X-Git-Url: https://git.mdrn.pl/librarian.git/blobdiff_plain/1fdd33e6b409ee5c034915d121f773f007b95b9f..318d79c6ec02163ec898a66723055da25837346a:/librarian/epub.py?ds=sidebyside
diff --git a/librarian/epub.py b/librarian/epub.py
index 5f017d4..be9488a 100644
--- a/librarian/epub.py
+++ b/librarian/epub.py
@@ -3,13 +3,13 @@
# This file is part of Librarian, licensed under GNU Affero GPLv3 or later.
# Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information.
#
-from __future__ import with_statement
+from __future__ import print_function, unicode_literals
import os
import os.path
import re
import subprocess
-from StringIO import StringIO
+from six import BytesIO
from copy import deepcopy
from mimetypes import guess_type
@@ -30,7 +30,7 @@ functions.reg_lang_code_3to2()
def squeeze_whitespace(s):
- return re.sub(r'\s+', ' ', s)
+ return re.sub(b'\\s+', b' ', s)
def set_hyph_language(source_tree):
@@ -38,7 +38,7 @@ def set_hyph_language(source_tree):
result = ''
text = ''.join(text)
with open(get_resource('res/ISO-639-2_8859-1.txt'), 'rb') as f:
- for line in f:
+ for line in f.read().decode('latin1').split('\n'):
list = line.strip().split('|')
if list[0] == text:
result = list[2]
@@ -77,12 +77,12 @@ def hyphenate_and_fix_conjunctions(source_tree, hyph):
def inner_xml(node):
""" returns node's text and children as a string
- >>> print inner_xml(etree.fromstring('xyz'))
+ >>> print(inner_xml(etree.fromstring('xyz')))
xyz
"""
nt = node.text if node.text is not None else ''
- return ''.join([nt] + [etree.tostring(child) for child in node])
+ return ''.join([nt] + [etree.tostring(child, encoding='unicode') for child in node])
def set_inner_xml(node, text):
@@ -90,7 +90,7 @@ def set_inner_xml(node, text):
>>> e = etree.fromstring('bxx')
>>> set_inner_xml(e, 'xyz')
- >>> print etree.tostring(e)
+ >>> print(etree.tostring(e, encoding='unicode'))
xyz
"""
@@ -102,7 +102,7 @@ def set_inner_xml(node, text):
def node_name(node):
""" Find out a node's name
- >>> print node_name(etree.fromstring('XYZ'))
+ >>> print(node_name(etree.fromstring('XYZ')))
XYZ
"""
@@ -122,7 +122,7 @@ def xslt(xml, sheet, **kwargs):
xml = etree.ElementTree(xml)
with open(sheet) as xsltf:
transform = etree.XSLT(etree.parse(xsltf))
- params = dict((key, transform.strparam(value)) for key, value in kwargs.iteritems())
+ params = dict((key, transform.strparam(value)) for key, value in kwargs.items())
return transform(xml, **params)
@@ -172,8 +172,8 @@ class Stanza(object):
>>> s = etree.fromstring("a c c/\\nbx/\\nyc/ \\nd")
>>> Stanza(s).versify()
- >>> print etree.tostring(s)
- a c cbx/
+ >>> print(etree.tostring(s, encoding='unicode'))
+ a ccbx/
ycd
"""
@@ -190,7 +190,7 @@ class Stanza(object):
tail = self.stanza.tail
self.stanza.clear()
self.stanza.tail = tail
- self.stanza.extend(self.verses)
+ self.stanza.extend(verse for verse in self.verses if verse.text or len(verse) > 0)
def open_normal_verse(self):
self.open_verse = self.stanza.makeelement("wers_normalny")
@@ -325,8 +325,8 @@ class TOC(object):
return "\n".join(texts)
def html(self):
- with open(get_resource('epub/toc.html')) as f:
- t = unicode(f.read(), 'utf-8')
+ with open(get_resource('epub/toc.html'), 'rb') as f:
+ t = f.read().decode('utf-8')
return t % self.html_part()
@@ -535,7 +535,10 @@ def transform(wldoc, verbose=False, style=None, html_toc=False,
functions.reg_mathml_epub(zip)
if os.path.isdir(ilustr_path):
+ ilustr_elements = set(ilustr.get('src') for ilustr in document.edoc.findall('//ilustr'))
for i, filename in enumerate(os.listdir(ilustr_path)):
+ if filename not in ilustr_elements:
+ continue
file_path = os.path.join(ilustr_path, filename)
zip.write(file_path, os.path.join('OPS', filename))
image_id = 'image%s' % i
@@ -546,16 +549,16 @@ def transform(wldoc, verbose=False, style=None, html_toc=False,
mime = zipfile.ZipInfo()
mime.filename = 'mimetype'
mime.compress_type = zipfile.ZIP_STORED
- mime.extra = ''
- zip.writestr(mime, 'application/epub+zip')
+ mime.extra = b''
+ zip.writestr(mime, b'application/epub+zip')
zip.writestr(
'META-INF/container.xml',
- ''
- ''
- ''
- ''
+ b''
+ b''
+ b''
+ b''
)
zip.write(get_resource('res/wl-logo-small.png'),
os.path.join('OPS', 'logo_wolnelektury.png'))
@@ -569,7 +572,7 @@ def transform(wldoc, verbose=False, style=None, html_toc=False,
if cover is True:
cover = make_cover
- cover_file = StringIO()
+ cover_file = BytesIO()
bound_cover = cover(document.book_info)
bound_cover.save(cover_file)
cover_name = 'cover.%s' % bound_cover.ext()
@@ -602,12 +605,12 @@ def transform(wldoc, verbose=False, style=None, html_toc=False,
annotations = etree.Element('annotations')
toc_file = etree.fromstring(
- ''
- ''
- ''
+ b''
+ b''
+ b''
)
nav_map = toc_file[-1]
@@ -645,7 +648,7 @@ def transform(wldoc, verbose=False, style=None, html_toc=False,
' '))
spine.append(etree.fromstring(
''))
- html_string = open(get_resource('epub/support.html')).read()
+ html_string = open(get_resource('epub/support.html'), 'rb').read()
chars.update(used_chars(etree.fromstring(html_string)))
zip.writestr('OPS/support.html', squeeze_whitespace(html_string))
@@ -679,7 +682,7 @@ def transform(wldoc, verbose=False, style=None, html_toc=False,
os.path.join(tmpdir, fname)]
env = {"PERL_USE_UNSAFE_INC": "1"}
if verbose:
- print "Running font-optimizer"
+ print("Running font-optimizer")
subprocess.check_call(optimizer_call, env=env)
else:
dev_null = open(os.devnull, 'w')