Implement simple xml import using xsl and load xml

7 files changed, 257 insertions, 0 deletions

diff --git a/.gitignore b/.gitignore
new file mode 100644
index 0000000..b9f3884
--- /dev/null
+++ b/.gitignore
@@ -0,0 +1,2 @@
+LIKAT Publications.xml
+dbxml.xml
diff --git a/convert-style-to-html.xsl b/convert-style-to-html.xsl
new file mode 100644
index 0000000..9858964
--- /dev/null
+++ b/convert-style-to-html.xsl
@@ -0,0 +1,107 @@
+<?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" encoding="UTF-8"  />
+
+	<!-- Convert endnote style definitions declarations to html tags -->
+	<!-- TODO: <author>Editors,</author> bzw <author>Editor,</author> -->
+	<xsl:template match="style" name="style">
+
+		<xsl:param name="oldface" select="@face"/>
+		<xsl:param name="exclude" select="''" />
+
+		<xsl:variable name="face">
+			<xsl:call-template name="string-replace-all">
+				<xsl:with-param name="text" select="$oldface" />
+				<xsl:with-param name="replace" select="$exclude" />
+				<xsl:with-param name="by" select="''" />
+			</xsl:call-template>
+		</xsl:variable>
+
+		<xsl:choose>
+			<xsl:when test="contains($face, 'subscript')">
+				<sub>
+					<xsl:call-template name="style">
+						<xsl:with-param name="oldface" select="$face"/>
+						<xsl:with-param name="exclude" select="'subscript'"/>
+					</xsl:call-template>
+				</sub>
+			</xsl:when>
+			<xsl:when test="contains($face, 'superscript')">
+				<sub>
+					<xsl:call-template name="style">
+						<xsl:with-param name="oldface" select="$face"/>
+						<xsl:with-param name="exclude" select="'superscript'"/>
+					</xsl:call-template>
+				</sub>
+			</xsl:when>
+			<xsl:when test="contains($face, 'italic')">
+				<em>
+					<xsl:call-template name="style">
+						<xsl:with-param name="oldface" select="$face"/>
+						<xsl:with-param name="exclude" select="'italic'"/>
+					</xsl:call-template>
+				</em>
+			</xsl:when>
+			<xsl:when test="contains($face, 'boldface')">
+				<strong>
+					<xsl:call-template name="style">
+						<xsl:with-param name="oldface" select="$face"/>
+						<xsl:with-param name="exclude" select="'boldface'"/>
+					</xsl:call-template>
+				</strong>
+			</xsl:when>
+			<xsl:when test="contains($face, 'normal')">
+				<xsl:call-template name="style">
+					<xsl:with-param name="oldface" select="$face"/>
+					<xsl:with-param name="exclude" select="'normal'"/>
+				</xsl:call-template>
+			</xsl:when>
+
+			<xsl:otherwise>
+				<xsl:apply-templates />
+			</xsl:otherwise>
+
+		</xsl:choose>
+	</xsl:template>
+
+	<!-- Drop style definitions from author-tags -->
+	<xsl:template match="author/style">
+		<xsl:apply-templates />
+	</xsl:template>
+
+	<xsl:template match="*">
+		<xsl:copy>
+			<xsl:copy-of select="@*"/>
+			<xsl:apply-templates />
+		</xsl:copy>
+	</xsl:template>
+
+	<xsl:template name="string-replace-all">
+		<xsl:param name="text" />
+		<xsl:param name="replace" />
+		<xsl:param name="by" />
+		<xsl:choose>
+			<xsl:when test="$replace = ''">
+				<xsl:value-of select="$text" />
+			</xsl:when>
+			<xsl:when test="contains($text, $replace)">
+				<xsl:value-of select="substring-before($text,$replace)" />
+				<xsl:value-of select="$by" />
+				<xsl:call-template name="string-replace-all">
+					<xsl:with-param name="text" select="substring-after($text,$replace)" />
+					<xsl:with-param name="replace" select="$replace" />
+					<xsl:with-param name="by" select="$by" />
+				</xsl:call-template>
+			</xsl:when>
+			<xsl:otherwise>
+				<xsl:value-of select="$text" />
+			</xsl:otherwise>
+		</xsl:choose>
+	</xsl:template>
+
+	<xsl:template match="/">
+		<xsl:apply-templates />
+	</xsl:template>
+
+</xsl:stylesheet>
diff --git a/endnote-to-dbxml.xsl b/endnote-to-dbxml.xsl
new file mode 100644
index 0000000..24d3be3
--- /dev/null
+++ b/endnote-to-dbxml.xsl
@@ -0,0 +1,72 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform" version="1.0">
+
+	<xsl:output method="xml" indent="yes" encoding="UTF-8"  />
+
+	<xsl:include href="util.xsl"/>
+
+	<xsl:template match="record">
+		<publication>
+			<uid>
+				<xsl:value-of select="rec-number" />
+			</uid>
+			<title>
+				<xsl:apply-templates select="titles/title" mode="copy-html" />
+			</title>
+			<secondary_title>
+				<xsl:apply-templates select="titles/secondary-title" mode="copy-html" />
+				<xsl:text></xsl:text>
+			</secondary_title>
+			<full_title>
+				<xsl:apply-templates select="periodical/full-title" mode="copy-html" />
+				<xsl:text></xsl:text>
+			</full_title>
+			<type>
+				<xsl:value-of select="ref-type/@name" />
+			</type>
+			<year>
+				<xsl:apply-templates select="dates/year" mode="content" />
+			</year>
+
+			<authors>
+				<!--<xsl:apply-templates select="authors" mode="references" />-->
+				<xsl:for-each select="contributors/authors/author">
+					<xsl:if test="position() &gt; 1">
+						<xsl:text>|</xsl:text>
+					</xsl:if>
+					<xsl:value-of select="." />
+				</xsl:for-each>
+			</authors>
+		</publication>
+	</xsl:template>
+
+	<xsl:template match="/">
+		<tx_likat_pubs_domain_model_publications>
+			<xsl:apply-templates select="xml/records/record" />
+		</tx_likat_pubs_domain_model_publications>
+	</xsl:template>
+
+	<xsl:template name="string-replace-all">
+		<xsl:param name="text" />
+		<xsl:param name="replace" />
+		<xsl:param name="by" />
+		<xsl:choose>
+			<xsl:when test="$replace = ''">
+				<xsl:value-of select="$text" />
+			</xsl:when>
+			<xsl:when test="contains($text, $replace)">
+				<xsl:value-of select="substring-before($text,$replace)" />
+				<xsl:value-of select="$by" />
+				<xsl:call-template name="string-replace-all">
+					<xsl:with-param name="text" select="substring-after($text,$replace)" />
+					<xsl:with-param name="replace" select="$replace" />
+					<xsl:with-param name="by" select="$by" />
+				</xsl:call-template>
+			</xsl:when>
+			<xsl:otherwise>
+				<xsl:value-of select="$text" />
+			</xsl:otherwise>
+		</xsl:choose>
+	</xsl:template>
+
+</xsl:stylesheet>
diff --git a/fix-empty-tags.sed b/fix-empty-tags.sed
new file mode 100644
index 0000000..f958206
--- /dev/null
+++ b/fix-empty-tags.sed
@@ -0,0 +1,3 @@
+# LOAD XML has bugs (NULL columns) when there are empty, self closing tags like
+# <title/> Replace those with <title></title>
+s/<\([^>]*\)\/>/<\1><\/\1>/g
diff --git a/update.sh b/update.sh
new file mode 100755
index 0000000..6ad1214
--- /dev/null
+++ b/update.sh
@@ -0,0 +1,13 @@
+#!/bin/sh
+
+cd $(dirname $0)
+cat "LIKAT Publications.xml" \
+	| xsltproc convert-style-to-html.xsl - \
+	| xsltproc endnote-to-dbxml.xsl - \
+	| sed -f fix-empty-tags.sed \
+	> dbxml.xml
+
+#	sed -f cdata.sed | \
+#	xmllint --pretty 1 - \
+
+mysql -v -utypo3 -ptypo3 typo3 < update.sql
diff --git a/update.sql b/update.sql
new file mode 100644
index 0000000..c745643
--- /dev/null
+++ b/update.sql
@@ -0,0 +1,6 @@
+START TRANSACTION;
+DELETE FROM tx_likat_pubs_domain_model_publications;
+LOAD XML LOCAL INFILE './dbxml.xml'
+	INTO TABLE tx_likat_pubs_domain_model_publications
+	ROWS IDENTIFIED BY '<publication>';
+COMMIT;
diff --git a/util.xsl b/util.xsl
new file mode 100644
index 0000000..19e6cf9
--- /dev/null
+++ b/util.xsl
@@ -0,0 +1,54 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform" version="1.0">
+
+
+	<xsl:template match="*" mode="copy-cdata">
+		<cdata>
+		<xsl:apply-templates />
+		</cdata>
+	</xsl:template>
+
+	<xsl:template match="*" mode="copy-html">
+		<xsl:apply-templates select="*|text()" mode="verb" />
+	</xsl:template>
+
+	<!--
+	     Copy nodeset as escaped XML
+	     http://logit.yudichev.net/2007/11/xslt-output-xml-escaped-copy-of-source.html
+	-->
+	<xsl:template match="*|@*" mode="verb">
+		<xsl:variable name="node-type">
+			<xsl:call-template name="node-type"/>
+		</xsl:variable>
+		<xsl:choose>
+			<xsl:when test="$node-type='element'"> <!-- element -->
+				<xsl:text>&lt;</xsl:text>
+				<xsl:value-of select="name()"/>
+				<xsl:apply-templates select="@*" mode="verb"/>
+				<xsl:text>&gt;</xsl:text>
+				<xsl:apply-templates mode="verb"/>
+				<xsl:text>&lt;/</xsl:text>
+				<xsl:value-of select="name()"/>
+				<xsl:text>&gt;</xsl:text>
+			</xsl:when>
+			<xsl:when test="$node-type='text'"> <!-- text -->
+				<xsl:value-of select="self::text()"/>
+			</xsl:when>
+			<xsl:when test="$node-type='attribute'"> <!--any attribute-->
+				<xsl:text> </xsl:text>
+				<xsl:value-of select="name()"/>
+				<xsl:text>="</xsl:text>
+				<xsl:value-of select="."/>
+				<xsl:text>"</xsl:text>
+			</xsl:when>
+		</xsl:choose>
+	</xsl:template>
+	<xsl:template name="node-type">
+		<xsl:param name="node" select="."/>
+		<xsl:apply-templates mode="nodetype" select="$node"/>
+	</xsl:template>
+	<xsl:template mode="nodetype" match="*">element</xsl:template>
+	<xsl:template mode="nodetype" match="@*">attribute</xsl:template>
+	<xsl:template mode="nodetype" match="text()">text</xsl:template>
+
+</xsl:stylesheet>