[post-instead-of-get] use utf8

Just print the subheading.

.gitignore

@@ -9,3 +9,4 @@
 wiki.log
 .prove
 TAGS
+/.vscode/

README.creole

@@ -1,197 +0,0 @@
-This is the README file distributed together with the
-[[https://oddmuse.org/|Oddmuse]] script.
-
-== Installing Oddmuse on a Debian System running Apache
-
-The following instructions require a number of tools. You can make sure
-they're all installed by issuing the following command as {{{root}}}:
-
-{{{
-apt-get install coreutils apache2 sudo wget w3m perl \
-    libwww-perl libxml-rss-perl diffutils
-}}}
-
-You probably created an account for yourself. You might have to add this
-user to the {{{sudo}}} group. Here's how I created my own user as
-{{{root}}}:
-
-{{{
-adduser alex
-usermod -a -G sudo alex
-}}}
-
-Now you can login as {{{alex}}} and do everything else using {{{sudo}}}.
-
-You need to copy wiki.pl into your cgi-bin directory, and you need to
-make the script executable. You might also have to change its owner to
-an appropriate user on your system.
-
-{{{
-sudo wget -O /usr/lib/cgi-bin/wiki.pl \
-  http://git.savannah.gnu.org/cgit/oddmuse.git/plain/wiki.pl
-sudo chmod +x /usr/lib/cgi-bin/wiki.pl
-sudo chown www-data.www-data /usr/lib/cgi-bin/wiki.pl
-}}}
-
-If you're on SUSE, the user might not be {{{www-data}}} but
-{{{wwwrun}}} without appropriate group:
-
-{{{
-sudo chown wwwrun.root /usr/lib/cgi-bin/wiki.pl
-}}}
-
-You should be able to test it right now! Visit
-{{{http://localhost/cgi-bin/wiki.pl}}}. If your site is available from
-the outside, you will be able to use a normal browser. If don't have a
-domain name yet, you'll probably have to use a text browser like
-{{{w3m}}}.
-
-{{{
-w3m http://localhost/cgi-bin/wiki.pl
-}}}
-
-If you create pages in this wiki, these will get stored in a temporary
-directory. You need change the data directory from {{{"/tmp/oddmuse"}}}
-to like {{{"/var/local/oddmuse"}}}. The best way to do this without
-changing {{{wiki.pl}}} is by editing
-{{{/etc/apache2/sites-available/default}}}. Add the following line:
-
-{{{
-SetEnv WikiDataDir /var/local/oddmuse
-}}}
-
-Enable the default site by calling the following command:
-
-{{{
-sudo a2ensite default
-}}}
-
-Reload the Apache configuration by calling the following command:
-
-{{{
-sudo service apache2 reload
-}}}
-
-You need to create the new data directory. You webserver runs CGI
-scripts as {{{www-data}}}. Thus, you need to change the owner and group
-of the directory to {{{www-data}}}.
-
-{{{
-sudo mkdir -p /var/local/oddmuse
-sudo chown www-data.www-data /var/local/oddmuse
-}}}
-
-Done! Visit your wiki and start editing. Click on the edit link (the
-first link below the navigation bar, at the bottom of the page). This
-will allow you to enter some text for this page. Click the Save button
-and you are done.
-
-To add new pages, edit the homepage and add links to new pages by
-putting their names in {{{[[double square brackets]]}}}.
-
-Enjoy your wiki experience.
-
-Visit https://www.oddmuse.org/ to learn more about the translation
-files and modules that are part of this package.
-
-== Checking the Apache Setup
-
-If you think this information doesn't work for you, here are some things
-to check.
-
-Apache's config directory is {{{/etc/apache2/apache2.conf}}}. This is
-where we get the {{{www-data}}} username from. It says:
-
-{{{
-# These need to be set in /etc/apache2/envvars
-User ${APACHE_RUN_USER}
-Group ${APACHE_RUN_GROUP}
-}}}
-
-Checking {{{/etc/apache2/envvars}}} we see the following:
-
-{{{
-export APACHE_RUN_USER=www-data
-export APACHE_RUN_GROUP=www-data
-}}}
-
-So that's what we're using in the {{{chown}}} command in our
-instructions above.
-
-The default site is configured in
-{{{/etc/apache2/sites-available/default}}}. In order for it to be
-//enabled//, there must be a symlink from a file in
-{{{/etc/apache2/sites-enabled}}} to the file in
-{{{sites-available}}}. You can enable it using the following command:
-
-{{{
-sudo a2ensite default
-}}}
-
-This file also lists the directories we've used in our instructions
-above.
-
-{{{
-ScriptAlias /cgi-bin/ /usr/lib/cgi-bin/
-}}}
-
-This means that {{{http://localhost/cgi-bin/wiki.pl}}} will call
-{{{/usr/lib/cgi-bin/wiki.pl}}}
-
-Don't forget to reload the Apache configuration as shown above, or
-simply restart it all:
-
-{{{
-sudo service apache2 graceful
-}}}
-
-== Using just Perl
-
-You can use Mojolicious as your web server. There is a simple
-##server.pl## which you can use. Here's how you might start it:
-
-{{{
-mkdir ~/oddmuse
-WikiDataDir=$HOME/oddmuse perl server.pl daemon
-}}}
-
-This makes the server available on {{{http://localhost:3000/wiki}}}.
-Make sure you create the directory before starting the server!
-If you don't, you'll get a strange error:
-`STDERR: : No such file or directory at ... perl5/Mojolicious/Plugin/CGI.pm`.
-
-If it works, feel free to upgrade to Hypnotoad.
-
-{{{
-WikiDataDir=$HOME/oddmuse hypnotoad server.pl
-}}}
-
-Note: Hypnotoad uses a different default port. The above makes the
-server available on {{{http://localhost:8080/wiki}}}. Hypnotoad will
-keep forking new processes. To stop it, use the {{{-s}}} flag.
-
-{{{
-hypnotoad -s server.pl
-}}}
-
-== License
-
-Permission is granted to copy, distribute and/or modify this document
-under the terms of the GNU Free Documentation License, Version 1.1 or
-any later version published by the Free Software Foundation.
-
-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 2 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.
-
-Both the GNU Free Documentation License, and the GNU General Public
-License are distributed together with this script. See the files
-[[https://github.com/kensanata/oddmuse/blob/master/FDL|FDL]] and
-[[https://github.com/kensanata/oddmuse/blob/master/GPL|GPL]],
-respectively.

README.md

@@ -0,0 +1,161 @@
+This is the README file distributed together with the
+[Oddmuse](https://oddmuse.org/) script.
+
+## Installing Oddmuse on a Debian System running Apache
+
+The following instructions require a number of tools. You can make sure
+they're all installed by issuing the following command as `root`:
+
+    apt-get install coreutils apache2 sudo wget w3m perl \
+        libwww-perl libxml-rss-perl diffutils
+
+You probably created an account for yourself. You might have to add this
+user to the `sudo` group. Here's how I created my own user as `root`:
+
+    adduser alex
+    usermod -a -G sudo alex
+
+Now you can login as `alex` and do everything else using `sudo`.
+
+You need to copy wiki.pl into your cgi-bin directory, and you need to
+make the script executable. You might also have to change its owner to
+an appropriate user on your system.
+
+    sudo wget -O /usr/lib/cgi-bin/wiki.pl \
+      http://git.savannah.gnu.org/cgit/oddmuse.git/plain/wiki.pl
+    sudo chmod +x /usr/lib/cgi-bin/wiki.pl
+    sudo chown www-data.www-data /usr/lib/cgi-bin/wiki.pl
+
+If you're on SUSE, the user might not be `www-data` but `wwwrun` without
+appropriate group:
+
+    sudo chown wwwrun.root /usr/lib/cgi-bin/wiki.pl
+
+You should be able to test it right now! Visit
+`http://localhost/cgi-bin/wiki.pl`. If your site is available from the
+outside, you will be able to use a normal browser. If don't have a
+domain name yet, you'll probably have to use a text browser like `w3m`.
+
+    w3m http://localhost/cgi-bin/wiki.pl
+
+If you create pages in this wiki, these will get stored in a temporary
+directory. You need change the data directory from `"/tmp/oddmuse"` to
+like `"/var/local/oddmuse"`. The best way to do this without changing
+`wiki.pl` is by editing `/etc/apache2/sites-available/default`. Add the
+following line:
+
+    SetEnv WikiDataDir /var/local/oddmuse
+
+Enable the default site by calling the following command:
+
+    sudo a2ensite default
+
+Reload the Apache configuration by calling the following command:
+
+    sudo service apache2 reload
+
+You need to create the new data directory. You webserver runs CGI
+scripts as `www-data`. Thus, you need to change the owner and group of
+the directory to `www-data`.
+
+    sudo mkdir -p /var/local/oddmuse
+    sudo chown www-data.www-data /var/local/oddmuse
+
+Done! Visit your wiki and start editing. Click on the edit link (the
+first link below the navigation bar, at the bottom of the page). This
+will allow you to enter some text for this page. Click the Save button
+and you are done.
+
+To add new pages, edit the homepage and add links to new pages by
+putting their names in `[[double square brackets]]`.
+
+Enjoy your wiki experience.
+
+Visit <https://www.oddmuse.org/> to learn more about the translation
+files and modules that are part of this package.
+
+## Checking the Apache Setup
+
+If you think this information doesn't work for you, here are some things
+to check.
+
+Apache's config directory is `/etc/apache2/apache2.conf`. This is where
+we get the `www-data` username from. It says:
+
+    # These need to be set in /etc/apache2/envvars
+    User ${APACHE_RUN_USER}
+    Group ${APACHE_RUN_GROUP}
+
+Checking `/etc/apache2/envvars` we see the following:
+
+    export APACHE_RUN_USER=www-data
+    export APACHE_RUN_GROUP=www-data
+
+So that's what we're using in the `chown` command in our instructions
+above.
+
+The default site is configured in
+`/etc/apache2/sites-available/default`. In order for it to be *enabled*,
+there must be a symlink from a file in `/etc/apache2/sites-enabled` to
+the file in `sites-available`. You can enable it using the following
+command:
+
+    sudo a2ensite default
+
+This file also lists the directories we've used in our instructions
+above.
+
+    ScriptAlias /cgi-bin/ /usr/lib/cgi-bin/
+
+This means that `http://localhost/cgi-bin/wiki.pl` will call
+`/usr/lib/cgi-bin/wiki.pl`
+
+Don't forget to reload the Apache configuration as shown above, or
+simply restart it all:
+
+    sudo service apache2 graceful
+
+## Using just Perl
+
+You can use Mojolicious as your web server. There is a simple
+`server.pl` which you can use. Here's how you might start it:
+
+    mkdir ~/oddmuse
+    WikiDataDir=$HOME/oddmuse perl server.pl daemon
+
+This makes the server available on `http://localhost:3000/wiki`. Make
+sure you create the directory before starting the server! If you don't,
+you'll get a strange error: \`STDERR: : No such file or directory at ...
+perl5/Mojolicious/Plugin/CGI.pm\`.
+
+If it works, feel free to upgrade to Hypnotoad.
+
+    WikiDataDir=$HOME/oddmuse hypnotoad server.pl
+
+Note: Hypnotoad uses a different default port. The above makes the
+server available on `http://localhost:8080/wiki`. Hypnotoad will keep
+forking new processes. To stop it, use the `-s` flag.
+
+    hypnotoad -s server.pl
+
+## License
+
+Permission is granted to copy, distribute and/or modify this document
+under the terms of the GNU Free Documentation License, Version 1.1 or
+any later version published by the Free Software Foundation.
+
+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 2 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.
+
+Both the GNU Free Documentation License, and the GNU General Public
+License are distributed together with this script. See the files
+[FDL](https://github.com/kensanata/oddmuse/blob/master/FDL) and
+[GPL](https://github.com/kensanata/oddmuse/blob/master/GPL),
+respectively.

css/green.css

@@ -1,32 +1,51 @@
 /* Authors: Murray Altheim (2004), Alex Schroeder (2004, 2005, 2006,
-   2009), Bayle Shanks (2006), Lion Kimbro (2006).
+   2009, 2020), Bayle Shanks (2006), Lion Kimbro (2006).
 
 This file is in the public domain.
 
 */
 
-html, body { /* hue 84 */
+html {
+    /* background-color:#becc92; */
+    background-color:#def4b5;
+    text-align: center;
+}
+
+body { /* hue 84 */
     /* Default color for text. It doesn't appear that many places,
        but you'll see it in Recent Changes summary comments, and
        you'll see it in the languages at the bottom of the place;
        It seeps out, here and there. */
     color:#000;
     /* This is the main light green background color. */
-    background-color:#def4b5;
     margin:0;
     padding:0;
+    font-size: 14pt;
+    max-width: 80ex;
+    display: inline-block;
+    text-align: left;
 }
- 
-/* This is not used in all that many places. */
-body.gray {
-    background-color:#d5e0c5;
-}
- 
-/* The next section includes some funky selectors.
-   See http://www.w3.org/TR/REC-CSS2/selector.html for more. */
 
-div.content, div.rc, body > form, div.footnotes, div.edit text {
-    margin: 1em;
+.wrapper {
+    padding: 1ex;
+}
+
+.languages {
+    display: none;
+}
+
+@media only screen and (max-width: 1200px) {
+    body {
+	padding: 5pt;
+	font-size: 15pt;
+    }
+}
+
+@media only screen and (max-width: 800px) {
+    body {
+	padding: 3pt;
+	font-size: 18pt;
+    }
 }
 
 /* The following statement hides the result count at the end
@@ -70,53 +89,16 @@ body.arrows a.near:before, body.arrows a.outside:before { content:"\2197"; }
 body.arrows a.near, body.arrows a.outside { text-decoration:none; }
 
 /* add every specific a here */
-div.header h1 a:hover, h1 a:hover, h2 a:hover, h3 a:hover { color: #fbb; }
+header h1 a:hover, h1 a:hover, h2 a:hover, h3 a:hover { color: #fbb; }
+header h1 a:visited { color: #fff; }
 a.definition:hover, a.near:hover, a:hover { color:#f00; }
 
 dl.irc dt { width:20ex; float:left; text-align:right; clear:left; }
 dl.irc dt span.time { float:left; }
 dl.irc dd { margin-left:22ex; }
 
-div.header {
-    background-color: #becc92;
-    margin:                  0;
-    padding:                 0;
-    height: 6em;
-}
-
-div.sidebar {
-    float:    right;
-    width:    15%;
-    padding:  0 0.5em 0 1em;
-    margin:   1em 0 1em 5em;
-    font-size: x-small;
-    border:   3px solid #000;
-    text-align:  left;
-    background-color: #dea;
-}
-
-div.sidebar h2 {
-    font-weight: bold;
-    font-size:   small;
-    color:       #000;
-    background-color: #dea;
-    padding:     0;
-    margin-right: 7%;
-    border-bottom: 1px solid #ab7;
-}
-
-div.sidebar ul, div.sidebar li {
-    display:block;
-    margin:0;
-    padding:0;
-}
-
-div.sidebar a:before {
-    content:"";
-}
-
-div.sidebar a {
-    font-weight: normal;
+header {
+    margin: 0;
 }
 
 h3 {
@@ -128,52 +110,49 @@ h3 {
     padding:0.3ex;
 }
 
-div.header span.gotobar {
+header span.gotobar {
     display:block;
     padding:1ex;
 }
-div.message {
-    position: absolute;
-    top: 1.5em;
-    left:0;
-    right:0;
-    z-index: 5;
-}
-div.message p {
-    display:inline;
-}
-div.message, div.question {
-    background-color:#fee;
-    color:#f00;
-    border:solid #f00;
-    font-weight:bold;
-    padding:0.1em 0 0.1em 1em;
+.message p, div.question {
+    font-size: smaller;
+    margin: 0;
+    padding: 0 0 0.5ex 1ex;
 }
 
-div.header h1 {
-    position:   absolute;
-    top: 1.5em;
-    left:0;
-    right:0;
+header h1 {
     background-color:#517005;
     font-family: "Tahoma", "Arial", "Helvetica", sans-serif;
-    font-size:xx-large;
-    border-bottom:2px dotted #87a036;
-    margin-top: 0;
-    padding: 0.125em 0.5em;
+    font-size: xx-large;
+    margin: 0;
+    padding: 0.1em 0.5ex;
 }
 
-div.header h1 a {
+header h1 a {
     text-decoration:none; 
     color: #ffffff;
 }
 
-img.logo { 
-    position:absolute;
-    top:1ex;
-    right:1ex;
+img.logo {
+    float: right;
+    height: 4em;
     border:none;
-    z-index:10;
+}
+
+@media only screen and (max-width: 1200px) {
+    img.logo {
+	height: 3em;
+    }
+}
+
+@media only screen and (max-width: 800px) {
+    img.logo {
+	height: 2em;
+    }
+}
+
+.fit {
+    max-width: 100%;
 }
 
 .right {
@@ -182,6 +161,7 @@ img.logo {
 
 .left {
     float:left;
+    margin-right: 8px;
 }
 
 .top {
@@ -221,7 +201,7 @@ a.small img {
 
 h1 {
     font-weight:bold;
-    font-size:larger;
+    font-size:150%;
     clear:left;
     color:#fff;
     background:#69aa00;
@@ -237,17 +217,23 @@ span.specialdays {
 
 h2 {
     font-weight:bold;
-    font-size:larger;
+    font-size:130%;
     color:#fff;
     background:#69aa00;
     padding:0.7ex;
+    clear:left;
 }
-h2 a, div.journal h1 a {
+/* Links in page titles */
+h1 a, h2 a, div.journal h1 a {
     text-decoration:none; color:#fff;
 }
 h3 {
-    font-weight:bold; font-size:medium; clear:left;
-    color:#fff; background:#84d600; padding:0.7ex;
+    font-weight:bold;
+    font-size: 110%;
+    clear:left;
+    color:#fff;
+    background:#84d600;
+    padding:0.7ex;
 }
 h3 a, div.journal h2 a {
     text-decoration:none; color:#fff;
@@ -277,8 +263,8 @@ div.footnotes hr + p {
     font-weight:bold;
 }
 
-div.footer {
-    background-color:#becc92;
+footer {
+    background-color: #cd9;
     border-bottom:solid;
     clear:    both;
     margin:  3em 0 0 0;
@@ -287,13 +273,89 @@ div.footer {
     color:black;
 }
 
-div.footer hr {
+footer hr {
     display:none;
 }
 
+/* License, definitions, near links */
+.note, .more {
+    font-size: 80%;
+}
+.more {
+    margin: 1ex;
+    margin-bottom: 3em;
+}
+div.near, div.definition {
+    display: none;
+}
+#toggle_more:checked ~ div.near, #toggle_more:checked ~ div.definition {
+    display: block;
+}
+
+/* Recent Changes */
+
 div.rc { margin-top:4ex; }
 div.rc hr { display:none; }
 
+div.rc {
+    overflow: hidden;
+}
+div.rc li + li {
+    margin-top: 1em;
+}
+div.rc li strong, table.history strong, strong.description {
+    font-family: inherit;
+    font-weight: inherit;
+}
+
+/* Colour flags for anonymous edits */
+
+.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.old { background-color:#ffd; }
 div.new { background-color:#dfd; }
 div.diff {
@@ -306,6 +368,9 @@ div.diff {
 div.diff + hr {
     display: none;
 }
+
+/* Referrers */
+
 div.refer {
     padding-left:5%;
     padding-right:5%;
@@ -313,10 +378,9 @@ div.refer {
 div.refer hr {
     display: none;
 }
-div.rss { background-color:#ce9; }
-body.gray div.rss {
-    background-color:#dec;
-}
+
+/* Sister Sites */
+
 div.sister {
     float:left;
     margin-right:1ex;
@@ -326,7 +390,9 @@ div.sister p { padding:1ex; margin:0; }
 div.sister hr { display:none; }
 div.near, div.definition { padding:1ex; margin:0; }
 div.near p, div.definition p { margin: 0; }
-div.footer + hr { display:none; }
+footer + hr { display:none; }
+
+/* Headers in Journal Pages (e.g. Blog) */
 
 div.journal hr { display:none; }
 div.journal h1, div.journal h2, div.journal h3, div.journal h4 {
@@ -343,23 +409,65 @@ div.include {
 span.description { font-weight:bold; }
 span.new { display:inline; font-weight:bold; }
 
-table.user { border-collapse:collapse; border:thin dotted; padding:1ex;
-    margin-bottom:1ex; width:inherit; margin:0 5%; }
-table.user tr td { padding: 0.5ex 1em; border: thin dotted; text-align:left; }
+/* Tables in wiki content */
+
+table.user {
+    border-collapse:collapse;
+    border: none;
+    padding:1ex;
+    margin-bottom:1ex;
+    width:inherit;
+    margin:0 5%;
+    background-color: #efd;
+}
+/* table.user .even { */
+/*     background-color: #efd; */
+/* } */
+/* table.user .odd { */
+/*     background-color: #efe; */
+/* } */
+/* table.user .first { */
+/*     background-color: #eff; */
+/* } */
+table.user td, table.user th {
+    padding: 0.5ex 1em;
+    border: none;
+    text-align:left;
+}
+table.user th {
+    padding: 1ex 1em;
+    text-align: center;
+}
+
+/* Tables in page history */
 
 table.history td[colspan="3"] {
     padding: 1em 0;
 }
 
+/* Lists (definitions list being used by IRC logs?) */
+
 dt { font-weight:bold; }
-dd, li {
+dd {
     margin-bottom: 0.5ex;
     margin-left: 2em;
 }
+li {
+    margin-bottom: 0.5ex;
+    margin-left: 0;
+}
 dl, ol, ul { margin-left:0em; }
 
-textarea#text { width:75%; height:70%; }
-textarea#summary { width:75%; height:10%; }
+/* textarea, summary */
+
+textarea {
+    box-sizing: border-box;
+    width:100%;
+    padding:5pt;
+    font-size: inherit;
+}
+textarea#text { height:70%; min-height: 20ex; }
+textarea#summary { height:10%; min-height: 2ex; }
 
 /* links to change from text to file and back */
 
@@ -375,12 +483,11 @@ form.edit a.svg, form.edit a.upload {
 /* images */
 
 img { border:0; }
-pre, img.upload {
-    border: #777 1px solid; padding: 0.5em;
-    margin-left: 1em; margin-right: 2em;
+pre, img.portrait, img.upload {
+    border: #777 1px solid; padding: 8px;
     white-space: pre;
     background-color: #fff; color: black;
-    overflow: hidden;
+    overflow: scroll;
 }
 a.smiley img.upload {
     border:none;
@@ -388,32 +495,21 @@ a.smiley img.upload {
     padding:0;
     background-color:inherit;
 }
+
+.color { min-height: 60px; }
 img.portrait {
-    float:left; clear:left;
-    background-color:#fff;
-    border:#999 1px solid;
-    padding:10px;
-    margin:10px;
-}
-div.portrait {
-    float:left; clear:left;
-    font-size:xx-small;
-    padding-left:10px;
-}
-div.portrait img.portrait {
-    float:none;
-    margin:10px 10px 0 0;
-}
-div.portrait a {
-    text-decoration:none;
-    color:#999;
-}
-div.color {
-    clear: left;
-    min-height:105px;
-    margin: 0;
-    padding: 0;
+    float: left;
+    margin-right: 8px;
 }
+div.portrait br { display: none }
+div.portrait, div.portrait p { display: inline}
+div.portrait p:after { content: ": " }
+div.color > p:first-of-type { display: inline }
+div.color { padding: 1ex 0.5ex 1em 0.5ex; }
+
+.half img { max-width: 50%; }
+
+/* indentation */
 div.one {
     background-color: #efb;
 }
@@ -434,101 +530,6 @@ hr {
     clear: left;
 }
 
-div.month {
-    float:left;
-    margin:3ex;
-    height:15ex;
-}
-div.month pre {
-    background-color:inherit;
-    border:none;
-    padding:0;
-    margin:0;
-}
-div.month a.edit { 
-    font-weight:normal;
-    color:#000;
-}
-
-rss {
-    color:#000;
-    margin:0;
-    padding:0;
-    background-color:#def4b5;
-}
-docs {
-    position: absolute;
-    top:0;
-    left:0;
-    right:0;
-    font-size: xx-large;
-    height: 1.5em;
-    color: #becc92; /* invisible */
-    background-color: #becc92;
-}
-
-channel * {
-    display: block;
-}
-
-/* if IE can't parse this, no problem */
-channel > title {
-    font-family: "Tahoma", "Arial", "Helvetica", sans-serif;
-}
-    
-title {
-    background-color:#517005;
-    font-size:xx-large;
-    font-weight:bold;
-    margin-top: 1.5em;
-    padding: 0.125em 0.5em;
-    border-bottom:2px dotted #87a036;
-    color:#fff;
-}
-item title {
-    background-color:#69aa00;
-    font-size: medium;
-    margin: 0 0 0 1em;
-    padding:0.7ex 0.5em;
-}
-
-copyright {
-    font-size: smaller;
-    margin: 1em 4em;
-}
-channel > link:before {
-    font-size: x-large;
-    display: block;
-    margin: 1em;
-    padding: 0.5em;
-    content: "This is an RSS feed, designed to be read in a feed reader.";
-    color: red;
-    border: 1px solid red;
-}
-link, license {
-    font-size: smaller;
-    margin: 1em 2em;
-}
-username, description, generator, interwiki { margin: 1em; }
-username:before { content: "Last edited by "; }
-username:after { content: "."; }
-generator:before { content: "Feed generated by "; }
-generator:after { content: "."; }
-channel description {
-        font-weight: bold;
-}
-item description {
-    font-style: italic;
-    font-weight: normal;
-    margin: 1em;
-}
-language,
-pubDate, lastBuildDate, ttl, guid, category, comments,
-image title, image link,
-status, version, diff, history, importance {
-        display: none;
-}
-
 /* 
 
 == Printing ==
@@ -542,10 +543,10 @@ when the page is printed (or during print preview).  More information:
 
 @media print {
     /* When printing, turn off a bunch of stuff. */
-    div.header span.gotobar,
+    header span.gotobar,
     span.specialdays,
     div.refer,
-    div.footer,
+    footer,
     div.near,
     div.definition,
     div.sister,
@@ -592,19 +593,16 @@ span[lang=pt], .pt { background-color:#bfb; }
 span[lang=es], .es { background-color:#fec; }
 span[lang=sv], .sv { background-color:#adf; }
 
-body.simple div.footer p.note,
-body.simple div.footer span.gotobar + br,
-body.simple div.footer span.gotobar,
+body.simple footer p.note,
+body.simple footer span.gotobar + br,
+body.simple footer span.gotobar,
 body.simple div.sister,
 body.simple div.near,
 body.simple div.definition,
 body.simple div.languages { display:none; }
 
 body.explicit a.near[title=MeatBall]:before { content:"MeatBall:"; }
-body.explicit a.near[title=WikiFeatures]:before { content:"WikiFeatures:"; }
-body.explicit a.near[title=CraoWiki]:before { content:"CraoWiki:"; }
-body.explicit a.near[title=InterWiki]:before { content:"InterWiki:"; }
-body.explicit a.near[title=OpenMeatballWiki]:before { content:"OpenMeatballWiki:"; }
+body.explicit a.near[title=Wiki]:before { content:"Wiki:"; }
 
 body.nolang span[lang] { background-color:inherit; }
 

modules/calendar.pl

@@ -1,4 +1,4 @@
-# Copyright (C) 2004, 2005, 2006  Alex Schroeder <alex@emacswiki.org>
+# Copyright (C) 2004–2023  Alex Schroeder <alex@gnu.org>
 # Copyright (C) 2006  Ingo Belka
 #
 # This program is free software; you can redistribute it and/or modify
@@ -112,9 +112,7 @@ sub DoCollect {
   my $search = GetParam('search', '');
   ReportError(T('The match parameter is missing.')) unless $match or $search;
   print GetHeader('', Ts('Page Collection for %s', $match||$search), '');
-  my @pages = (grep(/$match/, $search
-		    ? SearchTitleAndBody($search)
-		    : AllPagesList()));
+  my @pages = Matched($match, $search ? SearchTitleAndBody($search) : AllPagesList());
   if (!$CollectingJournal) {
     $CollectingJournal = 1;
     # Now save information required for saving the cache of the current page.

modules/index.pl

@@ -1,4 +1,4 @@
-# Copyright (C) 2004, 2007  Alex Schroeder <alex@emacswiki.org>
+# Copyright (C) 2004–2023  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
@@ -47,8 +47,5 @@ sub PrintableIndexPages {
   push(@pages, AllPagesList()) if GetParam('pages', 1);
   push(@pages, keys %PermanentAnchors) if GetParam('permanentanchors', 1);
   push(@pages, keys %NearSource) if GetParam('near', 0);
-  my $match = GetParam('match', '');
-  @pages = grep /$match/i, @pages if $match;
-  @pages = sort @pages;
-  return @pages;
+  return sort Matched(GetParam('match'), @pages);
 }

modules/journal-rss.pl

@@ -1,4 +1,4 @@
-# Copyright (C) 2004–2021  Alex Schroeder <alex@gnu.org>
+# Copyright (C) 2004–2023  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
@@ -31,6 +31,7 @@ sub DoJournalRss {
   local $CollectingJournal = 1;
   # Fake the result of GetRcLines()
   local *GetRcLines = \&JournalRssGetRcLines;
+  local *RcSelfWebsite = \&JournalRssSelfWebsite;
   local *RcSelfAction = \&JournalRssSelfAction;
   local *RcPreviousAction = \&JournalRssPreviousAction;
   local *RcLastAction = \&JournalRssLastAction;
@@ -55,6 +56,15 @@ sub JournalRssParameters {
   return $more;
 }
 
+sub JournalRssSelfWebsite {
+  my $more = '';
+  my $search = GetParam('rcfilteronly', '');
+  $more .= ";search=" . UrlEncode($search) if $search;
+  my $match = GetParam('match', '');
+  $more .= ";match=" . UrlEncode($match) if $match;
+  return $more;
+}
+
 sub JournalRssSelfAction {
   return "action=journal" . JournalRssParameters(qw(offset));
 }
@@ -76,7 +86,7 @@ sub JournalRssGetRcLines {
   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()));
+  my @pages = sort JournalSort (Matched($match, $search ? SearchTitleAndBody($search) : AllPagesList()));
   if ($monthly and not $match) {
     my ($sec, $min, $hour, $mday, $mon, $year, $wday, $yday) = gmtime();
     $match = '^' . sprintf("%04d-%02d", $year+1900, $mon+1) . '-\d\d';

modules/mail.pl

@@ -1,4 +1,4 @@
-# Copyright (C) 2009–2020  Alex Schroeder <alex@gnu.org>
+# Copyright (C) 2009–2022  Alex Schroeder <alex@gnu.org>
 # Copyright (C) 2015 Aleks-Daniel Jakimenko <alex.jakimenko@gmail.com>
 #
 # This program is free software; you can redistribute it and/or modify it under
@@ -505,9 +505,9 @@ sub MailUnsubscribe {
 
 =head1 Migrate
 
-The mailmigrate action will migrate your subscription list from the
-old format to the new format. This is necessary because these days
-because the keys and values of the DB_File are URL encoded.
+The mailmigrate action will migrate your subscription list from the old format
+to the new format. This is necessary because these days the keys and values of
+the DB_File are URL encoded.
 
 =cut
 

modules/markdown-rule.pl

@@ -1,5 +1,5 @@
 #! /usr/bin/perl
-# Copyright (C) 2014–2019  Alex Schroeder <alex@gnu.org>
+# Copyright (C) 2014–2022  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
@@ -194,12 +194,20 @@ sub MarkdownRule {
     return OpenHtmlEnvironment('pre',1) . $str; # always level 1
   }
   # link: [an example](http://example.com/ "Title")
-  elsif (m/\G\[((?:[^]\n]+\n?)+)\]\($FullUrlPattern(\s+"(.+?)")?\)/cg) {
+  elsif (m/\G\[((?:[^]\n]+\n?)+)\]\((\S+)(\s+"(.+?)")?\)/cg) {
     my ($text, $url, $title) = ($1, $2, $4);
-    $url =~ /^($UrlProtocols)/;
     my %params;
     $params{-href} = $url;
-    $params{-class} = "url $1";
+    $params{-class} = "url";
+    $params{-title} = $title if $title;
+    return $q->a(\%params, $text);
+  }
+  # link: [an example](#foo "Title")
+  elsif (m/\G\[((?:[^]\n]+\n?)+)\]\((#\S)+(\s+"(.+?)")?\)/cg) {
+    my ($text, $url, $title) = ($1, $2, $4);
+    my %params;
+    $params{-href} = $url;
+    $params{-class} = "named-anchor";
     $params{-title} = $title if $title;
     return $q->a(\%params, $text);
   }

modules/namespaces.pl

@@ -1,4 +1,4 @@
-# Copyright (C) 2004, 2005, 2006, 2007, 2008, 2009, 2012  Alex Schroeder <alex@gnu.org>
+# Copyright (C) 2004–2022  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

modules/network-blocker.pl

@@ -0,0 +1,198 @@
+# -*- mode: perl -*-
+# Copyright (C) 2023  Alex Schroeder <alex@gnu.org>
+
+# This program is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Affero 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 Affero General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Affero General Public License along
+# with this program. If not, see <https://www.gnu.org/licenses/>.
+
+=encoding utf8
+
+=head1 NAME
+
+Oddmuse Network Blocker
+
+=head1 DESCRIPTION
+
+This module hooks into regular Oddmuse Surge Protection. It adds the following
+features:
+
+Repeated offenders are blocked for increasingly longer times.
+
+For every offender, we record the CIDR their IP number belongs to. Everytime an
+IP number is blocked, all the CIDRs of the other blocked IPs are checked: if
+there are three or more blocked IP numbers sharing the same CIDRs, the CIDR
+itself is blocked.
+
+CIDR blocking works the same way: Repeated offenders are blocked for
+increasingly longer times.
+
+=head2 Behind a reverse proxy
+
+Make sure your config file copies the IP number to the correct environment
+variable:
+
+    $ENV{REMOTE_ADDR} = $ENV{HTTP_X_FORWARDED_FOR};
+
+=head1 SEE ALSO
+
+<Oddmuse Surge Protection|https://oddmuse.org/wiki/Surge_Protection>
+
+=cut
+
+use strict;
+use v5.10;
+use Net::IP qw(:PROC);
+use Net::DNS qw(rr);
+
+AddModuleDescription('network-blocker.pl', 'Network Blocker Extension');
+
+our ($Now, $DataDir, $SurgeProtectionViews, $SurgeProtectionTime);
+
+{
+  no warnings 'redefine';
+  *OldNetworkBlockerDelayRequired = \&DelayRequired;
+  *DelayRequired = \&NewNetworkBlockerDelayRequired;
+}
+
+# Block for at least this many seconds.
+my $NetworkBlockerMinimumPeriod = 30;
+
+# Every violation doubles the current period until this maximum is reached (four weeks).
+my $NetworkBlockerMaximumPeriod = 60 * 60 * 24 * 7 * 4;
+
+# All the blocked networks. Maps CIDR to an array [expiry timestamp, expiry
+# period].
+my %NetworkBlockerList;
+
+# Candidates are remembered for this many seconds.
+my $NetworkBlockerCachePeriod = 600;
+
+# All the candidate networks for a block. Maps IP to an array [ts, cidr, ...].
+# Candidates are removed after $NetworkBlockerCachePeriod.
+my %NetworkBlockerCandidates;
+
+sub NetworkBlockerRead {
+  my ($status, $data) = ReadFile("$DataDir/network-blocks");
+  return unless $status;
+  my @lines = split(/\n/, $data);
+  while ($_ = shift(@lines)) {
+    my @items = split(/,/);
+    $NetworkBlockerList{shift(@items)} = \@items;
+  }
+  # an empty line separates the two sections
+  while ($_ = shift(@lines)) {
+    my @items = split(/,/);
+    $NetworkBlockerCandidates{shift(@items)} = \@items;
+  }
+  return 1;
+}
+
+sub NetworkBlockerWrite {
+  RequestLockDir('network-blocks') or return '';
+  WriteStringToFile(
+    "$DataDir/network-blocks",
+    join("\n\n",
+         join("\n", map {
+           join(",", $_, @{$NetworkBlockerList{$_}})
+              } keys %NetworkBlockerList),
+         join("\n", map {
+           join(",", $_, @{$NetworkBlockerCandidates{$_}})
+              } keys %NetworkBlockerCandidates)));
+  ReleaseLockDir('network-blocks');
+}
+
+sub NewNetworkBlockerDelayRequired {
+  my $ip = shift;
+  # If $ip is a name and not an IP number, parsing fails. In this case, run the
+  # regular code.
+  my $ob = new Net::IP($ip);
+  return OldNetworkBlockerDelayRequired($ip) unless $ob;
+  # Read the file. If the file does not exist, no problem.
+  NetworkBlockerRead();
+  # See if the current IP number is one of the blocked CIDR ranges.
+  for my $cidr (keys %NetworkBlockerList) {
+    # Perhaps this CIDR block can be expired.
+    if ($NetworkBlockerList{$cidr}->[0] < $Now) {
+      delete $NetworkBlockerList{$cidr};
+      next;
+    }
+    # Forget the CIDR if it cannot be turned into a range.
+    my $range = new Net::IP($cidr);
+    if (not $range) {
+      warn "CIDR $cidr is blocked but has no range: " . Net::IP::Error();
+      delete $NetworkBlockerList{$cidr};
+      next;
+    }
+    # If the CIDR overlaps with the remote IP number, it's a block.
+    warn "Checking whether $ip is in $cidr\n";
+    my $overlap = $range->overlaps($ob);
+    # $IP_PARTIAL_OVERLAP (ranges overlap) $IP_NO_OVERLAP (no overlap)
+    # $IP_A_IN_B_OVERLAP (range2 contains range1) $IP_B_IN_A_OVERLAP (range1
+    # contains range2) $IP_IDENTICAL (ranges are identical) undef (problem)
+    if (defined $overlap and $overlap != $IP_NO_OVERLAP) {
+      # Double the block period unless it has reached $NetworkBlockerMaximumPeriod.
+      if ($NetworkBlockerList{$cidr}->[1] < $NetworkBlockerMaximumPeriod / 2) {
+        $NetworkBlockerList{$cidr}->[1] *= 2;
+      } else {
+        $NetworkBlockerList{$cidr}->[1] = $NetworkBlockerMaximumPeriod;
+      }
+      $NetworkBlockerList{$cidr}->[0] = $Now + $NetworkBlockerList{$cidr}->[1];
+      # And we're done!
+      NetworkBlockerWrite();
+      ReportError(Ts('Too many connections by %s', $cidr)
+		  . ': ' . Tss('Please do not fetch more than %1 pages in %2 seconds.',
+			       $SurgeProtectionViews, $SurgeProtectionTime),
+		  '503 SERVICE UNAVAILABLE');
+    }
+  }
+  # If the CIDR isn't blocked, let's see if Surge Protection wants to block it.
+  my $result = OldNetworkBlockerDelayRequired($ip);
+  warn "$ip was blocked\n" if $result;
+  # If the IP is to be blocked, determine its CIDRs and put them on a list. Sadly,
+  # routeviews does not support IPv6 at the moment!
+  if ($result and not ip_is_ipv6($ip) and not $NetworkBlockerCandidates{$ip}) {
+    my $reverse = $ob->reverse_ip();
+    $reverse =~ s/in-addr\.arpa\.$/asn.routeviews.org/;
+    my @candidates;
+    for my $rr (rr($reverse, "TXT")) {
+      next unless $rr->type eq "TXT";
+      my @data = $rr->txtdata;
+      push(@candidates, join("/", @data[1..2]));
+    }
+    warn "$ip is in @candidates\n";
+    $NetworkBlockerCandidates{$ip} = [$Now, @candidates];
+    # Expire any of the other candidates
+    for my $other_ip (keys %NetworkBlockerCandidates) {
+      if ($NetworkBlockerCandidates{$other_ip}->[0] < $Now - $NetworkBlockerCachePeriod) {
+        delete $NetworkBlockerCandidates{$other_ip};
+      }
+    }
+    # Determine if any of the CIDRs is to be blocked.
+    my $save;
+    for my $cidr (@candidates) {
+      # Count how often the candidate CIDRs show up for other IP numbers.
+      my $count = 0;
+      for my $other_ip (keys %NetworkBlockerCandidates) {
+        my @data = $NetworkBlockerCandidates{$other_ip};
+        for my $other_cidr (@data[1 .. $#data]) {
+          $count++ if $cidr eq $other_cidr;
+        }
+      }
+      if ($count >= 3) {
+        $NetworkBlockerList{$cidr} = [$Now + $NetworkBlockerMinimumPeriod, $NetworkBlockerMinimumPeriod];
+        $save = 1;
+      }
+    }
+    NetworkBlockerWrite() if $save;
+  }
+  return $result;
+}

modules/nosearch.pl

@@ -22,21 +22,18 @@ our ($q, @MyAdminCode);
 *OldGetSearchLink = \&GetSearchLink;
 *GetSearchLink = \&NewGetSearchLink;
 sub NewGetSearchLink {
-  my ($text, $class, $name, $title) = @_;
-  $name = UrlEncode($name);
-  $text =~ s/_/ /g;
-  return $q->span({-class=>$class}, $text);
+  my ($id, $class, $name, $title) = @_;
+  return NormalToFree($id);
 }
 
 push(@MyAdminCode, \&BacklinksMenu);
 sub BacklinksMenu {
   my ($id, $menuref, $restref) = @_;
   if ($id) {
-      my $text = T('Backlinks');
-      my $class = 'backlinks';
-      my $name = 'backlinks';
-      my $title = T('Click to search for references to this page');
-      my $link = ScriptLink('search=' . $id, $text, $class, $name, $title);
-      push(@$menuref, $link);
+    my $form = GetFormStart(undef, 'post', 'search');
+    $form .= $q->input({-type=>'hidden', -name=>'search', -value=>'"'.NormalToFree($id).'"'});
+    $form .= $q->p(T('Click to search for references to this page'));
+    $form .= $q->submit('search', T('Go!')) . $q->end_form;
+    push(@$menuref, $form);
   }
 }

modules/post-instead-of-get.pl

@@ -0,0 +1,166 @@
+#! /usr/bin/perl
+# Copyright (C) 2025  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/>.
+
+use strict;
+use v5.10;
+use utf8;
+
+AddModuleDescription('post-instead-of-get.pl', 'POST instead of GET extension');
+
+our ($q, $Now, $LastUpdate, %Action, @RcDays, $RcDefault, $ShowRollbacks, $ShowAll,
+     $ShowEdits, %Languages, $FullUrl);
+
+# You should install nosearch.pl, too.
+
+# Change the search from GET to POST
+
+*PostOldGetSearchForm=*GetSearchForm;
+*GetSearchForm=*PostNewGetSearchForm;
+
+sub PostNewGetSearchForm {
+  my $html = PostOldGetSearchForm(@_);
+  $html =~ s/method="get"/method="post"/;
+  return $html;
+}
+
+# Change the index filter from GET to POST
+
+*PostOldDoIndex=*DoIndex;
+*DoIndex=*PostNewDoIndex;
+# Update action hash as well!
+$Action{index} = \&DoIndex;
+
+sub PostNewDoIndex {
+  # Must capture the output.
+  my $html = ToString(\&PostOldDoIndex);
+  $html =~ s/method="get"/method="post"/;
+  print $html;
+}
+
+# Disable links in the Recent Changes menu
+
+*PostOldRcHeader=*RcHeader;
+*RcHeader=*PostNewRcHeader;
+
+sub PostNewRcHeader {
+  my ($from, $upto, $html) = (GetParam('from', 0), GetParam('upto', 0), '');
+  my $days = GetParam('days') + 0 || $RcDefault; # force numeric $days
+  my $all = GetParam('all', $ShowAll);
+  if ($from) {
+    $html .= $q->h2(Ts('Updates since %s', TimeToText(GetParam('from', 0))) . ' '
+		    . ($upto ? Ts('up to %s', TimeToText($upto)) : ''));
+  } else {
+    $html .= $q->h2((GetParam('days', $RcDefault) != 1)
+		    ? Ts('Updates in the last %s days', $days)
+		    : Ts('Updates in the last day'));
+  }
+  $html .= $q->p({-class => 'documentation'}, T('Using the ｢rollback｣ button on this page will reset the wiki to that particular point in time, undoing any later changes to all of the pages.')) if UserIsAdmin() and $all;
+  return $html;
+}
+
+# Change the More... link
+
+*PostOldRcHtml=*RcHtml;
+*RcHtml=*PostNewRcHtml;
+
+sub PostNewRcHtml {
+  my $html = PostOldRcHtml(@_);
+  # Based on RcPreviousAction
+  my $form = GetFormStart(undef, 'post', 'more');
+  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);
+  $form .= $q->input({-type=>'hidden', -name=>'action', -value=>'rc'});
+  $form .= $q->input({-type=>'hidden', -name=>'from', -value=>$from});
+  $form .= $q->input({-type=>'hidden', -name=>'upto', -value=>$to});
+  # Based on RcOtherParameters
+  foreach (qw(days page diff full all showedit rollback rcidonly rcuseronly rchostonly rcclusteronly rcfilteronly match lang followup)) {
+    my $val = GetParam($_, '');
+    $form .= $q->input({-type=>'hidden', -name=>$_, -value=>$val}) if $val;
+  }
+  $form .= $q->submit('more', T('More...'));
+  $form .= $q->end_form();
+  $html =~ s/<p class="more">.*?<\/p>//;
+  return $html . $form;
+}
+
+# Change Recent Changes filter form to represent all options.
+
+*PostOldGetFilterForm=*GetFilterForm;
+*GetFilterForm=*PostNewGetFilterForm;
+
+sub PostNewGetFilterForm {
+  my $all = GetParam('all', $ShowAll);
+  my $showedit = GetParam('showedit', $ShowEdits);
+  my $rollback = GetParam('rollback', $ShowRollbacks);
+  my $lang = GetParam('lang', '');
+  my $form = GetFormStart(undef, 'post', 'filter') . $q->h2(T('Filters'));
+  $form .= $q->input({-type=>'hidden', -name=>'action', -value=>'rc'});
+  $form .= $q->radio_group(-name=>'days', -values=>\@RcDays, -default=> $RcDefault) . ' ' . T('days') . $q->br();
+  $form .= $q->input({-type=>'checkbox', -id=>'all', -name=>'all', -value=>1, $all && '-checked'});
+  $form .= $q->label({-for=>'all'}, ' ' . T('List all changes')) . $q->br();
+  $form .= $q->input({-type=>'checkbox', -id=>'showedit', -name=>'showedit', -value=>1, $showedit && '-checked'});
+  $form .= $q->label({-for=>'showedit'}, ' ' . T('Include minor changes')) . $q->br();
+  $form .= $q->input({-type=>'checkbox', -id=>'rollback', -name=>'rollback', -value=>1, $rollback && '-checked'});
+  $form .= $q->label({-for=>'rollback'}, ' ' . T('Include rollbacks')) . $q->br();
+  foreach my $h (['match' => T('Title:')], ['rcfilteronly' => T('Title and Body:')],
+     ['rcuseronly' => T('Username:')], ['rchostonly' => T('Host:')], ['followup' => T('Follow up to:')]) {
+    $form .= $q->label({-for=>$h->[0], -style=>'width:20ch; display:inline-block'}, $h->[1]);
+    $form .= $q->textfield(-name=>$h->[0], -id=>$h->[0], -size=>20);
+    $form .= $q->br();
+  }
+  if (%Languages) {
+    $form .= $q->label({-for=>'rclang', -style=>'width:20ch; display:inline-block'}, T('Language:'));
+    $form .= $q->textfield(-name=>'lang', -id=>'rclang', -size=>20, -default=>$lang);
+  }
+  $form .= $q->br() . $q->submit('dofilter', T('Go!')) . $q->end_form;
+  $form .= GetFormStart(undef, 'post', 'later');
+  $form .= $q->input({-type=>'hidden', -name=>'action', -value=>'rc'});
+  $form .= $q->input({-type=>'hidden', -name=>'all', -value=>1}) if $all;
+  $form .= $q->input({-type=>'hidden', -name=>'showedit', -value=>1}) if $showedit;
+  $form .= $q->input({-type=>'hidden', -name=>'from', -value=>$LastUpdate+1});
+  $form .= $q->p(T('List later changes') . ' ' . $q->submit('dofilter', T('Go!')));
+  $form .= $q->end_form;
+  $form .= $q->p({-class => 'documentation'}, T('Using the ｢rollback｣ button on this page will reset the wiki to that particular point in time, undoing any later changes to all of the pages.')) if UserIsAdmin() and $all;
+  return $form;
+}
+
+# History page with new form
+
+*PostOldGetFooterLinks=*GetFooterLinks;
+*GetFooterLinks=*PostNewGetFooterLinks;
+
+sub PostNewGetFooterLinks {
+  my $html = PostOldGetFooterLinks(@_);
+  my ($id, $rev) = @_;
+  if ($Action{history} and $rev ne '') {
+    my $label = T('View all changes');
+    my $unwanted = quotemeta(GetRCLink($id, $label));
+    my $form = qq{
+<form style="display: inline" method="POST" action="$FullUrl">
+  <input type="hidden" name="action" value="rc"/>
+  <input type="hidden" name="all" value="1"/>
+  <input type="hidden" name="showedit" value="1"/>
+  <input type="hidden" name="from" value="1"/>
+  <input type="hidden" name="rcidonly" value="$id"/>
+  <input type="submit" name="dobacklinks" value="$label">
+</form>};
+    $html =~ s/$unwanted/$form/;
+  }
+  return $html;
+}
+
+1;

modules/rename-pages.pl

@@ -1,4 +1,4 @@
-# Copyright (C) 2019  Alex Schroeder <alex@gnu.org>
+# Copyright (C) 2019–2023  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
@@ -65,6 +65,7 @@ sub RenamePageMenu {
 	 . GetHiddenValue('id', $id)
 	 . $q->textfield(-name=>'to', -size=>20)
 	 . ' '
-	 . $q->submit('Do it'));
+	 . $q->submit('Do it')
+         . $q->end_form());
   }
 }

modules/search-list.pl

@@ -1,4 +1,4 @@
-# Copyright (C) 2006, 2007, 2008  Alex Schroeder <alex@emacswiki.org>
+# Copyright (C) 2006–2023  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
@@ -36,18 +36,12 @@ sub SearchListRule {
       $term = GetId();
     }
     local ($OpenPageName, %Page);
-    my %hash = ();
+    my @found;
     if ($variation eq 'list') {
-      foreach my $id (SearchTitleAndBody($term)) {
-        $hash{$id} = 1 unless $id eq $original; # skip the page with the query
-      }
+      @found = grep { $_ ne $original } SearchTitleAndBody($term);
+    } elsif ($variation eq 'titlelist') {
+      @found = grep { $_ ne $original } Matched($term, AllPagesList());
     }
-    if ($variation eq 'titlelist') {
-      foreach my $id (grep(/$term/, AllPagesList())) {
-        $hash{$id} = 1 unless $id eq $original; # skip the page with the query
-      }
-    }
-    my @found = keys %hash;
     if (defined &PageSort) {
       @found = sort PageSort @found;
     } else {
@@ -63,32 +57,24 @@ sub SearchListRule {
   return;
 }
 
-
 # Add a new action list
 
 $Action{list} = \&DoList;
 
 sub DoList {
-my $id = shift;
-my $match = GetParam('match', '');
-my $search = GetParam('search', '');
+  my $id = shift;
+  my $match = GetParam('match', '');
+  my $search = GetParam('search', '');
   ReportError(T('The search parameter is missing.')) unless $match or $search;
   print GetHeader('', Ts('Page list for %s', $match||$search), '');
   local (%Page, $OpenPageName);
-    my %hash = ();
-    foreach my $id (grep(/$match/, $search
-                    ? SearchTitleAndBody($search)
-                    : AllPagesList()))  {
-      $hash{$id} = 1;
-    }
-    my @found = keys %hash;
-    if (defined &PageSort) {
-      @found = sort PageSort @found;
-    } else {
-      @found = sort(@found);
-    }
-    @found = map { $q->li(GetPageLink($_)) } @found;
-    print $q->start_div({-class=>'search list'}),
-      $q->ul(@found), $q->end_div;
+  my @found = Matched($match, $search ? SearchTitleAndBody($search) : AllPagesList());
+  if (defined &PageSort) {
+    @found = sort PageSort @found;
+  } else {
+    @found = sort(@found);
+  }
+  @found = map { $q->li(GetPageLink($_)) } @found;
+  print $q->start_div({-class=>'search list'}), $q->ul(@found), $q->end_div;
   PrintFooter();
 }

modules/sistersites.pl

@@ -1,4 +1,4 @@
-# Copyright (C) 2007–2014  Alex Schroeder <alex@gnu.org>
+# Copyright (C) 2007–2023  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
@@ -36,8 +36,7 @@ sub SisterPages {
   push(@pages, AllPagesList()) if GetParam('pages', 1);
   push(@pages, keys %PermanentAnchors) if GetParam('permanentanchors', 1);
   push(@pages, keys %NearSource) if GetParam('near', 0);
-  my $match = GetParam('match', '');
-  @pages = grep /$match/i, @pages if $match;
+  @pages = Matched(GetParam('match', ''), @pages);
   @pages = sort @pages;
   return @pages;
 }

modules/translations/new-utf8.pl

@@ -1667,15 +1667,7 @@ Upgrade complete. Please remove $ModuleDir/upgade.pl, now.
 ################################################################################
 # modules/usemod.pl
 ################################################################################
-http://search.barnesandnoble.com/booksearch/isbninquiry.asp?ISBN=%s
-
-http://www.amazon.com/exec/obidos/ISBN=%s
-
-alternate
-
-http://www.pricescan.com/books/BookDetail.asp?isbn=%s
-
-search
+https://en.wikipedia.org/wiki/Special:BookSources/%s
 
 ################################################################################
 # modules/wanted.pl

modules/usemod.pl

@@ -213,13 +213,8 @@ sub ISBN {
   $num =~ s/[- ]//g;
   my $len = length($num);
   return "ISBN $rawnum" unless $len == 10 or $len == 13 or $len = 14; # be prepared for 2007-01-01
-  my $first  = $q->a({-href => Ts('http://search.barnesandnoble.com/booksearch/isbninquiry.asp?ISBN=%s', $num)},
+  my $html = $q->a({-href => Ts("https://en.wikipedia.org/wiki/Special:BookSources/%s", $num)},
       "ISBN " . $rawprint);
-  my $second = $q->a({-href => Ts('http://www.amazon.com/exec/obidos/ISBN=%s', $num)},
-      T('alternate'));
-  my $third  = $q->a({-href => Ts('http://www.pricescan.com/books/BookDetail.asp?isbn=%s', $num)},
-      T('search'));
-  my $html = "$first ($second, $third)";
   $html .= ' '  if ($rawnum =~ / $/);  # Add space if old ISBN had space.
   return $html;
 }

modules/wordcount.pl

@@ -23,7 +23,7 @@
 use strict;
 use v5.10;
 
-our (@MyInitVariables, $HtmlHeaders);
+our (@MyInitVariables, $HtmlHeaders, $EditNote);
 
 AddModuleDescription('wordcount.pl', 'Word Count Extension');
 
@@ -57,3 +57,5 @@ sub WordcountAddScript {
 	}
 	</script>";
 }
+
+$EditNote = "Words: <span id='textWordCount'></span>" . $EditNote;

scripts/tarballs.pl

@@ -85,7 +85,9 @@ versions of Oddmuse.</p>
 <ul>
 % for my $tarball (@$tarballs) {
 <li>
+% if ($tarball ne 'latest') {
 <a href="https://oddmuse.org/releases/<%= $tarball %>.tar.gz"><%= $tarball %>.tar.gz</a>
+% }
 (files for <%= link_to release => {tarball => $tarball} => begin %>\
 <%= $tarball =%><%= end %>)
 </li>

stuff/gemini-server.pl

@@ -675,10 +675,10 @@ sub gemini_text {
     $block =~ s/\[\[tag:([^]|]+)\]\]/push(@links, $self->gemini_link("tag\/$1", $1)); $1/ge;
     $block =~ s/\[\[tag:([^]|]+)\|([^\]|]+)\]\]/push(@links, $self->gemini_link("tag\/$1", $2)); $2/ge;
     $block =~ s/<journal search tag:(\S+)>\n*/push(@links, $self->gemini_link("tag\/$1", "Explore the $1 tag")); ""/ge;
-    $block =~ s/\[\[image:([^]|]+)\]\]/push(@links, $self->gemini_link($1, "$1 (image)")); "$1"/ge;
-    $block =~ s/\[\[image:([^]|]+)\|([^\]|]+)\]\]/push(@links, $self->gemini_link($1, "$2 (image)")); "$2"/ge;
-    $block =~ s/\[\[image:([^]|]+)\|([^\]|]*)\|([^\]|]+)\]\]/push(@links, $self->gemini_link($1, "$2 (image)"), $self->gemini_link($3, "$2 (follow-up)")); "$2"/ge;
-    $block =~ s/\[\[image:([^]|]+)\|([^\]|]*)\|([^\]|]*)\|([^\]|]+)\]\]/push(@links, $self->gemini_link($1, "$2 (image)"), $self->gemini_link($3, "$4 (follow-up)")); "$2"/ge;
+    $block =~ s/\[\[image(?:\/right)?:([^]|]+)\]\]/push(@links, $self->gemini_link($1, "$1 (image)")); "$1"/ge;
+    $block =~ s/\[\[image(?:\/right)?:([^]|]+)\|([^\]|]+)\]\]/push(@links, $self->gemini_link($1, "$2 (image)")); "$2"/ge;
+    $block =~ s/\[\[image(?:\/right)?:([^]|]+)\|([^\]|]*)\|([^\]|]+)\]\]/push(@links, $self->gemini_link($1, "$2 (image)"), $self->gemini_link($3, "$2 (follow-up)")); "$2"/ge;
+    $block =~ s/\[\[image(?:\/right)?:([^]|]+)\|([^\]|]*)\|([^\]|]*)\|([^\]|]+)\]\]/push(@links, $self->gemini_link($1, "$2 (image)"), $self->gemini_link($3, "$4 (follow-up)")); "$2"/ge;
     $block =~ s/\[\[$FreeLinkPattern\|([^\]|]+)\]\]/push(@links, $self->gemini_link($1, $2)); $2/ge;
     $block =~ s/\[\[$FreeLinkPattern\]\]/push(@links, $self->gemini_link($1)); $1/ge;
     $block =~ s/\[color=([^]]+)\]/color($1)/ge;

t/markdown-rule.t

@@ -137,13 +137,13 @@ EOT
 
 xpath_run_tests(split(/\n/,<<'EOT'));
 [example](http://example.com/)
-//a[@class="url http"][@href="http://example.com/"][text()="example"]
+//a[@class="url"][@href="http://example.com/"][text()="example"]
 [an example](http://example.com/)
-//a[@class="url http"][@href="http://example.com/"][text()="an example"]
+//a[@class="url"][@href="http://example.com/"][text()="an example"]
 [an example](http://example.com/ "Title")
-//a[@class="url http"][@href="http://example.com/"][@title="Title"][text()="an example"]
+//a[@class="url"][@href="http://example.com/"][@title="Title"][text()="an example"]
 [an\nexample](http://example.com/)
-//a[@class="url http"][@href="http://example.com/"][text()="an\nexample"]
+//a[@class="url"][@href="http://example.com/"][text()="an\nexample"]
 \n[an\n\nexample](http://example.com/)
 //p[text()="[an"]/following-sibling::p//text()[contains(string(),"example](")]
 EOT

t/post-instead-of-get.t

@@ -0,0 +1,26 @@
+# Copyright (C) 2025  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 => 6;
+
+add_module('nosearch.pl');
+add_module('post-instead-of-get.pl');
+
+like(get_page('HomePage'), qr/<h1>HomePage<\/h1>/, "no link in the title");
+my $page = get_page('RecentChanges');
+for my $day (@RcDays) {
+  like($page, qr/$day/, "$day days found");
+}

t/rollback-hang.t

@@ -0,0 +1,58 @@
+# Copyright (C) 2006–2023  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;
+
+# Reproduce a particular bug from alexschroeder.ch with the rc.log provided.
+WriteStringToFile($RcFile, <<'EOT');
+16853910992023-05-29_Net_newsHow to IRCAnonymousAlex2en
+16854004152023-05-29_Net_newsHow to IRCAnonymousAlex3en
+1685430599[[rollback]]1685400415Anonymous
+16855185032023-05-29_Net_newsAnonymousAlex4en
+EOT
+
+local $SIG{ALRM} = sub { fail "timeout!"; kill 'KILL', $$; };
+alarm 3;
+# this is recent changes from between the rollback and the page before it, so there are no pages to roll back
+my $page = get_page("action=rss full=1 short=0 from=1685413682");
+alarm 0;
+test_page($page, '2023-05-29 Net news');
+test_page_negative($page, 'rollback');
+
+# Reproduce a follow-up bug. First, rolling back just Test works as intended.
+WriteStringToFile($RcFile, <<'EOT');
+1691499987Testham127.0.0.1Berta1
+1691499988Mustuff127.0.0.1Chris1
+1691499989Testspam127.0.0.1Spammer2
+1691499990Test0Rollback to 2023-08-08 13:06 UTC127.0.0.1Alex3
+1691499990[[rollback]]1691499987Test
+EOT
+
+my $feed = get_page('action=rc raw=1 from=1691499900'); # need from or the result is empty
+test_page($feed, 'title: Test');
+
+# Rolling back all of the wiki doesn't work.
+WriteStringToFile($RcFile, <<'EOT');
+1691499987Testham127.0.0.1Berta1
+1691499988Mustuff127.0.0.1Chris1
+1691499989Testspam127.0.0.1Spammer2
+1691499990Test0Rollback to 2023-08-08 13:06 UTC127.0.0.1Alex3
+1691499990[[rollback]]1691499987
+EOT
+
+$feed = get_page('action=rc raw=1 from=1691499900'); # need from or the result is empty
+test_page($feed, 'title: Test');

t/translation-links.t

@@ -129,8 +129,8 @@ test_page(update_page('Testing', 'This is spam.'), 'This page does not exist');
 test_page(update_page('Spam', 'Trying again.'), 'This page does not exist');
 test_page(get_page('action=translate id=Spam target=Harmless translation=en'),
 	  'Edit Denied',
-	  'Regular expression "spam" matched on this page');
+	  'Regular expression "spam" matched "Spam" on this page');
 test_page(get_page('Spam'), 'This page does not exist');
 test_page(get_page('action=translate id=Harmless target=Spam translation=en'),
 	  'Edit Denied',
-	  'Regular expression "spam" matched on this page');
+	  'Regular expression "spam" matched "Spam" on this page');

wiki.pl

@@ -1,5 +1,5 @@
 #! /usr/bin/env perl
-# Copyright (C) 2001-2020
+# Copyright (C) 2001-2023
 #     Alex Schroeder <alex@gnu.org>
 # Copyright (C) 2014-2015
 #     Alex Jakimenko <alex.jakimenko@gmail.com>
@@ -391,7 +391,7 @@ sub InitLinkPatterns {
   my $EndChars = '[-a-zA-Z0-9/@=+$_~*]'; # no punctuation at the end of the url.
   $UrlPattern = "((?:$UrlProtocols):$UrlChars+$EndChars)";
   $FullUrlPattern="((?:$UrlProtocols):$UrlChars+)"; # when used in square brackets
-  $ImageExtensions = '(gif|jpg|jpeg|png|bmp|svg)';
+  $ImageExtensions = '(gif|jpg|jpeg|png|bmp|svg|webp)';
 }
 
 sub Clean {
@@ -514,7 +514,7 @@ sub ApplyRules {
 	Clean(CloseHtmlEnvironments() . AddHtmlEnvironment('p')); # another one like this further up
       } elsif (m/\G&amp;([A-Za-z]+|#[0-9]+|#x[A-Za-f0-9]+);/cg) { # entity references
 	Clean("&$1;");
-      } elsif (m/\G\s+/cg) {
+      } elsif (m/\G[ \t\r\n]+/cg) { # don't use \s because we want to honor NO-BREAK SPACE etc
 	Clean(' ');
       } elsif (m/\G([A-Za-z\x{0080}-\x{fffd}]+([ \t]+[a-z\x{0080}-\x{fffd}]+)*[ \t]+)/cg
 	       or m/\G([A-Za-z\x{0080}-\x{fffd}]+)/cg or m/\G(\S)/cg) {
@@ -1313,7 +1313,7 @@ sub GetId {
       SetParam($p, 1);    # script/p/q -> p=1
     }
   }
-  return $id;
+  return FreeToNormal($id);
 }
 
 sub DoBrowseRequest {
@@ -1543,28 +1543,34 @@ sub LatestChanges {
 sub StripRollbacks {
   my @result = @_;
   if (not (GetParam('all', $ShowAll) or GetParam('rollback', $ShowRollbacks))) { # strip rollbacks
-    my (%rollback);
+    my (%rollback); # used for single-page rollbacks
     for (my $i = $#result; $i >= 0; $i--) {
       # some fields have a different meaning if looking at rollbacks
       my ($ts, $id, $target_ts, $target_id) = @{$result[$i]};
+      # if this is a rollback marker
       if ($id eq '[[rollback]]') {
+        # if this is a single page rollback marker, strip it
 	if ($target_id) {
-	  $rollback{$target_id} = $target_ts; # single page rollback
-	  splice(@result, $i, 1);             # strip marker
+          # if this page is not already being rolled back, remember the target
+          # id and target ts so that those lines can be stripped below
+          if (not $rollback{$target_id} or $target_ts < $rollback{$target_id}) {
+            $rollback{$target_id} = $target_ts;
+          }
+          # the marker is always stripped
+	  splice(@result, $i, 1);
 	} else {
+          # if this is a global rollback, things are different: we're going to
+          # find the correct timestamp and strip all of those lines immediately
 	  my $end = $i;
-	  while ($ts > $target_ts and $i > 0) {
-	    $i--; # quickly skip all these lines
-	    $ts = $result[$i][0];
-	  }
-	  splice(@result, $i + 1, $end - $i);
-	  $i++; # compensate $i-- in for loop
+	  $i-- while $i > 0 and $target_ts < $result[$i-1][0];
+          # splice the lines found
+          splice(@result, $i, $end - $i + 1);
 	}
       } elsif ($rollback{$id} and $ts > $rollback{$id}) {
 	splice(@result, $i, 1); # strip rolled back single pages
       }
     }
-  } else {		  # just strip the marker left by DoRollback()
+  } else { # if rollbacks are not not shown, just strip the markers
     for (my $i = $#result; $i >= 0; $i--) {
       splice(@result, $i, 1) if $result[$i][1] eq '[[rollback]]'; # id
     }
@@ -1712,6 +1718,11 @@ sub RcOtherParameters {
   return $more;
 }
 
+sub RcSelfWebsite {
+  my $action = 'rc';
+  return "action=$action" . RcOtherParameters(qw(from upto days));
+}
+
 sub RcSelfAction {
   my $action = GetParam('action', 'rc');
   return "action=$action" . RcOtherParameters(qw(from upto days));
@@ -1898,7 +1909,7 @@ sub GetRcRss {
 };
   my $title = QuoteHtml($SiteName) . ': ' . GetParam('title', QuoteHtml(NormalToFree($HomePage)));
   $rss .= "<title>$title</title>\n";
-  $rss .= "<link>$ScriptName?" . RcSelfAction() . "</link>\n";
+  $rss .= "<link>$ScriptName?" . RcSelfWebsite() . "</link>\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};
@@ -1918,7 +1929,7 @@ sub GetRcRss {
     $rss .= "<image>\n";
     $rss .= "<url>$RssImageUrl</url>\n";
     $rss .= "<title>$title</title>\n";    # the same as the channel
-    $rss .= "<link>$ScriptName?" . RcSelfAction() . "</link>\n"; # the same as the channel
+    $rss .= "<link>$ScriptName?" . RcSelfWebsite() . "</link>\n"; # the same as the channel
     $rss .= "</image>\n";
   }
   my $limit = GetParam("rsslimit", 15); # Only take the first 15 entries
@@ -2552,23 +2563,30 @@ sub GetFormStart {
 }
 
 sub GetSearchForm {
-  my $html = GetFormStart(undef, 'get', 'search') . $q->start_p;
-  $html .= $q->label({-for=>'search'}, T('Search:')) . ' '
-      . $q->textfield(-name=>'search', -id=>'search', -size=>15, -accesskey=>T('f')) . ' ';
-  if (GetParam('search') ne '' and UserIsAdmin()) { # see DoBrowseRequest
-    $html .= $q->label({-for=>'replace'}, T('Replace:')) . ' '
-	. $q->textfield(-name=>'replace', -id=>'replace', -size=>20) . ' '
-        . $q->label({-for=>'delete', -title=>'If you want to replace matches with the empty string'}, T('Delete')) . ' '
-	. $q->input({-type=>'checkbox', -name=>'delete'})
-	. $q->submit('preview', T('Preview'));
+  my $html = GetFormStart(undef, 'get', 'search');
+  my $replacing = (GetParam('search') ne '' and UserIsAdmin());
+  $html .= $q->start_p({-class => ($replacing ? 'replace' : 'search')});
+  $html .= $q->span({-class=>'search'},
+                    $q->label({-for=>'search'}, T('Search:')) . ' '
+                    . $q->textfield(-name=>'search', -id=>'search', -size=>15, -accesskey=>T('f'))) . ' ';
+  if ($replacing) { # see DoBrowseRequest
+    $html .= $q->span({-class=>'replace'},
+                      $q->label({-for=>'replace'}, T('Replace:')) . ' '
+                      . $q->textfield(-name=>'replace', -id=>'replace', -size=>20)) . ' '
+        . $q->span({-class=>'delete'},
+                   $q->label({-for=>'delete', -title=>'If you want to replace matches with the empty string'}, T('Delete')) . ' '
+                   . $q->input({-type=>'checkbox', -name=>'delete'})) . ' '
+	. $q->submit('preview', T('Preview')) . ' ';
   }
   if (GetParam('matchingpages', $MatchingPages)) {
-    $html .= $q->label({-for=>'matchingpage'}, T('Filter:')) . ' '
-	. $q->textfield(-name=>'match', -id=>'matchingpage', -size=>15) . ' ';
+    $html .= $q->span({-class=>'match'},
+                      $q->label({-for=>'matchingpage'}, T('Filter:')) . ' '
+                      . $q->textfield(-name=>'match', -id=>'matchingpage', -size=>15)) . ' ';
   }
   if (%Languages) {
-    $html .= $q->label({-for=>'searchlang'}, T('Language:')) . ' '
-	. $q->textfield(-name=>'lang', -id=>'searchlang', -size=>5, -default=>GetParam('lang', '')) . ' ';
+    $html .= $q->span({-class=>'lang'},
+                      $q->label({-for=>'searchlang'}, T('Language:')) . ' '
+                      . $q->textfield(-name=>'lang', -id=>'searchlang', -size=>5, -default=>GetParam('lang', ''))) . ' ';
   }
   $html .= $q->submit('dosearch', T('Go!')) . $q->end_p . $q->end_form;
   return $html;
@@ -3356,7 +3374,6 @@ sub SortIndex {
 
 sub DoIndex {
   my $raw = GetParam('raw', 0);
-  my $match = GetParam('match', '');
   my $limit = GetParam('n', '');
   my @pages = ();
   my @menu = ($q->label({-for=>'indexmatch'}, T('Filter:')) . ' '
@@ -3368,7 +3385,7 @@ sub DoIndex {
     push(@pages, $sub->()) if $value;
     push(@menu, $q->checkbox(-name=>$option, -checked=>$value, -label=>$text));
   }
-  @pages = grep /$match/i, @pages if $match;
+  @pages = Matched(GetParam('match', ''), @pages);
   @pages = sort SortIndex @pages;
   @pages = @pages[0 .. $limit - 1] if $limit;
   if ($raw) {
@@ -3541,11 +3558,24 @@ sub SearchTitleAndBody {
   return @found;
 }
 
-sub Filtered { # this is overwriten in extensions such as tags.pl
+# Filter the pages to be searched for $string. The default implementation
+# ignores $string and uses $match instead, just in case the user used both
+# search and match parameters. This is overwritten in extensions such as tags.pl
+# which extract tags from $string and use that to filter the pages.
+sub Filtered {
   my ($string, @pages) = @_;
-  my $match = GetParam('match', '');
-  @pages = grep /$match/i, @pages if $match;
-  return @pages;
+  return Matched(GetParam('match', ''), @pages);
+}
+
+sub Matched { # strictly for page titles
+  my ($string, @pages) = @_;
+  return @pages unless $string;
+  my @terms = grep { $_ } split(/[ _]+/, $string);
+  return grep {
+    my $id = $_;
+    for (@terms) { return unless $id =~ /$_/i }
+    return $id;
+  } @pages;
 }
 
 sub SearchString {