# Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information.
#
import os
+import re
import urllib
from copy import deepcopy
from mimetypes import guess_type
from urllib2 import urlopen
from lxml import etree
-from librarian import OPFNS, NCXNS, XHTMLNS, DCNS
+from librarian import OPFNS, NCXNS, XHTMLNS, DCNS, BuildError
from librarian import core
from librarian.formats import Format
from librarian.formats.cover.evens import EvensCover
yield wrapper
+class NaturalText(EpubRenderer):
+ def render_text(self, text, ctx):
+ root, inner = self.text_container()
+ chunks = re.split('(?<=\s\w) ', text)
+ inner.text = chunks[0]
+ for chunk in chunks[1:]:
+ x = etree.Entity("nbsp")
+ x.tail = chunk
+ inner.append(x)
+ return root
+
+
+class Silent(EpubRenderer):
+ def render_text(self, text, ctx):
+ root, inner = self.text_container()
+ return root
+
+
class Footnotes(object):
def __init__(self):
self.counter = 0
# Renderers
-class AsideR(EpubRenderer):
+class AsideR(NaturalText):
def render(self, element, ctx):
outputs = list(super(AsideR, self).render(element, ctx))
anchor = ctx.footnotes.append(outputs)
yield wrapper
EpubFormat.renderers.register(core.Aside, None, AsideR('div'))
+EpubFormat.renderers.register(core.Aside, 'comment', Silent())
-class DivR(EpubRenderer):
+
+class DivR(NaturalText):
def container(self, ctx):
root, inner = super(DivR, self).container(ctx)
if getattr(ctx, 'inline', False):
inner.set('style', 'display: block;')
return root, inner
EpubFormat.renderers.register(core.Div, None, DivR('div'))
+EpubFormat.renderers.register(core.Div, 'p', NaturalText('p'))
+
+EpubFormat.renderers.register(core.Div, 'list', NaturalText('ul'))
+EpubFormat.renderers.register(core.Div, 'list.enum', NaturalText('ol'))
+EpubFormat.renderers.register(core.Div, 'item', NaturalText('li'))
+EpubFormat.renderers.register(core.Span, 'item', NaturalText('li'))
class DivImageR(EpubRenderer):
def render(self, element, ctx):
src = element.attrib.get('src', '')
ctx.images.append(src)
+ if '/' not in src:
+ raise BuildError('Bad image URL')
src = src.rsplit('/', 1)[1]
return super(DivImageR, self).render(element, Context(ctx, src=src))
EpubFormat.renderers.register(core.Div, 'img', DivImageR('img'))
-class HeaderR(EpubRenderer):
+class DivVideoR(Silent):
+ def render(self, element, ctx):
+ src = 'https://www.youtube.com/watch?v=%s' % element.attrib.get('videoid', '')
+ return super(DivVideoR, self).render(element, Context(ctx, src=src))
+
+ def container(self, ctx):
+ root, inner = super(DivVideoR, self).container(ctx)
+ src = getattr(ctx, 'src', '')
+ link = etree.Element('a', {'href': src})
+ link.text = src
+ inner.append(link)
+ return root, inner
+EpubFormat.renderers.register(core.Div, 'video', DivVideoR('p'))
+
+
+class HeaderR(NaturalText):
def subcontext(self, element, ctx):
return Context(ctx, inline=True)
EpubFormat.renderers.register(core.Header, None, HeaderR('h1'))
-class SectionR(EpubRenderer):
+class SectionR(NaturalText):
epub_separate = True
def render(self, element, ctx):
EpubFormat.renderers.register(core.Section, None, SectionR())
-class SpanR(EpubRenderer):
+class SpanR(NaturalText):
pass
EpubFormat.renderers.register(core.Span, None, SpanR('span'))
+EpubFormat.renderers.register(core.Span, 'cite', SpanR('i'))
+EpubFormat.renderers.register(core.Span, 'emp', SpanR('b'))
+EpubFormat.renderers.register(core.Span, 'emph', SpanR('i'))
+
+
+class SpanLink(EpubRenderer):
+ def render(self, element, ctx):
+ parts = super(SpanLink, self).render(element, ctx)
+ for part in parts:
+ src = element.attrib.get('href', '')
+ if src.startswith('file://'):
+ src = ctx.files_path + src[7:]
+ part[0].attrib['href'] = src
+ yield part
+EpubFormat.renderers.register(core.Span, 'link', SpanLink('a'))