def anchorNCXUpdates(data, ncx_bookpath, originating_bookpath, keylist, valuelist): data = _remove_xml_header(data) # lxml on a Mac does not seem to handle full unicode properly, so encode as utf-8 data = data.encode('utf-8') # rebuild serialized lookup dictionary id_dict = OrderedDict() for i in range(0, len(keylist)): id_dict[ keylist[i] ] = valuelist[i] startdir = startingDir(ncx_bookpath) xmlbuilder = LXMLTreeBuilderForXML(parser=None, empty_element_tags=ebook_xml_empty_tags) soup = BeautifulSoup(data, features=None, from_encoding="utf-8", builder=xmlbuilder) for tag in soup.find_all("content"): if "src" in tag.attrs: src = tag["src"] if src.find(":") == -1: parts = src.split('#') apath = urldecodepart(parts[0]) # convert this path to its target bookpath target_bookpath = buildBookPath(apath, startdir) if (parts is not None) and (len(parts) > 1) and (target_bookpath == originating_bookpath) and (parts[1] != ""): fragment_id = urldecodepart(parts[1]) if fragment_id in id_dict: target_bookpath = id_dict[fragment_id] attribute_value = urlencodepart(buildRelativePath(ncx_bookpath, target_bookpath)) attribute_value = attribute_value + "#" + urlencodepart(fragment_id) tag["src"] = attribute_value; newdata = soup.decodexml(indent_level=0, formatter='minimal', indent_chars=" ") return newdata
def performPageMapUpdates(data, newbkpath, oldbkpath, keylist, valuelist): data = _remove_xml_header(data) # lxml on a Mac does not seem to handle full unicode properly, so encode as utf-8 data = data.encode('utf-8') # rebuild serialized lookup dictionary of xml_updates properly adjusted updates = OrderedDict() for i in range(0, len(keylist)): updates[ keylist[i] ] = valuelist[i] xml_empty_tags = ["page"] xmlbuilder = LXMLTreeBuilderForXML(parser=None, empty_element_tags=xml_empty_tags) soup = BeautifulSoup(data, features=None, from_encoding="utf-8", builder=xmlbuilder) for tag in soup.find_all(["page"]): for att in ["href"]: if att in tag.attrs : ref = tag[att] if ref.find(":") == -1 : parts = ref.split('#') apath = urldecodepart(parts[0]) fragment = "" if len(parts) > 1: fragment = urldecodepart(parts[1]) oldtarget = buildBookPath(apath, startingDir(oldbkpath)) newtarget = updates.get(oldtarget, oldtarget) attribute_value = urlencodepart(buildRelativePath(newbkpath, newtarget)) if fragment != "": attribute_value = attribute_value + "#" + urlencodepart(fragment) tag[att] = attribute_value newdata = soup.decodexml(indent_level=0, formatter='minimal', indent_chars=" ") return newdata
def anchorNCXUpdatesAfterMerge(data, ncx_bookpath, sink_bookpath, merged_bookpaths): data = _remove_xml_header(data) startdir = startingDir(ncx_bookpath) # lxml on a Mac does not seem to handle full unicode properly, so encode as utf-8 data = data.encode('utf-8') xmlbuilder = LXMLTreeBuilderForXML(parser=None, empty_element_tags=ebook_xml_empty_tags) soup = BeautifulSoup(data, features=None, from_encoding="utf-8", builder=xmlbuilder) for tag in soup.find_all("content"): if "src" in tag.attrs: src = tag["src"] if src.find(":") == -1: parts = src.split('#') if parts is not None: apath = urldecodepart(parts[0]) target_bookpath = buildBookPath(apath, startdir) if target_bookpath in merged_bookpaths: attribute_value = urlencodepart( buildRelativePath(ncx_bookpath, sink_bookpath)) if len(parts) > 1 and parts[1] != "": fragment = urldecodepart(parts[1]) attribute_value += "#" + urlencodepart(parts[1]) tag["src"] = attribute_value newdata = soup.decodexml(indent_level=0, formatter='minimal', indent_chars=" ") return newdata
def _parseData(self): cnt = 0 for prefix, tname, tattr, tcontent in self._opf_tag_iter(): # package if tname == "package": ver = tattr.pop("version", "2.0") uid = tattr.pop("unique-identifier", "bookid") if self.ns_remap: if "xmlns:opf" in tattr: tattr.pop("xmlns:opf") tattr["xmlns"] = "http://www/idpf.org/2007/opf" self.package = (ver, uid, tattr) continue # metadata if tname == "metadata": if self.ns_remap: if not "xmlns:opf" in tattr: tattr["xmlns:opf"] = "http://www/idpf.org/2007/opf" self.metadata_attr = tattr continue if tname in ["meta", "link" ] or tname.startswith("dc:") and "metadata" in prefix: self.metadata.append((tname, tcontent, tattr)) continue # manifest if tname == "item" and "manifest" in prefix: nid = "xid%03d" % cnt cnt += 1 id = tattr.pop("id", nid) # must keep all hrefs in encoded) form # if relative, then no fragments so decode and then encode for safety href = tattr.pop("href", "") if href.find(':') == -1: href = urldecodepart(href) href = urlencodepart(href) mtype = tattr.pop("media-type", "") self.manifest.append((id, href, mtype, tattr)) continue # spine if tname == "spine": self.spine_attr = tattr continue if tname == "itemref" and "spine" in prefix: idref = tattr.pop("idref", "") self.spine.append((idref, tattr)) continue # guide if tname == "reference" and "guide" in prefix: type = tattr.pop("type", "") title = tattr.pop("title", "") # must keep all hrefs in quoted (encoded) form href = tattr.pop("href", "") self.guide.append((type, title, href)) continue # bindings if tname in ["mediaType", "mediatype"] and "bindings" in prefix: mtype = tattr.pop("media-type", "") handler = tattr.pop("handler", "") self.bindings.append((mtype, handler)) continue
def build_manifest_xml(self): manout = [] manout.append(' <manifest>\n') for id in sorted(self.id_to_mime): href = self.id_to_href[id] # relative manifest hrefs must have no fragments if href.find(':') == -1: href = urlencodepart(href) mime = self.id_to_mime[id] props = '' properties = self.id_to_props[id] if properties is not None: props = ' properties="%s"' % properties fall = '' fallback = self.id_to_fall[id] if fallback is not None: fall = ' fallback="%s"' % fallback over = '' overlay = self.id_to_over[id] if overlay is not None: over = ' media-overlay="%s"' % overlay manout.append( ' <item id="%s" href="%s" media-type="%s"%s%s%s />\n' % (id, href, mime, props, fall, over)) manout.append(' </manifest>\n') return "".join(manout)
def parse_nav(qp, navdata, navbkpath, newdir): qp.setContent(navdata) toclist = [] pagelist = [] landmarks = [] lvl = 0 pgcnt = 0 maxlvl = -1 nav_type = None href = None title = "" play = 0 navdir = startingDir(navbkpath) for txt, tp, tname, ttype, tattr in qp.parse_iter(): if txt is not None: if ".a." in tp or tp.endswith(".a"): title = title + txt else: title = "" else: if tname == "nav": if ttype == "begin": nav_type = tattr.get("epub:type", None) if ttype == "end": nav_type = None continue if tname == "ol" and nav_type is not None and nav_type in ( "toc", "page-list", "landmarks"): if ttype == "begin": lvl += 1 if nav_type == "toc": if lvl > maxlvl: maxlvl = lvl if ttype == "end": lvl -= 1 continue if tname == "a" and ttype == "begin": # get the raw href (urlencoded) href = tattr.get("href", "") if href.find(":") == -1: # first strip off any fragment fragment = "" if href.find("#") != -1: href, fragment = href.split("#") # find destination bookpath href = urldecodepart(href) fragment = urldecodepart(fragment) if href.startswith("./"): href = href[2:] if href == "": destbkpath = navbkpath else: destbkpath = buildBookPath(href, navdir) # create relative path to destbkpath from newdir href = relativePath(destbkpath, newdir) href = urlencodepart(href) fragment = urlencodepart(fragment) if fragment != "": href = href + "#" + fragment epubtype = tattr.get("epub:type", None) continue if tname == "a" and ttype == "end": if nav_type == "toc": play += 1 toclist.append((play, lvl, href, title)) elif nav_type == "page-list": pgcnt += 1 pagelist.append((pgcnt, href, title)) elif nav_type == "landmarks": if epubtype is not None: gtype = _epubtype_guide_map.get(epubtype, None) landmarks.append((gtype, href, title)) title = "" continue return toclist, pagelist, landmarks, maxlvl, pgcnt
def launch(self): script_module = self.script_module script_type = self.script_type container = self.container sys.stdout = SavedStream(sys.stdout, 'stdout', self) sys.stderr = SavedStream(sys.stderr, 'stderr', self) try: target_script = __import__(script_module) self.exitcode = target_script.run(container) sys.stdout = sys.stdout.stream sys.stderr = sys.stderr.stream except Exception as e: sys.stderr.write(traceback.format_exc()) sys.stderr.write("Error: %s\n" % e) sys.stdout = sys.stdout.stream sys.stderr = sys.stderr.stream self.exitcode = -1 pass if self.exitcode != 0: self.wrapout.append(_XML_HEADER) self.wrapout.append( '<wrapper type="%s">\n<result>failed</result>\n<changes/>\n' % script_type) return self.exitcode if script_type == "edit": # write out the final updated opf to the outdir container._w.write_opf() # save the wrapper results to a file before exiting the thread # Note: the hrefs generated for the result xml are all relative paths # with no fragments since they must refer to entire files self.wrapout.append(_XML_HEADER) self.wrapout.append('<wrapper type="%s">\n' % script_type) self.wrapout.append('<result>success</result>\n') self.wrapout.append('<changes>\n') if script_type == "edit": for ftype, id, href in container._w.deleted: if ftype == 'manifest': bookhref = href mime = container._w.getmime(bookhref) else: bookhref = id id = "" mime = container._w.getmime(bookhref) self.wrapout.append( '<deleted href="%s" id="%s" media-type="%s" />\n' % (urlencodepart(bookhref), id, mime)) if script_type in ['input', 'edit']: for id in container._w.added: if id in container._w.id_to_bookpath: bookhref = container._w.id_to_bookpath[id] mime = container._w.id_to_mime[id] else: bookhref = id id = "" mime = container._w.getmime(bookhref) self.wrapout.append( '<added href="%s" id="%s" media-type="%s" />\n' % (urlencodepart(bookhref), id, mime)) if script_type == 'edit': for id in container._w.modified: if id in container._w.id_to_bookpath: bookhref = container._w.id_to_bookpath[id] mime = container._w.id_to_mime[id] else: bookhref = id id = "" mime = container._w.getmime(bookhref) self.wrapout.append( '<modified href="%s" id="%s" media-type="%s" />\n' % (urlencodepart(bookhref), id, mime)) if script_type == 'validation': for vres in container.results: self.wrapout.append( '<validationresult type="%s" bookpath="%s" linenumber="%s" charoffset="%s" message="%s" />\n' % (vres.restype, vres.bookpath, vres.linenumber, vres.charoffset, vres.message)) self.wrapout.append('</changes>\n') self.exitcode = 0 return