chore(build): Setup actual regeneration rules for Unicode XML to Lua …

…conversion

Makefile-fonts

@@ -2,12 +2,12 @@
 
 if FONT_DOWNLOAD_TOOLS
 
+# Target defined in Makefile.am, just adding a dependency here
+.sources: fonttooling
+
 .fonts: fonttooling
 	[ -h .fonts ] || mkdir -p $@
 
-.sources: fonttooling
-	[ -h .sources ] || mkdir -p $@
-
 fonttooling:
 	$(if $(BSDTAR),,$(error Please set BSDTAR with path or `./configure --enable-developer-mode`))
 	$(if $(CURL),,$(error Please set CURL with path or `./configure --enable-developer-mode`))

Makefile.am

@@ -76,6 +76,7 @@ dist_license_DATA = LICENSE.md
 EXTRA_DIST = spec tests documentation sile-dev-1.rockspec fontconfig.conf
 EXTRA_DIST += build-aux/action-updater.js build-aux/cargo-updater.js build-aux/config.ld build-aux/decore-automake.sh build-aux/git-version-gen
 EXTRA_DIST += Dockerfile build-aux/docker-bootstrap.sh build-aux/docker-fontconfig.conf hooks/build
+EXTRA_DIST += build-aux/xml-entities-to-lua.xsl
 EXTRA_DIST += default.nix flake.nix flake.lock shell.nix build-aux/pkg.nix
 EXTRA_DIST += package.json # imported by both Nix and Docker
 EXTRA_DIST += $(FIGURES)
@@ -94,6 +95,10 @@ else !SHARED
 EXTRA_RUNTIME_DEPS =
 endif
 
+MATHML_ENTITIES = packages/math/mathml-entities.lua
+EXTRA_DIST += $(MATHML_ENTITIES)
+BUILT_SOURCES += $(MATHML_ENTITIES)
+
 CLEANFILES = $(MANUAL)
 
 DISTCLEANFILES = @AMINCLUDE@
@@ -312,6 +317,18 @@ patterndeps = $(_FORCED) $(_TEST_DEPS) $(_DOCS_DEPS) | $(bin_PROGRAMS) $(EXTRA_R
 	$(DOT) -Tpdf $< -o $@.gs
 	$(GS) -q -sDEVICE=pdfwrite -dCompatibilityLevel=1.5 -o $@ $@.gs
 
+XML_ENTITIES = .sources/unicode.xml
+XML_ENTITIES_COMMIT = 77acf14428202e4e1dba54ff1e5ed43fe5ab474f
+
+.sources:
+	[ -h .sources ] || mkdir -p $@
+
+$(XML_ENTITIES):
+	$(CURL) https://raw.githubusercontent.com/w3c/xml-entities/$(XML_ENTITIES_COMMIT)/unicode.xml -o $@
+
+$(MATHML_ENTITIES): build-aux/xml-entities-to-lua.xsl
+	$(XSLTPROC) $< $(XML_ENTITIES) | $(or $(STYLUA),cat) - > $@
+
 .PHONY: force
 force: ;
 

build-aux/pkg.nix

@@ -24,6 +24,7 @@
   icu,
   fontconfig,
   libiconv,
+  libxslt,
   stylua,
   taplo,
   typos,
@@ -125,6 +126,7 @@ stdenv.mkDerivation (finalAttrs: {
       icu
       fontconfig
       libiconv
+      libxslt
       stylua
       taplo
       typos

packages/math/tools/unicode-xml-to-sile.xsl → build-aux/xml-entities-to-lua.xsl

@@ -1,10 +1,4 @@
 <xsl:stylesheet version="1.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
-<!--
-  Stylesheet to convert the unicode.xml file to a SILE Lua file:
-  xsltproc unicode-xml-to-sile.xsl unicode.xml > ../packages/math/unicode-symbols-generated.lua
-  Where unicode.xml is:
-  https://raw.githubusercontent.com/w3c/xml-entities/gh-pages/unicode.xml
--->
 <xsl:output method="text" indent="no"/>
 
 <xsl:template name="format-value">
@@ -96,16 +90,16 @@
 --    Originally designed by Sebastian Rahtz in conjunction with Barbara Beeton for the STIX project
 --
 
-local atoms = require("packages/math/atoms")
+local atoms = require("packages.math.atoms")
 
 --- Transform a list of codepoints into a string
 local function U (...)
-  local t = { ... }
-  local str = ""
-  for i = 1, #t do
-    str = str .. luautf8.char(t[i])
-  end
-  return str
+   local t = { ... }
+   local str = ""
+   for i = 1, #t do
+      str = str .. luautf8.char(t[i])
+   end
+   return str
 end
 
 local symbols = {}
@@ -118,33 +112,33 @@ local operatorDict = {}
 -- @tparam string _         Unicode name of the symbol (informative)
 -- @tparam table  ops       List of operator forms and their properties
 local function addSymbol (str, shortatom, mathlatex, _, ops)
-  if mathlatex then
-    SU.debug("math.symbols", "Registering symbol", str, "as", mathlatex)
-    symbols[mathlatex] = str
-  end
-  local op = {}
-  op.atom = atoms.types[shortatom]
-  if ops then
-    op.forms = {}
-    for _, v in pairs(ops) do
-      if v.form then
-        -- NOTE: At this point the mu unit is not yet defined, so keep it as a string.
-        v.lspace = v.lspace and ("%smu"):format(v.lspace) or "0mu"
-        v.rspace = v.rspace and ("%smu"):format(v.rspace) or "0mu"
-        op.forms[v.form] = v
-      else
-        SU.warn("No form for operator " .. str .. " (operator dictionary is probably incomplete)")
+   if mathlatex then
+      SU.debug("math.symbols", "Registering symbol", str, "as", mathlatex)
+      symbols[mathlatex] = str
+   end
+   local op = {}
+   op.atom = atoms.types[shortatom]
+   if ops then
+      op.forms = {}
+      for _, v in pairs(ops) do
+         if v.form then
+            -- NOTE: At this point the mu unit is not yet defined, so keep it as a string.
+            v.lspace = v.lspace and ("%smu"):format(v.lspace) or "0mu"
+            v.rspace = v.rspace and ("%smu"):format(v.rspace) or "0mu"
+            op.forms[v.form] = v
+         else
+            SU.warn("No form for operator " .. str .. " (operator dictionary is probably incomplete)")
+         end
       end
-    end
-  end
-  operatorDict[str] = op
+   end
+   operatorDict[str] = op
 end
 
 <xsl:apply-templates select="charlist/character" />
 
 return {
-  operatorDict = operatorDict,
-  symbols = symbols
+   operatorDict = operatorDict,
+   symbols = symbols,
 }
 </xsl:template>
 
@@ -160,42 +154,46 @@ return {
     </xsl:call-template>
   </xsl:variable>
   <xsl:if test="$atom != 'ord' or $mathlatex or operator-dictionary">
-    <xsl:text>
-addSymbol(</xsl:text>
-    <!-- Codepoints -->
+    <xsl:text>addSymbol(</xsl:text>
+      <!-- Codepoints -->
     <xsl:call-template name="format-codepoint">
       <xsl:with-param name="codepoint" select="@id" />
     </xsl:call-template>
     <!-- Atom type -->
-    <xsl:text>, "</xsl:text><xsl:value-of select="$atom" /><xsl:text>", </xsl:text>
+    <xsl:text>,"</xsl:text><xsl:value-of select="$atom" /><xsl:text>",</xsl:text>
     <!-- Math latex name or nil -->
     <xsl:call-template name="format-mathlatex">
       <xsl:with-param name="mathlatex" select="$mathlatex" />
     </xsl:call-template>
     <!-- Description -->
-    <xsl:text>, "</xsl:text><xsl:value-of select="description" /><xsl:text>"</xsl:text>
+    <xsl:text>,"</xsl:text><xsl:value-of select="description" /><xsl:text>"</xsl:text>
     <!-- Operator dictionary or nil -->
     <xsl:choose>
       <xsl:when test="operator-dictionary">
-        <xsl:text>, {</xsl:text>
-          <xsl:apply-templates select="operator-dictionary">
-            <xsl:sort select="@priority" data-type="number" order="descending" /><!-- sort by @priority -->
-          </xsl:apply-templates>
+        <xsl:text>,{</xsl:text>
+        <xsl:apply-templates select="operator-dictionary">
+          <!-- sort by @priority -->
+          <xsl:sort select="@priority" data-type="number" order="descending" />
+        </xsl:apply-templates>
         <xsl:text>}</xsl:text>
       </xsl:when>
-      <xsl:otherwise><xsl:text>, nil</xsl:text></xsl:otherwise>
+      <xsl:otherwise><xsl:text>,nil</xsl:text></xsl:otherwise>
     </xsl:choose>
     <xsl:text>)</xsl:text>
   </xsl:if>
 </xsl:template>
 
 <xsl:template match="operator-dictionary">
-  { <xsl:for-each select="@*">
+  {
+  <xsl:for-each select="@*">
     <xsl:sort select="name()" />
-    <xsl:value-of select="name()" /> = <xsl:call-template name="format-value">
-      <xsl:with-param name="value" select="." />
-    </xsl:call-template><xsl:if test="position() != last()">, </xsl:if>
-  </xsl:for-each> }<xsl:if test="position() != last()">,</xsl:if>
+    <xsl:value-of select="name()" />
+    =
+      <xsl:call-template name="format-value">
+        <xsl:with-param name="value" select="." />
+      </xsl:call-template>,
+    </xsl:for-each>
+  },
 </xsl:template>
 
 </xsl:stylesheet>

configure.ac

@@ -175,8 +175,8 @@ AM_CONDITIONAL([ICU], [test "x$with_icu" = "xyes"])
 # Required for downloading fonts for the manual and for tests
 # Since the source tarball includes a prebuilt manual we only need this for Git source builds
 AM_COND_IF([FONT_DOWNLOAD_TOOLS], [
-    QUE_PROGVAR([curl])
     QUE_PROGVAR([bsdtar])
+    QUE_PROGVAR([curl])
 ])
 
 AM_COND_IF([DEVELOPER_MODE], [
@@ -197,6 +197,7 @@ AM_COND_IF([DEVELOPER_TOOLS], [
     QUE_PROGVAR([taplo])
     QUE_PROGVAR([tr])
     QUE_PROGVAR([typos])
+    QUE_PROGVAR([xsltproc])
 ])
 
 AX_PROG_LUA([5.1], [], [], [], [prefer])

packages/math/unicode-symbols.lua

@@ -1,7 +1,7 @@
 --- Extend the unicode-symbols-generated symbols with additional aliases.
-local gen = require("packages/math/unicode-symbols-generated")
-local symbols = gen.symbols
-local operatorDict = gen.operatorDict
+local mathml_entities = require("packages.math.mathml-entities")
+local symbols = mathml_entities.symbols
+local operatorDict = mathml_entities.operatorDict
 
 --- Add aliases for symbols that have multiple names.
 -- We check that the alias is already defined in the generated dictionary,