Focus on source XML manipulation, introduce indexes

Instead of XHTML or Atom, so resulting files get generated out of
custom XML structures now.

This was needed to get indexes that won't depend on Atom (and to avoid
the duplication).

Now indexes can be included with something like the following:

<notes xmlns="" number="5"
       src="../build/notes-by-modification-date.xml" />

12 files changed, 236 insertions, 213 deletions

diff --git a/tools/README b/tools/README
index 43dfea4..bbd251d 100644
--- a/tools/README
+++ b/tools/README
@@ -7,8 +7,8 @@ works for now, and is pretty simple.
 
 Editing:
 
-- skeleton.xml is the primary skeleton to start with, for *.xhtml
-  files in ~/homepage/src/.
+- skeleton.xml is the primary skeleton to start with, for *.xhtml files
+  in ~/homepage/src/.
 
 This stage is mostly needed to get unified styles, metadata
 properties, and other bits in case if I'll change them. The same can
@@ -19,21 +19,30 @@ Editing is done with the emacs html-mode, with
 <https://github.com/defanor/html-wysiwyg>.
 
 
+Indexing:
+
+- xml-notes-dump.xsl (with some help from publish.sh) is used to
+  create notes-dump.xml in the build directory.
+- xml-notes-sort.xsl sorts that to create
+  notes-by-publication-date.xml and notes-by-modification-date.xml.
+- xml-notes-limit.xsl is used to extract the most recent notes.
+
+
 HTML:
 
-- xml-to-html.sh translates source files into actual (X)HTML in
-  ~/homepage/ by applying xml-to-html.xsl to them.
+- xml-to-html.xsl is used to translate source files into actual (X)HTML
+  in ~/homepage/.
+
+Indexes can be included into those using the <notes> element (so it
+should run after those are built, if they are used).
 
 
 Atom feed:
 
-- html-to-atom-dump.xsl dumps HTML files into a single atom feed.
-- atom-sort.xsl sorts those by modification date.
-- atom-limit.xsl limits them, taking the newest entries.
-- html-to-atom.sh finds source files, and pipes them through the
-  XSLTs.
+- xml-notes-to-atom.xsl translates a dump of notes (such as
+  notes-by-modification-date.xml) into an atom feed.
 
 
 Publishing:
 
-- publish.sh runs the above two and uploads the files with rsync.
+- publish.sh runs all of the above and uploads the files with rsync.
diff --git a/tools/atom-limit.xsl b/tools/atom-limit.xsl
deleted file mode 100644
index 2b6a285..0000000
--- a/tools/atom-limit.xsl
+++ /dev/null
@@ -1,27 +0,0 @@
-<?xml version="1.0" encoding="UTF-8"?>
-
-<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
-                xmlns:atom="http://www.w3.org/2005/Atom"
-                xmlns:xhtml="http://www.w3.org/1999/xhtml"
-                xmlns="http://www.w3.org/2005/Atom"
-                version="1.0">
-  <xsl:output method="xml" indent="yes"/>
-
-  <xsl:template match="/">
-    <xsl:apply-templates />
-  </xsl:template>
-
-  <xsl:template match="atom:feed">
-    <feed>
-      <xsl:apply-templates select="node()[not(self::atom:entry)]" />
-      <!-- Take the top 10 entries. -->
-      <xsl:apply-templates select="atom:entry[position()&lt;10]" />
-    </feed>
-  </xsl:template>
-
-  <!-- Copy everything else -->
-  <xsl:template match="node()" priority="0">
-    <xsl:copy-of select="." />
-  </xsl:template>
-
-</xsl:stylesheet>
diff --git a/tools/atom-sort.xsl b/tools/atom-sort.xsl
deleted file mode 100644
index a00b599..0000000
--- a/tools/atom-sort.xsl
+++ /dev/null
@@ -1,29 +0,0 @@
-<?xml version="1.0" encoding="UTF-8"?>
-
-<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
-                xmlns:atom="http://www.w3.org/2005/Atom"
-                xmlns:xhtml="http://www.w3.org/1999/xhtml"
-                xmlns="http://www.w3.org/2005/Atom"
-                version="1.0">
-  <xsl:output method="xml" indent="yes"/>
-
-  <xsl:template match="/">
-    <xsl:apply-templates />
-  </xsl:template>
-
-  <xsl:template match="atom:feed">
-    <feed>
-      <xsl:apply-templates select="node()[not(self::atom:entry)]" />
-      <!-- Sort by modification date, descending. -->
-      <xsl:apply-templates select="atom:entry">
-        <xsl:sort select="atom:updated" order="descending"/>
-      </xsl:apply-templates>
-    </feed>
-  </xsl:template>
-
-  <!-- Copy everything else -->
-  <xsl:template match="node()" priority="0">
-    <xsl:copy-of select="." />
-  </xsl:template>
-
-</xsl:stylesheet>
diff --git a/tools/html-to-atom-dump.xsl b/tools/html-to-atom-dump.xsl
deleted file mode 100644
index fce78a2..0000000
--- a/tools/html-to-atom-dump.xsl
+++ /dev/null
@@ -1,110 +0,0 @@
-<?xml version="1.0" encoding="UTF-8"?>
-
-<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
-                xmlns:atom="http://www.w3.org/2005/Atom"
-                xmlns:xhtml="http://www.w3.org/1999/xhtml"
-                xmlns="http://www.w3.org/2005/Atom"
-                version="1.0">
-  <xsl:output method="xml" indent="yes"/>
-  <xsl:variable name="baseIRI">
-    <xsl:text>https://defanor.uberspace.net/</xsl:text>
-  </xsl:variable>
-
-  <!-- Feed -->
-  <xsl:template match="/">
-    <feed>
-      <title>defanor's notes</title>
-      <link rel="self" href="{$baseIRI}atom.xml" />
-      <link rel="alternate" href="{$baseIRI}" />
-      <id>
-        <xsl:value-of select="$baseIRI" />
-      </id>
-      <!--
-          Setting <updated> statically, since it applies to some
-          meaningful feed updates. Perhaps should be modified on
-          metadata changes.
-      -->
-      <updated>2018-04-14T10:00:00Z</updated>
-      <xsl:apply-templates select="/list/entry" />
-    </feed>
-  </xsl:template>
-
-  <!-- Turns index entries into Atom entries -->
-  <xsl:template match="entry">
-    <xsl:variable name="localPath">
-      <xsl:text>../</xsl:text>
-      <xsl:value-of select="@name" />
-    </xsl:variable>
-    <xsl:variable name="doc" select="document($localPath)" />
-    <entry>
-      <link rel="alternate" href="{$baseIRI}{@name}" />
-      <id>
-        <xsl:text>https://defanor.uberspace.net/</xsl:text>
-        <xsl:value-of select="@name" />
-      </id>
-      <!-- Setting this statically for now. -->
-      <author>
-        <name>defanor</name>
-      </author>
-      <xsl:apply-templates mode="atom-entry" select="$doc" />
-    </entry>
-  </xsl:template>
-
-  <xsl:template mode="atom-entry" match="/ | xhtml:head | xhtml:html">
-    <xsl:apply-templates mode="atom-entry" />
-  </xsl:template>
-
-  <!-- Title -->
-  <xsl:template mode="atom-entry" match="xhtml:title">
-    <title>
-      <xsl:value-of select="." />
-    </title>
-  </xsl:template>
-
-  <!-- Description/summary -->
-  <xsl:template mode="atom-entry"
-                match="xhtml:meta[@name='description' or
-                       @property='dc:description schema:description']">
-    <summary>
-      <xsl:value-of select="@content" />
-    </summary>
-  </xsl:template>
-
-  <!--
-      Atom is rather strict in its date-time requriements. It's not
-      always convenient or useful to be that strict in HTML metadata,
-      so might be better to introduce some check and conversion.
-  -->
-  <!-- Publication date -->
-  <xsl:template mode="atom-entry"
-                match="xhtml:meta[@property='dc:issued schema:datePublished']">
-    <published>
-      <xsl:value-of select="@content" />
-    </published>
-  </xsl:template>
-  
-  <!-- Modification date -->
-  <xsl:template mode="atom-entry"
-                match="xhtml:meta[@property='dc:modified schema:dateModified']">
-    <updated>
-      <xsl:value-of select="@content" />
-    </updated>
-  </xsl:template>
-
-  <!-- Body/content -->
-  <xsl:template mode="atom-entry" match="xhtml:body">
-    <content type="xhtml">
-      <!--
-          "the content of atom:content MUST be a single XHTML div
-          element"
-      -->
-      <xhtml:div>
-        <xsl:copy-of select="*" />
-      </xhtml:div>
-    </content>
-  </xsl:template>
-
-  <!-- Skip everything else -->
-  <xsl:template mode="atom-entry" match="node()" priority="0" />
-
-</xsl:stylesheet>
diff --git a/tools/html-to-atom.sh b/tools/html-to-atom.sh
deleted file mode 100755
index 3bebf45..0000000
--- a/tools/html-to-atom.sh
+++ /dev/null
@@ -1,20 +0,0 @@
-#!/bin/sh
-
-# Composes an atom feed out of XHTML files.
-
-BASEDIR=~/homepage
-TOOLS="${BASEDIR}/tools"
-SOURCES="${BASEDIR}/src"
-ATOM="${BASEDIR}/atom.xml"
-NOTES="${BASEDIR}/notes"
-
-(echo '<?xml version="1.0" encoding="UTF-8"?>' &&
-     echo '<list>' &&
-     find "${NOTES}" -name '*.xhtml' |
-         # grep -Ev "^(${TOOLS}|${SOURCES}).*" |
-         sed -e "sS^${BASEDIR}/\(.*\)S<entry name=\"\\1\" />S" &&
-     echo '</list>') |
-    xsltproc "${TOOLS}/html-to-atom-dump.xsl" - |
-    xsltproc "${TOOLS}/atom-sort.xsl" - |
-    xsltproc "${TOOLS}/atom-limit.xsl" - |
-    xmllint --format --nsclean - > "${ATOM}"
diff --git a/tools/publish.sh b/tools/publish.sh
index 5a6cbea..c962063 100755
--- a/tools/publish.sh
+++ b/tools/publish.sh
@@ -1,8 +1,40 @@
 #!/bin/sh
 
-TOOLS=~/homepage/tools
+BASEDIR=~/homepage
+TOOLS="${BASEDIR}/tools"
+BUILD="${BASEDIR}/build"
+ATOM="${BASEDIR}/atom.xml"
+SOURCES="${BASEDIR}/src"
+NOTES="${BASEDIR}/notes"
 
-${TOOLS}/xml-to-html.sh
-${TOOLS}/html-to-atom.sh
-rsync --exclude '.*' --exclude '*.org' --exclude 'src/' --exclude 'tools/' \
+
+# dump all the notes into a single file
+(echo '<?xml version="1.0" encoding="UTF-8"?>' &&
+     echo '<notes>' &&
+     find "${NOTES}" -name '*.xhtml' |
+         sed -e "sS^${BASEDIR}/\(.*\)S  <note src=\"\\1\" />S" &&
+     echo '</notes>') |
+    xsltproc -o "${BUILD}/notes-dump.xml" "${TOOLS}/xml-notes-dump.xsl" -
+
+# sort by publication and modification dates
+xsltproc -o "${BUILD}/notes-by-publication-date.xml" \
+         "${TOOLS}/xml-notes-sort.xsl" "${BUILD}/notes-dump.xml"
+xsltproc -o "${BUILD}/notes-by-modification-date.xml" \
+         --stringparam sortBy modified \
+         "${TOOLS}/xml-notes-sort.xsl" "${BUILD}/notes-dump.xml"
+
+# create an atom feed with the most recently modified 10 entries
+xsltproc --param number 10 "${TOOLS}/xml-notes-limit.xsl" \
+         "${BUILD}/notes-by-modification-date.xml" |
+    xsltproc -o "${ATOM}" "${TOOLS}/xml-notes-to-atom.xsl" -
+
+# convert sources into XHTML
+find "${SOURCES}" -name '*.xhtml' |
+    sed -e "sS^${SOURCES}SS" |
+    xargs -Ifile xsltproc -o "${BASEDIR}/file" \
+          "${TOOLS}/xml-to-html.xsl" "${SOURCES}file"
+
+# upload
+rsync --exclude '.*' --exclude '*.org' --exclude 'src/' \
+      --exclude 'tools/' --exclude 'build/' \
       -avz . uberspace.net:homepage/web/
diff --git a/tools/xml-notes-dump.xsl b/tools/xml-notes-dump.xsl
new file mode 100644
index 0000000..79b9606
--- /dev/null
+++ b/tools/xml-notes-dump.xsl
@@ -0,0 +1,26 @@
+<?xml version="1.0" encoding="UTF-8"?>
+
+<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
+                version="1.0">
+  <xsl:output method="xml" indent="yes"/>
+
+  <xsl:template match="note">
+    <xsl:variable name="docPath">
+      <xsl:text>../src/</xsl:text>
+      <xsl:value-of select="@src" />
+    </xsl:variable>
+    <xsl:variable name="doc" select="document($docPath)" />
+    <xsl:copy>
+      <xsl:copy-of select="@*" />
+      <xsl:copy-of select="$doc" />
+    </xsl:copy>
+  </xsl:template>
+
+  <xsl:template match="node()" priority="0">
+    <xsl:copy>
+      <xsl:copy-of select="@*" />
+      <xsl:apply-templates />
+    </xsl:copy>
+  </xsl:template>
+
+</xsl:stylesheet>
diff --git a/tools/xml-notes-limit.xsl b/tools/xml-notes-limit.xsl
new file mode 100644
index 0000000..149cbd9
--- /dev/null
+++ b/tools/xml-notes-limit.xsl
@@ -0,0 +1,22 @@
+<?xml version="1.0" encoding="UTF-8"?>
+
+<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
+                version="1.0">
+  <xsl:output method="xml" indent="yes"/>
+  <xsl:param name="number" />
+
+  <xsl:template match="notes">
+    <xsl:copy>
+      <xsl:copy-of select="@*" />
+      <xsl:apply-templates select="note[position()&lt;=$number]" />
+    </xsl:copy>
+  </xsl:template>
+
+  <xsl:template match="node()" priority="0">
+    <xsl:copy>
+      <xsl:copy-of select="@*" />
+      <xsl:apply-templates />
+    </xsl:copy>
+  </xsl:template>
+
+</xsl:stylesheet>
diff --git a/tools/xml-notes-sort.xsl b/tools/xml-notes-sort.xsl
new file mode 100644
index 0000000..9d6ac70
--- /dev/null
+++ b/tools/xml-notes-sort.xsl
@@ -0,0 +1,32 @@
+<?xml version="1.0" encoding="UTF-8"?>
+
+<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
+                version="1.0">
+  <xsl:output method="xml" indent="yes"/>
+  <xsl:param name="sortBy" />
+
+  <xsl:template match="notes">
+    <xsl:copy>
+      <xsl:copy-of select="@*" />
+      <xsl:choose>
+        <xsl:when test="$sortBy='modified'">
+          <xsl:apply-templates select="note">
+            <xsl:sort select="document/@modified" order="descending"/>
+          </xsl:apply-templates>
+        </xsl:when>
+        <xsl:otherwise>
+          <xsl:apply-templates select="note">
+            <xsl:sort select="document/@published" order="descending"/>
+          </xsl:apply-templates>
+        </xsl:otherwise>
+      </xsl:choose>
+    </xsl:copy>
+  </xsl:template>
+
+  <xsl:template match="node()" priority="0">
+    <xsl:copy>
+      <xsl:copy-of select="@*" />
+      <xsl:apply-templates />
+    </xsl:copy>
+  </xsl:template>
+</xsl:stylesheet>
diff --git a/tools/xml-notes-to-atom.xsl b/tools/xml-notes-to-atom.xsl
new file mode 100644
index 0000000..fd18072
--- /dev/null
+++ b/tools/xml-notes-to-atom.xsl
@@ -0,0 +1,73 @@
+<?xml version="1.0" encoding="UTF-8"?>
+
+<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
+                xmlns:atom="http://www.w3.org/2005/Atom"
+                xmlns:xhtml="http://www.w3.org/1999/xhtml"
+                xmlns="http://www.w3.org/2005/Atom"
+                version="1.0">
+  <xsl:output method="xml" indent="yes"/>
+  <xsl:variable name="baseIRI">
+    <xsl:text>https://defanor.uberspace.net/</xsl:text>
+  </xsl:variable>
+
+  <xsl:template match="notes">
+    <feed>
+      <title>defanor's notes</title>
+      <link rel="self" href="{$baseIRI}atom.xml" />
+      <link rel="alternate" href="{$baseIRI}" />
+      <id>
+        <xsl:value-of select="$baseIRI" />
+      </id>
+      <!--
+          Setting <updated> statically, since it applies to some
+          meaningful feed updates. Perhaps should be modified on
+          metadata changes.
+      -->
+      <updated>2018-05-01T01:00:00Z</updated>
+      <xsl:copy-of select="@*" />
+      <xsl:apply-templates  />
+    </feed>
+  </xsl:template>
+
+  <xsl:template match="note">
+    <entry>
+      <link rel="alternate" href="{$baseIRI}{@src}" />
+      <id>
+        <xsl:text>https://defanor.uberspace.net/</xsl:text>
+        <xsl:value-of select="@src" />
+      </id>
+      <author>
+        <name>defanor</name>
+      </author>
+      <title>
+        <xsl:value-of select="document/@title" />
+      </title>
+      <summary>
+        <xsl:value-of select="document/@description" />
+      </summary>
+      <published>
+        <xsl:value-of select="document/@published" />
+      </published>
+      <updated>
+        <xsl:value-of select="document/@modified" />
+      </updated>
+      <content type="xhtml">
+        <!--
+            "the content of atom:content MUST be a single XHTML div
+            element"
+        -->
+        <xhtml:div xmlns="http://www.w3.org/1999/xhtml">
+          <xsl:copy-of select="document/xhtml:body/*" />
+        </xhtml:div>
+      </content>
+    </entry>
+  </xsl:template>
+
+  <xsl:template match="node()" priority="0">
+    <xsl:copy>
+      <xsl:copy-of select="@*" />
+      <xsl:apply-templates />
+    </xsl:copy>
+  </xsl:template>
+
+</xsl:stylesheet>
diff --git a/tools/xml-to-html.sh b/tools/xml-to-html.sh
deleted file mode 100755
index c05d749..0000000
--- a/tools/xml-to-html.sh
+++ /dev/null
@@ -1,12 +0,0 @@
-#!/bin/sh
-
-# Translates simplified documents from src/
-
-BASEDIR=~/homepage
-TOOLS="${BASEDIR}/tools"
-SOURCES="${BASEDIR}/src"
-
-find "${SOURCES}" -name '*.xhtml' |
-    sed -e "sS^${SOURCES}SS" |
-    xargs -Ifile xsltproc -o "${BASEDIR}/file" \
-          "${TOOLS}/xml-to-html.xsl" "${SOURCES}file"
diff --git a/tools/xml-to-html.xsl b/tools/xml-to-html.xsl
index 97e8431..1b92318 100644
--- a/tools/xml-to-html.xsl
+++ b/tools/xml-to-html.xsl
@@ -77,7 +77,7 @@
               href="/xhtml-rdfa-light.css" />
         <meta name="robots" content="noarchive" />
       </head>
-      <xsl:copy-of select="xhtml:body" />
+      <xsl:apply-templates mode="body" select="xhtml:body" />
     </html>
   </xsl:template>
 
@@ -92,4 +92,31 @@
     <xsl:copy-of select="." />
   </xsl:template>
 
+  <xsl:template mode="body" match="notes">
+    <xsl:variable name="notes" select="document(@src)" />
+    <xsl:variable name="number" select="@number" />
+    <dl>
+      <xsl:apply-templates mode="note-index"
+                           select="$notes/notes/note[$number=0 or position()&lt;=$number]" />
+    </dl>
+  </xsl:template>
+
+  <xsl:template mode="note-index" match="note">
+    <dt>
+      <a href="{@src}">
+        <xsl:value-of select="document/@title" />
+      </a>
+    </dt>
+    <dd>
+      <xsl:value-of select="document/@description" />
+    </dd>
+  </xsl:template>
+
+  <xsl:template mode="body" match="node()" priority="0">
+    <xsl:copy>
+      <xsl:copy-of select="@*" />
+      <xsl:apply-templates mode="body" />
+    </xsl:copy>
+  </xsl:template>
+
 </xsl:stylesheet>