X-Git-Url: https://git.mdrn.pl/librarian.git/blobdiff_plain/bddf82ce5ec2a7c34b6b14d3bdb6c4cf24af7388..5f92f5a341c58a30c0a5c0d64ba3ed498bb9db85:/librarian/epub.py?ds=inline
diff --git a/librarian/epub.py b/librarian/epub.py
index e52e524..bf58a9f 100644
--- a/librarian/epub.py
+++ b/librarian/epub.py
@@ -26,6 +26,7 @@ from librarian.hyphenator import Hyphenator
functions.reg_person_name()
functions.reg_lang_code_3to2()
+
def set_hyph_language(source_tree):
def get_short_lng_code(text):
result = ''
@@ -34,18 +35,21 @@ def set_hyph_language(source_tree):
for line in f:
list = line.strip().split('|')
if list[0] == text:
- result=list[2]
+ result = list[2]
if result == '':
return text
else:
return result
- bibl_lng = etree.XPath('//dc:language//text()', namespaces = {'dc':str(DCNS)})(source_tree)
- short_lng = get_short_lng_code(bibl_lng[0])
+ bibl_lng = etree.XPath('//dc:language//text()',
+ namespaces={'dc': str(DCNS)})(source_tree)
+ short_lng = get_short_lng_code(bibl_lng[0])
try:
- return Hyphenator(get_resource('res/hyph-dictionaries/hyph_' + short_lng + '.dic'))
+ return Hyphenator(get_resource('res/hyph-dictionaries/hyph_' +
+ short_lng + '.dic'))
except:
pass
-
+
+
def hyphenate_and_fix_conjunctions(source_tree, hyph):
if hyph is not None:
texts = etree.XPath('/utwor/*[2]//text()')(source_tree)
@@ -54,13 +58,14 @@ def hyphenate_and_fix_conjunctions(source_tree, hyph):
newt = ''
wlist = re.compile(r'\w+|[^\w]', re.UNICODE).findall(t)
for w in wlist:
- newt += hyph.inserted(w, u'\u00AD')
+ 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
-
+
+
def inner_xml(node):
""" returns node's text and children as a string
@@ -71,6 +76,7 @@ def inner_xml(node):
nt = node.text if node.text is not None else ''
return ''.join([nt] + [etree.tostring(child) for child in node])
+
def set_inner_xml(node, text):
""" sets node's text and children from a string
@@ -134,7 +140,7 @@ def find_annotations(annotations, source, part_no):
for child in source:
if child.tag in ('pe', 'pa', 'pt', 'pr'):
annotation = deepcopy(child)
- number = str(len(annotations)+1)
+ number = str(len(annotations) + 1)
annotation.set('number', number)
annotation.set('part', str(part_no))
annotation.tail = ''
@@ -159,7 +165,7 @@ class Stanza(object):
>>> print etree.tostring(s)
a c cbx/
ycd
-
+
"""
def __init__(self, stanza_elem):
self.stanza = stanza_elem
@@ -221,18 +227,17 @@ def add_to_manifest(manifest, partno):
""" Adds a node to the manifest section in content.opf file """
partstr = 'part%d' % partno
- e = manifest.makeelement(OPFNS('item'), attrib={
- 'id': partstr,
- 'href': partstr + '.html',
- 'media-type': 'application/xhtml+xml',
- })
+ e = manifest.makeelement(
+ OPFNS('item'), attrib={'id': partstr, 'href': partstr + '.html',
+ 'media-type': 'application/xhtml+xml'}
+ )
manifest.append(e)
def add_to_spine(spine, partno):
""" Adds a node to the spine section in content.opf file """
- e = spine.makeelement(OPFNS('itemref'), attrib={'idref': 'part%d' % partno});
+ e = spine.makeelement(OPFNS('itemref'), attrib={'idref': 'part%d' % partno})
spine.append(e)
@@ -246,7 +251,7 @@ class TOC(object):
def add(self, name, part_href, level=0, is_part=True, index=None):
assert level == 0 or index is None
if level > 0 and self.children:
- return self.children[-1].add(name, part_href, level-1, is_part)
+ return self.children[-1].add(name, part_href, level - 1, is_part)
else:
t = TOC(name)
t.part_href = part_href
@@ -304,7 +309,7 @@ class TOC(object):
texts.append(
"
" %
(depth, child.href(), child.name))
- texts.append(child.html_part(depth+1))
+ texts.append(child.html_part(depth + 1))
return "\n".join(texts)
def html(self):
@@ -327,10 +332,10 @@ def chop(main_text):
# prepare a container for each chunk
part_xml = etree.Element('utwor')
etree.SubElement(part_xml, 'master')
- main_xml_part = part_xml[0] # master
+ main_xml_part = part_xml[0] # master
last_node_part = False
-
+
# the below loop are workaround for a problem with epubs in drama ebooks without acts
is_scene = False
is_act = False
@@ -340,7 +345,7 @@ def chop(main_text):
is_scene = True
elif name == 'naglowek_akt':
is_act = True
-
+
for one_part in main_text:
name = one_part.tag
if is_act is False and is_scene is True:
@@ -364,7 +369,7 @@ def chop(main_text):
main_xml_part[:] = [deepcopy(one_part)]
else:
main_xml_part.append(deepcopy(one_part))
- last_node_part = False
+ last_node_part = False
yield part_xml
@@ -390,11 +395,12 @@ def transform_chunk(chunk_xml, chunk_no, annotations, empty=False, _empty_html_s
replace_by_verse(chunk_xml)
html_tree = xslt(chunk_xml, get_resource('epub/xsltScheme.xsl'))
chars = used_chars(html_tree.getroot())
- output_html = etree.tostring(html_tree, pretty_print = True,
- xml_declaration = True,
- encoding = "utf-8",
- doctype='')
+ output_html = etree.tostring(
+ html_tree, pretty_print=True, xml_declaration=True,
+ encoding="utf-8",
+ doctype=''
+ )
return output_html, toc, chars
@@ -412,11 +418,10 @@ def transform(wldoc, verbose=False,
""" processes one input file and proceeds to its children """
replace_characters(wldoc.edoc.getroot())
-
+
hyphenator = set_hyph_language(wldoc.edoc.getroot())
hyphenate_and_fix_conjunctions(wldoc.edoc.getroot(), hyphenator)
-
-
+
# every input file will have a TOC entry,
# pointing to starting chunk
toc = TOC(wldoc.book_info.title, "part%d.html" % chunk_counter)
@@ -425,12 +430,15 @@ def transform(wldoc, verbose=False,
# write book title page
html_tree = xslt(wldoc.edoc, get_resource('epub/xsltTitle.xsl'))
chars = used_chars(html_tree.getroot())
- zip.writestr('OPS/title.html',
- etree.tostring(html_tree, pretty_print = True,
- xml_declaration = True,
- encoding = "utf-8",
- doctype=''))
+ zip.writestr(
+ 'OPS/title.html',
+ etree.tostring(
+ html_tree, pretty_print=True, xml_declaration=True,
+ encoding="utf-8",
+ doctype=''
+ )
+ )
# add a title page TOC entry
toc.add(u"Strona tytuÅowa", "title.html")
elif wldoc.book_info.parts:
@@ -441,12 +449,12 @@ def transform(wldoc, verbose=False,
else:
html_tree = xslt(wldoc.edoc, get_resource('epub/xsltChunkTitle.xsl'))
chars = used_chars(html_tree.getroot())
- html_string = etree.tostring(html_tree,
- pretty_print = True,
- xml_declaration = True,
- encoding = "utf-8",
- doctype='')
+ html_string = etree.tostring(
+ html_tree, pretty_print=True, xml_declaration=True,
+ encoding="utf-8",
+ doctype=''
+ )
zip.writestr('OPS/part%d.html' % chunk_counter, html_string)
add_to_manifest(manifest, chunk_counter)
add_to_spine(spine, chunk_counter)
@@ -486,7 +494,6 @@ def transform(wldoc, verbose=False,
return toc, chunk_counter, chars, sample
-
document = deepcopy(wldoc)
del wldoc
@@ -495,8 +502,10 @@ def transform(wldoc, verbose=False,
document.edoc.getroot().set(flag, 'yes')
# add editors info
- document.edoc.getroot().set('editors', u', '.join(sorted(
- editor.readable() for editor in document.editors())))
+ editors = document.editors()
+ if editors:
+ document.edoc.getroot().set('editors', u', '.join(sorted(
+ editor.readable() for editor in editors)))
if document.book_info.funders:
document.edoc.getroot().set('funders', u', '.join(
document.book_info.funders))
@@ -511,19 +520,27 @@ 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)
+
# write static elements
mime = zipfile.ZipInfo()
mime.filename = 'mimetype'
mime.compress_type = zipfile.ZIP_STORED
mime.extra = ''
zip.writestr(mime, 'application/epub+zip')
- zip.writestr('META-INF/container.xml', '' \
- '' \
- '')
- zip.write(get_resource('res/wl-logo-small.png'), os.path.join('OPS', 'logo_wolnelektury.png'))
- zip.write(get_resource('res/jedenprocent.png'), os.path.join('OPS', 'jedenprocent.png'))
+ zip.writestr(
+ 'META-INF/container.xml',
+ ''
+ ''
+ ''
+ ''
+ )
+ zip.write(get_resource('res/wl-logo-small.png'),
+ os.path.join('OPS', 'logo_wolnelektury.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'))
@@ -542,9 +559,11 @@ def transform(wldoc, verbose=False,
cover_tree = etree.parse(get_resource('epub/cover.html'))
cover_tree.find('//' + XHTMLNS('img')).set('src', cover_name)
zip.writestr('OPS/cover.html', etree.tostring(
- cover_tree, pretty_print = True, xml_declaration = True, encoding = "utf-8",
- doctype=''))
+ cover_tree, pretty_print=True, xml_declaration=True,
+ encoding="utf-8",
+ doctype=''
+ ))
if bound_cover.uses_dc_cover:
if document.book_info.cover_by:
@@ -560,14 +579,16 @@ def transform(wldoc, verbose=False,
opf.getroot()[0].append(etree.fromstring(''))
guide.append(etree.fromstring(''))
-
annotations = etree.Element('annotations')
- toc_file = etree.fromstring('' \
- '' \
- '')
+ toc_file = etree.fromstring(
+ ''
+ ''
+ ''
+ )
nav_map = toc_file[-1]
if html_toc:
@@ -593,11 +614,11 @@ def transform(wldoc, verbose=False,
html_tree = xslt(annotations, get_resource('epub/xsltAnnotations.xsl'))
chars = chars.union(used_chars(html_tree.getroot()))
zip.writestr('OPS/annotations.html', etree.tostring(
- html_tree, pretty_print = True,
- xml_declaration = True,
- encoding = "utf-8",
- doctype=''))
+ html_tree, pretty_print=True, xml_declaration=True,
+ encoding="utf-8",
+ doctype=''
+ ))
toc.add("Wesprzyj Wolne Lektury", "support.html")
manifest.append(etree.fromstring(
@@ -616,11 +637,11 @@ def transform(wldoc, verbose=False,
html_tree = xslt(document.edoc, get_resource('epub/xsltLast.xsl'))
chars.update(used_chars(html_tree.getroot()))
zip.writestr('OPS/last.html', etree.tostring(
- html_tree, pretty_print = True,
- xml_declaration = True,
- encoding = "utf-8",
- doctype=''))
+ html_tree, pretty_print=True, xml_declaration=True,
+ encoding="utf-8",
+ doctype=''
+ ))
if not flags or not 'without-fonts' in flags:
# strip fonts
@@ -632,8 +653,10 @@ def transform(wldoc, verbose=False,
os.chdir(os.path.join(os.path.dirname(os.path.realpath(__file__)), 'font-optimizer'))
for fname in 'DejaVuSerif.ttf', 'DejaVuSerif-Bold.ttf', 'DejaVuSerif-Italic.ttf', 'DejaVuSerif-BoldItalic.ttf':
- optimizer_call = ['perl', 'subset.pl', '--chars', ''.join(chars).encode('utf-8'),
- get_resource('fonts/' + fname), os.path.join(tmpdir, fname)]
+ optimizer_call = ['perl', 'subset.pl', '--chars',
+ ''.join(chars).encode('utf-8'),
+ get_resource('fonts/' + fname),
+ os.path.join(tmpdir, fname)]
if verbose:
print "Running font-optimizer"
subprocess.check_call(optimizer_call)
@@ -645,9 +668,8 @@ def transform(wldoc, verbose=False,
rmtree(tmpdir)
if cwd is not None:
os.chdir(cwd)
- zip.writestr('OPS/content.opf', etree.tostring(opf, pretty_print = True,
- xml_declaration = True,
- encoding = "utf-8"))
+ zip.writestr('OPS/content.opf', etree.tostring(opf, pretty_print=True,
+ xml_declaration=True, encoding="utf-8"))
title = document.book_info.title
attributes = "dtb:uid", "dtb:depth", "dtb:totalPageCount", "dtb:maxPageNumber"
for st in attributes:
@@ -664,9 +686,8 @@ def transform(wldoc, verbose=False,
toc.add(u"Spis treÅci", "toc.html", index=1)
zip.writestr('OPS/toc.html', toc.html().encode('utf-8'))
toc.write_to_xml(nav_map)
- zip.writestr('OPS/toc.ncx', etree.tostring(toc_file, pretty_print = True,
- xml_declaration = True,
- encoding = "utf-8"))
+ zip.writestr('OPS/toc.ncx', etree.tostring(toc_file, pretty_print=True,
+ xml_declaration=True, encoding="utf-8"))
zip.close()
return OutputFile.from_filename(output_file.name)