Improve XMLittré formatting

diff --git a/XMLittre-entree.xslt b/XMLittre-entree.xslt
index 97c1b1e42a48adc313fe1c0132c47f972e272bc9..b9cb9a744417ad466e40443f9a449ba54255e64b 100644 (file)
--- a/XMLittre-entree.xslt
+++ b/XMLittre-entree.xslt
@@ -1,8 +1,10 @@
 <?xml version="1.0" encoding="utf-8"?>
 <xsl:stylesheet version="1.0"
-               xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
+               xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
+               xmlns:my="xmlittre"
+               extension-element-prefixes="my">
   <xsl:output method="xml" version="1.0" encoding="UTF-8"
-             indent="yes" standalone="no" omit-xml-declaration="yes"/>
+             indent="no" standalone="no" omit-xml-declaration="yes"/>
 
   <xsl:template match="@*|node()">
     <xsl:copy>
@@ -10,7 +12,7 @@
     </xsl:copy>
   </xsl:template>
 
-  <xsl:template match="entree">
+  <xsl:template match="entree"><div>
     <xsl:for-each select="entete/nature | entete/indent | entete/cit | entete/a | entete/semantique">
       <xsl:apply-templates select="."/>
     </xsl:for-each>
@@ -49,17 +51,16 @@
 
     <xsl:if test="entete/prononciation">
       <p><b>PRONONCIATION</b></p>
-      <div>
-       <xsl:apply-templates select="entete/prononciation/node()" />
-      </div>
+      <p><xsl:apply-templates select="entete/prononciation/node()" /></p>
     </xsl:if>
-  </xsl:template>
+  </div></xsl:template>
 
   <xsl:template match="nature">
     <i><xsl:value-of select="text()"/></i>
   </xsl:template>
+
   <xsl:template match="indent">
-    <div style="padding-left: 1em"><xsl:apply-templates select="node()"/></div>
+    <div style="padding-left: 2em"><xsl:apply-templates select="node()"/></div>
   </xsl:template>
 
   <xsl:template match="corps|résumé" mode="plusieurs_variantes">
@@ -71,7 +72,7 @@
   </xsl:template>
 
   <xsl:template match="corps|résumé" mode="unique_variante">
-    <div><xsl:apply-templates select="node()"/></div>
+    <div><xsl:apply-templates select="variante/node()"/></div>
   </xsl:template>
 
   <xsl:template match="cit">
@@ -96,4 +97,6 @@
     <xsl:apply-templates select="node()"/>
   </xsl:template>
 
+  <xsl:template match="a"><my:massagelink/></xsl:template>
+
 </xsl:stylesheet>

diff --git a/XMLittre-stoplist.txt b/XMLittre-stoplist.txt
index 10082219304ac937ca3561835a9f2c4ae1567555..159bedddf3cc9f4bae88ae4d455cd1f3ae0b8d0c 100644 (file)
--- a/XMLittre-stoplist.txt
+++ b/XMLittre-stoplist.txt
@@ -43,3 +43,4 @@ ecte
 uite
 une
 simplement
+a

diff --git a/XMLittre_to_tab_separated b/XMLittre_to_tab_separated
index 0c041d99e76727ce033511953eae791458a13cb3..71a42bab4465ca43141df3b347b8bdc8a2494603 100755 (executable)
--- a/XMLittre_to_tab_separated
+++ b/XMLittre_to_tab_separated
@@ -2,8 +2,25 @@
 
 from glob import glob
 from lxml import etree
+import re
+from copy import deepcopy
 
-xslt_entree = etree.XSLT(etree.parse("XMLittre-entree.xslt"))
+class MassageLink(etree.XSLTExtension):
+    def execute(self, context, self_node, input_node, output_parent):
+        link = etree.Element("a")
+
+        ref = input_node.attrib["ref"]
+        term = re.match("^[^#. ]+", ref).group(0)
+        if input_node.text == "ce mot":
+            link.text = term.lower()
+        else:
+            link.text = input_node.text.lower()
+        link.attrib["style"] = "font-variant: small-caps"
+        link.attrib["href"] = f"q://d?&{term.upper()}"
+        output_parent.append(link)
+
+xslt_entree = etree.XSLT(etree.parse("XMLittre-entree.xslt"),
+                         extensions={("xmlittre", "massagelink"): MassageLink()})
 
 class Parser:
     def __init__(self):
@@ -15,32 +32,82 @@ class Parser:
         for entree in root.getchildren():
             assert entree.tag == "entree"
             terme = entree.attrib["terme"]
+            termes, sous_titre = massage_terme(terme)
+
             entree_html = xslt_entree(entree)
             entree_text = str(entree_html)
-            entree_text = entree_text.replace("\n", "")
-            if terme not in self.entrees:
-                self.entrees[terme] = []
-            self.entrees[terme].append(entree_text)
+            # entree_text = entree_text.replace("\n", "")
+            entree_text = re.sub(r"\s+|\n", " ", entree_text)
+            assert entree_text.startswith("<div>") and entree_text.endswith("</div>")
+            entree_text = entree_text[5:]
+            entree_text = entree_text[:-6]
+
+            if termes not in self.entrees:
+                self.entrees[termes] = []
+
+            self.entrees[termes].append((sous_titre, entree_text))
 
     def writeout(self, fname):
         with open(fname, "w") as f:
-            for terme in self.entrees:
-                f.write(terme)
+            for termes in self.entrees:
+                f.write(", ".join(termes))
                 f.write("\t")
-                if len(self.entrees[terme]) > 1:
-                    f.write("<ol>")
-                    for entree in self.entrees[terme]:
-                        f.write("<li>")
+                if len(self.entrees[termes]) > 1:
+                    for i, (sous_titre, entree) in enumerate(self.entrees[termes]):
+                        if sous_titre:
+                            f.write(f"<h2>{i+1}. {sous_titre}</h2>")
+                        else:
+                            f.write(f"<h2>{i+1}. {','.join(termes)}</h2>")
                         f.write(entree)
-                        f.write("</li>")
-                    f.write("</ol>")
                 else:
-                    f.write(self.entrees[terme][0])
+                    sous_titre, entree = self.entrees[termes][0]
+                    if sous_titre:
+                        f.write(f"<h2>{sous_titre}</h2>")
+                    f.write(entree)
                 f.write("\n")
 
+def massage_terme(terme):
+    vedettes = re.split(r"[ ,](?:[A-Z]?[a-zéêèàç-]+[ ,']*)+", terme)
+    vedettes = list(filter(None, vedettes))
+    sous_terme = None
+    if len(vedettes) > 1:
+        sous_terme = terme
+    retval = []
+    for v in vedettes:
+        m = re.match(r"^(.*) \((.*)\)$", v)
+        if m:
+            if m.group(2)[-1] == "'":
+                space = ""
+            else:
+                space = " "
+            v = f"{m.group(2)}{space}{m.group(1)}"
+            if sous_terme:
+                sous_terme = sous_terme.replace(m.group(0), v)
+        formes = v.split(",")
+        if len(formes) > 1:
+            v = formes[0]
+            sous_terme = terme
+        retval.append(v)
+    return (tuple(retval), sous_terme)
+
+def test_massage_terme():
+    assert massage_terme("ÉTANCHE") == (("ÉTANCHE",), None)
+    assert massage_terme("ÉREINTÉ, ÉE") == (("ÉREINTÉ",), "ÉREINTÉ, ÉE")
+    assert massage_terme("CLÉ ou CLEF") == (("CLÉ", "CLEF"), "CLÉ ou CLEF")
+    assert massage_terme("HÉMORRAGIE, suivant le dictionnaire de l'Académie, mais mieux HÉMORRHAGIE") == \
+        (("HÉMORRAGIE", "HÉMORRHAGIE"), "HÉMORRAGIE, suivant le dictionnaire de l'Académie, mais mieux HÉMORRHAGIE")
+    assert massage_terme("QUINDÉCEMVIRS ou mieux QUINDÉCIMVIRS puisque le latin est quindecimviri") \
+        == (("QUINDÉCEMVIRS", "QUINDÉCIMVIRS"), "QUINDÉCEMVIRS ou mieux QUINDÉCIMVIRS puisque le latin est quindecimviri")
+    assert massage_terme("DÉVOUEMENT, ou comme quelques-uns écrivent, dit l'Académie, DÉVOÛMENT") == (("DÉVOUEMENT", "DÉVOÛMENT"), "DÉVOUEMENT, ou comme quelques-uns écrivent, dit l'Académie, DÉVOÛMENT")
+    assert massage_terme("HÉMA- ou HÉMO- ou HÉMATO-") == (("HÉMA-", "HÉMO-", "HÉMATO-"), "HÉMA- ou HÉMO- ou HÉMATO-")
+    assert massage_terme("DIVINIS (A)") == (("A DIVINIS",), None)
+    assert massage_terme("BORDEAUX (VIN DE) ou vulgairement et en termes de commerce BORDEAUX") == \
+        (("VIN DE BORDEAUX", "BORDEAUX"), "VIN DE BORDEAUX ou vulgairement et en termes de commerce BORDEAUX")
+
 def main():
     p = Parser()
-    for fname in glob("../xmlittre-data/?.xml"):
+    for fname in sorted(glob("../xmlittre-data/?.xml")):
+        print(f"Running on {fname}")
         p.parse_file(fname)
     p.writeout("XMLittre.tab_separated")