arta tech

[librarian.git] / librarian / epub.py

diff --git a/librarian/epub.py b/librarian/epub.py
index a87e1d8..348df0c 100644 (file)
--- a/librarian/epub.py
+++ b/librarian/epub.py
@@ -5,6 +5,7 @@
 #
 from __future__ import with_statement
 
 #
 from __future__ import with_statement
 

+from copy import deepcopy

 import os
 import os.path
 import subprocess
 import os
 import os.path
 import subprocess


@@ -17,8 +18,9 @@ from shutil import rmtree
 
 import sys
 
 
 import sys
 

-from librarian import XMLNamespace, RDFNS, DCNS, WLNS, NCXNS, OPFNS, NoDublinCore
+from librarian import XMLNamespace, RDFNS, DCNS, WLNS, NCXNS, OPFNS, XHTMLNS, NoDublinCore

 from librarian.dcparser import BookInfo
 from librarian.dcparser import BookInfo

+from librarian.cover import ImageCover

 
 from librarian import functions, get_resource
 
 
 from librarian import functions, get_resource
 


@@ -84,13 +86,14 @@ def replace_characters(node):
                    .replace(",,", u"\u201E")\
                    .replace('"', u"\u201D")\
                    .replace("'", u"\u2019")
                    .replace(",,", u"\u201E")\
                    .replace('"', u"\u201D")\
                    .replace("'", u"\u2019")

-    if node.tag == 'extra':
+    if node.tag in ('uwaga', 'extra'):
+        t = node.tail

         node.clear()
         node.clear()

-    else:
-        node.text = replace_chars(node.text)
-        node.tail = replace_chars(node.tail)
-        for child in node:
-            replace_characters(child)
+        node.tail = t
+    node.text = replace_chars(node.text)
+    node.tail = replace_chars(node.tail)
+    for child in node:
+        replace_characters(child)

 
 
 def find_annotations(annotations, source, part_no):
 
 
 def find_annotations(annotations, source, part_no):


@@ -106,7 +109,7 @@ def find_annotations(annotations, source, part_no):
             child.clear()
             child.tail = tail
             child.text = number
             child.clear()
             child.tail = tail
             child.text = number

-        if child.tag not in ('extra',):
+        if child.tag not in ('extra', 'uwaga'):

             find_annotations(annotations, child, part_no)
 
 
             find_annotations(annotations, child, part_no)
 
 


@@ -266,7 +269,7 @@ def transform_chunk(chunk_xml, chunk_no, annotations, empty=False, _empty_html_s
 
 
 def transform(provider, slug=None, file_path=None, output_file=None, output_dir=None, make_dir=False, verbose=False,
 
 
 def transform(provider, slug=None, file_path=None, output_file=None, output_dir=None, make_dir=False, verbose=False,

-              sample=None, cover_fn=None, flags=None):
+              sample=None, cover=None, flags=None):

     """ produces a EPUB file
 
     provider: a DocProvider
     """ produces a EPUB file
 
     provider: a DocProvider


@@ -275,8 +278,8 @@ def transform(provider, slug=None, file_path=None, output_file=None, output_dir=
     output_dir: path to directory to save output file to; either this or output_file must be present
     make_dir: writes output to <output_dir>/<author>/<slug>.epub instead of <output_dir>/<slug>.epub
     sample=n: generate sample e-book (with at least n paragraphs)
     output_dir: path to directory to save output file to; either this or output_file must be present
     make_dir: writes output to <output_dir>/<author>/<slug>.epub instead of <output_dir>/<slug>.epub
     sample=n: generate sample e-book (with at least n paragraphs)

-    cover_fn: function(author, title) -> cover image
-    flags: less-advertising,
+    cover: a cover.Cover object
+    flags: less-advertising, images, not-wl

     """
 
     def transform_file(input_xml, chunk_counter=1, first=True, sample=None):
     """
 
     def transform_file(input_xml, chunk_counter=1, first=True, sample=None):


@@ -381,6 +384,10 @@ def transform(provider, slug=None, file_path=None, output_file=None, output_dir=
         else:
             output_file = open(os.path.join(output_dir, os.path.splitext(os.path.basename(file_path))[0] + '.epub'), 'w')
 
         else:
             output_file = open(os.path.join(output_dir, os.path.splitext(os.path.basename(file_path))[0] + '.epub'), 'w')
 

+    opf = xslt(metadata, get_resource('epub/xsltContent.xsl'))
+    manifest = opf.find('.//' + OPFNS('manifest'))
+    spine = opf.find('.//' + OPFNS('spine'))
+

     zip = zipfile.ZipFile(output_file, 'w', zipfile.ZIP_DEFLATED)
 
     # write static elements
     zip = zipfile.ZipFile(output_file, 'w', zipfile.ZIP_DEFLATED)
 
     # write static elements


@@ -395,26 +402,52 @@ def transform(provider, slug=None, file_path=None, output_file=None, output_dir=
                        'media-type="application/oebps-package+xml" />' \
                        '</rootfiles></container>')
     zip.write(get_resource('epub/style.css'), os.path.join('OPS', 'style.css'))
                        'media-type="application/oebps-package+xml" />' \
                        '</rootfiles></container>')
     zip.write(get_resource('epub/style.css'), os.path.join('OPS', 'style.css'))

-    zip.write(get_resource('res/wl-logo-small.png'), os.path.join('OPS', 'logo_wolnelektury.png'))
-
-    opf = xslt(metadata, get_resource('epub/xsltContent.xsl'))
-    manifest = opf.find('.//' + OPFNS('manifest'))
-    spine = opf.find('.//' + OPFNS('spine'))
+    if not flags or 'not-wl' not in flags:
+        manifest.append(etree.fromstring(
+            '<item id="logo_wolnelektury" href="logo_wolnelektury.png" media-type="image/png" />'))
+        zip.write(get_resource('res/wl-logo-small.png'), os.path.join('OPS', 'logo_wolnelektury.png'))
+
+    if cover:
+        cover_file = StringIO()
+        c = cover(book_info.author.readable(), book_info.title)
+        c.save(cover_file)
+        c_name = 'cover.%s' % c.ext()
+        zip.writestr(os.path.join('OPS', c_name), cover_file.getvalue())
+        del cover_file
+
+        cover_tree = etree.parse(get_resource('epub/cover.html'))
+        cover_tree.find('//' + XHTMLNS('img')).set('src', c_name)
+        zip.writestr('OPS/cover.html', etree.tostring(
+                        cover_tree, method="html", pretty_print=True))

 
 

-    if cover_fn:
-        cover = StringIO()
-        cover_fn(book_info.author.readable(), book_info.title).save(cover, format='JPEG')
-        zip.writestr(os.path.join('OPS', 'cover.jpg'), cover.getvalue())
-        del cover
-        zip.writestr('OPS/cover.html', open(get_resource('epub/cover.html')).read())

         manifest.append(etree.fromstring(
             '<item id="cover" href="cover.html" media-type="application/xhtml+xml" />'))
         manifest.append(etree.fromstring(
         manifest.append(etree.fromstring(
             '<item id="cover" href="cover.html" media-type="application/xhtml+xml" />'))
         manifest.append(etree.fromstring(

-            '<item id="cover-image" href="cover.jpg" media-type="image/jpeg" />'))
+            '<item id="cover-image" href="%s" media-type="%s" />' % (c_name, c.mime_type())))

         spine.insert(0, etree.fromstring('<itemref idref="cover" />'))
         opf.getroot()[0].append(etree.fromstring('<meta name="cover" content="cover-image"/>'))
         opf.getroot().append(etree.fromstring('<guide><reference href="cover.html" type="cover" title="Okładka"/></guide>'))
 
         spine.insert(0, etree.fromstring('<itemref idref="cover" />'))
         opf.getroot()[0].append(etree.fromstring('<meta name="cover" content="cover-image"/>'))
         opf.getroot().append(etree.fromstring('<guide><reference href="cover.html" type="cover" title="Okładka"/></guide>'))
 

+    if flags and 'images' in flags:
+        for ilustr in input_xml.findall('//ilustr'):
+            src = ilustr.get('src')
+            mime = ImageCover(src)().mime_type()
+            zip.write(src, os.path.join('OPS', src))
+            manifest.append(etree.fromstring(
+                '<item id="%s" href="%s" media-type="%s" />' % (src, src, mime)))
+            # get it up to master
+            after = ilustr
+            while after.getparent().tag not in ['powiesc', 'opowiadanie', 'liryka_l', 'liryka_lp', 'dramat_wierszowany_l', 'dramat_wierszowany_lp', 'dramat_wspolczesny']:
+                after = after.getparent()
+            if not(after is ilustr):
+                moved = deepcopy(ilustr)
+                ilustr.tag = 'extra'
+                ilustr.text = None
+                moved.tail = None
+                after.addnext(moved)
+    else:
+        for ilustr in input_xml.findall('//ilustr'):
+            ilustr.tag = 'extra'

 
     annotations = etree.Element('annotations')
 
 
     annotations = etree.Element('annotations')
 


@@ -456,7 +489,13 @@ def transform(provider, slug=None, file_path=None, output_file=None, output_dir=
         '<item id="last" href="last.html" media-type="application/xhtml+xml" />'))
     spine.append(etree.fromstring(
         '<itemref idref="last" />'))
         '<item id="last" href="last.html" media-type="application/xhtml+xml" />'))
     spine.append(etree.fromstring(
         '<itemref idref="last" />'))

-    html_tree = xslt(input_xml, get_resource('epub/xsltLast.xsl'))
+    stopka = input_xml.find('//stopka')
+    if stopka is not None:
+        stopka.tag = 'stopka_'
+        replace_by_verse(stopka)
+        html_tree = xslt(stopka, get_resource('epub/xsltScheme.xsl'))
+    else:
+        html_tree = xslt(input_xml, get_resource('epub/xsltLast.xsl'))

     chars.update(used_chars(html_tree.getroot()))
     zip.writestr('OPS/last.html', etree.tostring(
                         html_tree, method="html", pretty_print=True))
     chars.update(used_chars(html_tree.getroot()))
     zip.writestr('OPS/last.html', etree.tostring(
                         html_tree, method="html", pretty_print=True))