def get_download_filename_from_response(response):
from polyglot.urllib import unquote, urlparse
filename = last_part_name = ''
try:
purl = urlparse(response.geturl())
last_part_name = unquote(purl.path.split('/')[-1])
disposition = response.info().get('Content-disposition', '')
if isinstance(disposition, bytes):
disposition = disposition.decode('utf-8', 'replace')
for p in disposition.split(';'):
if 'filename' in p:
if '*=' in disposition:
parts = disposition.split('*=')[-1]
filename = parts.split('\'')[-1]
else:
filename = disposition.split('=')[-1]
if filename[0] in ('\'', '"'):
filename = filename[1:]
if filename[-1] in ('\'', '"'):
filename = filename[:-1]
filename = unquote(filename)
break
except Exception:
import traceback
traceback.print_exc()
return filename or last_part_name
def get_download_filename_from_response(response):
from polyglot.urllib import unquote, urlparse
filename = last_part_name = ''
try:
purl = urlparse(response.geturl())
last_part_name = unquote(purl.path.split('/')[-1])
disposition = response.info().get('Content-disposition', '')
if isinstance(disposition, bytes):
disposition = disposition.decode('utf-8', 'replace')
for p in disposition.split(';'):
if 'filename' in p:
if '*=' in disposition:
parts = disposition.split('*=')[-1]
filename = parts.split('\'')[-1]
else:
filename = disposition.split('=')[-1]
if filename[0] in ('\'', '"'):
filename = filename[1:]
if filename[-1] in ('\'', '"'):
filename = filename[:-1]
filename = unquote(filename)
break
except Exception:
import traceback
traceback.print_exc()
return filename or last_part_name
def __init__(self, href_or_path, basedir=os.getcwdu(), is_path=True):
self._href = None
self._basedir = basedir
self.path = None
self.fragment = ''
try:
self.mime_type = guess_type(href_or_path)[0]
except:
self.mime_type = None
if self.mime_type is None:
self.mime_type = 'application/octet-stream'
if is_path:
path = href_or_path
if not os.path.isabs(path):
path = os.path.abspath(os.path.join(basedir, path))
if isinstance(path, bytes):
path = path.decode(sys.getfilesystemencoding())
self.path = path
else:
url = urlparse(href_or_path)
if url[0] not in ('', 'file'):
self._href = href_or_path
else:
pc = url[2]
if isinstance(pc, unicode_type):
pc = pc.encode('utf-8')
pc = unquote(pc).decode('utf-8')
self.path = os.path.abspath(os.path.join(basedir, pc.replace('/', os.sep)))
self.fragment = unquote(url[-1])
def __init__(self, href_or_path, basedir=os.getcwdu(), is_path=True):
self._href = None
self._basedir = basedir
self.path = None
self.fragment = ''
try:
self.mime_type = guess_type(href_or_path)[0]
except:
self.mime_type = None
if self.mime_type is None:
self.mime_type = 'application/octet-stream'
if is_path:
path = href_or_path
if not os.path.isabs(path):
path = os.path.abspath(os.path.join(basedir, path))
if isinstance(path, bytes):
path = path.decode(sys.getfilesystemencoding())
self.path = path
else:
url = urlparse(href_or_path)
if url[0] not in ('', 'file'):
self._href = href_or_path
else:
pc = url[2]
if isinstance(pc, unicode_type):
pc = pc.encode('utf-8')
pc = unquote(pc).decode('utf-8')
self.path = os.path.abspath(
os.path.join(basedir, pc.replace('/', os.sep)))
self.fragment = unquote(url[-1])
def process_navpoint(np, dest):
try:
play_order = int(get_attr(np, 1))
except:
play_order = 1
href = fragment = text = None
nd = dest
nl = nl_path(np)
if nl:
nl = nl[0]
text = ''
for txt in txt_path(nl):
text += etree.tostring(txt,
method='text',
encoding='unicode',
with_tail=False)
content = content_path(np)
if content and text:
content = content[0]
# if get_attr(content, attr='src'):
purl = urlparse(content.get('src'))
href, fragment = unquote(purl[2]), unquote(purl[5])
nd = dest.add_item(href, fragment, text)
nd.play_order = play_order
for c in np_path(np):
process_navpoint(c, nd)
def process_navpoint(np, dest):
try:
play_order = int(get_attr(np, 1))
except:
play_order = 1
href = fragment = text = None
nd = dest
nl = nl_path(np)
if nl:
nl = nl[0]
text = u''
for txt in txt_path(nl):
text += etree.tostring(txt, method='text',
encoding='unicode', with_tail=False)
content = content_path(np)
if content and text:
content = content[0]
# if get_attr(content, attr='src'):
purl = urlparse(content.get('src'))
href, fragment = unquote(purl[2]), unquote(purl[5])
nd = dest.add_item(href, fragment, text)
nd.play_order = play_order
for c in np_path(np):
process_navpoint(c, nd)
def parse_uri(uri, parse_query=True):
scheme, authority, path = parse_request_uri(uri)
if path is None:
raise HTTPSimpleResponse(http_client.BAD_REQUEST, "No path component")
if b'#' in path:
raise HTTPSimpleResponse(http_client.BAD_REQUEST,
"Illegal #fragment in Request-URI.")
if scheme:
try:
scheme = scheme.decode('ascii')
except ValueError:
raise HTTPSimpleResponse(http_client.BAD_REQUEST,
'Un-decodeable scheme')
path, qs = path.partition(b'?')[::2]
if parse_query:
try:
query = MultiDict.create_from_query_string(qs)
except Exception:
raise HTTPSimpleResponse(http_client.BAD_REQUEST,
'Unparseable query string')
else:
query = None
try:
path = '%2F'.join(
unquote(x).decode('utf-8') for x in quoted_slash.split(path))
except ValueError as e:
raise HTTPSimpleResponse(http_client.BAD_REQUEST, as_unicode(e))
path = tuple(filter(None,
(x.replace('%2F', '/') for x in path.split('/'))))
return scheme, path, query
def parse_uri(uri, parse_query=True):
scheme, authority, path = parse_request_uri(uri)
if path is None:
raise HTTPSimpleResponse(http_client.BAD_REQUEST, "No path component")
if b'#' in path:
raise HTTPSimpleResponse(http_client.BAD_REQUEST, "Illegal #fragment in Request-URI.")
if scheme:
try:
scheme = scheme.decode('ascii')
except ValueError:
raise HTTPSimpleResponse(http_client.BAD_REQUEST, 'Un-decodeable scheme')
path, qs = path.partition(b'?')[::2]
if parse_query:
try:
query = MultiDict.create_from_query_string(qs)
except Exception:
raise HTTPSimpleResponse(http_client.BAD_REQUEST, 'Unparseable query string')
else:
query = None
try:
path = '%2F'.join(unquote(x).decode('utf-8') for x in quoted_slash.split(path))
except ValueError as e:
raise HTTPSimpleResponse(http_client.BAD_REQUEST, as_unicode(e))
path = tuple(filter(None, (x.replace('%2F', '/') for x in path.split('/'))))
return scheme, path, query
def dnd_get_image(md, image_exts=None):
'''
Get the image in the QMimeData object md.
:return: None, None if no image is found
QPixmap, None if an image is found, the pixmap is guaranteed not null
url, filename if a URL that points to an image is found
'''
if md.hasImage():
for x in md.formats():
x = unicode_type(x)
if x.startswith('image/'):
cdata = bytes(md.data(x))
pmap = QPixmap()
pmap.loadFromData(cdata)
if not pmap.isNull():
return pmap, None
break
if md.hasFormat('application/octet-stream'):
cdata = bytes(md.data('application/octet-stream'))
pmap = QPixmap()
pmap.loadFromData(cdata)
if not pmap.isNull():
return pmap, None
if image_exts is None:
image_exts = image_extensions()
# No image, look for an URL pointing to an image
urls = urls_from_md(md)
paths = [path_from_qurl(u) for u in urls]
# First look for a local file
images = [
xi for xi in paths
if posixpath.splitext(unquote(xi))[1][1:].lower() in image_exts
]
images = [xi for xi in images if os.path.exists(xi)]
p = QPixmap()
for path in images:
try:
with open(path, 'rb') as f:
p.loadFromData(f.read())
except Exception:
continue
if not p.isNull():
return p, None
# No local images, look for remote ones
# First, see if this is from Firefox
rurl, fname = get_firefox_rurl(md, image_exts)
if rurl and fname:
return rurl, fname
# Look through all remaining URLs
for remote_url, filename in remote_urls_from_qurl(urls, image_exts):
return remote_url, filename
return None, None
def path_from_qurl(qurl):
raw = bytes(qurl.toEncoded(
QUrl.PreferLocalFile | QUrl.RemoveScheme | QUrl.RemovePassword | QUrl.RemoveUserInfo |
QUrl.RemovePort | QUrl.RemoveAuthority | QUrl.RemoveQuery | QUrl.RemoveFragment))
ans = as_unicode_polyglot(unquote(raw), errors='replace')
if iswindows and ans.startswith('/'):
ans = ans[1:]
return ans
def dnd_get_image(md, image_exts=None):
'''
Get the image in the QMimeData object md.
:return: None, None if no image is found
QPixmap, None if an image is found, the pixmap is guaranteed not null
url, filename if a URL that points to an image is found
'''
if md.hasImage():
for x in md.formats():
x = unicode_type(x)
if x.startswith('image/'):
cdata = bytes(md.data(x))
pmap = QPixmap()
pmap.loadFromData(cdata)
if not pmap.isNull():
return pmap, None
break
if md.hasFormat('application/octet-stream'):
cdata = bytes(md.data('application/octet-stream'))
pmap = QPixmap()
pmap.loadFromData(cdata)
if not pmap.isNull():
return pmap, None
if image_exts is None:
image_exts = image_extensions()
# No image, look for an URL pointing to an image
urls = urls_from_md(md)
paths = [path_from_qurl(u) for u in urls]
# First look for a local file
images = [xi for xi in paths if
posixpath.splitext(unquote(xi))[1][1:].lower() in
image_exts]
images = [xi for xi in images if os.path.exists(xi)]
p = QPixmap()
for path in images:
try:
with open(path, 'rb') as f:
p.loadFromData(f.read())
except Exception:
continue
if not p.isNull():
return p, None
# No local images, look for remote ones
# First, see if this is from Firefox
rurl, fname = get_firefox_rurl(md, image_exts)
if rurl and fname:
return rurl, fname
# Look through all remaining URLs
for remote_url, filename in remote_urls_from_qurl(urls, image_exts):
return remote_url, filename
return None, None
def download_one(tdir, timeout, progress_report, data_uri_map, url):
try:
purl = urlparse(url)
data_url_key = None
with NamedTemporaryFile(dir=tdir, delete=False) as df:
if purl.scheme == 'file':
path = unquote(purl.path)
if iswindows and path.startswith('/'):
path = path[1:]
src = lopen(path, 'rb')
filename = os.path.basename(path)
sz = (src.seek(0, os.SEEK_END), src.tell(), src.seek(0))[1]
elif purl.scheme == 'data':
prefix, payload = purl.path.split(',', 1)
parts = prefix.split(';')
if parts and parts[-1].lower() == 'base64':
payload = re.sub(r'\s+', '', payload)
payload = from_base64_bytes(payload)
else:
payload = payload.encode('utf-8')
seen_before = data_uri_map.get(payload)
if seen_before is not None:
return True, (url, filename, seen_before,
guess_type(seen_before))
data_url_key = payload
src = BytesIO(payload)
sz = len(payload)
ext = 'unknown'
for x in parts:
if '=' not in x and '/' in x:
exts = mimetypes.guess_all_extensions(x)
if exts:
ext = exts[0]
break
filename = 'data-uri.' + ext
else:
src = urlopen(url, timeout=timeout)
filename = get_filename(purl, src)
sz = get_content_length(src)
progress_report(url, 0, sz)
dest = ProgressTracker(df, url, sz, progress_report)
with closing(src):
shutil.copyfileobj(src, dest)
if data_url_key is not None:
data_uri_map[data_url_key] = dest.name
filename = sanitize_file_name(filename)
mt = guess_type(filename)
if mt in OEB_DOCS:
raise ValueError(
'The external resource {} looks like a HTML document ({})'.
format(url, filename))
if not mt or mt == 'application/octet-stream' or '.' not in filename:
raise ValueError(
'The external resource {} is not of a known type'.format(
url))
return True, (url, filename, dest.name, mt)
except Exception as err:
return False, (url, as_unicode(err))
def _cover_from_html(self, hcover):
from calibre.ebooks import render_html_svg_workaround
with TemporaryDirectory('_html_cover') as tdir:
writer = OEBWriter()
writer(self.oeb, tdir)
path = os.path.join(tdir, unquote(hcover.href))
data = render_html_svg_workaround(path, self.logger)
if not data:
data = b''
id, href = self.oeb.manifest.generate('cover', 'cover.jpg')
item = self.oeb.manifest.add(id, href, JPEG_MIME, data=data)
return item
def add_links(self):
for link in self.links:
path, href, frag = link[0]
page, rect = link[1:]
combined_path = os.path.normcase(
os.path.abspath(
os.path.join(os.path.dirname(path),
*unquote(href).split('/'))))
is_local = not href or combined_path in self.anchors
annot = Dictionary({
'Type': Name('Annot'),
'Subtype': Name('Link'),
'Rect': rect,
'Border': Array([0, 0, 0]),
})
if self.mark_links:
annot.update({
'Border': Array([16, 16, 1]),
'C': Array([1.0, 0, 0])
})
if is_local:
path = combined_path if href else path
try:
annot['Dest'] = self.anchors[path][frag]
except KeyError:
try:
annot['Dest'] = self.anchors[path][None]
except KeyError:
pass
else:
url = href + (('#' + frag) if frag else '')
try:
purl = urlparse(url)
except Exception:
self.pdf.debug('Ignoring unparseable URL: %r' % url)
continue
if purl.scheme and purl.scheme != 'file':
action = Dictionary({
'Type': Name('Action'),
'S': Name('URI'),
})
# Do not try to normalize/quote/unquote this URL as if it
# has a query part, it will get corrupted
action['URI'] = String(url)
annot['A'] = action
if 'A' in annot or 'Dest' in annot:
if 'Annots' not in page:
page['Annots'] = Array()
page['Annots'].append(self.pdf.objects.add(annot))
else:
self.pdf.debug(
'Could not find destination for link: %s in file %s' %
(href, path))
def _cover_from_html(self, hcover):
from calibre.ebooks import render_html_svg_workaround
with TemporaryDirectory('_html_cover') as tdir:
writer = OEBWriter()
writer(self.oeb, tdir)
path = os.path.join(tdir, unquote(hcover.href))
data = render_html_svg_workaround(path, self.logger)
if not data:
data = ''
id, href = self.oeb.manifest.generate('cover', 'cover.jpg')
item = self.oeb.manifest.add(id, href, JPEG_MIME, data=data)
return item
def convert(self, oeb_book, output_path, input_plugin, opts, log):
from polyglot.urllib import unquote
from lxml import etree
self.log, self.opts = log, opts
if not os.path.exists(output_path):
os.makedirs(output_path)
from calibre.ebooks.oeb.base import OPF_MIME, NCX_MIME, PAGE_MAP_MIME, OEB_STYLES
from calibre.ebooks.oeb.normalize_css import condense_sheet
with CurrentDir(output_path):
results = oeb_book.to_opf2(page_map=True)
for key in (OPF_MIME, NCX_MIME, PAGE_MAP_MIME):
href, root = results.pop(key, [None, None])
if root is not None:
if key == OPF_MIME:
try:
self.workaround_nook_cover_bug(root)
except:
self.log.exception(
'Something went wrong while trying to'
' workaround Nook cover bug, ignoring')
try:
self.workaround_pocketbook_cover_bug(root)
except:
self.log.exception(
'Something went wrong while trying to'
' workaround Pocketbook cover bug, ignoring')
self.migrate_lang_code(root)
self.adjust_mime_types(root)
raw = etree.tostring(root,
pretty_print=True,
encoding='utf-8',
xml_declaration=True)
if key == OPF_MIME:
# Needed as I can't get lxml to output opf:role and
# not output <opf:metadata> as well
raw = re.sub(br'(<[/]{0,1})opf:', br'\1', raw)
with lopen(href, 'wb') as f:
f.write(raw)
for item in oeb_book.manifest:
if (not self.opts.expand_css and item.media_type in OEB_STYLES
and hasattr(item.data, 'cssText')
and 'nook' not in self.opts.output_profile.short_name):
condense_sheet(item.data)
path = os.path.abspath(unquote(item.href))
dir = os.path.dirname(path)
if not os.path.exists(dir):
os.makedirs(dir)
with lopen(path, 'wb') as f:
f.write(item.bytes_representation)
item.unload_data_from_memory(memory=path)
def get_toc(self):
self.stream.seek(24)
toc_offset = self.read_i32()
self.stream.seek(toc_offset)
pages = self.read_i32()
toc = RBToc()
for i in range(pages):
name = unquote(self.stream.read(32).strip(b'\x00'))
size, offset, flags = self.read_i32(), self.read_i32(), self.read_i32()
toc.append(RBToc.Item(name=name, size=size, offset=offset, flags=flags))
return toc
def get_toc(self):
self.stream.seek(24)
toc_offset = self.read_i32()
self.stream.seek(toc_offset)
pages = self.read_i32()
toc = RBToc()
for i in range(pages):
name = unquote(self.stream.read(32).strip(b'\x00'))
size, offset, flags = self.read_i32(), self.read_i32(), self.read_i32()
toc.append(RBToc.Item(name=name, size=size, offset=offset, flags=flags))
return toc
def read_from_opf(self, opfreader):
toc = opfreader.soup.find('spine', toc=True)
if toc is not None:
toc = toc['toc']
if toc is None:
try:
toc = opfreader.soup.find('guide').find('reference',
attrs={'type':
'toc'})['href']
except:
for item in opfreader.manifest:
if 'toc' in item.href().lower():
toc = item.href()
break
if toc is not None:
if toc.lower() not in ('ncx', 'ncxtoc'):
toc = urlparse(unquote(toc))[2]
toc = toc.replace('/', os.sep)
if not os.path.isabs(toc):
toc = os.path.join(self.base_path, toc)
try:
if not os.path.exists(toc):
bn = os.path.basename(toc)
bn = bn.replace('_top.htm',
'_toc.htm') # Bug in BAEN OPF files
toc = os.path.join(os.path.dirname(toc), bn)
self.read_html_toc(toc)
except:
print(
'WARNING: Could not read Table of Contents. Continuing anyway.'
)
else:
path = opfreader.manifest.item(toc.lower())
path = getattr(path, 'path', path)
if path and os.access(path, os.R_OK):
try:
self.read_ncx_toc(path)
except Exception as err:
print('WARNING: Invalid NCX file:', err)
return
cwd = os.path.abspath(self.base_path)
m = glob.glob(os.path.join(cwd, '*.ncx'))
if m:
toc = m[0]
self.read_ncx_toc(toc)
def convert(self, oeb_book, output_path, input_plugin, opts, log):
from polyglot.urllib import unquote
from lxml import etree
self.log, self.opts = log, opts
if not os.path.exists(output_path):
os.makedirs(output_path)
from calibre.ebooks.oeb.base import OPF_MIME, NCX_MIME, PAGE_MAP_MIME, OEB_STYLES
from calibre.ebooks.oeb.normalize_css import condense_sheet
with CurrentDir(output_path):
results = oeb_book.to_opf2(page_map=True)
for key in (OPF_MIME, NCX_MIME, PAGE_MAP_MIME):
href, root = results.pop(key, [None, None])
if root is not None:
if key == OPF_MIME:
try:
self.workaround_nook_cover_bug(root)
except:
self.log.exception('Something went wrong while trying to'
' workaround Nook cover bug, ignoring')
try:
self.workaround_pocketbook_cover_bug(root)
except:
self.log.exception('Something went wrong while trying to'
' workaround Pocketbook cover bug, ignoring')
self.migrate_lang_code(root)
raw = etree.tostring(root, pretty_print=True,
encoding='utf-8', xml_declaration=True)
if key == OPF_MIME:
# Needed as I can't get lxml to output opf:role and
# not output <opf:metadata> as well
raw = re.sub(br'(<[/]{0,1})opf:', br'\1', raw)
with open(href, 'wb') as f:
f.write(raw)
for item in oeb_book.manifest:
if (
not self.opts.expand_css and item.media_type in OEB_STYLES and hasattr(
item.data, 'cssText') and 'nook' not in self.opts.output_profile.short_name):
condense_sheet(item.data)
path = os.path.abspath(unquote(item.href))
dir = os.path.dirname(path)
if not os.path.exists(dir):
os.makedirs(dir)
with open(path, 'wb') as f:
f.write(item.bytes_representation)
item.unload_data_from_memory(memory=path)
def insert_cover(self):
from calibre.ebooks.oeb.base import urldefrag
g, m = self.oeb.guide, self.oeb.manifest
item = None
if 'titlepage' not in g:
if 'cover' in g:
href = g['cover'].href
else:
href = self.default_cover()
if href is None:
return
width, height = self.inspect_cover(href)
if width == -1 or height == -1:
self.log.warning('Failed to read cover dimensions')
width, height = 600, 800
# if self.preserve_aspect_ratio:
# width, height = 600, 800
self.svg_template = self.svg_template.replace(
'__viewbox__', '0 0 %d %d' % (width, height))
self.svg_template = self.svg_template.replace(
'__width__', unicode_type(width))
self.svg_template = self.svg_template.replace(
'__height__', unicode_type(height))
if href is not None:
templ = self.non_svg_template if self.no_svg_cover \
else self.svg_template
tp = templ % unquote(href)
id, href = m.generate('titlepage', 'titlepage.xhtml')
item = m.add(id,
href,
guess_type('t.xhtml')[0],
data=etree.fromstring(tp))
else:
item = self.oeb.manifest.hrefs[urldefrag(
self.oeb.guide['titlepage'].href)[0]]
if item is not None:
self.oeb.spine.insert(0, item, True)
if 'cover' not in self.oeb.guide.refs:
self.oeb.guide.add('cover', 'Title Page', 'a')
self.oeb.guide.refs['cover'].href = item.href
if 'titlepage' in self.oeb.guide.refs:
self.oeb.guide.refs['titlepage'].href = item.href
titem = getattr(self.oeb.toc, 'item_that_refers_to_cover', None)
if titem is not None:
titem.href = item.href
def parse_html_toc(data):
from html5_parser import parse
from calibre.utils.cleantext import clean_xml_chars
from lxml import etree
if isinstance(data, bytes):
data = xml_to_unicode(data, strip_encoding_pats=True, resolve_entities=True)[0]
root = parse(clean_xml_chars(data), maybe_xhtml=True, keep_doctype=False, sanitize_names=True)
for a in root.xpath('//*[@href and local-name()="a"]'):
purl = urlparse(unquote(a.get('href')))
href, fragment = purl[2], purl[5]
if not fragment:
fragment = None
else:
fragment = fragment.strip()
href = href.strip()
txt = etree.tostring(a, method='text', encoding='unicode')
yield href, fragment, txt
def insert_cover(self):
from calibre.ebooks.oeb.base import urldefrag
g, m = self.oeb.guide, self.oeb.manifest
item = None
if 'titlepage' not in g:
if 'cover' in g:
href = g['cover'].href
else:
href = self.default_cover()
if href is None:
return
width, height = self.inspect_cover(href)
if width == -1 or height == -1:
self.log.warning('Failed to read cover dimensions')
width, height = 600, 800
# if self.preserve_aspect_ratio:
# width, height = 600, 800
self.svg_template = self.svg_template.replace('__viewbox__',
'0 0 %d %d'%(width, height))
self.svg_template = self.svg_template.replace('__width__',
str(width))
self.svg_template = self.svg_template.replace('__height__',
str(height))
if href is not None:
templ = self.non_svg_template if self.no_svg_cover \
else self.svg_template
tp = templ%unquote(href)
id, href = m.generate('titlepage', u'titlepage.xhtml')
item = m.add(id, href, guess_type('t.xhtml')[0],
data=etree.fromstring(tp))
else:
item = self.oeb.manifest.hrefs[
urldefrag(self.oeb.guide['titlepage'].href)[0]]
if item is not None:
self.oeb.spine.insert(0, item, True)
if 'cover' not in self.oeb.guide.refs:
self.oeb.guide.add('cover', 'Title Page', 'a')
self.oeb.guide.refs['cover'].href = item.href
if 'titlepage' in self.oeb.guide.refs:
self.oeb.guide.refs['titlepage'].href = item.href
titem = getattr(self.oeb.toc, 'item_that_refers_to_cover', None)
if titem is not None:
titem.href = item.href
def build_node(current_node, parent=None):
if parent is None:
parent = etree.Element('ul')
elif len(current_node.nodes):
parent = element(parent, ('ul'))
for node in current_node.nodes:
point = element(parent, 'li')
href = relpath(abspath(unquote(node.href)), dirname(ref_url))
if isinstance(href, bytes):
href = href.decode('utf-8')
link = element(point, 'a', href=clean_xml_chars(href))
title = node.title
if isinstance(title, bytes):
title = title.decode('utf-8')
if title:
title = re.sub(r'\s+', ' ', title)
link.text = clean_xml_chars(title)
build_node(node, point)
return parent
def build_node(current_node, parent=None):
if parent is None:
parent = etree.Element('ul')
elif len(current_node.nodes):
parent = element(parent, ('ul'))
for node in current_node.nodes:
point = element(parent, 'li')
href = relpath(abspath(unquote(node.href)), dirname(ref_url))
if isinstance(href, bytes):
href = href.decode('utf-8')
link = element(point, 'a', href=clean_xml_chars(href))
title = node.title
if isinstance(title, bytes):
title = title.decode('utf-8')
if title:
title = re.sub(r'\s+', ' ', title)
link.text = clean_xml_chars(title)
build_node(node, point)
return parent
def read_from_opf(self, opfreader):
toc = opfreader.soup.find('spine', toc=True)
if toc is not None:
toc = toc['toc']
if toc is None:
try:
toc = opfreader.soup.find('guide').find('reference', attrs={'type':'toc'})['href']
except:
for item in opfreader.manifest:
if 'toc' in item.href().lower():
toc = item.href()
break
if toc is not None:
if toc.lower() not in ('ncx', 'ncxtoc'):
toc = urlparse(unquote(toc))[2]
toc = toc.replace('/', os.sep)
if not os.path.isabs(toc):
toc = os.path.join(self.base_path, toc)
try:
if not os.path.exists(toc):
bn = os.path.basename(toc)
bn = bn.replace('_top.htm', '_toc.htm') # Bug in BAEN OPF files
toc = os.path.join(os.path.dirname(toc), bn)
self.read_html_toc(toc)
except:
print('WARNING: Could not read Table of Contents. Continuing anyway.')
else:
path = opfreader.manifest.item(toc.lower())
path = getattr(path, 'path', path)
if path and os.access(path, os.R_OK):
try:
self.read_ncx_toc(path)
except Exception as err:
print('WARNING: Invalid NCX file:', err)
return
cwd = os.path.abspath(self.base_path)
m = glob.glob(os.path.join(cwd, '*.ncx'))
if m:
toc = m[0]
self.read_ncx_toc(toc)
def _build_manifest(self):
states = ['linear', 'nonlinear', 'css', 'images']
manifest = dict((state, []) for state in states)
for item in self._oeb.manifest.values():
if item.spine_position is not None:
key = 'linear' if item.linear else 'nonlinear'
manifest[key].append(item)
elif item.media_type in OEB_STYLES:
manifest['css'].append(item)
elif item.media_type in LIT_IMAGES:
manifest['images'].append(item)
data = io.BytesIO()
data.write(pack('<Bc', 1, b'\\'))
offset = 0
for state in states:
items = sorted(manifest[state], key=attrgetter('sort_key'))
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 = unquote(item.href)
item.offset = offset \
if state in ('linear', 'nonlinear') else 0
data.write(pack('<I', item.offset))
entry = [
codepoint_to_chr(len(id)),
unicode_type(id),
codepoint_to_chr(len(href)),
unicode_type(href),
codepoint_to_chr(len(media_type)),
unicode_type(media_type)
]
for value in entry:
data.write(value.encode('utf-8'))
data.write(b'\0')
offset += item.size
self._add_file('/manifest', data.getvalue())
def dnd_get_local_image_and_pixmap(md, image_exts=None):
if md.hasImage():
for x in md.formats():
x = unicode_type(x)
if x.startswith('image/'):
cdata = bytes(md.data(x))
pmap = QPixmap()
pmap.loadFromData(cdata)
if not pmap.isNull():
return pmap, cdata
if md.hasFormat('application/octet-stream'):
cdata = bytes(md.data('application/octet-stream'))
pmap = QPixmap()
pmap.loadFromData(cdata)
if not pmap.isNull():
return pmap, cdata
if image_exts is None:
image_exts = image_extensions()
# No image, look for an URL pointing to an image
urls = urls_from_md(md)
paths = [path_from_qurl(u) for u in urls]
# Look for a local file
images = [
xi for xi in paths
if posixpath.splitext(unquote(xi))[1][1:].lower() in image_exts
]
images = [xi for xi in images if os.path.exists(xi)]
for path in images:
try:
with open(path, 'rb') as f:
cdata = f.read()
except Exception:
continue
p = QPixmap()
p.loadFromData(cdata)
if not p.isNull():
return p, cdata
return None, None
def rewrite_links(self, url):
href, frag = urldefrag(url)
try:
href = self.current_item.abshref(href)
except ValueError:
# Unparsable URL
return url
try:
href = urlnormalize(href)
except ValueError:
# href has non utf-8 quoting
return url
if href in self.map:
anchor_map = self.map[href]
nhref = anchor_map[frag if frag else None]
nhref = self.current_item.relhref(nhref)
if frag:
nhref = '#'.join((unquote(nhref), frag))
return nhref
return url
def dnd_get_files(md, exts, allow_all_extensions=False, filter_exts=()):
'''
Get the file in the QMimeData object md with an extension that is one of
the extensions in exts.
:return: None, None if no file is found
[paths], None if a local file is found
[urls], [filenames] if URLs that point to a files are found
'''
# Look for a URL pointing to a file
urls = urls_from_md(md)
# First look for a local file
local_files = [path_from_qurl(x) for x in urls]
def is_ok(path):
ext = posixpath.splitext(path)[1][1:].lower()
if allow_all_extensions and ext and ext not in filter_exts:
return True
return ext in exts and ext not in filter_exts
local_files = [p for p in local_files if is_ok(unquote(p))]
local_files = [x for x in local_files if os.path.exists(x)]
if local_files:
return local_files, None
# No local files, look for remote ones
# First, see if this is from Firefox
rurl, fname = get_firefox_rurl(md, exts)
if rurl and fname:
return [rurl], [fname]
# Look through all remaining URLs
rurls, filenames = [], []
for rurl, fname in remote_urls_from_qurl(urls, exts):
rurls.append(rurl), filenames.append(fname)
if rurls:
return rurls, filenames
return None, None
def dnd_get_files(md, exts, allow_all_extensions=False, filter_exts=()):
'''
Get the file in the QMimeData object md with an extension that is one of
the extensions in exts.
:return: None, None if no file is found
[paths], None if a local file is found
[urls], [filenames] if URLs that point to a files are found
'''
# Look for a URL pointing to a file
urls = urls_from_md(md)
# First look for a local file
local_files = [path_from_qurl(x) for x in urls]
def is_ok(path):
ext = posixpath.splitext(path)[1][1:].lower()
if allow_all_extensions and ext and ext not in filter_exts:
return True
return ext in exts and ext not in filter_exts
local_files = [p for p in local_files if is_ok(unquote(p))]
local_files = [x for x in local_files if os.path.exists(x)]
if local_files:
return local_files, None
# No local files, look for remote ones
# First, see if this is from Firefox
rurl, fname = get_firefox_rurl(md, exts)
if rurl and fname:
return [rurl], [fname]
# Look through all remaining URLs
rurls, filenames = [], []
for rurl, fname in remote_urls_from_qurl(urls, exts):
rurls.append(rurl), filenames.append(fname)
if rurls:
return rurls, filenames
return None, None
def read_html_toc(self, toc):
self.base_path = os.path.dirname(toc)
soup = BeautifulSoup(open(toc, 'rb').read(), convertEntities=BeautifulSoup.HTML_ENTITIES)
for a in soup.findAll('a'):
if not a.has_key('href'): # noqa
continue
purl = urlparse(unquote(a['href']))
href, fragment = purl[2], purl[5]
if not fragment:
fragment = None
else:
fragment = fragment.strip()
href = href.strip()
txt = ''.join([unicode_type(s).strip() for s in a.findAll(text=True)])
add = True
for i in self.flat():
if i.href == href and i.fragment == fragment:
add = False
break
if add:
self.add_item(href, fragment, txt)
def __init__(self, title=_('Choose Files'),
filters=[],
add_all_files_filter=True,
parent=None,
modal=True,
name='',
mode=QFileDialog.ExistingFiles,
default_dir=u'~',
no_save_dir=False,
combine_file_and_saved_dir=False
):
from calibre.gui2 import dynamic, sanitize_env_vars
QObject.__init__(self)
ftext = ''
if filters:
for filter in filters:
text, extensions = filter
extensions = ['*'+(i if i.startswith('.') else '.'+i) for i in
extensions]
ftext += '%s (%s);;'%(text, ' '.join(extensions))
if add_all_files_filter or not ftext:
ftext += 'All files (*)'
if ftext.endswith(';;'):
ftext = ftext[:-2]
self.dialog_name = dialog_name(name, title)
self.selected_files = None
self.fd = None
if combine_file_and_saved_dir:
bn = os.path.basename(default_dir)
prev = dynamic.get(self.dialog_name,
expanduser(u'~'))
if os.path.exists(prev):
if os.path.isfile(prev):
prev = os.path.dirname(prev)
else:
prev = expanduser(u'~')
initial_dir = os.path.join(prev, bn)
elif no_save_dir:
initial_dir = expanduser(default_dir)
else:
initial_dir = dynamic.get(self.dialog_name,
expanduser(default_dir))
if not isinstance(initial_dir, string_or_bytes):
initial_dir = expanduser(default_dir)
if not initial_dir or (not os.path.exists(initial_dir) and not (
mode == QFileDialog.AnyFile and (no_save_dir or combine_file_and_saved_dir))):
initial_dir = select_initial_dir(initial_dir)
self.selected_files = []
use_native_dialog = 'CALIBRE_NO_NATIVE_FILEDIALOGS' not in os.environ
with sanitize_env_vars():
opts = QFileDialog.Option()
if not use_native_dialog:
opts |= QFileDialog.DontUseNativeDialog
if mode == QFileDialog.AnyFile:
f = QFileDialog.getSaveFileName(parent, title,
initial_dir, ftext, "", opts)
if f and f[0]:
self.selected_files.append(f[0])
elif mode == QFileDialog.ExistingFile:
f = QFileDialog.getOpenFileName(parent, title,
initial_dir, ftext, "", opts)
if f and f[0] and os.path.exists(f[0]):
self.selected_files.append(f[0])
elif mode == QFileDialog.ExistingFiles:
fs = QFileDialog.getOpenFileNames(parent, title, initial_dir,
ftext, "", opts)
if fs and fs[0]:
for f in fs[0]:
f = unicode_type(f)
if not f:
continue
if not os.path.exists(f):
# QFileDialog for some reason quotes spaces
# on linux if there is more than one space in a row
f = unquote(f)
if f and os.path.exists(f):
self.selected_files.append(f)
else:
if mode == QFileDialog.Directory:
opts |= QFileDialog.ShowDirsOnly
f = unicode_type(QFileDialog.getExistingDirectory(parent, title, initial_dir, opts))
if os.path.exists(f):
self.selected_files.append(f)
if self.selected_files:
self.selected_files = [unicode_type(q) for q in self.selected_files]
saved_loc = self.selected_files[0]
if os.path.isfile(saved_loc):
saved_loc = os.path.dirname(saved_loc)
if not no_save_dir:
dynamic[self.dialog_name] = saved_loc
self.accepted = bool(self.selected_files)
def __init__(
self, title=_('Choose Files'),
filters=[],
add_all_files_filter=True,
parent=None,
modal=True,
name='',
mode=QFileDialog.ExistingFiles,
default_dir=u'~',
no_save_dir=False,
combine_file_and_saved_dir=False
):
from calibre.gui2 import dynamic, sanitize_env_vars
from calibre.gui2.ui import get_gui
gui = get_gui()
adapt_menubar = gui.bars_manager.adapt_menu_bar_for_dialog if gui is not None else Dummy()
QObject.__init__(self)
ftext = ''
if filters:
for filter in filters:
text, extensions = filter
extensions = ['*'+(i if i.startswith('.') else '.'+i) for i in
extensions]
ftext += '%s (%s);;'%(text, ' '.join(extensions))
if add_all_files_filter or not ftext:
ftext += 'All files (*)'
if ftext.endswith(';;'):
ftext = ftext[:-2]
self.dialog_name = dialog_name(name, title)
self.selected_files = None
self.fd = None
if combine_file_and_saved_dir:
bn = os.path.basename(default_dir)
prev = dynamic.get(self.dialog_name,
os.path.expanduser(u'~'))
if os.path.exists(prev):
if os.path.isfile(prev):
prev = os.path.dirname(prev)
else:
prev = os.path.expanduser(u'~')
initial_dir = os.path.join(prev, bn)
elif no_save_dir:
initial_dir = os.path.expanduser(default_dir)
else:
initial_dir = dynamic.get(self.dialog_name,
os.path.expanduser(default_dir))
if not isinstance(initial_dir, string_or_bytes):
initial_dir = os.path.expanduser(default_dir)
if not initial_dir or (not os.path.exists(initial_dir) and not (
mode == QFileDialog.AnyFile and (no_save_dir or combine_file_and_saved_dir))):
initial_dir = select_initial_dir(initial_dir)
self.selected_files = []
use_native_dialog = 'CALIBRE_NO_NATIVE_FILEDIALOGS' not in os.environ
with sanitize_env_vars():
opts = QFileDialog.Option()
if not use_native_dialog:
opts |= QFileDialog.DontUseNativeDialog
if mode == QFileDialog.AnyFile:
with adapt_menubar:
f = QFileDialog.getSaveFileName(parent, title,
initial_dir, ftext, "", opts)
if f and f[0]:
self.selected_files.append(f[0])
elif mode == QFileDialog.ExistingFile:
with adapt_menubar:
f = QFileDialog.getOpenFileName(parent, title,
initial_dir, ftext, "", opts)
if f and f[0] and os.path.exists(f[0]):
self.selected_files.append(f[0])
elif mode == QFileDialog.ExistingFiles:
with adapt_menubar:
fs = QFileDialog.getOpenFileNames(parent, title, initial_dir,
ftext, "", opts)
if fs and fs[0]:
for f in fs[0]:
f = unicode_type(f)
if not f:
continue
if not os.path.exists(f):
# QFileDialog for some reason quotes spaces
# on linux if there is more than one space in a row
f = unquote(f)
if f and os.path.exists(f):
self.selected_files.append(f)
else:
if mode == QFileDialog.Directory:
opts |= QFileDialog.ShowDirsOnly
with adapt_menubar:
f = unicode_type(QFileDialog.getExistingDirectory(parent, title, initial_dir, opts))
if os.path.exists(f):
self.selected_files.append(f)
if self.selected_files:
self.selected_files = [unicode_type(q) for q in self.selected_files]
saved_loc = self.selected_files[0]
if os.path.isfile(saved_loc):
saved_loc = os.path.dirname(saved_loc)
if not no_save_dir:
dynamic[self.dialog_name] = saved_loc
self.accepted = bool(self.selected_files)
def convert(self, oeb_book, output_path, input_plugin, opts, log):
from lxml import etree
from calibre.utils import zipfile
from templite import Templite
from polyglot.urllib import unquote
from calibre.ebooks.html.meta import EasyMeta
# read template files
if opts.template_html_index is not None:
template_html_index_data = open(opts.template_html_index, 'rb').read()
else:
template_html_index_data = P('templates/html_export_default_index.tmpl', data=True)
if opts.template_html is not None:
template_html_data = open(opts.template_html, 'rb').read()
else:
template_html_data = P('templates/html_export_default.tmpl', data=True)
if opts.template_css is not None:
template_css_data = open(opts.template_css, 'rb').read()
else:
template_css_data = P('templates/html_export_default.css', data=True)
template_html_index_data = template_html_index_data.decode('utf-8')
template_html_data = template_html_data.decode('utf-8')
template_css_data = template_css_data.decode('utf-8')
self.log = log
self.opts = opts
meta = EasyMeta(oeb_book.metadata)
tempdir = os.path.realpath(PersistentTemporaryDirectory())
output_file = os.path.join(tempdir,
basename(re.sub(r'\.zip', '', output_path)+'.html'))
output_dir = re.sub(r'\.html', '', output_file)+'_files'
if not exists(output_dir):
os.makedirs(output_dir)
css_path = output_dir+os.sep+'calibreHtmlOutBasicCss.css'
with open(css_path, 'wb') as f:
f.write(template_css_data.encode('utf-8'))
with open(output_file, 'wb') as f:
html_toc = self.generate_html_toc(oeb_book, output_file, output_dir)
templite = Templite(template_html_index_data)
nextLink = oeb_book.spine[0].href
nextLink = relpath(output_dir+os.sep+nextLink, dirname(output_file))
cssLink = relpath(abspath(css_path), dirname(output_file))
tocUrl = relpath(output_file, dirname(output_file))
t = templite.render(has_toc=bool(oeb_book.toc.count()),
toc=html_toc, meta=meta, nextLink=nextLink,
tocUrl=tocUrl, cssLink=cssLink,
firstContentPageLink=nextLink)
if isinstance(t, unicode_type):
t = t.encode('utf-8')
f.write(t)
with CurrentDir(output_dir):
for item in oeb_book.manifest:
path = abspath(unquote(item.href))
dir = dirname(path)
if not exists(dir):
os.makedirs(dir)
if item.spine_position is not None:
with open(path, 'wb') as f:
pass
else:
with open(path, 'wb') as f:
f.write(item.bytes_representation)
item.unload_data_from_memory(memory=path)
for item in oeb_book.spine:
path = abspath(unquote(item.href))
dir = dirname(path)
root = item.data.getroottree()
# get & clean HTML <HEAD>-data
head = root.xpath('//h:head', namespaces={'h': 'http://www.w3.org/1999/xhtml'})[0]
head_content = etree.tostring(head, pretty_print=True, encoding='unicode')
head_content = re.sub(r'\<\/?head.*\>', '', head_content)
head_content = re.sub(re.compile(r'\<style.*\/style\>', re.M|re.S), '', head_content)
head_content = re.sub(r'<(title)([^>]*)/>', r'<\1\2></\1>', head_content)
# get & clean HTML <BODY>-data
body = root.xpath('//h:body', namespaces={'h': 'http://www.w3.org/1999/xhtml'})[0]
ebook_content = etree.tostring(body, pretty_print=True, encoding='unicode')
ebook_content = re.sub(r'\<\/?body.*\>', '', ebook_content)
ebook_content = re.sub(r'<(div|a|span)([^>]*)/>', r'<\1\2></\1>', ebook_content)
# generate link to next page
if item.spine_position+1 < len(oeb_book.spine):
nextLink = oeb_book.spine[item.spine_position+1].href
nextLink = relpath(abspath(nextLink), dir)
else:
nextLink = None
# generate link to previous page
if item.spine_position > 0:
prevLink = oeb_book.spine[item.spine_position-1].href
prevLink = relpath(abspath(prevLink), dir)
else:
prevLink = None
cssLink = relpath(abspath(css_path), dir)
tocUrl = relpath(output_file, dir)
firstContentPageLink = oeb_book.spine[0].href
# render template
templite = Templite(template_html_data)
toc = lambda: self.generate_html_toc(oeb_book, path, output_dir)
t = templite.render(ebookContent=ebook_content,
prevLink=prevLink, nextLink=nextLink,
has_toc=bool(oeb_book.toc.count()), toc=toc,
tocUrl=tocUrl, head_content=head_content,
meta=meta, cssLink=cssLink,
firstContentPageLink=firstContentPageLink)
# write html to file
with open(path, 'wb') as f:
f.write(t.encode('utf-8'))
item.unload_data_from_memory(memory=path)
zfile = zipfile.ZipFile(output_path, "w")
zfile.add_dir(output_dir, basename(output_dir))
zfile.write(output_file, basename(output_file), zipfile.ZIP_DEFLATED)
if opts.extract_to:
if os.path.exists(opts.extract_to):
shutil.rmtree(opts.extract_to)
os.makedirs(opts.extract_to)
zfile.extractall(opts.extract_to)
self.log('Zip file extracted to', opts.extract_to)
zfile.close()
# cleanup temp dir
shutil.rmtree(tempdir)
def convert(self, oeb_book, output_path, input_plugin, opts, log):
from lxml import etree
from calibre.utils import zipfile
from templite import Templite
from polyglot.urllib import unquote
from calibre.ebooks.html.meta import EasyMeta
# read template files
if opts.template_html_index is not None:
template_html_index_data = open(opts.template_html_index,
'rb').read()
else:
template_html_index_data = P(
'templates/html_export_default_index.tmpl', data=True)
if opts.template_html is not None:
template_html_data = open(opts.template_html, 'rb').read()
else:
template_html_data = P('templates/html_export_default.tmpl',
data=True)
if opts.template_css is not None:
template_css_data = open(opts.template_css, 'rb').read()
else:
template_css_data = P('templates/html_export_default.css',
data=True)
template_html_index_data = template_html_index_data.decode('utf-8')
template_html_data = template_html_data.decode('utf-8')
template_css_data = template_css_data.decode('utf-8')
self.log = log
self.opts = opts
meta = EasyMeta(oeb_book.metadata)
tempdir = os.path.realpath(PersistentTemporaryDirectory())
output_file = os.path.join(
tempdir, basename(re.sub(r'\.zip', '', output_path) + '.html'))
output_dir = re.sub(r'\.html', '', output_file) + '_files'
if not exists(output_dir):
os.makedirs(output_dir)
css_path = output_dir + os.sep + 'calibreHtmlOutBasicCss.css'
with open(css_path, 'wb') as f:
f.write(template_css_data.encode('utf-8'))
with open(output_file, 'wb') as f:
html_toc = self.generate_html_toc(oeb_book, output_file,
output_dir)
templite = Templite(template_html_index_data)
nextLink = oeb_book.spine[0].href
nextLink = relpath(output_dir + os.sep + nextLink,
dirname(output_file))
cssLink = relpath(abspath(css_path), dirname(output_file))
tocUrl = relpath(output_file, dirname(output_file))
t = templite.render(has_toc=bool(oeb_book.toc.count()),
toc=html_toc,
meta=meta,
nextLink=nextLink,
tocUrl=tocUrl,
cssLink=cssLink,
firstContentPageLink=nextLink)
if isinstance(t, unicode_type):
t = t.encode('utf-8')
f.write(t)
with CurrentDir(output_dir):
for item in oeb_book.manifest:
path = abspath(unquote(item.href))
dir = dirname(path)
if not exists(dir):
os.makedirs(dir)
if item.spine_position is not None:
with open(path, 'wb') as f:
pass
else:
with open(path, 'wb') as f:
f.write(str(item))
item.unload_data_from_memory(memory=path)
for item in oeb_book.spine:
path = abspath(unquote(item.href))
dir = dirname(path)
root = item.data.getroottree()
# get & clean HTML <HEAD>-data
head = root.xpath(
'//h:head',
namespaces={'h': 'http://www.w3.org/1999/xhtml'})[0]
head_content = etree.tostring(head,
pretty_print=True,
encoding='utf-8')
head_content = re.sub(r'\<\/?head.*\>', '', head_content)
head_content = re.sub(
re.compile(r'\<style.*\/style\>', re.M | re.S), '',
head_content)
head_content = re.sub(r'<(title)([^>]*)/>', r'<\1\2></\1>',
head_content)
# get & clean HTML <BODY>-data
body = root.xpath(
'//h:body',
namespaces={'h': 'http://www.w3.org/1999/xhtml'})[0]
ebook_content = etree.tostring(body,
pretty_print=True,
encoding='utf-8')
ebook_content = re.sub(r'\<\/?body.*\>', '', ebook_content)
ebook_content = re.sub(r'<(div|a|span)([^>]*)/>',
r'<\1\2></\1>', ebook_content)
# generate link to next page
if item.spine_position + 1 < len(oeb_book.spine):
nextLink = oeb_book.spine[item.spine_position + 1].href
nextLink = relpath(abspath(nextLink), dir)
else:
nextLink = None
# generate link to previous page
if item.spine_position > 0:
prevLink = oeb_book.spine[item.spine_position - 1].href
prevLink = relpath(abspath(prevLink), dir)
else:
prevLink = None
cssLink = relpath(abspath(css_path), dir)
tocUrl = relpath(output_file, dir)
firstContentPageLink = oeb_book.spine[0].href
# render template
templite = Templite(template_html_data)
toc = lambda: self.generate_html_toc(oeb_book, path, output_dir
)
t = templite.render(ebookContent=ebook_content,
prevLink=prevLink,
nextLink=nextLink,
has_toc=bool(oeb_book.toc.count()),
toc=toc,
tocUrl=tocUrl,
head_content=head_content,
meta=meta,
cssLink=cssLink,
firstContentPageLink=firstContentPageLink)
# write html to file
with open(path, 'wb') as f:
f.write(t)
item.unload_data_from_memory(memory=path)
zfile = zipfile.ZipFile(output_path, "w")
zfile.add_dir(output_dir, basename(output_dir))
zfile.write(output_file, basename(output_file), zipfile.ZIP_DEFLATED)
if opts.extract_to:
if os.path.exists(opts.extract_to):
shutil.rmtree(opts.extract_to)
os.makedirs(opts.extract_to)
zfile.extractall(opts.extract_to)
self.log('Zip file extracted to', opts.extract_to)
zfile.close()
# cleanup temp dir
shutil.rmtree(tempdir)
def add_links(self):
for link in self.links:
path, href, frag = link[0]
page, rect = link[1:]
combined_path = os.path.normcase(os.path.abspath(os.path.join(os.path.dirname(path), *unquote(href).split('/'))))
is_local = not href or combined_path in self.anchors
annot = Dictionary({
'Type':Name('Annot'), 'Subtype':Name('Link'),
'Rect':rect, 'Border':Array([0,0,0]),
})
if self.mark_links:
annot.update({'Border':Array([16, 16, 1]), 'C':Array([1.0, 0,
0])})
if is_local:
path = combined_path if href else path
try:
annot['Dest'] = self.anchors[path][frag]
except KeyError:
try:
annot['Dest'] = self.anchors[path][None]
except KeyError:
pass
else:
url = href + (('#'+frag) if frag else '')
try:
purl = urlparse(url)
except Exception:
self.pdf.debug('Ignoring unparseable URL: %r' % url)
continue
if purl.scheme and purl.scheme != 'file':
action = Dictionary({
'Type':Name('Action'), 'S':Name('URI'),
})
# Do not try to normalize/quote/unquote this URL as if it
# has a query part, it will get corrupted
action['URI'] = String(url)
annot['A'] = action
if 'A' in annot or 'Dest' in annot:
if 'Annots' not in page:
page['Annots'] = Array()
page['Annots'].append(self.pdf.objects.add(annot))
else:
self.pdf.debug('Could not find destination for link: %s in file %s'%
(href, path))
