X-Git-Url: https://git.mdrn.pl/librarian.git/blobdiff_plain/98a3fbdcedc1b8c4da602b8bb8ee304830a6de1b..076474d2a928649c934b4bb12031d39cdea2c77d:/librarian/pyhtml.py

diff --git a/librarian/pyhtml.py b/librarian/pyhtml.py
index f41c583..6db1701 100644
--- a/librarian/pyhtml.py
+++ b/librarian/pyhtml.py
@@ -52,11 +52,14 @@ class EduModule(Xmill):
     handle_tytul_dziela = tag('em', 'title')
     handle_slowo_obce = tag('em', 'foreign')
 
+    def naglowek_to_anchor(self, naglowek):
+        return re.sub(r" +", " ", naglowek.text.strip())
+
     def handle_nazwa_utworu(self, element):
         toc = []
         for naglowek in element.getparent().findall('.//naglowek_rozdzial'):
             a = etree.Element("a")
-            a.attrib["href"] = "#" + naglowek.text
+            a.attrib["href"] = "#" + self.naglowek_to_anchor(naglowek)
             a.text = naglowek.text
             atxt = etree.tostring(a, encoding=unicode)
             toc.append("<li>%s</li>" % atxt)
@@ -66,7 +69,7 @@ class EduModule(Xmill):
 
     @tagged("h2")
     def handle_naglowek_rozdzial(self, element):
-        return "", "".join(tag_open_close("a", name=element.text))
+        return "", "".join(tag_open_close("a", name=self.naglowek_to_anchor(element)))
 
     def handle_uwaga(self, _e):
         return None
@@ -79,7 +82,10 @@ class EduModule(Xmill):
             }
         submill = EduModule(dict(self.options.items() + {'sub_gen': True}.items()))
 
-        opis = submill.generate(element.xpath('opis')[0])
+        if element.xpath('opis'):
+            opis = submill.generate(element.xpath('opis')[0])
+        else:
+            opis = ''
 
         n = element.xpath('wskazowki')
         if n: wskazowki = submill.generate(n[0])
@@ -138,7 +144,6 @@ u"""%(wskazowki)s
 
         typ = element.attrib['typ']
         self.exercise_counter += 1
-        print self.exercise_counter
         self.options = {'exercise_counter': self.exercise_counter}
         handler = exercise_handlers[typ](self.options)
         return handler.generate(element)
@@ -149,7 +154,7 @@ u"""%(wskazowki)s
         if ltype == 'slowniczek':
             surl = element.attrib.get('src', None)
             if surl is None:
-                print '** missing src on <slowniczek>, setting default'
+                # print '** missing src on <slowniczek>, setting default'
                 surl = 'http://edukacjamedialna.edu.pl/slowniczek'
             sxml = None
             if surl:
@@ -225,8 +230,9 @@ u"""%(wskazowki)s
             return tag('a', href=element.attrib['url'])(self, element)
         elif 'material' in element.attrib:
             material_err = u' [BRAKUJĄCY MATERIAŁ]'
+            slug = element.attrib['material']
             make_url = lambda f: self.options['urlmapper'] \
-              .url_for_material(element.attrib['material'], f)
+              .url_for_material(slug, f)
 
             if 'format' in element.attrib:
                 formats = re.split(r"[, ]+",
@@ -234,18 +240,20 @@ u"""%(wskazowki)s
             else:
                 formats = [None]
 
+            formats = self.options['urlmapper'].materials(slug)
+
             try:
-                def_href = make_url(formats[0])
+                def_href = make_url(formats[0][0])
                 def_err = u""
-            except self.options['urlmapper'].MaterialNotFound:
+            except (IndexError, self.options['urlmapper'].MaterialNotFound):
                 def_err = material_err
                 def_href = u""
             fmt_links = []
             for f in formats[1:]:
                 try:
-                    fmt_links.append(u'<a href="%s">%s</a>' % (make_url(f), f.upper()))
+                    fmt_links.append(u'<a href="%s">%s</a>' % (make_url(f[0]), f[0].upper()))
                 except self.options['urlmapper'].MaterialNotFound:
-                    fmt_links.append(u'<a>%s%s</a>' % (f.upper(), material_err))
+                    fmt_links.append(u'<a>%s%s</a>' % (f[0].upper(), material_err))
             more_links = u' (%s)' % u', '.join(fmt_links) if fmt_links else u''
 
             return u"<a href='%s'>" % def_href, u'%s</a>%s' % (def_err, more_links)
@@ -492,7 +500,7 @@ class PrawdaFalsz(Exercise):
 
 
 class EduModuleFormat(Format):
-    DEFAULT_MATERIAL_FORMAT = 'odt'
+    PRIMARY_MATERIAL_FORMATS = ('pdf', 'odt')
 
     class MaterialNotFound(BaseException):
         pass
@@ -501,19 +509,33 @@ class EduModuleFormat(Format):
         super(EduModuleFormat, self).__init__(wldoc, **kwargs)
 
     def build(self):
+        # Sort materials by slug.
+        self.materials_by_slug = {}
+        for name, att in self.wldoc.source.attachments.items():
+            parts = name.rsplit('.', 1)
+            if len(parts) == 1:
+                continue
+            slug, ext = parts
+            if slug not in self.materials_by_slug:
+                self.materials_by_slug[slug] = {}
+            self.materials_by_slug[slug][ext] = att
+
         edumod = EduModule({'provider': self.wldoc.provider, 'urlmapper': self, 'wldoc': self.wldoc})
 
         html = edumod.generate(self.wldoc.edoc.getroot())
 
         return IOFile.from_string(html.encode('utf-8'))
 
-    def url_for_material(self, slug, fmt=None):
-        if fmt is None:
-            fmt = self.DEFAULT_MATERIAL_FORMAT
-        # No briliant idea for an API here.
-        if fmt:
-            return "%s.%s" % (slug, fmt)
-        return slug
+    def materials(self, slug):
+        """Returns a list of pairs: (ext, iofile)."""
+        order = dict(reversed(k) for k in enumerate(self.PRIMARY_MATERIAL_FORMATS))
+        mats = self.materials_by_slug.get(slug, {}).items()
+        if not mats:
+            print "!! Material missing: '%s'" % slug
+        return sorted(mats, key=lambda (x, y): order.get(x, x))
+
+    def url_for_material(self, slug, fmt):
+        return "%s.%s" % (slug, fmt)
 
 
 def transform(wldoc, stylesheet='edumed', options=None, flags=None):