Fonts used in reports
[librarian.git] / librarian / epub.py
index bbeb3d7..de136ea 100644 (file)
@@ -7,6 +7,7 @@ from __future__ import with_statement
 
 import os
 import os.path
 
 import os
 import os.path
+import re
 import subprocess
 from StringIO import StringIO
 from copy import deepcopy
 import subprocess
 from StringIO import StringIO
 from copy import deepcopy
@@ -14,10 +15,11 @@ from lxml import etree
 import zipfile
 from tempfile import mkdtemp, NamedTemporaryFile
 from shutil import rmtree
 import zipfile
 from tempfile import mkdtemp, NamedTemporaryFile
 from shutil import rmtree
+from mimetypes import guess_type
 
 from librarian import RDFNS, WLNS, NCXNS, OPFNS, XHTMLNS, OutputFile
 
 from librarian import RDFNS, WLNS, NCXNS, OPFNS, XHTMLNS, OutputFile
-from librarian.cover import WLCover
-
+from librarian.cover import WLCover, FutureOfCopyrightCover
+from librarian.latex import LatexFragment
 from librarian import functions, get_resource
 
 functions.reg_person_name()
 from librarian import functions, get_resource
 
 functions.reg_person_name()
@@ -109,31 +111,74 @@ def find_annotations(annotations, source, part_no):
             find_annotations(annotations, child, part_no)
 
 
             find_annotations(annotations, child, part_no)
 
 
+class Stanza(object):
+    """
+    Converts / verse endings into verse elements in a stanza.
+
+    Slashes may only occur directly in the stanza. Any slashes in subelements
+    will be ignored, and the subelements will be put inside verse elements.
+
+    >>> s = etree.fromstring("<strofa>a <b>c</b> <b>c</b>/\\nb<x>x/\\ny</x>c/ \\nd</strofa>")
+    >>> Stanza(s).versify()
+    >>> print etree.tostring(s)
+    <strofa><wers_normalny>a <b>c</b> <b>c</b></wers_normalny><wers_normalny>b<x>x/
+    y</x>c</wers_normalny><wers_normalny>d</wers_normalny></strofa>
+    
+    """
+    def __init__(self, stanza_elem):
+        self.stanza = stanza_elem
+        self.verses = []
+        self.open_verse = None
+
+    def versify(self):
+        self.push_text(self.stanza.text)
+        for elem in self.stanza:
+            self.push_elem(elem)
+            self.push_text(elem.tail)
+        tail = self.stanza.tail
+        self.stanza.clear()
+        self.stanza.tail = tail
+        self.stanza.extend(self.verses)
+
+    def open_normal_verse(self):
+        self.open_verse = self.stanza.makeelement("wers_normalny")
+        self.verses.append(self.open_verse)
+
+    def get_open_verse(self):
+        if self.open_verse is None:
+            self.open_normal_verse()
+        return self.open_verse
+
+    def push_text(self, text):
+        if not text:
+            return
+        for i, verse_text in enumerate(re.split(r"/\s*\n", text)):
+            if i:
+                self.open_normal_verse()
+            verse = self.get_open_verse()
+            if len(verse):
+                verse[-1].tail = (verse[-1].tail or "") + verse_text
+            else:
+                verse.text = (verse.text or "") + verse_text
+
+    def push_elem(self, elem):
+        if elem.tag.startswith("wers"):
+            verse = deepcopy(elem)
+            verse.tail = None
+            self.verses.append(verse)
+            self.open_verse = verse
+        else:
+            appended = deepcopy(elem)
+            appended.tail = None
+            self.get_open_verse().append(appended)
+
+
 def replace_by_verse(tree):
     """ Find stanzas and create new verses in place of a '/' character """
 
     stanzas = tree.findall('.//' + WLNS('strofa'))
 def replace_by_verse(tree):
     """ Find stanzas and create new verses in place of a '/' character """
 
     stanzas = tree.findall('.//' + WLNS('strofa'))
-    for node in stanzas:
-        for child_node in node:
-            if child_node.tag in ('slowo_obce', 'wyroznienie'):
-                foreign_verses = inner_xml(child_node).split('/\n')
-                if len(foreign_verses) > 1:
-                    new_foreign = ''
-                    for foreign_verse in foreign_verses:
-                        if foreign_verse.startswith('<wers'):
-                            new_foreign += foreign_verse
-                        else:
-                            new_foreign += ''.join(('<wers_normalny>', foreign_verse, '</wers_normalny>'))
-                    set_inner_xml(child_node, new_foreign)
-        verses = inner_xml(node).split('/\n')
-        if len(verses) > 1:
-            modified_inner_xml = ''
-            for verse in verses:
-                if verse.startswith('<wers') or verse.startswith('<extra'):
-                    modified_inner_xml += verse
-                else:
-                    modified_inner_xml += ''.join(('<wers_normalny>', verse, '</wers_normalny>'))
-            set_inner_xml(node, modified_inner_xml)
+    for stanza in stanzas:
+        Stanza(stanza).versify()
 
 
 def add_to_manifest(manifest, partno):
 
 
 def add_to_manifest(manifest, partno):
@@ -285,9 +330,33 @@ def transform_chunk(chunk_xml, chunk_no, annotations, empty=False, _empty_html_s
     return output_html, toc, chars
 
 
     return output_html, toc, chars
 
 
+def flatten_image_paths(wldoc):
+    root = wldoc.edoc.getroot()
+    for node in root.findall(".//ilustr"):
+        node.attrib['src'] = os.path.basename(node.attrib['src'])
+    return wldoc
+
+def render_latex(wldoc, prefix="latex"):
+    """
+    Renders <latex>CODE</latex> as images and returns
+    (changed_wldoc, [ (epub_filepath1, latexfragment_object1), ... ]
+"""
+    root = wldoc.edoc.getroot()
+    latex_nodes = root.findall(".//latex")
+    images = []
+    for ln in latex_nodes:
+        fragment = LatexFragment(ln.text, resize=40)
+        images.append((os.path.join(prefix, fragment.filename), fragment))
+        ln.tag = "img"
+        ln.text = os.path.join(prefix, fragment.filename)
+        
+    return wldoc, images
+        
+
 def transform(wldoc, verbose=False,
               style=None, html_toc=False,
 def transform(wldoc, verbose=False,
               style=None, html_toc=False,
-              sample=None, cover=None, flags=None):
+              sample=None, cover=None, flags=None, resources=None,
+              intro_file=None, cover_file=None):
     """ produces a EPUB file
 
     sample=n: generate sample e-book (with at least n paragraphs)
     """ produces a EPUB file
 
     sample=n: generate sample e-book (with at least n paragraphs)
@@ -311,7 +380,7 @@ def transform(wldoc, verbose=False,
             zip.writestr('OPS/title.html',
                  etree.tostring(html_tree, method="html", pretty_print=True))
             # add a title page TOC entry
             zip.writestr('OPS/title.html',
                  etree.tostring(html_tree, method="html", pretty_print=True))
             # add a title page TOC entry
-            toc.add(u"Strona tytułowa", "title.html")
+            toc.add(u"Tytuł", "title.html")
         elif wldoc.book_info.parts:
             # write title page for every parent
             if sample is not None and sample <= 0:
         elif wldoc.book_info.parts:
             # write title page for every parent
             if sample is not None and sample <= 0:
@@ -335,6 +404,8 @@ def transform(wldoc, verbose=False,
             if main_text.tag == RDFNS('RDF'):
                 main_text = None
 
             if main_text.tag == RDFNS('RDF'):
                 main_text = None
 
+        flatten_image_paths(wldoc)
+
         if main_text is not None:
             for chunk_xml in chop(main_text):
                 empty = False
         if main_text is not None:
             for chunk_xml in chop(main_text):
                 empty = False
@@ -392,28 +463,48 @@ def transform(wldoc, verbose=False,
                        'media-type="application/oebps-package+xml" />' \
                        '</rootfiles></container>')
     zip.write(get_resource('res/wl-logo-small.png'), os.path.join('OPS', 'logo_wolnelektury.png'))
                        'media-type="application/oebps-package+xml" />' \
                        '</rootfiles></container>')
     zip.write(get_resource('res/wl-logo-small.png'), os.path.join('OPS', 'logo_wolnelektury.png'))
+    zip.write(get_resource('res/logo.png'), os.path.join('OPS', 'logo.png'))
     zip.write(get_resource('res/jedenprocent.png'), os.path.join('OPS', 'jedenprocent.png'))
     if not style:
         style = get_resource('epub/style.css')
     zip.write(style, os.path.join('OPS', 'style.css'))
 
     zip.write(get_resource('res/jedenprocent.png'), os.path.join('OPS', 'jedenprocent.png'))
     if not style:
         style = get_resource('epub/style.css')
     zip.write(style, os.path.join('OPS', 'style.css'))
 
+    document, latex_images = render_latex(document)
+    for image in latex_images:
+        zip.write(image[1].path, os.path.join('OPS', image[0]))
+        image[1].remove()
+
+    if resources:
+        if os.path.isdir(resources):
+            for dp, dirs, files in os.walk(resources):
+                for fname in files:
+                    fpath  = os.path.join(dp, fname)
+                    if os.path.isfile(fpath):
+                        zip.write(fpath, os.path.join('OPS', fname))
+                        manifest.append(etree.fromstring(
+                                '<item id="%s" href="%s" media-type="%s" />' % (os.path.splitext(fname)[0], fname, guess_type(fpath)[0])))
+
+        else:
+            print "resources path %s is not directory" % resources
+                
+
     if cover:
         if cover is True:
     if cover:
         if cover is True:
-            cover = WLCover
+            cover = FutureOfCopyrightCover
 
         cover_file = StringIO()
 
         cover_file = StringIO()
-        bound_cover = cover(document.book_info)
-        bound_cover.save(cover_file)
-        cover_name = 'cover.%s' % bound_cover.ext()
-        zip.writestr(os.path.join('OPS', cover_name), cover_file.getvalue())
+        c = cover(document.book_info)
+        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'))
         del cover_file
 
         cover_tree = etree.parse(get_resource('epub/cover.html'))
-        cover_tree.find('//' + XHTMLNS('img')).set('src', cover_name)
+        cover_tree.find('//' + XHTMLNS('img')).set('src', c_name)
         zip.writestr('OPS/cover.html', etree.tostring(
                         cover_tree, method="html", pretty_print=True))
 
         zip.writestr('OPS/cover.html', etree.tostring(
                         cover_tree, method="html", pretty_print=True))
 
-        if bound_cover.uses_dc_cover:
+        if c.uses_dc_cover:
             if document.book_info.cover_by:
                 document.edoc.getroot().set('data-cover-by', document.book_info.cover_by)
             if document.book_info.cover_source:
             if document.book_info.cover_by:
                 document.edoc.getroot().set('data-cover-by', document.book_info.cover_by)
             if document.book_info.cover_source:
@@ -422,11 +513,12 @@ def transform(wldoc, verbose=False,
         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="%s" media-type="%s" />' % (cover_name, bound_cover.mime_type())))
+            '<item id="cover-image" href="%s" media-type="%s" />' % (c_name, c.mime_type())))
         spine.insert(0, etree.fromstring('<itemref idref="cover" linear="no" />'))
         opf.getroot()[0].append(etree.fromstring('<meta name="cover" content="cover-image"/>'))
         guide.append(etree.fromstring('<reference href="cover.html" type="cover" title="Okładka"/>'))
         spine.insert(0, etree.fromstring('<itemref idref="cover" linear="no" />'))
         opf.getroot()[0].append(etree.fromstring('<meta name="cover" content="cover-image"/>'))
         guide.append(etree.fromstring('<reference href="cover.html" type="cover" title="Okładka"/>'))
-
+    
+              
 
     annotations = etree.Element('annotations')
 
 
     annotations = etree.Element('annotations')
 
@@ -437,17 +529,36 @@ def transform(wldoc, verbose=False,
                                '</navMap></ncx>')
     nav_map = toc_file[-1]
 
                                '</navMap></ncx>')
     nav_map = toc_file[-1]
 
+    manifest.append(etree.fromstring(
+        '<item id="first" href="first.html" media-type="application/xhtml+xml" />'))
+    spine.append(etree.fromstring(
+        '<itemref idref="first" />'))
+    html_tree = xslt(document.edoc, get_resource('epub/xsltFirst.xsl'))
+#    chars.update(used_chars(html_tree.getroot()))
+    zip.writestr('OPS/first.html', etree.tostring(
+                        html_tree, method="html", pretty_print=True))
+
+    if intro_file:
+        manifest.append(etree.fromstring(
+                '<item id="intro" href="intro.html" media-type="application/xhtml+xml" />'))
+        spine.append(etree.fromstring(
+                '<itemref idref="intro" />'))
+        zip.writestr('OPS/intro.html', open(intro_file or get_resource('epub/intro.html')).read())
+
+
     if html_toc:
         manifest.append(etree.fromstring(
             '<item id="html_toc" href="toc.html" media-type="application/xhtml+xml" />'))
         spine.append(etree.fromstring(
             '<itemref idref="html_toc" />'))
     if html_toc:
         manifest.append(etree.fromstring(
             '<item id="html_toc" href="toc.html" media-type="application/xhtml+xml" />'))
         spine.append(etree.fromstring(
             '<itemref idref="html_toc" />'))
-        guide.append(etree.fromstring('<reference href="toc.html" type="toc" title="Spis treści"/>'))
+        guide.append(etree.fromstring('<reference href="toc.html" type="toc" title="Table of Contents"/>'))
 
     toc, chunk_counter, chars, sample = transform_file(document, sample=sample)
 
 
     toc, chunk_counter, chars, sample = transform_file(document, sample=sample)
 
+    toc.add("Informacje redakcyjne", "first.html", index=0)
+
     if len(toc.children) < 2:
     if len(toc.children) < 2:
-        toc.add(u"Początek utworu", "part1.html")
+        toc.add(u"Początek książki", "part1.html")
 
     # Last modifications in container files and EPUB creation
     if len(annotations) > 0:
 
     # Last modifications in container files and EPUB creation
     if len(annotations) > 0:
@@ -462,7 +573,16 @@ def transform(wldoc, verbose=False,
         zip.writestr('OPS/annotations.html', etree.tostring(
                             html_tree, method="html", pretty_print=True))
 
         zip.writestr('OPS/annotations.html', etree.tostring(
                             html_tree, method="html", pretty_print=True))
 
-    toc.add("Strona redakcyjna", "last.html")
+    # toc.add("Weprzyj Wolne Lektury", "support.html")
+    # manifest.append(etree.fromstring(
+    #     '<item id="support" href="support.html" media-type="application/xhtml+xml" />'))
+    # spine.append(etree.fromstring(
+    #     '<itemref idref="support" />'))
+    # html_string = open(get_resource('epub/support.html')).read()
+    # chars.update(used_chars(etree.fromstring(html_string)))
+    # zip.writestr('OPS/support.html', html_string)
+
+    toc.add("Informacje redakcyjne", "last.html")
     manifest.append(etree.fromstring(
         '<item id="last" href="last.html" media-type="application/xhtml+xml" />'))
     spine.append(etree.fromstring(
     manifest.append(etree.fromstring(
         '<item id="last" href="last.html" media-type="application/xhtml+xml" />'))
     spine.append(etree.fromstring(