def _init_read(self):
"""Get content from existing epub file"""
# Read container.xml to get OPF xml file path
xmlstring = self.read('META-INF/container.xml')
container_xml = minidom.parseString(xmlstring).documentElement
for element in container_xml.getElementsByTagName('rootfile'):
if element.getAttribute('media-type') == MIMETYPE_OPF:
# Only take the first full-path available
self.opf_path = element.getAttribute('full-path')
break
# Read OPF xml file
xml_string = self.read(self.opf_path)
self.opf = opf.parse_opf(xml_string)
uids = [x for x in self.opf.metadata.identifiers
if x[1] == self.opf.uid_id]
if uids:
self.uid = uids[0]
else:
self.uid = None
warnings.warn('The ePub does not define any uid', SyntaxWarning)
item_toc = self.get_item(self.opf.spine.toc)
# Inspect NCX toc file
self.toc = None
if item_toc is not None:
self.toc = ncx.parse_toc(self.read_item(item_toc))
else:
warnings.warn('The ePub does not define any NCX file',
SyntaxWarning)
self.toc = ncx.Ncx()
self.toc.uid = self.uid
def maj_meta(monepub, metaname, metac, metacal, copyr):
os.chdir(monepub.chemin)
job.log('Met à jour les métadonnées')
mesmeta = monepub.opf.as_xml_document()
if metaname :
job.log('\t il existe un metadata.opf')
calibre_meta = opf.parse_opf(os.path.normpath(metaname))
calibrexml = calibre_meta.as_xml_document()
metaxml = modifie_metas(calibrexml, metac, metacal)
else :
job.log('\t on utilise les metas internes')
metaxml = modifie_metas(mesmeta, metac, metacal)
for node in mesmeta.getElementsByTagName('package'):
metas = node.firstChild
package = metas.parentNode
monmanifest = metas.nextSibling
package.removeChild(metas)
package.insertBefore(metaxml, monmanifest)
path = os.path.normpath(os.path.join(monepub.chemin, monepub.opf_path))
f = open(path, mode='w', encoding='utf8', errors='strict')
f.write(mesmeta.toprettyxml())
f.close()
monepub.init_read()
metas_xhtml(monepub, copyr)
return
def maj_publication(monepub, copyright):
from PyQt4 import QtGui
try:
job.ProgressDialog.setLabelText('Récupération de la date de copyright')
QtGui.QApplication.processEvents()
except:
pass
job.log('Récupération du copyright')
pubdate = sorted(copyright)[0] + '-01-01'
file = os.path.normpath(monepub.chemin + '\\' + monepub.opf_path)
parser = etree.XMLParser(attribute_defaults=False, load_dtd=False, remove_comments=False, ns_clean=True)
tree = etree.parse(file, parser)
xpdate = xpath(tree, '//dc:date')
for child in xpdate :
if child.get("{http://www.idpf.org/2007/opf}event") == 'publication' :
child.text = pubdate
elif child.get("{http://www.idpf.org/2007/opf}event") is None:
child.text = pubdate
texte = etree.tostring(tree, encoding='utf-8', xml_declaration=True, pretty_print=True)
with open(file, mode='wb') as f:
f.write(texte)
opf.parse_opf(file)
return
def init_read(self):
# Read container.xml to get OPF xml file path
xmlstring = 'META-INF/container.xml'
container_xml = minidom.parse(xmlstring).documentElement
for e in container_xml.getElementsByTagName('rootfile'):
if e.getAttribute('media-type') == MIMETYPE_OPF:
# Only take the first full-path available
self.opf_path = e.getAttribute('full-path') # OEBPS\content.opf ou content.opf
if os.path.isfile(os.path.join(self.chemin, self.opf_path)):
self.content_path = os.path.dirname(self.opf_path) # OEBPS ou ''
break
else :
self.opf_path = self.cherche_opf(self.chemin)
if self.opf_path:
self.content_path = os.path.dirname(self.opf_path) # OEBPS ou ''
else:
job.log('Epub invalide : pas de content.opf')
break
# Read OPF xml file
xml_string = (self.opf_path)
self.opf = opf.parse_opf(xml_string)
try:
self.uid = [x for x in self.opf.metadata.identifiers
if x[1] == self.opf.uid_id][0]
except:
self.uid = None
self.item_toc = self.get_item(self.opf.spine.toc)
if self.item_toc is None:
for identifier in self.opf.manifest:
item = self.opf.manifest[identifier]
if 'ncx' in item.href :
self.item_toc = os.path.join(self.content_path, item.href)
if self.opf.metadata.creators:
self.auteur = self.opf.metadata.creators[0][0]
else :
self.auteur = ''
if self.opf.metadata.titles :
self.titre = self.opf.metadata.titles[0][0]
else:
self.titre = ''
# Get the css files (on en profite pour implémenter la liste des fichiers)
self.css = []
self.liste_items = []
self.liste_fichiers = []
for identifier in self.opf.manifest:
item = self.opf.manifest[identifier]
self.liste_items.append(item.href)
self.liste_fichiers.append(self.normalize(self.relat_to_abs(item.href)))
if 'css' in item.href :
self.css.append(os.path.join(self.content_path, item.href)) # OEBPS\Style\styles.css ou stylesheet.css
# Inspect NCX toc file
try:
self.toc = ncx.parse_toc(self.read_item(self.item_toc))
except (UnicodeError, IOError) :
self.toc = None
# recherche la page et l'image de couverture
if self.pagedecouv is None:
self.set_cover()
def modif_metas_tatoo(monepub):
# crée un nouvel identifiant uuid
monepub.uid = 'urn:uuid:' + '%s' % uuid.uuid4()
parser = etree.XMLParser(attribute_defaults=False, load_dtd=False, remove_comments=False, ns_clean=True) # , recover=True)
# traitement de toc.ncx
mytoc = os.path.normpath(monepub.chemin + '\\' + monepub.content_path + '\\' + monepub.item_toc.href)
tree = etree.parse(mytoc, parser)
# on remplace l'identifiant par le nouveau
meta_uid = XPath('//dt:meta[@name="dtb:uid"]')
for key, value in enumerate(meta_uid(tree)):
value.set('content', monepub.uid)
texte = etree.tostring(tree, encoding='utf-8', xml_declaration=True, pretty_print=True)
with open(mytoc, mode='wb') as f:
f.write(texte)
# traitement du .opf
myopf = os.path.join(monepub.chemin, monepub.opf_path)
tree = etree.parse(myopf, parser)
# on récupère l'id de l'identifiant unique
pack = XPath('//opf:package')
unique = str(pack(tree)[0].get('unique-identifier'))
# on remplace l'identifiant par le nouveau
ident = XPath('//dc:identifier[@id="{}"]'.format(unique))
if len(ident(tree)):
ident(tree)[0].text = monepub.uid
# puis on change l'id et le opf:scheme
ident(tree)[0].set('id', 'BookId')
pack(tree)[0].set('unique-identifier', 'BookId')
job.logtat('remplacement de l\'identifiant unique par : ' + monepub.uid)
ident_scheme = xpath(tree, '//dc:identifier')
for child in ident_scheme:
child.set("{http://www.idpf.org/2007/opf}scheme", 'uuid')
# on supprime les autres identifiants
autres_ident = XPath('//dc:identifier')
for id in autres_ident(tree):
if id.get('id') != 'BookId':
if (id.get('id') == 'ean') : # or (id.get('opf:scheme') in ('ISBN', 'isbn')) :
id.text = id.text[0:13]
else :
job.logtat('suppression de l\'identifiant : "' + id.text + '"')
id.getparent().remove(id)
# supprimer les dc:rights
rights = XPath('//dc:rights')
for right in rights(tree):
if right.text is not None:
job.logtat('suppression du dc:rights : "' + str(right.text) + '" ')
right.getparent().remove(right)
# supprimer les description contenant code et date de la version
desc = XPath('//dc:description')
for com in desc(tree):
if re.search('Version ([0-9]{1,6})', str(com.text)):
job.logtat('suppression de la description : "' + str(com.text) + '"')
com.text = str(com.text).replace(str(com.text), '')
# supprimer les commentaires
for elem in tree.iter(tag=etree.Comment):
if elem.text:
job.logtat('suppression des commentaires dans le opf : "' + str(elem.text) + '"')
elem.getparent().remove(elem)
# réécrire le fichier
texte = etree.tostring(tree, encoding='utf-8', xml_declaration=True, pretty_print=True)
with open(myopf, mode='wb') as f:
f.write(texte)
opf.parse_opf(myopf)
return
