mirror of
https://github.com/kovidgoyal/calibre.git
synced 2025-07-09 03:04:10 -04:00
IGN:...
This commit is contained in:
Kovid Goyal
commit
072a063be3
@ -21,6 +21,9 @@ mimetypes.add_type('application/epub+zip', '.epub')
|
||||
mimetypes.add_type('text/x-sony-bbeb+xml', '.lrs')
|
||||
mimetypes.add_type('application/x-sony-bbeb', '.lrf')
|
||||
mimetypes.add_type('application/x-dtbncx+xml', '.ncx')
|
||||
mimetypes.add_type('application/adobe-page-template+xml', '.xpgt')
|
||||
mimetypes.add_type('application/x-font-opentype', '.otf')
|
||||
mimetypes.add_type('application/x-font-truetype', '.ttf')
|
||||
|
||||
def to_unicode(raw, encoding='utf-8', errors='strict'):
|
||||
if isinstance(raw, unicode):
|
||||
|
||||
@ -23,7 +23,7 @@ from urllib import unquote as urlunquote
|
||||
from lxml import etree
|
||||
from calibre.ebooks.lit.reader import DirectoryEntry
|
||||
import calibre.ebooks.lit.maps as maps
|
||||
from calibre.ebooks.oeb.base import OEB_DOCS, OEB_STYLES, OEB_CSS_MIME, \
|
||||
from calibre.ebooks.oeb.base import OEB_DOCS, XHTML_MIME, OEB_STYLES, \
|
||||
CSS_MIME, OPF_MIME, XML_NS, XML
|
||||
from calibre.ebooks.oeb.base import namespace, barename, prefixname, \
|
||||
urlnormalize, xpath
|
||||
@ -474,7 +474,7 @@ class LitWriter(object):
|
||||
name = '/data/' + item.id
|
||||
data = item.data
|
||||
secnum = 0
|
||||
if not isinstance(data, basestring):
|
||||
if isinstance(data, etree._Element):
|
||||
self._add_folder(name)
|
||||
rebin = ReBinary(data, item, self._oeb, map=HTML_MAP)
|
||||
self._add_file(name + '/ahc', rebin.ahc, 0)
|
||||
@ -483,6 +483,8 @@ class LitWriter(object):
|
||||
data = rebin.content
|
||||
name = name + '/content'
|
||||
secnum = 1
|
||||
elif isinstance(data, unicode):
|
||||
data = data.encode('utf-8')
|
||||
self._add_file(name, data, secnum)
|
||||
item.size = len(data)
|
||||
|
||||
@ -493,7 +495,7 @@ class LitWriter(object):
|
||||
if item.spine_position is not None:
|
||||
key = 'linear' if item.linear else 'nonlinear'
|
||||
manifest[key].append(item)
|
||||
elif item.media_type == CSS_MIME:
|
||||
elif item.media_type in OEB_STYLES:
|
||||
manifest['css'].append(item)
|
||||
elif item.media_type in LIT_IMAGES:
|
||||
manifest['images'].append(item)
|
||||
@ -506,6 +508,11 @@ class LitWriter(object):
|
||||
data.write(pack('<I', len(items)))
|
||||
for item in items:
|
||||
id, media_type = item.id, item.media_type
|
||||
if media_type in OEB_DOCS:
|
||||
# Needs to have 'html' in media-type
|
||||
media_type = XHTML_MIME
|
||||
elif media_type in OEB_STYLES:
|
||||
media_type = CSS_MIME
|
||||
href = urlunquote(item.href)
|
||||
item.offset = offset \
|
||||
if state in ('linear', 'nonlinear') else 0
|
||||
@ -525,7 +532,12 @@ class LitWriter(object):
|
||||
pb3 = StringIO()
|
||||
pb3cur = 0
|
||||
bits = 0
|
||||
linear = []
|
||||
nonlinear = []
|
||||
for item in self._oeb.spine:
|
||||
dest = linear if item.linear else nonlinear
|
||||
dest.append(item)
|
||||
for item in chain(linear, nonlinear):
|
||||
page_breaks = copy.copy(item.page_breaks)
|
||||
if not item.linear:
|
||||
page_breaks.insert(0, (0, []))
|
||||
|
||||
@ -62,6 +62,16 @@ def SVG(name): return '{%s}%s' % (SVG_NS, name)
|
||||
def XLINK(name): return '{%s}%s' % (XLINK_NS, name)
|
||||
def CALIBRE(name): return '{%s}%s' % (CALIBRE_NS, name)
|
||||
|
||||
def LINK_SELECTORS():
|
||||
results = []
|
||||
for expr in ('h:head/h:link/@href', 'h:body//h:a/@href',
|
||||
'h:body//h:img/@src', 'h:body//h:object/@data',
|
||||
'h:body//*/@xl:href', '//ncx:content/@src',
|
||||
'o2:page/@href'):
|
||||
results.append(etree.XPath(expr, namespaces=XPNSMAP))
|
||||
return results
|
||||
LINK_SELECTORS = LINK_SELECTORS()
|
||||
|
||||
EPUB_MIME = 'application/epub+zip'
|
||||
XHTML_MIME = 'application/xhtml+xml'
|
||||
CSS_MIME = 'text/css'
|
||||
@ -89,6 +99,10 @@ COLLAPSE_RE = re.compile(r'[ \t\r\n\v]+')
|
||||
QNAME_RE = re.compile(r'^[{][^{}]+[}][^{}]+$')
|
||||
PREFIXNAME_RE = re.compile(r'^[^:]+[:][^:]+')
|
||||
XMLDECL_RE = re.compile(r'^\s*<[?]xml.*?[?]>')
|
||||
CSSURL_RE = re.compile(r'''url[(](?P<q>["']?)(?P<url>[^)]+)(?P=q)[)]''')
|
||||
|
||||
RECOVER_PARSER = etree.XMLParser(recover=True)
|
||||
|
||||
|
||||
def element(parent, *args, **kwargs):
|
||||
if parent is not None:
|
||||
@ -140,14 +154,17 @@ def xml2str(root):
|
||||
return etree.tostring(root, encoding='utf-8', xml_declaration=True)
|
||||
|
||||
ASCII_CHARS = set(chr(x) for x in xrange(128))
|
||||
URL_SAFE = set(u'ABCDEFGHIJKLMNOPQRSTUVWXYZ'
|
||||
u'abcdefghijklmnopqrstuvwxyz'
|
||||
u'0123456789' u'_.-/~')
|
||||
URL_UNSAFE = ASCII_CHARS - URL_SAFE
|
||||
UNIBYTE_CHARS = set(chr(x) for x in xrange(256))
|
||||
URL_SAFE = set('ABCDEFGHIJKLMNOPQRSTUVWXYZ'
|
||||
'abcdefghijklmnopqrstuvwxyz'
|
||||
'0123456789' '_.-/~')
|
||||
URL_UNSAFE = [ASCII_CHARS - URL_SAFE, UNIBYTE_CHARS - URL_SAFE]
|
||||
def urlquote(href):
|
||||
result = []
|
||||
unsafe = 0 if isinstance(href, unicode) else 1
|
||||
unsafe = URL_UNSAFE[unsafe]
|
||||
for char in href:
|
||||
if char in URL_UNSAFE:
|
||||
if char in unsafe:
|
||||
char = "%%%02x" % ord(char)
|
||||
result.append(char)
|
||||
return ''.join(result)
|
||||
@ -185,7 +202,7 @@ class AbstractContainer(object):
|
||||
|
||||
class DirContainer(AbstractContainer):
|
||||
def __init__(self, rootdir):
|
||||
self.rootdir = rootdir
|
||||
self.rootdir = unicode(rootdir)
|
||||
|
||||
def read(self, path):
|
||||
path = os.path.join(self.rootdir, path)
|
||||
@ -205,16 +222,23 @@ class DirContainer(AbstractContainer):
|
||||
return os.path.isfile(urlunquote(path))
|
||||
|
||||
class DirWriter(object):
|
||||
def __init__(self, version=2.0):
|
||||
def __init__(self, version='2.0', page_map=False):
|
||||
self.version = version
|
||||
self.page_map = page_map
|
||||
|
||||
def dump(self, oeb, path):
|
||||
version = int(self.version[0])
|
||||
if not os.path.isdir(path):
|
||||
os.mkdir(path)
|
||||
output = DirContainer(path)
|
||||
for item in oeb.manifest.values():
|
||||
output.write(item.href, str(item))
|
||||
metadata = oeb.to_opf2() if self.version == 2 else oeb.to_opf1()
|
||||
if version == 1:
|
||||
metadata = oeb.to_opf1()
|
||||
elif version == 2:
|
||||
metadata = oeb.to_opf2(page_map=self.page_map)
|
||||
else:
|
||||
raise OEBError("Unrecognized OPF version %r" % self.version)
|
||||
for href, data in metadata.values():
|
||||
output.write(href, xml2str(data))
|
||||
return
|
||||
@ -455,7 +479,6 @@ class Manifest(object):
|
||||
# Convert to Unicode and normalize line endings
|
||||
data = self.oeb.decode(data)
|
||||
data = XMLDECL_RE.sub('', data)
|
||||
data = data.replace('\r\n', '\n').replace('\r', '\n')
|
||||
# Handle broken XHTML w/ SVG (ugh)
|
||||
if 'svg:' in data and SVG_NS not in data:
|
||||
data = data.replace(
|
||||
@ -480,7 +503,10 @@ class Manifest(object):
|
||||
if elem.text:
|
||||
elem.text = elem.text.strip('-')
|
||||
data = etree.tostring(data, encoding=unicode)
|
||||
try:
|
||||
data = etree.fromstring(data)
|
||||
except etree.XMLSyntaxError:
|
||||
data = etree.fromstring(data, parser=RECOVER_PARSER)
|
||||
# Force into the XHTML namespace
|
||||
if barename(data.tag) != 'html':
|
||||
raise OEBError(
|
||||
@ -536,6 +562,8 @@ class Manifest(object):
|
||||
data = self._force_xhtml(data)
|
||||
elif self.media_type[-4:] in ('+xml', '/xml'):
|
||||
data = etree.fromstring(data)
|
||||
elif self.media_type in OEB_STYLES:
|
||||
data = self.oeb.decode(data)
|
||||
self._data = data
|
||||
return data
|
||||
def fset(self, value):
|
||||
@ -549,6 +577,8 @@ class Manifest(object):
|
||||
data = self.data
|
||||
if isinstance(data, etree._Element):
|
||||
return xml2str(data)
|
||||
if isinstance(data, unicode):
|
||||
return data.encode('utf-8')
|
||||
return str(data)
|
||||
|
||||
def __eq__(self, other):
|
||||
@ -572,7 +602,9 @@ class Manifest(object):
|
||||
return cmp(skey, okey)
|
||||
|
||||
def relhref(self, href):
|
||||
if '/' not in self.href or ':' in href:
|
||||
if urlparse(href).scheme:
|
||||
return href
|
||||
if '/' not in self.href:
|
||||
return href
|
||||
base = os.path.dirname(self.href).split('/')
|
||||
target, frag = urldefrag(href)
|
||||
@ -588,7 +620,12 @@ class Manifest(object):
|
||||
return relhref
|
||||
|
||||
def abshref(self, href):
|
||||
if '/' not in self.href or ':' in href:
|
||||
if urlparse(href).scheme:
|
||||
return href
|
||||
path, frag = urldefrag(href)
|
||||
if not path:
|
||||
return '#'.join((self.href, frag))
|
||||
if '/' not in self.href:
|
||||
return href
|
||||
dirname = os.path.dirname(self.href)
|
||||
href = os.path.join(dirname, href)
|
||||
@ -615,13 +652,15 @@ class Manifest(object):
|
||||
if item in self.oeb.spine:
|
||||
self.oeb.spine.remove(item)
|
||||
|
||||
def generate(self, id, href):
|
||||
href = urlnormalize(href)
|
||||
def generate(self, id=None, href=None):
|
||||
if id is not None:
|
||||
base = id
|
||||
index = 1
|
||||
while id in self.ids:
|
||||
id = base + str(index)
|
||||
index += 1
|
||||
if href is not None:
|
||||
href = urlnormalize(href)
|
||||
base, ext = os.path.splitext(href)
|
||||
index = 1
|
||||
while href in self.hrefs:
|
||||
@ -996,13 +1035,11 @@ class OEBBook(object):
|
||||
metadata = etree.SubElement(nroot, OPF('metadata'), nsmap=nsmap)
|
||||
ignored = (OPF('dc-metadata'), OPF('x-metadata'))
|
||||
for elem in xpath(opf, 'o2:metadata//*'):
|
||||
if elem.tag in ignored:
|
||||
continue
|
||||
if namespace(elem.tag) in DC_NSES:
|
||||
tag = barename(elem.tag).lower()
|
||||
elem.tag = '{%s}%s' % (DC11_NS, tag)
|
||||
for name in elem.attrib:
|
||||
if name in ('role', 'file-as', 'scheme', 'event'):
|
||||
nsname = '{%s}%s' % (OPF2_NS, name)
|
||||
elem.attrib[nsname] = elem.attrib.pop(name)
|
||||
metadata.append(elem)
|
||||
for element in xpath(opf, 'o2:metadata//o2:meta'):
|
||||
metadata.append(element)
|
||||
@ -1015,7 +1052,6 @@ class OEBBook(object):
|
||||
data = self.container.read(opfpath)
|
||||
data = self.decode(data)
|
||||
data = XMLDECL_RE.sub('', data)
|
||||
data = data.replace('\r\n', '\n').replace('\r', '\n')
|
||||
try:
|
||||
opf = etree.fromstring(data)
|
||||
except etree.XMLSyntaxError:
|
||||
@ -1078,6 +1114,45 @@ class OEBBook(object):
|
||||
self.logger.warn('Title not specified')
|
||||
metadata.add('title', self.translate(__('Unknown')))
|
||||
|
||||
def _manifest_add_missing(self):
|
||||
manifest = self.manifest
|
||||
known = set(manifest.hrefs)
|
||||
unchecked = set(manifest.values())
|
||||
while unchecked:
|
||||
new = set()
|
||||
for item in unchecked:
|
||||
if (item.media_type in OEB_DOCS or
|
||||
item.media_type[-4:] in ('/xml', '+xml')) and \
|
||||
item.data is not None:
|
||||
hrefs = [sel(item.data) for sel in LINK_SELECTORS]
|
||||
for href in chain(*hrefs):
|
||||
href, _ = urldefrag(href)
|
||||
if not href:
|
||||
continue
|
||||
href = item.abshref(urlnormalize(href))
|
||||
scheme = urlparse(href).scheme
|
||||
if not scheme and href not in known:
|
||||
new.add(href)
|
||||
elif item.media_type in OEB_STYLES:
|
||||
for match in CSSURL_RE.finditer(item.data):
|
||||
href, _ = urldefrag(match.group('url'))
|
||||
href = item.abshref(urlnormalize(href))
|
||||
scheme = urlparse(href).scheme
|
||||
if not scheme and href not in known:
|
||||
new.add(href)
|
||||
unchecked.clear()
|
||||
for href in new:
|
||||
known.add(href)
|
||||
if not self.container.exists(href):
|
||||
self.logger.warn('Referenced file %r not found' % href)
|
||||
continue
|
||||
self.logger.warn('Referenced file %r not in manifest' % href)
|
||||
id, _ = manifest.generate(id='added')
|
||||
guessed = mimetypes.guess_type(href)[0]
|
||||
media_type = guessed or BINARY_MIME
|
||||
added = manifest.add(id, href, media_type)
|
||||
unchecked.add(added)
|
||||
|
||||
def _manifest_from_opf(self, opf):
|
||||
self.manifest = manifest = Manifest(self)
|
||||
for elem in xpath(opf, '/o2:package/o2:manifest/o2:item'):
|
||||
@ -1100,6 +1175,40 @@ class OEBBook(object):
|
||||
self.logger.warn(u'Duplicate manifest id %r' % id)
|
||||
id, href = manifest.generate(id, href)
|
||||
manifest.add(id, href, media_type, fallback)
|
||||
self._manifest_add_missing()
|
||||
|
||||
def _spine_add_extra(self):
|
||||
manifest = self.manifest
|
||||
spine = self.spine
|
||||
unchecked = set(spine)
|
||||
selector = XPath('h:body//h:a/@href')
|
||||
extras = set()
|
||||
while unchecked:
|
||||
new = set()
|
||||
for item in unchecked:
|
||||
if item.media_type not in OEB_DOCS:
|
||||
# TODO: handle fallback chains
|
||||
continue
|
||||
for href in selector(item.data):
|
||||
href, _ = urldefrag(href)
|
||||
if not href:
|
||||
continue
|
||||
href = item.abshref(urlnormalize(href))
|
||||
if href not in manifest.hrefs:
|
||||
continue
|
||||
found = manifest.hrefs[href]
|
||||
if found.media_type not in OEB_DOCS or \
|
||||
found in spine or found in extras:
|
||||
continue
|
||||
new.add(found)
|
||||
extras.update(new)
|
||||
unchecked = new
|
||||
version = int(self.version[0])
|
||||
for item in sorted(extras):
|
||||
if version >= 2:
|
||||
self.logger.warn(
|
||||
'Spine-referenced file %r not in spine' % item.href)
|
||||
spine.add(item, linear=False)
|
||||
|
||||
def _spine_from_opf(self, opf):
|
||||
self.spine = spine = Spine(self)
|
||||
@ -1110,16 +1219,9 @@ class OEBBook(object):
|
||||
continue
|
||||
item = self.manifest[idref]
|
||||
spine.add(item, elem.get('linear'))
|
||||
extras = []
|
||||
for item in self.manifest.values():
|
||||
if item.media_type in OEB_DOCS \
|
||||
and item not in spine:
|
||||
extras.append(item)
|
||||
extras.sort()
|
||||
for item in extras:
|
||||
spine.add(item, False)
|
||||
if len(spine) == 0:
|
||||
raise OEBError("Spine is empty")
|
||||
self._spine_add_extra()
|
||||
|
||||
def _guide_from_opf(self, opf):
|
||||
self.guide = guide = Guide(self)
|
||||
@ -1189,12 +1291,11 @@ class OEBBook(object):
|
||||
href = site.get('href')
|
||||
if not title or not href:
|
||||
continue
|
||||
href = item.abshref(urlnormalize(href))
|
||||
path, _ = urldefrag(href)
|
||||
path, _ = urldefrag(urlnormalize(href))
|
||||
if path not in self.manifest.hrefs:
|
||||
self.logger.warn('TOC reference %r not found' % href)
|
||||
continue
|
||||
id = child.get('id')
|
||||
id = site.get('id')
|
||||
toc.add(title, href, id=id)
|
||||
return True
|
||||
|
||||
@ -1217,12 +1318,12 @@ class OEBBook(object):
|
||||
order = []
|
||||
for anchor in xpath(html, './/h:a[@href]'):
|
||||
href = anchor.attrib['href']
|
||||
href = item.abshref(urlnormalize(href))
|
||||
path, frag = urldefrag(href)
|
||||
if not path:
|
||||
href = '#'.join((itempath, frag))
|
||||
if path not in self.manifest.hrefs:
|
||||
continue
|
||||
title = ' '.join(xpath(anchor, './/text()'))
|
||||
title = COLLAPSE_RE.sub(' ', title.strip())
|
||||
href = urlnormalize(href)
|
||||
if href not in titles:
|
||||
order.append(href)
|
||||
titles[href].append(title)
|
||||
@ -1313,7 +1414,12 @@ class OEBBook(object):
|
||||
continue
|
||||
name = COLLAPSE_RE.sub(' ', name.strip())
|
||||
href = item.abshref(urlnormalize(href))
|
||||
pages.add(name, href)
|
||||
type = 'normal'
|
||||
if not name:
|
||||
type = 'special'
|
||||
elif name.lower().strip('ivxlcdm') == '':
|
||||
type = 'front'
|
||||
pages.add(name, href, type=type)
|
||||
return True
|
||||
|
||||
def _pages_from_opf(self, opf, item):
|
||||
@ -1337,8 +1443,10 @@ class OEBBook(object):
|
||||
if self.metadata.cover:
|
||||
id = str(self.metadata.cover[0])
|
||||
item = self.manifest.ids.get(id, None)
|
||||
if item is not None:
|
||||
if item is not None and item.media_type in OEB_IMAGES:
|
||||
return item
|
||||
else:
|
||||
self.logger.warn('Invalid cover image @id %r' % id)
|
||||
hcover = self.spine[0]
|
||||
if 'cover' in self.guide:
|
||||
href = self.guide['cover'].href
|
||||
@ -1376,6 +1484,7 @@ class OEBBook(object):
|
||||
self.metadata.add('cover', cover.id)
|
||||
|
||||
def _all_from_opf(self, opf):
|
||||
self.version = opf.get('version', '1.2')
|
||||
self._metadata_from_opf(opf)
|
||||
self._manifest_from_opf(opf)
|
||||
self._spine_from_opf(opf)
|
||||
@ -1408,6 +1517,8 @@ class OEBBook(object):
|
||||
except UnicodeDecodeError:
|
||||
pass
|
||||
data, _ = xml_to_unicode(data)
|
||||
data = data.replace('\r\n', '\n')
|
||||
data = data.replace('\r', '\n')
|
||||
return data
|
||||
|
||||
def to_opf1(self):
|
||||
@ -1447,7 +1558,8 @@ class OEBBook(object):
|
||||
next += 1
|
||||
selector = XPath('ncx:content/@src')
|
||||
for elem in xpath(ncx, '//*[@playOrder and ./ncx:content[@src]]'):
|
||||
order = playorder[selector(elem)[0]]
|
||||
href = selector(elem)[0]
|
||||
order = playorder.get(href, 0)
|
||||
elem.attrib['playOrder'] = str(order)
|
||||
return
|
||||
|
||||
|
||||
@ -172,9 +172,8 @@ class Stylizer(object):
|
||||
if path not in hrefs:
|
||||
return (None, None)
|
||||
data = hrefs[path].data
|
||||
data = self.oeb.decode(data)
|
||||
data = XHTML_CSS_NAMESPACE + data
|
||||
return (None, data)
|
||||
return ('utf-8', data)
|
||||
|
||||
def flatten_rule(self, rule, href, index):
|
||||
results = []
|
||||
|
||||
@ -13,13 +13,9 @@ from urlparse import urldefrag
|
||||
from lxml import etree
|
||||
import cssutils
|
||||
from calibre.ebooks.oeb.base import XPNSMAP, CSS_MIME, OEB_DOCS
|
||||
from calibre.ebooks.oeb.base import LINK_SELECTORS, CSSURL_RE
|
||||
from calibre.ebooks.oeb.base import urlnormalize
|
||||
|
||||
LINK_SELECTORS = []
|
||||
for expr in ('//h:link/@href', '//h:img/@src', '//h:object/@data',
|
||||
'//*/@xl:href'):
|
||||
LINK_SELECTORS.append(etree.XPath(expr, namespaces=XPNSMAP))
|
||||
|
||||
class ManifestTrimmer(object):
|
||||
def transform(self, oeb, context):
|
||||
oeb.logger.info('Trimming unused files from manifest...')
|
||||
@ -53,15 +49,13 @@ class ManifestTrimmer(object):
|
||||
if found not in used:
|
||||
new.add(found)
|
||||
elif item.media_type == CSS_MIME:
|
||||
def replacer(uri):
|
||||
absuri = item.abshref(urlnormalize(uri))
|
||||
if absuri in oeb.manifest.hrefs:
|
||||
for match in CSSURL_RE.finditer(item.data):
|
||||
href = match.group('url')
|
||||
href = item.abshref(urlnormalize(href))
|
||||
if href in oeb.manifest.hrefs:
|
||||
found = oeb.manifest.hrefs[href]
|
||||
if found not in used:
|
||||
new.add(found)
|
||||
return uri
|
||||
sheet = cssutils.parseString(item.data, href=item.href)
|
||||
cssutils.replaceUrls(sheet, replacer)
|
||||
used.update(new)
|
||||
unchecked = new
|
||||
for item in oeb.manifest.values():
|
||||
|
||||
@ -840,7 +840,9 @@ def _readUrl(url, fetcher=None, overrideEncoding=None, parentEncoding=None):
|
||||
|
||||
try:
|
||||
# encoding may still be wrong if encoding *is lying*!
|
||||
if content is not None:
|
||||
if isinstance(content, unicode):
|
||||
decodedCssText = content
|
||||
elif content is not None:
|
||||
decodedCssText = codecs.lookup("css")[1](content, encoding=encoding)[0]
|
||||
else:
|
||||
decodedCssText = None
|
||||
|
||||
Loading…
x
Reference in New Issue
Block a user