pastokes/UnicodeTableGenerator.xsl

## UnicodeTableGenerator.xsl
<?xml version="1.0" encoding="UTF-8"?>
<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform" xmlns:xs="http://www.w3.org/2001/XMLSchema" exclude-result-prefixes="xs"
    version="2.0">

    <!--
        Script to extract content from Unicode 13.0 Character Code Charts - Scripts (https://www.unicode.org/charts/#scripts)
        The ultimate goal is to get a more or less machine-readable table of code points for all scripts in ISO15924.
        This information must surely be online already, but I haven't yet managed to find it.
        The XSLT is interesting for use of tail recursion with the so-called 'Kayessian' method for structuring flat sequences of elements
        by their attributes.
        Peter A. Stokes, École Pratique des Hautes Études – Université PSL, May 2020.
        -->

    <xsl:output method="html" encoding="utf-8" indent="yes" version="5"/>

    <xsl:template match="/">
        <html>
            <head>
                <title>Unicode Points by Script</title>
            </head>
            <body>
                <table>
                    <thead>
                        <tr>
                            <th>Script</th>
                            <th>Start</th>
                            <th>End</th>
                        </tr>
                    </thead>
                    <tbody>
                        <xsl:apply-templates select="//td"/>
                    </tbody>
                </table>
            </body>
        </html>
    </xsl:template>


    <!-- td elements are used as the main division structure in the table (!),
        then the 'rows' of the table are in fact a flat sequence of HTML p elements
        so we use the Kayessian method, which means keeping track of our position.
        I do this using tail recursion and a counter parameter, but I'm sure there
        are other ways using position() or similar.
        -->
    <xsl:template match="td">
        <xsl:call-template name="p_mb">
            <xsl:with-param name="n" select="1"/>
            <xsl:with-param name="root" select="."/>
        </xsl:call-template>
    </xsl:template>


    <xsl:template name="p_mb">
        <xsl:param name="n"/>
        <xsl:param name="root"/>

        <xsl:variable name="currnode" select="$root//p[@class = 'mb'][$n]"/>
        <xsl:variable name="script" select="$currnode/a/."/>

        <xsl:apply-templates select="$currnode/a[@title]"/>

        <xsl:for-each
            select="$root//p[@class = 'mb'][$n]/following-sibling::p[(@class = 'pb' or @class = 'sb') and count(preceding-sibling::p[@class = 'mb']) = $n]">
            <xsl:apply-templates select="a[@title]" mode="subrow"/>
        </xsl:for-each>


        <xsl:if test="$root//p[@class = 'mb'][$n + 1]">
            <xsl:call-template name="p_mb">
                <xsl:with-param name="n" select="$n + 1"/>
                <xsl:with-param name="root" select="$root"/>
            </xsl:call-template>
        </xsl:if>
    </xsl:template>

    <xsl:template match="a">
        <tr>
            <td>
                <xsl:value-of select="."/>
            </td>
            <xsl:apply-templates select="@title"/>
        </tr>
    </xsl:template>

    <xsl:template match="a" mode="subrow">
        <tr>
            <td>
                <xsl:value-of select="concat('–– ', .)"/>
            </td>
            <xsl:choose>
                <xsl:when test="@title">
                    <xsl:apply-templates select="@title"/>
                </xsl:when>
                <!-- Shouldn't ever get here... -->
                <xsl:otherwise>
                    <td>––</td>
                    <td>––</td>
                </xsl:otherwise>
            </xsl:choose>
        </tr>
    </xsl:template>

    <!-- Doesn't really need to be in a named template, but structure is a bit clearer this way I think, since then
        each p has its own template -->
    <xsl:template name="sub_p">
        <xsl:param name="script"/>
        <xsl:param name="root"/>

        <!-- Careful: sometimes we have more than one a node in a p, and sometimes it's a cross-ref, not a range (so no title) -->
        <xsl:apply-templates select="$root/a[@title]" mode="subrow"/>
    </xsl:template>

    <xsl:template match="a/@title">
        <!-- Careful:
            - Sometimes the title uses a hypen and sometimes an en-dash
            - Sometimes there is only a single code point
            TODO: The following cases are not properly covered:
            - Sometimes there are multiple ranges (ranges after first simply included in second column)
            - Sometimes there is no @title but there is a start-point in the URL (currently simply omitted)
            - Sometimes the link is a cross-reference, not a range (currently simply omitted)
        -->
        <xsl:choose>
            <!-- Hyphen -->
            <xsl:when test="contains(., '-')">
                <td>
                    <xsl:value-of select="substring-before(., '-')"/>
                </td>
                <td>
                    <xsl:value-of select="substring-after(., '-')"/>
                </td>
            </xsl:when>
            <!-- En-dash -->
            <xsl:when test="contains(., '–')">
                <td>
                    <xsl:value-of select="substring-before(., '–')"/>
                </td>
                <td>
                    <xsl:value-of select="substring-after(., '–')"/>
                </td>
            </xsl:when>
            <xsl:otherwise>
                <td>
                    <xsl:value-of select="."/>
                </td>
                <td>
                    ––
                </td>
            </xsl:otherwise>
        </xsl:choose>

    </xsl:template>
</xsl:stylesheet>
	<?xml version="1.0" encoding="UTF-8"?>
	<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform" xmlns:xs="http://www.w3.org/2001/XMLSchema" exclude-result-prefixes="xs"
	version="2.0">

	<!--
	Script to extract content from Unicode 13.0 Character Code Charts - Scripts (https://www.unicode.org/charts/#scripts)
	The ultimate goal is to get a more or less machine-readable table of code points for all scripts in ISO15924.
	This information must surely be online already, but I haven't yet managed to find it.
	The XSLT is interesting for use of tail recursion with the so-called 'Kayessian' method for structuring flat sequences of elements
	by their attributes.
	Peter A. Stokes, École Pratique des Hautes Études – Université PSL, May 2020.
	-->

	<xsl:output method="html" encoding="utf-8" indent="yes" version="5"/>

	<xsl:template match="/">
	<html>
	<head>
	<title>Unicode Points by Script</title>
	</head>
	<body>
	<table>
	<thead>
	<tr>
	<th>Script</th>
	<th>Start</th>
	<th>End</th>
	</tr>
	</thead>
	<tbody>
	<xsl:apply-templates select="//td"/>
	</tbody>
	</table>
	</body>
	</html>
	</xsl:template>


	<!-- td elements are used as the main division structure in the table (!),
	then the 'rows' of the table are in fact a flat sequence of HTML p elements
	so we use the Kayessian method, which means keeping track of our position.
	I do this using tail recursion and a counter parameter, but I'm sure there
	are other ways using position() or similar.
	-->
	<xsl:template match="td">
	<xsl:call-template name="p_mb">
	<xsl:with-param name="n" select="1"/>
	<xsl:with-param name="root" select="."/>
	</xsl:call-template>
	</xsl:template>


	<xsl:template name="p_mb">
	<xsl:param name="n"/>
	<xsl:param name="root"/>

	<xsl:variable name="currnode" select="$root//p[@class = 'mb'][$n]"/>
	<xsl:variable name="script" select="$currnode/a/."/>

	<xsl:apply-templates select="$currnode/a[@title]"/>

	<xsl:for-each
	select="$root//p[@class = 'mb'][$n]/following-sibling::p[(@class = 'pb' or @class = 'sb') and count(preceding-sibling::p[@class = 'mb']) = $n]">
	<xsl:apply-templates select="a[@title]" mode="subrow"/>
	</xsl:for-each>


	<xsl:if test="$root//p[@class = 'mb'][$n + 1]">
	<xsl:call-template name="p_mb">
	<xsl:with-param name="n" select="$n + 1"/>
	<xsl:with-param name="root" select="$root"/>
	</xsl:call-template>
	</xsl:if>
	</xsl:template>

	<xsl:template match="a">
	<tr>
	<td>
	<xsl:value-of select="."/>
	</td>
	<xsl:apply-templates select="@title"/>
	</tr>
	</xsl:template>

	<xsl:template match="a" mode="subrow">
	<tr>
	<td>
	<xsl:value-of select="concat('–– ', .)"/>
	</td>
	<xsl:choose>
	<xsl:when test="@title">
	<xsl:apply-templates select="@title"/>
	</xsl:when>
	<!-- Shouldn't ever get here... -->
	<xsl:otherwise>
	<td>––</td>
	<td>––</td>
	</xsl:otherwise>
	</xsl:choose>
	</tr>
	</xsl:template>

	<!-- Doesn't really need to be in a named template, but structure is a bit clearer this way I think, since then
	each p has its own template -->
	<xsl:template name="sub_p">
	<xsl:param name="script"/>
	<xsl:param name="root"/>

	<!-- Careful: sometimes we have more than one a node in a p, and sometimes it's a cross-ref, not a range (so no title) -->
	<xsl:apply-templates select="$root/a[@title]" mode="subrow"/>
	</xsl:template>

	<xsl:template match="a/@title">
	<!-- Careful:
	- Sometimes the title uses a hypen and sometimes an en-dash
	- Sometimes there is only a single code point
	TODO: The following cases are not properly covered:
	- Sometimes there are multiple ranges (ranges after first simply included in second column)
	- Sometimes there is no @title but there is a start-point in the URL (currently simply omitted)
	- Sometimes the link is a cross-reference, not a range (currently simply omitted)
	-->
	<xsl:choose>
	<!-- Hyphen -->
	<xsl:when test="contains(., '-')">
	<td>
	<xsl:value-of select="substring-before(., '-')"/>
	</td>
	<td>
	<xsl:value-of select="substring-after(., '-')"/>
	</td>
	</xsl:when>
	<!-- En-dash -->
	<xsl:when test="contains(., '–')">
	<td>
	<xsl:value-of select="substring-before(., '–')"/>
	</td>
	<td>
	<xsl:value-of select="substring-after(., '–')"/>
	</td>
	</xsl:when>
	<xsl:otherwise>
	<td>
	<xsl:value-of select="."/>
	</td>
	<td>
	––
	</td>
	</xsl:otherwise>
	</xsl:choose>

	</xsl:template>
	</xsl:stylesheet>