def remapLinks(self, webpage):
for img in webpage.tree.findall('.//img'):
img_fn = webpage.images.get(img.attrib['src'])
if img_fn:
zip_rel_path = os.path.join(config.img_rel_path, os.path.basename(img_fn))
img.attrib['src'] = zip_rel_path
else:
remove_node(img)
target_ids = [safe_xml_id(_id) for _id in webpage.tree.xpath('.//@id')]
for a in webpage.tree.findall('.//a'):
href = a.get('href')
if not href: # this link is probably just an anchor
continue
if href.startswith('#'):
target_id = safe_xml_id(href)[1:]
if target_id not in target_ids:
a.set('id', target_id)
target_ids.append(target_id)
a.set('href', '#'+target_id)
else:
url = clean_url(urlparse.urljoin(webpage.url, href))
linked_wp = webpage.coll.url2webpage.get(url)
if linked_wp:
a.set('href', linked_wp.id + '.xhtml')
else:
a.set('href', url)
def remapLinks(self, webpage):
for img in webpage.tree.findall('.//img'):
img_fn = webpage.images.get(img.attrib['src'])
if img_fn:
zip_rel_path = os.path.join(config.img_rel_path,
os.path.basename(img_fn))
img.attrib['src'] = zip_rel_path
else:
remove_node(img)
target_ids = [safe_xml_id(_id) for _id in webpage.tree.xpath('.//@id')]
for a in webpage.tree.findall('.//a'):
href = a.get('href')
if not href: # this link is probably just an anchor
continue
if href.startswith('#'):
target_id = safe_xml_id(href)[1:]
if target_id not in target_ids:
a.set('id', target_id)
target_ids.append(target_id)
a.set('href', '#' + target_id)
else:
url = clean_url(urlparse.urljoin(webpage.url, href))
linked_wp = webpage.coll.url2webpage.get(url)
if linked_wp:
a.set('href', linked_wp.id + '.xhtml')
else:
a.set('href', url)
def writeOPF_manifest():
tree = E.manifest()
tree.extend([E.item({'id': article.id,
'href': article.path,
'media-type': 'application/xhtml+xml'})
for article in self.articles])
tree.append(E.item({'id':'ncx',
'href': os.path.basename(config.ncx_fn),
'media-type': 'application/x-dtbncx+xml'}))
#FIXME add missing resources:
# images
# css
for fn in self.added_files:
if fn.startswith('OPS/'):
fn = fn[4:]
mimetype, encoding = mimetypes.guess_type(fn)
if mimetype in ['text/css',
'image/png',
'image/jpeg',
'image/gif',
]:
_id = 'cover-image' if fn==self.cover_img_path else safe_xml_id(fn)
tree.append(E.item({'id': _id,
'href': fn,
'media-type': mimetype}))
return tree
def addArticle(self, webpage):
path = 'OPS/%s.xhtml' % webpage.id
self.add_file(path, webpage.xml)
self.articles.append(
ArticleInfo(id=safe_xml_id(webpage.id),
path=os.path.basename(path),
title=webpage.title,
type='article' if isinstance(
webpage, collection.WebPage) else 'chapter'))
if getattr(webpage, 'tree', False) != False:
used_images = [
src[len(config.img_rel_path):]
for src in webpage.tree.xpath('//img/@src')
]
else:
used_images = []
if getattr(webpage, 'images', False) != False:
for img_src, img_fn in webpage.images.items():
basename = os.path.basename(img_fn)
if basename not in used_images:
continue
zip_fn = os.path.join(config.img_abs_path, basename)
self.link_file(img_fn, zip_fn, compression=False)
def addArticle(self, webpage):
path = 'OPS/%s.xhtml' % webpage.id
self.add_file(path, webpage.xml)
self.articles.append(ArticleInfo(id=safe_xml_id(webpage.id),
path=os.path.basename(path),
title=webpage.title,
type='article' if isinstance(webpage, collection.WebPage) else 'chapter'))
if getattr(webpage, 'tree', False) != False:
used_images = [src[len(config.img_rel_path):] for src in webpage.tree.xpath('//img/@src')]
else:
used_images = []
if getattr(webpage, 'images', False) != False:
for img_src, img_fn in webpage.images.items():
basename = os.path.basename(img_fn)
if basename not in used_images:
continue
zip_fn = os.path.join(config.img_abs_path, basename)
self.link_file(img_fn, zip_fn, compression=False)
def writeOPF_manifest():
tree = E.manifest()
tree.extend([
E.item({
'id': article.id,
'href': article.path,
'media-type': 'application/xhtml+xml'
}) for article in self.articles
])
tree.append(
E.item({
'id': 'ncx',
'href': os.path.basename(config.ncx_fn),
'media-type': 'application/x-dtbncx+xml'
}))
#FIXME add missing resources:
# images
# css
for fn in self.added_files:
if fn.startswith('OPS/'):
fn = fn[4:]
mimetype, encoding = mimetypes.guess_type(fn)
if mimetype in [
'text/css',
'image/png',
'image/jpeg',
'image/gif',
]:
_id = 'cover-image' if fn == self.cover_img_path else safe_xml_id(
fn)
tree.append(
E.item({
'id': _id,
'href': fn,
'media-type': mimetype
}))
return tree