUMERAL FIFTY
+
+=cut
+
+use strict;
+use v5.10;
+
+use Unicode::UCD qw(charprop);
+
+AddModuleDescription('ban-mixed-scripts.pl', 'Ban Mixed Scripts Extension');
+
+*OldBanMixedScriptsBannedContent = \&BannedContent;
+*BannedContent = \&NewBanMixedScriptsBannedContent;
+
+sub NewBanMixedScriptsBannedContent {
+  my $rule = OldBanMixedScriptsBannedContent(@_);
+  $rule ||= BanMixedScript(@_);
+  return $rule;
+}
+
+sub BanMixedScript {
+  my $str = shift;
+  my @words = $str =~ m/\w+/g;
+  my %seen;
+  my %prop;
+  for my $word (@words) {
+    next if $seen{$word};
+    $seen{$word} = 1;
+    my $script;
+    for my $char (split(//, $word)) {
+      my $s = $prop{$char};
+      if (not $s) {
+	$s = charprop(ord($char), "Script_Extensions");
+	if ($s eq 'Hiragana') {
+	  $s = 'Han'; # this mixing is ok
+	}
+	$prop{$char} = $s;
+      }
+      next if $s eq "Common";
+      if (not $script) {
+	$script = $s;
+      } elsif ($script ne $s) {
+	return "Mixed scripts in $word ($script and $s, if not more)";
+      }
+    }
+  }
+}

modules/banned-regexps.pl

@@ -1,4 +1,4 @@
-# Copyright (C) 2012  Alex Schroeder <alex@gnu.org>
+# Copyright (C) 2012-2018  Alex Schroeder <alex@gnu.org>
 #
 # This program is free software; you can redistribute it and/or modify it under
 # the terms of the GNU General Public License as published by the Free Software
@@ -62,9 +62,10 @@ push(@MyInitVariables, sub {
 
 sub RegexpNewBannedContent {
   my $str = shift;
-  # remove URLs as they are controlled by $BannedContent
-  $str =~ s/$FullUrlPattern//g;
+  # check whether Banned Content complains
   my $rule = RegexpOldBannedContent($str, @_);
+  # remove URLs as they have been checked by $BannedContent
+  $str =~ s/$FullUrlPattern//g;
   if (not $rule) {
     foreach (split(/\n/, GetPageContent($BannedRegexps))) {
       next unless m/^\s*([^#]+?)\s*(#\s*(\d\d\d\d-\d\d-\d\d\s*)?(.*))?$/;

modules/journal-rss.pl

@@ -1,4 +1,4 @@
-# Copyright (C) 2004–2014  Alex Schroeder <alex@gnu.org>
+# Copyright (C) 2004–2018  Alex Schroeder <alex@gnu.org>
 #
 # This program is free software; you can redistribute it and/or modify
 # it under the terms of the GNU General Public License as published by
@@ -30,15 +30,43 @@ sub DoJournalRss {
   local $CollectingJournal = 1;
   # Fake the result of GetRcLines()
   local *GetRcLines = \&JournalRssGetRcLines;
+  local *RcSelfAction = \&JournalRssSelfAction;
+  local *RcPreviousAction = \&JournalRssPreviousAction;
+  local *RcLastAction = \&JournalRssLastAction;
+  SetParam('full', 1);
   print GetHttpHeader('application/xml') . GetRcRss();
 }
 
+sub JournalRssParameters {
+  my $more = '';
+  foreach (@_, qw(rsslimit match search reverse monthly)) {
+    my $val = GetParam($_, '');
+    $more .= ";$_=$val" if $val;
+  }
+  return $more;
+}
+
+sub JournalRssSelfAction {
+  return "action=journal" . JournalRssParameters(qw(offset));
+}
+
+sub JournalRssPreviousAction {
+  my $num = GetParam('rsslimit', 10);
+  my $offset = GetParam('offset', 0) + $num;
+  return "action=journal;offset=$offset" . JournalRssParameters();
+}
+
+sub JournalRssLastAction {
+  return "action=journal" . JournalRssParameters();
+}
+
 sub JournalRssGetRcLines {
   my $num = GetParam('rsslimit', 10);
   my $match = GetParam('match', '^\d\d\d\d-\d\d-\d\d');
   my $search = GetParam('search', '');
   my $reverse = GetParam('reverse', 0);
   my $monthly = GetParam('monthly', 0);
+  my $offset = GetParam('offset', 0);
   my @pages = sort JournalSort (grep(/$match/, $search ? SearchTitleAndBody($search) : AllPagesList()));
   if ($monthly and not $match) {
     my ($sec, $min, $hour, $mday, $mon, $year, $wday, $yday) = gmtime();
@@ -48,28 +76,19 @@ sub JournalRssGetRcLines {
     @pages = reverse @pages;
   }
   # FIXME: Missing 'future' and 'past' keywords.
-  # FIXME: Do we need 'offset'? I don't think so.
   my @result = ();
+  my $n = 0;
   foreach my $id (@pages) {
     # Now save information required for saving the cache of the current page.
     local %Page;
     local $OpenPageName = '';
     OpenPage($id);
-    # If this is a minor edit, ignore it. Load the last major revision
-    # instead, if you can.
-    if ($Page{minor}) {
-      # Perhaps the old kept revision is gone due to $KeepMajor=0 or
-      # admin.pl or because a page was created as a minor change and
-      # never edited. Reading kept revisions in this case results in
-      # an error.
-      eval {
- 	%Page = GetKeptRevision($Page{lastmajor});
-      };
-      next if $@;
-    }
     next if $Page{text} =~ /^\s*$/; # only whitespace is also to be deleted
     next if $DeletedPage && substr($Page{text}, 0, length($DeletedPage))
-      eq $DeletedPage; # no regexp
+	eq $DeletedPage; # no regexp
+    # OK, this is a candidate page
+    $n++;
+    next if $n <= $offset;
     # Generate artifical rows in the list to pass to GetRcRss. We need
     # to open every single page, because the meta-data ordinarily
     # available in the rc.log file is not available to us. This is why
@@ -79,7 +98,7 @@ sub JournalRssGetRcLines {
     push (@result, [$Page{ts}, $id, $Page{minor}, $Page{summary}, $Page{host},
 		    $Page{username}, $Page{revision}, \@languages,
 		    GetCluster($Page{text})]);
-    last if $num ne 'all' and $#result + 1 >= $num;
+    last if @result >= $num;
   }
   return @result;
 }

modules/tags.pl

@@ -239,6 +239,11 @@ sub TagFind {
   return @result;
 }
 
+sub TagsTerms {
+  my $string = shift;
+  return grep(/./, $string =~ /\"([^\"]+)\"|(\S+)/g);
+}
+
 *OldTagFiltered = \&Filtered;
 *Filtered = \&NewTagFiltered;
 
@@ -246,7 +251,7 @@ sub NewTagFiltered { # called within a lock!
   my ($string, @pages) = @_;
   my %page = map { $_ => 1 } @pages;
   # looking at all the "tag:SOME TERMS" and and tag:TERM
-  my @tagterms = map { FreeToNormal($_) } grep(/^-?tag:/, shift =~ /\"([^\"]+)\"|(\S+)/g);
+  my @tagterms = map { FreeToNormal($_) } grep(/^-?tag:/, TagsTerms($string));
   my @positives = map {substr($_, 4)} grep(/^tag:/, @tagterms);
   my @negatives = map {substr($_, 5)} grep(/^-tag:/, @tagterms);
   if (@positives) {
@@ -260,8 +265,9 @@ sub NewTagFiltered { # called within a lock!
   foreach my $id (TagFind(@negatives)) {
     delete $page{$id};
   }
-  # filter out the tags from the search string
-  $string = join(' ', grep(!/^-?tag:/, $string =~ /\"([^\"]+)\"|(\S+)/g));
+  # filter out the tags from the search string, and add quotes which might have
+  # been stripped
+  $string = join(' ', map { qq{"$_"} } grep(!/^-?tag:/, TagsTerms($string)));
   # run the old code for any remaining search terms
   return OldTagFiltered($string, sort keys %page);
 }
@@ -279,11 +285,12 @@ We're need to remove all tag terms (again) in order to not confuse it.
 *SearchString = \&NewTagSearchString;
 
 sub NewTagSearchString {
-  # filter out the negative tags from the search string
-  my $string = join(' ', map { NormalToFree($_) }
-		    grep(!/^-tag:/, shift =~ /\"([^\"]+)\"|(\S+)/g));
+  my ($string, @rest) = @_;
+  # filter out the negative tags from the search string, and add quotes which
+  # might have been stripped
+  $string = join(' ', map { NormalToFree($_) } map { qq{"$_"} } grep(!/^-tag:/, TagsTerms($string)));
   return 1 unless $string;
-  return OldTagSearchString($string, @_);
+  return OldTagSearchString($string, @rest);
 }
 
 =pod

stuff/gopher-server.pl

@@ -20,10 +20,13 @@ use base qw(Net::Server::Fork); # any personality will do
 use MIME::Base64;
 use Text::Wrap;
 use List::Util qw(first);
+use Socket;
 
-our($RunCGI, $DataDir, %IndexHash, @IndexList, $IndexFile, $TagFile, $q,
-    %Page, $OpenPageName, $MaxPost, $ShowEdits, %Locks, $CommentsPattern,
-    $CommentsPrefix, $EditAllowed, $NoEditFile, $SiteName, $ScriptName);
+our($RunCGI, $DataDir, %IndexHash, @IndexList, $IndexFile, $TagFile, $q, %Page,
+    $OpenPageName, $MaxPost, $ShowEdits, %Locks, $CommentsPattern,
+    $CommentsPrefix, $EditAllowed, $NoEditFile, $SiteName, $ScriptName, $Now,
+    %RecentVisitors, $SurgeProtection, $SurgeProtectionTime,
+    $SurgeProtectionViews);
 
 my $external_image_path = '/home/alex/alexschroeder.ch/pics/';
 
@@ -93,6 +96,12 @@ sub post_configure_hook {
   # make sure search is sorted newest first because NewTagFiltered resorts
   *OldGopherFiltered = \&Filtered;
   *Filtered = \&NewGopherFiltered;
+  *ReportError = sub {
+    my ($error, $status, $log, @html) = @_;
+    $self->print_error("Error: $error");
+    map { ReleaseLockDir($_); } keys %Locks;
+    exit 2;
+  };
 }
 
 my $usage = << 'EOT';
@@ -435,9 +444,21 @@ sub serve_text_page_menu {
   $self->serve_page_history_link($id, $revision);
 
   my $first = 1;
-  while ($page->{text} =~ /\[\[([^\]|]*)(?:\|([^\]]*))?\]\]|\[(https?:\/\/\S+)\s+([^\]]*)\]|\[gopher:\/\/([^:\/]*)(?::(\d+))?(?:\/(\d)(\S+))?\s+([^\]]+)\]/g) {
-    my ($title, $text, $url, $hostname, $port, $type, $selector)
-	= ($1, $2||$4||$9, $3, $5, $6||70, $7||1, $8);
+  while ($page->{text} =~ /
+	 \[\[ (?<title>[^\]|]*) (?:\|(?<text>[^\]]*))? \]\]
+	 | \[ (?<url>https?:\/\/\S+) \s+ (?<text>[^\]]*) \]
+	 | \[ (?<text>[^\]]*) \] \( (?<url>https?:\/\/\S+) \)
+	 | \[ gopher:\/\/ (?<hostname>[^:\/]*) (?::(?<port>\d+))?
+	      (?:\/(?<type>\d) (?<selector>\S+))?
+              \s+ (?<text>[^\]]+)\]
+	 | \[ (?<text>[^\]]+) \]
+           \( gopher:\/\/ (?<hostname>[^:\/]*) (?::(?<port>\d+))?
+	      (?:\/(?<type>\d) (?<selector>\S+))? \)
+	 /xg) {
+    my ($title, $text, $url, $hostname,
+	$port, $type, $selector)
+	= ($+{title}, $+{text}, $+{url}, $+{hostname},
+	   $+{port}||70, $+{type}||1, $+{selector});
     if ($first) {
       $self->print_info("");
       $self->print_info("Links leaving " . normal_to_free($id) . ":");
@@ -801,13 +822,41 @@ sub read_text {
   return $buf;
 }
 
-sub process_request {
+sub allow_deny_hook {
   my $self = shift;
+  my $client = shift;
 
-  # clear cookie and all that
+  # clear cookie, read config file
   $q = undef;
   Init();
 
+  # don't do surge protection if we're testing
+  return 1 unless $SurgeProtection;
+
+  # get the client IP number
+  my $peeraddr = $self->{server}->{'peeraddr'};
+
+  # implement standard surge protection using Oddmuse tools but without using
+  # ReportError and all that
+  $self->log(4, "Adding visitor $peeraddr");
+  ReadRecentVisitors();
+  AddRecentVisitor($peeraddr);
+  if (RequestLockDir('visitors')) { # not fatal
+    WriteRecentVisitors();
+    ReleaseLockDir('visitors');
+    my @entries = @{$RecentVisitors{$peeraddr}};
+    my $ts = $entries[$SurgeProtectionViews];
+    if (($Now - $ts) < $SurgeProtectionTime) {
+      $self->log(2, "Too many requests by $peeraddr");
+      return 0;
+    }
+  }
+  return 1;
+}
+
+sub process_request {
+  my $self = shift;
+
   # refresh list of pages
   if (IsFile($IndexFile) and ReadIndex()) {
     # we're good

stuff/release

@@ -33,7 +33,7 @@ EOT
 
 my $min = version->parse(shift || "2.3.0");
 
-my @tags = grep { /\d+\.\d+\.\d+/ and version->parse($_) >= $min } 
+my @tags = grep { /(\d+\.\d+\.\d+)/ and version->parse($1) >= $min } 
            split(/\n/, qx{git tag --list});
 
 unless (@tags) {

t/ban-mixed-scripts.t

@@ -0,0 +1,33 @@
+# Copyright (C) 2018  Alex Schroeder <alex@gnu.org>
+#
+# This program is free software: you can redistribute it and/or modify it under
+# the terms of the GNU General Public License as published by the Free Software
+# Foundation, either version 3 of the License, or (at your option) any later
+# version.
+#
+# This program is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License along with
+# this program. If not, see <http://www.gnu.org/licenses/>.
+
+require './t/test.pl';
+package OddMuse;
+use Test::More tests => 4;
+use utf8; # tests contain UTF-8 characters and it matters
+
+add_module('ban-mixed-scripts.pl');
+
+# ordinary page editing still works
+test_page(update_page('Test', 'This is a test'),
+	  'This is a test');
+test_page(update_page('Test', '🙇🏽‍ 本当にごめんね – I am really sorry.'),
+	  'I am really sorry');
+
+# mixed scripts are not ok
+test_page(update_page('Test', "It's diffіcult to find knowledgeable people on this topic, but youu sound like you know wgat you're taⅼkіng аboսt!"),
+	  'I am really sorry');
+
+# error message is shown
+test_page($redirect, "Mixed scripts");

t/ban.t

@@ -1,4 +1,4 @@
-# Copyright (C) 2006, 2007, 2010  Alex Schroeder <alex@gnu.org>
+# Copyright (C) 2006-2018  Alex Schroeder <alex@gnu.org>
 #
 # This program is free software: you can redistribute it and/or modify it under
 # the terms of the GNU General Public License as published by the Free Software
@@ -14,7 +14,7 @@
 
 require './t/test.pl';
 package OddMuse;
-use Test::More tests => 23;
+use Test::More tests => 27;
 
 $localhost = 'confusibombus';
 $ENV{'REMOTE_ADDR'} = $localhost;
@@ -96,3 +96,23 @@ test_page(get_page("action=rollback to=$ts id=CriminalPage username=Alex"),
 test_page(get_page("action=rollback to=$ts id=CriminalPage pwd=foo"),
 	  'Rolling back changes',
 	  'CriminalPage</a> rolled back');
+
+# make sure it also doesn't work in the homepage field for comments
+AppendStringToFile($ConfigFile, "\$CommentsPrefix = 'Comments on ';\n");
+
+# mafia is still banned
+test_page(get_page(join(' ', 'title=Comments_on_CriminalPage',
+			'aftertext=Innocent',
+			'username=Alex',
+			'homepage=http://mafia.example.com')),
+	  'Reason: crime');
+
+# but it still works!
+test_page(get_page(join(' ', 'title=Comments_on_CriminalPage',
+			'aftertext=Innocent',
+			'username=Alex',
+			'homepage=http://police.example.com')),
+	  'Status: 302');
+test_page(get_page('Comments_on_CriminalPage'),
+	  'Innocent',
+	  'http://police.example.com');

t/banned-regexps.t

@@ -1,4 +1,4 @@
-# Copyright (C) 2017  Alex Schroeder <alex@gnu.org>
+# Copyright (C) 2017-2018  Alex Schroeder <alex@gnu.org>
 #
 # This program is free software: you can redistribute it and/or modify it under
 # the terms of the GNU General Public License as published by the Free Software
@@ -14,7 +14,7 @@
 
 require './t/test.pl';
 package OddMuse;
-use Test::More tests => 13;
+use Test::More tests => 15;
 
 add_module('banned-regexps.pl');
 
@@ -63,3 +63,13 @@ test_page(update_page('Test2', 'Voldemort', 'one banned word'),
 
 # Make sure the underscores don't show up in the page link
 test_page(get_page('action=admin'), 'Local Banned Regexps');
+
+# Make sure it doesn't break BannedContent!
+
+# mafia is banned
+update_page('BannedContent', 'mafia', 'one banned word', 0, 1);
+test_page(update_page('CriminalPage', 'This is about http://mafia.example.com'),
+	  'This page does not exist');
+
+# error message is shown
+test_page($redirect, 'Edit Denied');

t/journal-rss.t

@@ -1,4 +1,4 @@
-# Copyright (C) 2008, 2009  Alex Schroeder <alex@gnu.org>
+# Copyright (C) 2008-2018  Alex Schroeder <alex@gnu.org>
 #
 # This program is free software: you can redistribute it and/or modify
 # it under the terms of the GNU General Public License as published by
@@ -15,33 +15,26 @@
 
 require './t/test.pl';
 package OddMuse;
-use Test::More tests => 48;
+use Test::More tests => 41;
 
 add_module('journal-rss.pl');
 
-# summaries eq page content since no summaries are provided
 update_page('2008-09-21', 'first page');
 update_page('2008-09-22', 'second page'); # major
 sleep(1);
-update_page('2008-09-22', 'third edit', 'third edit', 1); # minor
+update_page('2008-09-22', 'third edit content', 'third edit summary', 1); # minor
 update_page('unrelated', 'wrong page');
 
 my $page = get_page('action=journal');
 test_page($page,
 	  '2008-09-21', 'first page',
-	  # ignore minor edits are ignored: show last major edit
-	  # instead
-	  '2008-09-22', 'second page',
+	  # make sure we're showing full page content, not summaries
+	  '2008-09-22', 'third edit content',
 	  # reverse sort is the default
 	  '2008-09-22(.*\n)+.*2008-09-21');
 
 # make sure unrelated pages and minor edits don't show up
-test_page_negative($page, 'unrelated', 'wrong page',
-		   'third edit');
-
-# verify the order of pages
-test_page(get_page('action=journal'),
-	  '2008-09-22(.*\n)+.*2008-09-21');
+test_page_negative($page, 'unrelated', 'wrong page');
 
 # reverse the order
 test_page(get_page('action=journal reverse=1'),
@@ -52,21 +45,6 @@ $page = get_page('action=journal match=21');
 test_page($page, '2008-09-21', 'first page');
 test_page_negative($page, '2008-09-22', 'second page');
 
-# search parameter
-$page = get_page('action=journal search=second');
-
-# no pages found, since this is for an old revision!
-test_page_negative($page,
-		   '2008-09-21', 'first page',
-		   '2008-09-22', 'second page',
-		   'third edit');
-
-# strange but true: search returns a page based on the minor edit but
-# shows the latest major revision that doesn't actually match.
-$page = get_page('action=journal search=third');
-test_page($page, '2008-09-22', 'second page');
-test_page_negative($page, '2008-09-21', 'first page', 'third edit');
-
 # testing the limit default
 update_page('2008-09-05', 'page');
 update_page('2008-09-06', 'page');
@@ -98,17 +76,23 @@ $page = get_page('action=journal rsslimit=1');
 test_page($page, '2008-09-22');
 test_page_negative($page, '2008-09-21');
 
-$page = get_page('action=journal rsslimit=all');
-test_page($page, '2008-09-22', '2008-09-05');
+# make sure we start from a well-known point in time
+AppendStringToFile($ConfigFile, "push(\@MyInitVariables, sub { \$Now = '$Now' });\n");
 
-# Now let's show that we're using the timestamp of the last major
-# change if possible.
+# check default RSS
+xpath_test(get_page('action=journal'),
+	   '//atom:link[@rel="self"][@href="http://localhost/wiki.pl?action=journal"]',
+	   '//atom:link[@rel="last"][@href="http://localhost/wiki.pl?action=journal"]',
+	   '//atom:link[@rel="previous"][@href="http://localhost/wiki.pl?action=journal;offset=10"]');
 
-my @dates = get_page('action=rss showedit=1 all=1 match=2008-09-22')
-  =~ m!<pubDate>(.*?)</pubDate>!g;
-# $dates[0] is the channel pubDate
-my $date2 = $dates[1]; # revision 2 comes first
-my $date1 = $dates[2]; # revision 1 comes second
-my ($item) = $page =~ m!(<item>\n<title>2008-09-22</title>\n(.*\n)+?</item>\n)!;
-test_page($item, "<pubDate>$date1</pubDate>");
-test_page_negative($item, "<pubDate>$date2</pubDate>");
+# check next page
+xpath_test(get_page('action=journal offset=10'),
+	   '//atom:link[@rel="self"][@href="http://localhost/wiki.pl?action=journal;offset=10"]',
+	   '//atom:link[@rel="last"][@href="http://localhost/wiki.pl?action=journal"]',
+	   '//atom:link[@rel="previous"][@href="http://localhost/wiki.pl?action=journal;offset=20"]');
+
+# check next page but with a tag search
+xpath_test(get_page('action=journal search=tag:oddmuse'),
+	   '//atom:link[@rel="self"][@href="http://localhost/wiki.pl?action=journal;search=tag:oddmuse"]',
+	   '//atom:link[@rel="last"][@href="http://localhost/wiki.pl?action=journal;search=tag:oddmuse"]',
+	   '//atom:link[@rel="previous"][@href="http://localhost/wiki.pl?action=journal;offset=10;search=tag:oddmuse"]');

t/rss.t

@@ -1,4 +1,4 @@
-# Copyright (C) 2006–2015  Alex Schroeder <alex@gnu.org>
+# Copyright (C) 2006–2018  Alex Schroeder <alex@gnu.org>
 #
 # This program is free software; you can redistribute it and/or modify
 # it under the terms of the GNU General Public License as published by
@@ -15,7 +15,7 @@
 
 require './t/test.pl';
 package OddMuse;
-use Test::More tests => 126;
+use Test::More tests => 135;
 use utf8; # tests contain UTF-8 characters and it matters
 
 AppendStringToFile($ConfigFile, "\$CommentsPrefix = 'Comments on ';\n");
@@ -107,6 +107,36 @@ update_page('big', "mee too\n" x 2 . 'x' x 55000);
 test_page(get_page('action=rss full=1'), 'too big to send over RSS');
 test_page(get_page('action=rss full=1 diff=1'), 'mee too', 'too big to send over RSS');
 
+# pagination
+my $interval = $RcDefault * 24 * 60 * 60;
+my $t1 = $Now - $interval;
+my $t2 = $Now - 2 * $interval;
+my $t3 = $Now - 3 * $interval;
+my $action1 = " from=$t2 upto=$t1";
+my $window1 = ";from=$t2;upto=$t1";
+my $window2 = ";from=$t3;upto=$t2";
+
+# make sure we start from a well-known point in time
+AppendStringToFile($ConfigFile, "push(\@MyInitVariables, sub { \$Now = '$Now' });\n");
+
+# check default RSS
+xpath_test(get_page('action=rss'),
+	   '//atom:link[@rel="self"][@href="http://localhost/wiki.pl?action=rss"]',
+	   '//atom:link[@rel="last"][@href="http://localhost/wiki.pl?action=rss"]',
+	   '//atom:link[@rel="previous"][@href="http://localhost/wiki.pl?action=rss' . $window1 . '"]');
+
+# check next page
+xpath_test(get_page('action=rss' . $action1),
+	   '//atom:link[@rel="self"][@href="http://localhost/wiki.pl?action=rss' . $window1 . '"]',
+	   '//atom:link[@rel="last"][@href="http://localhost/wiki.pl?action=rss"]',
+	   '//atom:link[@rel="previous"][@href="http://localhost/wiki.pl?action=rss' . $window2 . '"]');
+
+# check next page but with full pages
+xpath_test(get_page('action=rss full=1' . $action1),
+	   '//atom:link[@rel="self"][@href="http://localhost/wiki.pl?action=rss' . $window1 . ';full=1"]',
+	   '//atom:link[@rel="last"][@href="http://localhost/wiki.pl?action=rss;full=1"]',
+	   '//atom:link[@rel="previous"][@href="http://localhost/wiki.pl?action=rss' . $window2 . ';full=1"]');
+
 SKIP: {
 
   eval {

t/tags.t

@@ -1,4 +1,4 @@
-# Copyright (C) 2006, 2009  Alex Schroeder <alex@gnu.org>
+# Copyright (C) 2006-2018  Alex Schroeder <alex@gnu.org>
 #
 # This program is free software; you can redistribute it and/or modify
 # it under the terms of the GNU General Public License as published by
@@ -15,12 +15,27 @@
 
 require './t/test.pl';
 package OddMuse;
-use Test::More tests => 70;
+use Test::More tests => 94;
 use utf8;
 
 add_module('tags.pl');
 InitVariables();
 
+# first, let's make sure the old search still works
+test_page(join(':', TagsTerms('a b "c d"')), 'a:b:c d');
+
+update_page('test', 'foo bar baz');
+test_page(get_page('search=foo'), 'test', 'Search for: foo', '<strong>foo</strong>');
+test_page(get_page('search=foo+bar'), 'test', 'Search for: foo bar', '<strong>foo</strong> <strong>bar</strong>');
+test_page(get_page('search=bar+foo'), 'test', 'Search for: bar foo', '<strong>foo</strong> <strong>bar</strong>');
+test_page(get_page('search=foo+baz'), 'test', 'Search for: foo baz', '<strong>foo</strong> bar <strong>baz</strong>');
+$page = get_page('search=%22foo+baz%22');
+test_page($page, 'Search for: "foo baz"');
+test_page_negative($page, 'test');
+test_page(get_page('search=%22foo+bar%22'), 'test', 'Search for: "foo bar"', '<strong>foo bar</strong>');
+
+# testing other stuff
+
 $TagFeedIcon = 'http://www.example.org/pics/rss.png';
 
 xpath_run_tests(split('\n',<<'EOT'));
@@ -160,6 +175,12 @@ $page = update_page('Podcasts', '<journal "." search tag:podcast>');
 test_page($page, qw(Pödgecäst´s Brilliant Sons));
 test_page_negative($page, qw(Alex Foo));
 
+# check recent changes
+test_page(get_page('action=rc'), qw(Brilliant Alex Jeff));
+$page = get_page('action=rc rcfilteronly=tag:Podcast');
+test_page($page, qw(Brilliant));
+test_page_negative($page, qw(Alex Jeff));
+
 # check the tag cloud
 xpath_test(get_page('action=tagcloud'),
 	   '//h1[text()="Tag Cloud"]',

wiki.pl

@@ -1700,15 +1700,35 @@ sub RcHeader {
   return $html;
 }
 
-sub GetScriptUrlWithRcParameters {
-  my $url = "$ScriptName?action=rss";
-  foreach my $param (qw(from upto days all showedit rollback rcidonly rcuseronly
-			rchostonly rcclusteronly rcfilteronly match lang
-			followup page diff full)) {
-    my $val = GetParam($param, undef);
-    $url .= ";$param=$val" if defined $val;
+sub RcOtherParameters {
+  my $more = '';
+  foreach (@_, qw(page diff full all showedit rollback rcidonly rcuseronly rchostonly rcclusteronly rcfilteronly match lang followup)) {
+    my $val = GetParam($_, '');
+    $more .= ";$_=$val" if $val;
   }
-  return $url;
+  return $more;
+}
+
+sub RcSelfAction {
+  my $action = GetParam('action', 'rc');
+  return "action=$action" . RcOtherParameters(qw(from upto days));
+}
+
+sub RcPreviousAction {
+  my $action = GetParam('action', 'rc');
+  my $interval = GetParam('days', $RcDefault) * 86400;
+  # use delta between from and upto, or use days, whichever is available
+  my $to = GetParam('from', GetParam('upto', $Now - $interval));
+  my $from = $to - (GetParam('upto') ? GetParam('upto') - GetParam('from') : $interval);
+  return "action=$action;from=$from;upto=$to" . RcOtherParameters();
+}
+
+sub RcLastAction {
+  my $action = GetParam('action', 'rc');
+  my $more = "action=$action";
+  my $days = GetParam('days', $RcDefault);
+  $more .= ";days=$days" if $days != $RcDefault;
+  return $more . RcOtherParameters();
 }
 
 sub GetFilterForm {
@@ -1798,16 +1818,7 @@ sub RcHtml {
   };
   ProcessRcLines($printDailyTear, $printRCLine);
   $html .= '</ul>' if $inlist;
-  # use delta between from and upto, or use days, whichever is available
-  my $to = GetParam('from', GetParam('upto', $Now - GetParam('days', $RcDefault) * 86400));
-  my $from = $to - (GetParam('upto') ? GetParam('upto') - GetParam('from') : GetParam('days', $RcDefault) * 86400);
-  my $more = "action=rc;from=$from;upto=$to";
-  foreach (qw(all showedit rollback rcidonly rcuseronly rchostonly
-	      rcclusteronly rcfilteronly match lang followup)) {
-    my $val = GetParam($_, '');
-    $more .= ";$_=$val" if $val;
-  }
-  $html .= $q->p({-class=>'more'}, ScriptLink($more, T('More...'), 'more'));
+  $html .= $q->p({-class=>'more'}, ScriptLink(RcPreviousAction(), T('More...'), 'more'));
   return GetFormStart(undef, 'get', 'rc') . $html . $q->end_form;
 }
 
@@ -1884,8 +1895,9 @@ sub GetRcRss {
   my $title = QuoteHtml($SiteName) . ': ' . GetParam('title', QuoteHtml(NormalToFree($HomePage)));
   $rss .= "<title>$title</title>\n";
   $rss .= "<link>" . ScriptUrl($HomePage) . "</link>\n";
-  $rss .= qq{<atom:link href="} . GetScriptUrlWithRcParameters()
-    . qq{" rel="self" type="application/rss+xml" />\n};
+  $rss .= qq{<atom:link href="$ScriptName?} . RcSelfAction() . qq{" rel="self" type="application/rss+xml" />\n};
+  $rss .= qq{<atom:link href="$ScriptName?} . RcPreviousAction() . qq{" rel="previous" type="application/rss+xml" />\n};
+  $rss .= qq{<atom:link href="$ScriptName?} . RcLastAction() . qq{" rel="last" type="application/rss+xml" />\n};
   if ($SiteDescription) {
     $rss .= "<description>" . QuoteHtml($SiteDescription) . "</description>\n"
   }
@@ -4100,8 +4112,8 @@ sub TextIsFile { $_[0] =~ /^#FILE (\S+) ?(\S+)?\n/; }
 
 sub AddModuleDescription { # cannot use $q here because this is module init time
   my ($filename, $page, $dir, $tag) = @_;
-  my $src = "http://git.savannah.gnu.org/cgit/oddmuse.git/tree/modules/$dir" . UrlEncode($filename) . ($tag ? '?' . $tag : '');
-  my $doc = 'https://www.oddmuse.org/cgi-bin/oddmuse/' . UrlEncode(FreeToNormal($page));
+  my $src = "https://alexschroeder.ch/cgit/oddmuse/tree/modules/$dir" . UrlEncode($filename) . ($tag ? '?' . $tag : '');
+  my $doc = 'https://www.oddmuse.org/wiki/' . UrlEncode(FreeToNormal($page));
   $ModulesDescription .= "<p><a href=\"$src\">" . QuoteHtml($filename) . "</a>" . ($tag ? " ($tag)" : '');
   $ModulesDescription .= T(', see') . " <a href=\"$doc\">" . QuoteHtml($page) . "</a>" if $page;
   $ModulesDescription .= "</p>";