def run_checks(container):
errors = []
# Check parsing
xml_items, html_items, raster_images, stylesheets = [], [], [], []
for name, mt in container.mime_map.iteritems():
items = None
if mt in XML_TYPES:
items = xml_items
elif mt in OEB_DOCS:
items = html_items
elif mt in OEB_STYLES:
items = stylesheets
elif is_raster_image(mt):
items = raster_images
if items is not None:
items.append((name, mt, container.open(name, "rb").read()))
errors.extend(run_checkers(check_html_size, html_items))
errors.extend(run_checkers(check_xml_parsing, xml_items))
errors.extend(run_checkers(check_xml_parsing, html_items))
errors.extend(run_checkers(check_raster_images, raster_images))
for err in errors:
if err.level > WARN:
return errors
# cssutils is not thread safe
for name, mt, raw in stylesheets:
if not raw:
errors.append(EmptyFile(name))
continue
errors.extend(check_css_parsing(name, raw))
for name, mt, raw in html_items + xml_items:
errors.extend(check_encoding_declarations(name, container))
for name, mt, raw in html_items:
if not raw:
continue
root = container.parsed(name)
for style in root.xpath('//*[local-name()="style"]'):
if style.get("type", "text/css") == "text/css" and style.text:
errors.extend(check_css_parsing(name, style.text, line_offset=style.sourceline - 1))
for elem in root.xpath("//*[@style]"):
raw = elem.get("style")
if raw:
errors.extend(check_css_parsing(name, raw, line_offset=elem.sourceline - 1, is_declaration=True))
errors += check_mimetypes(container)
errors += check_links(container) + check_link_destinations(container)
errors += check_fonts(container)
errors += check_filenames(container)
errors += check_ids(container)
errors += check_markup(container)
errors += check_opf(container)
return errors
def run_checks(container):
errors = []
# Check parsing
xml_items, html_items, raster_images, stylesheets = [], [], [], []
for name, mt in container.mime_map.iteritems():
items = None
if mt in XML_TYPES:
items = xml_items
elif mt in OEB_DOCS:
items = html_items
elif mt in OEB_STYLES:
items = stylesheets
elif is_raster_image(mt):
items = raster_images
if items is not None:
items.append((name, mt, container.open(name, 'rb').read()))
errors.extend(run_checkers(check_html_size, html_items))
errors.extend(run_checkers(check_xml_parsing, xml_items))
errors.extend(run_checkers(check_xml_parsing, html_items))
errors.extend(run_checkers(check_raster_images, raster_images))
for err in errors:
if err.level > WARN:
return errors
# cssutils is not thread safe
for name, mt, raw in stylesheets:
if not raw:
errors.append(EmptyFile(name))
continue
errors.extend(check_css_parsing(name, raw))
for name, mt, raw in html_items + xml_items:
errors.extend(check_encoding_declarations(name, container))
for name, mt, raw in html_items:
if not raw:
continue
root = container.parsed(name)
for style in root.xpath('//*[local-name()="style"]'):
if style.get('type', 'text/css') == 'text/css' and style.text:
errors.extend(check_css_parsing(name, style.text, line_offset=style.sourceline - 1))
for elem in root.xpath('//*[@style]'):
raw = elem.get('style')
if raw:
errors.extend(check_css_parsing(name, raw, line_offset=elem.sourceline - 1, is_declaration=True))
errors += check_mimetypes(container)
errors += check_links(container) + check_link_destinations(container)
errors += check_fonts(container)
errors += check_ids(container)
errors += check_filenames(container)
errors += check_markup(container)
errors += check_opf(container)
return errors
def fix_opf(self, container):
spine_names = {n for n, l in container.spine_names}
spine = container.opf_xpath('//opf:spine')[0]
rmap = {v: k for k, v in iteritems(container.manifest_id_map)}
# Add unreferenced text files to the spine
for name, mt in iteritems(container.mime_map):
if mt in OEB_DOCS and name not in spine_names:
spine_names.add(name)
container.insert_into_xml(
spine, spine.makeelement(OPF('itemref'), idref=rmap[name]))
# Remove duplicate entries from spine
seen = set()
for item, name, linear in container.spine_iter:
if name in seen:
container.remove_from_xml(item)
seen.add(name)
# Remove the <guide> which is not needed in EPUB 3
for guide in container.opf_xpath('//*[local-name()="guide"]'):
guide.getparent().remove(guide)
# Ensure that the cover-image property is set
cover_id = rmap['_static/' + self.config.epub_cover[0]]
for item in container.opf_xpath(
'//opf:item[@id="{}"]'.format(cover_id)):
item.set('properties', 'cover-image')
for item in container.opf_xpath(
'//opf:item[@href="epub-cover.xhtml"]'):
item.set('properties', 'svg calibre:title-page')
for item in container.opf_xpath('//opf:package'):
prefix = item.get('prefix') or ''
if prefix:
prefix += ' '
item.set('prefix', prefix + 'calibre: https://calibre-ebook.com')
# Remove any <meta cover> tag as it is not needed in epub 3
for meta in container.opf_xpath('//opf:meta[@name="cover"]'):
meta.getparent().remove(meta)
# Remove unreferenced files
for error in check_links(container):
if error.__class__ is UnreferencedResource:
container.remove_item(error.name)
# Pretty print the OPF
pretty_opf(container.parsed(container.opf_name))
container.dirty(container.opf_name)
def run_checks(container):
errors = []
# Check parsing
xml_items, html_items, raster_images, stylesheets = [], [], [], []
for name, mt in container.mime_map.iteritems():
items = None
if mt in XML_TYPES:
items = xml_items
elif mt in OEB_DOCS:
items = html_items
elif mt in OEB_STYLES:
items = stylesheets
elif is_raster_image(mt):
items = raster_images
if items is not None:
items.append((name, mt, container.open(name, 'rb').read()))
errors.extend(run_checkers(check_xml_parsing, xml_items))
errors.extend(run_checkers(check_xml_parsing, html_items))
errors.extend(run_checkers(check_raster_images, raster_images))
# cssutils is not thread safe
for name, mt, raw in stylesheets:
errors.extend(check_css_parsing(name, raw))
for name, mt, raw in html_items:
root = container.parsed(name)
for style in root.xpath('//*[local-name()="style"]'):
if style.get('type', 'text/css') == 'text/css':
errors.extend(
check_css_parsing(name,
style.text,
line_offset=style.sourceline - 1))
for elem in root.xpath('//*[@style]'):
raw = elem.get('style')
if raw:
errors.extend(
check_css_parsing(name,
raw,
line_offset=elem.sourceline - 1,
is_declaration=True))
errors += check_links(container)
errors += check_fonts(container)
return errors
def run_checks(container):
errors = []
# Check parsing
xml_items, html_items, raster_images, stylesheets = [], [], [], []
for name, mt in container.mime_map.iteritems():
items = None
if mt in XML_TYPES:
items = xml_items
elif mt in OEB_DOCS:
items = html_items
elif mt in OEB_STYLES:
items = stylesheets
elif is_raster_image(mt):
items = raster_images
if items is not None:
items.append((name, mt, container.open(name, 'rb').read()))
errors.extend(run_checkers(check_html_size, html_items))
errors.extend(run_checkers(check_xml_parsing, xml_items))
errors.extend(run_checkers(check_xml_parsing, html_items))
errors.extend(run_checkers(check_raster_images, raster_images))
# cssutils is not thread safe
for name, mt, raw in stylesheets:
errors.extend(check_css_parsing(name, raw))
for name, mt, raw in html_items:
root = container.parsed(name)
for style in root.xpath('//*[local-name()="style"]'):
if style.get('type', 'text/css') == 'text/css' and style.text:
errors.extend(check_css_parsing(name, style.text, line_offset=style.sourceline - 1))
for elem in root.xpath('//*[@style]'):
raw = elem.get('style')
if raw:
errors.extend(check_css_parsing(name, raw, line_offset=elem.sourceline - 1, is_declaration=True))
errors += check_mimetypes(container)
errors += check_links(container) + check_link_destinations(container)
errors += check_fonts(container)
errors += check_filenames(container)
errors += check_ids(container)
errors += check_opf(container)
return errors
def fix_opf(self, container):
spine_names = {n for n, l in container.spine_names}
spine = container.opf_xpath('//opf:spine')[0]
rmap = {v: k for k, v in container.manifest_id_map.iteritems()}
# Add unreferenced text files to the spine
for name, mt in container.mime_map.iteritems():
if mt in OEB_DOCS and name not in spine_names:
spine_names.add(name)
container.insert_into_xml(
spine, spine.makeelement(OPF('itemref'), idref=rmap[name]))
# Remove duplicate entries from spine
seen = set()
for item, name, linear in container.spine_iter:
if name in seen:
container.remove_from_xml(item)
seen.add(name)
# Ensure that the meta cover tag is correct
cover_id = rmap['_static/' + self.config.epub_cover[0]]
for meta in container.opf_xpath('//opf:meta[@name="cover"]'):
meta.set('content', cover_id)
# Add description metadata
metadata = container.opf_xpath('//opf:metadata')[0]
container.insert_into_xml(metadata,
metadata.makeelement(DC('description')))
metadata[-1].text = 'Comprehensive documentation for calibre'
# Remove search.html since it is useless in EPUB
container.remove_item('search.html')
# Remove unreferenced files
for error in check_links(container):
if error.__class__ is UnreferencedResource:
container.remove_item(error.name)
# Pretty print the OPF
pretty_opf(container.parsed(container.opf_name))
container.dirty(container.opf_name)
def fix_opf(self, container):
spine_names = {n for n, l in container.spine_names}
spine = container.opf_xpath('//opf:spine')[0]
rmap = {v:k for k, v in container.manifest_id_map.iteritems()}
# Add unreferenced text files to the spine
for name, mt in container.mime_map.iteritems():
if mt in OEB_DOCS and name not in spine_names:
spine_names.add(name)
container.insert_into_xml(spine, spine.makeelement(OPF('itemref'), idref=rmap[name]))
# Remove duplicate entries from spine
seen = set()
for item, name, linear in container.spine_iter:
if name in seen:
container.remove_from_xml(item)
seen.add(name)
# Ensure that the meta cover tag is correct
cover_id = rmap['_static/' + self.config.epub_cover[0]]
for meta in container.opf_xpath('//opf:meta[@name="cover"]'):
meta.set('content', cover_id)
# Add description metadata
metadata = container.opf_xpath('//opf:metadata')[0]
container.insert_into_xml(metadata, metadata.makeelement(DC('description')))
metadata[-1].text = 'Comprehensive documentation for calibre'
# Remove search.html since it is useless in EPUB
container.remove_item('search.html')
# Remove unreferenced files
for error in check_links(container):
if error.__class__ is UnreferencedResource:
container.remove_item(error.name)
# Pretty print the OPF
pretty_opf(container.parsed(container.opf_name))
container.dirty(container.opf_name)
def fix_opf(self, container):
spine_names = {n for n, l in container.spine_names}
spine = container.opf_xpath('//opf:spine')[0]
rmap = {v:k for k, v in iteritems(container.manifest_id_map)}
# Add unreferenced text files to the spine
for name, mt in iteritems(container.mime_map):
if mt in OEB_DOCS and name not in spine_names:
spine_names.add(name)
container.insert_into_xml(spine, spine.makeelement(OPF('itemref'), idref=rmap[name]))
# Remove duplicate entries from spine
seen = set()
for item, name, linear in container.spine_iter:
if name in seen:
container.remove_from_xml(item)
seen.add(name)
# Remove the <guide> which is not needed in EPUB 3
for guide in container.opf_xpath('//*[local-name()="guide"]'):
guide.getparent().remove(guide)
# Ensure that the cover-image property is set
cover_id = rmap['_static/' + self.config.epub_cover[0]]
for item in container.opf_xpath('//opf:item[@id="{}"]'.format(cover_id)):
item.set('properties', 'cover-image')
# Remove any <meta cover> tag as it is not needed in epub 3
for meta in container.opf_xpath('//opf:meta[@name="cover"]'):
meta.getparent().remove(meta)
# Remove unreferenced files
for error in check_links(container):
if error.__class__ is UnreferencedResource:
container.remove_item(error.name)
# Pretty print the OPF
pretty_opf(container.parsed(container.opf_name))
container.dirty(container.opf_name)
