subscripts in WL XML

[librarian.git] / librarian / epub.py

diff --git a/librarian/epub.py b/librarian/epub.py
index 1ea2688..3e9056c 100644 (file)
--- a/librarian/epub.py
+++ b/librarian/epub.py
@@ -11,6 +11,8 @@ import re
 import subprocess
 from StringIO import StringIO
 from copy import deepcopy
+from mimetypes import guess_type
+
 from lxml import etree
 import zipfile
 from tempfile import mkdtemp, NamedTemporaryFile
@@ -51,19 +53,21 @@ def set_hyph_language(source_tree):
 
 
 def hyphenate_and_fix_conjunctions(source_tree, hyph):
-    if hyph is not None:
-        texts = etree.XPath('/utwor/*[2]//text()')(source_tree)
-        for t in texts:
-            parent = t.getparent()
+    texts = etree.XPath('/utwor/*[2]//text()')(source_tree)
+    for t in texts:
+        parent = t.getparent()
+        if hyph is not None:
             newt = ''
             wlist = re.compile(r'\w+|[^\w]', re.UNICODE).findall(t)
             for w in wlist:
                 newt += hyph.inserted(w, u'\u00AD')
-            newt = re.sub(r'(?<=\s\w)\s+', u'\u00A0', newt)
-            if t.is_text:
-                parent.text = newt
-            elif t.is_tail:
-                parent.tail = newt
+        else:
+            newt = t
+        newt = re.sub(r'(?<=\s\w)\s+', u'\u00A0', newt)
+        if t.is_text:
+            parent.text = newt
+        elif t.is_tail:
+            parent.tail = newt
 
 
 def inner_xml(node):
@@ -109,11 +113,13 @@ def node_name(node):
     return tempnode.text
 
 
-def xslt(xml, sheet):
+def xslt(xml, sheet, **kwargs):
     if isinstance(xml, etree._Element):
         xml = etree.ElementTree(xml)
     with open(sheet) as xsltf:
-        return xml.xslt(etree.parse(xsltf))
+        transform = etree.XSLT(etree.parse(xsltf))
+        params = dict((key, transform.strparam(value)) for key, value in kwargs.iteritems())
+        return transform(xml, **params)
 
 
 def replace_characters(node):
@@ -404,9 +410,8 @@ def transform_chunk(chunk_xml, chunk_no, annotations, empty=False, _empty_html_s
     return output_html, toc, chars
 
 
-def transform(wldoc, verbose=False,
-              style=None, html_toc=False,
-              sample=None, cover=None, flags=None):
+def transform(wldoc, verbose=False, style=None, html_toc=False,
+              sample=None, cover=None, flags=None, hyphenate=False, ilustr_path='', output_type='epub'):
     """ produces a EPUB file
 
     sample=n: generate sample e-book (with at least n paragraphs)
@@ -419,7 +424,7 @@ def transform(wldoc, verbose=False,
 
         replace_characters(wldoc.edoc.getroot())
 
-        hyphenator = set_hyph_language(wldoc.edoc.getroot())
+        hyphenator = set_hyph_language(wldoc.edoc.getroot()) if hyphenate else None
         hyphenate_and_fix_conjunctions(wldoc.edoc.getroot(), hyphenator)
 
         # every input file will have a TOC entry,
@@ -428,7 +433,7 @@ def transform(wldoc, verbose=False,
         chars = set()
         if first:
             # write book title page
-            html_tree = xslt(wldoc.edoc, get_resource('epub/xsltTitle.xsl'))
+            html_tree = xslt(wldoc.edoc, get_resource('epub/xsltTitle.xsl'), outputtype=output_type)
             chars = used_chars(html_tree.getroot())
             zip.writestr(
                 'OPS/title.html',
@@ -501,6 +506,9 @@ def transform(wldoc, verbose=False,
         for flag in flags:
             document.edoc.getroot().set(flag, 'yes')
 
+    document.clean_ed_note()
+    document.clean_ed_note('abstrakt')
+
     # add editors info
     editors = document.editors()
     if editors:
@@ -520,6 +528,16 @@ def transform(wldoc, verbose=False,
     output_file = NamedTemporaryFile(prefix='librarian', suffix='.epub', delete=False)
     zip = zipfile.ZipFile(output_file, 'w', zipfile.ZIP_DEFLATED)
 
+    functions.reg_mathml_epub(zip)
+
+    if os.path.isdir(ilustr_path):
+        for i, filename in enumerate(os.listdir(ilustr_path)):
+            file_path = os.path.join(ilustr_path, filename)
+            zip.write(file_path, os.path.join('OPS', filename))
+            image_id = 'image%s' % i
+            manifest.append(etree.fromstring(
+                '<item id="%s" href="%s" media-type="%s" />' % (image_id, filename, guess_type(file_path)[0])))
+
     # write static elements
     mime = zipfile.ZipInfo()
     mime.filename = 'mimetype'
@@ -632,7 +650,7 @@ def transform(wldoc, verbose=False,
         '<item id="last" href="last.html" media-type="application/xhtml+xml" />'))
     spine.append(etree.fromstring(
         '<itemref idref="last" />'))
-    html_tree = xslt(document.edoc, get_resource('epub/xsltLast.xsl'))
+    html_tree = xslt(document.edoc, get_resource('epub/xsltLast.xsl'), outputtype=output_type)
     chars.update(used_chars(html_tree.getroot()))
     zip.writestr('OPS/last.html', etree.tostring(
         html_tree, pretty_print=True, xml_declaration=True,
@@ -641,7 +659,7 @@ def transform(wldoc, verbose=False,
                 '"http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">'
     ))
 
-    if not flags or not 'without-fonts' in flags:
+    if not flags or 'without-fonts' not in flags:
         # strip fonts
         tmpdir = mkdtemp('-librarian-epub')
         try:
@@ -659,7 +677,8 @@ def transform(wldoc, verbose=False,
                 print "Running font-optimizer"
                 subprocess.check_call(optimizer_call)
             else:
-                subprocess.check_call(optimizer_call, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
+                dev_null = open(os.devnull, 'w')
+                subprocess.check_call(optimizer_call, stdout=dev_null, stderr=dev_null)
             zip.write(os.path.join(tmpdir, fname), os.path.join('OPS', fname))
             manifest.append(etree.fromstring(
                 '<item id="%s" href="%s" media-type="application/x-font-truetype" />' % (fname, fname)))