def getmime(self, href): href = unicode_str(href) href = unquoteurl(href) filename = os.path.basename(href) ext = os.path.splitext(filename)[1] ext = ext.lower() return ext_mime_map.get(ext, "")
def deleteotherfile(self, book_href): id = unicode_str(book_href) id = unquoteurl(id) if id is None: raise WrapperException('None is not a valid book hrefbook href') if id not in self.other and id in self.id_to_href: raise WrapperException( 'Incorrect interface routine - use deletefile') filepath = self.book_href_to_filepath.get(id, None) if filepath is None: raise WrapperException('Book href does not exist') if id in PROTECTED_FILES or id == self.opfbookpath: raise WrapperException('attempt to delete protected file') add_to_deleted = True # if file was added or modified delete file from outdir if id in self.added or id in self.modified: filepath = os.path.join(self.outdir, filepath) if unipath.exists(filepath) and unipath.isfile(filepath): os.remove(filepath) if id in self.added: self.added.remove(id) add_to_deleted = False if id in self.other: self.other.remove(id) if id in self.modified: del self.modified[id] if add_to_deleted: self.deleted.append(('other', id, book_href)) del self.book_href_to_filepath[id]
def readotherfile(self, book_href): id = unicode_str(book_href) id = unquoteurl(id) if id is None: raise WrapperException('None is not a valid book href') if id not in self.other and id in self.id_to_href: raise WrapperException( 'Incorrect interface routine - use readfile') # handle special case of trying to read the opf after it has been modified if id == self.opfbookpath: if id in self.modified: return self.build_opf() filepath = self.book_href_to_filepath.get(id, None) if filepath is None: raise WrapperException('Book href does not exist') basedir = self.ebook_root if id in self.added or id in self.modified: basedir = self.outdir filepath = os.path.join(basedir, filepath) if not unipath.exists(filepath): raise WrapperException('File Does Not Exist') basename = os.path.basename(filepath) ext = os.path.splitext(basename)[1] ext = ext.lower() mime = ext_mime_map.get(ext, "") data = b'' with open(filepath, 'rb') as fp: data = fp.read() if mime in TEXT_MIMETYPES: data = unicode_str(data) return data
def setguide(self, new_guide): guide = [] for (type, title, href) in new_guide: type = unicode_str(type) title = unicode_str(title) href = unicode_str(href) href = unquoteurl(href) if type not in _guide_types: type = "other." + type if title is None: title = 'title missing' thref = href.split('#')[0] if thref not in self.href_to_id: raise WrapperException('guide href not in manifest') guide.append((type, title, href)) self.guide = guide self.modified[self.opfbookpath] = 'file'
def getTOC(self): # parse the nav to get the table of contents navsrc = self.content toclist = [] qp = QuickXHTMLParser() qp.setContent(navsrc) lvl = 0 po = 0 title = "" nav_type = None href = None 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" and ttype == "begin": nav_type = tattr.get("epub:type", None) continue if tname == "nav" and ttype == "end": nav_type = None continue if nav_type is not None and nav_type == "toc": if tname == "ol": if ttype == "begin": lvl += 1 if ttype == "end": lvl -= 1 continue if tname == "a" and ttype == "begin": href = tattr.get("href", "") href = unquoteurl(href) continue if tname == "a" and ttype == "end": po += 1 title = xmldecode(title) toclist.append((po, lvl, href, title)) title = "" href = None continue return toclist
def addotherfile(self, book_href, data): id = unicode_str(book_href) id = unquoteurl(id) if id is None: raise WrapperException('None is not a valid book href') if id in self.other: raise WrapperException('Book href must be unique') desired_path = id.replace("/", os.sep) filepath = os.path.join(self.outdir, desired_path) if unipath.isfile(filepath): raise WrapperException('Desired path already exists') base = os.path.dirname(filepath) if not unipath.exists(base): os.makedirs(pathof(base)) if isinstance(data, text_type): data = utf8_str(data) with open(pathof(filepath), 'wb') as fp: fp.write(data) self.other.append(id) self.added.append(id) self.book_href_to_filepath[id] = desired_path
def writeotherfile(self, book_href, data): id = unicode_str(book_href) id = unquoteurl(id) if id is None: raise WrapperException('None is not a valid book href') if id not in self.other and id in self.id_to_href: raise WrapperException('Incorrect interface routine - use writefile') filepath = self.book_href_to_filepath.get(id, None) if filepath is None: raise WrapperException('Book href does not exist') if id in PROTECTED_FILES or id == self.opfbookpath: raise WrapperException('Attempt to modify protected file') filepath = os.path.join(self.outdir, filepath) base = os.path.dirname(filepath) if not unipath.exists(base): os.makedirs(base) if isinstance(data, text_type): data = utf8_str(data) with open(filepath,'wb') as fp: fp.write(data) self.modified[id] = 'file'
def getLandmarks(self): # parse the nav to get the landmarks navsrc = self.content landmarks = [] qp = QuickXHTMLParser() qp.setContent(navsrc) title = "" nav_type = None href = None epubtype = None 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" and ttype == "begin": nav_type = tattr.get("epub:type", None) continue if tname == "nav" and ttype == "end": nav_type = None continue if nav_type is not None and nav_type == "landmarks": if tname == "a" and ttype == "begin": href = tattr.get("href", "") href = unquoteurl(href) epubtype = tattr.get("epub:type", None) continue if tname == "a" and ttype == "end": if epubtype is not None: title = xmldecode(title) landmarks.append((epubtype, href, title)) title = "" epubtype = None href = None continue return landmarks
def getPageList(self): # parse the nav source to get the page-list navsrc = self.content pagelist = [] qp = QuickXHTMLParser() qp.setContent(navsrc) pgcnt = 0 nav_type = None href = None title = "" 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" and ttype == "begin": nav_type = tattr.get("epub:type", None) continue if tname == "nav" and ttype == "end": nav_type = None continue if nav_type is not None and nav_type == "page-list": if tname == "a" and ttype == "begin" and nav_type == "page-list": href = tattr.get("href", "") href = unquoteurl(href) continue if tname == "a" and ttype == "end": pgcnt += 1 title = xmldecode(title) pagelist.append((pgcnt, href, title)) title = "" continue return pagelist
def _parseData(self): cnt = 0 for prefix, tname, tattr, tcontent in self._opf_tag_iter(): if self._debug: print(" Parsing OPF: ", prefix, tname, tattr, tcontent) # package if tname == "package": ver = tattr.pop("version", "2.0") uid = tattr.pop("unique-identifier", "bookid") self.package = (ver, uid, tattr) continue # metadata if tname == "metadata": self.metadata_attr = tattr continue if tname in ["meta", "link" ] or tname.startswith("dc:") and "metadata" in prefix: self.metadata.append((tname, tattr, tcontent)) if tattr.get("name", "") == "cover": self.cover_id = tattr.get("content", None) continue # manifest if tname == "item" and "manifest" in prefix: nid = "xid%03d" % cnt cnt += 1 id = tattr.pop("id", nid) href = tattr.pop("href", '') mtype = tattr.pop("media-type", '') if mtype == "text/html": mtype = "application/xhtml+xml" if mtype not in mime_group_map: print("****Opf_Parser Warning****: Unknown MediaType: ", mtype) href = unquoteurl(href) properties = tattr.pop("properties", None) fallback = tattr.pop("fallback", None) overlay = tattr.pop("media-overlay", None) # external resources are now allowed in the opf under epub3 # we can ignore fragments here as these are links to files self.manifest_id_to_href[id] = href bookpath = "" if href.find(":") == -1: bookpath = buildBookPath(href, self.opf_dir) self.manifest_id_to_bookpath[id] = bookpath self.manifest_id_to_mime[id] = mtype # self.bookpaths.append(bookpath) group = mime_group_map.get(mtype, '') if bookpath != "" and group != "": folderlst = self.group_folder.get(group, []) countlst = self.group_count.get(group, []) sdir = startingDir(bookpath) if sdir not in folderlst: folderlst.append(sdir) countlst.append(1) else: pos = folderlst.index(sdir) countlst[pos] = countlst[pos] + 1 self.group_folder[group] = folderlst self.group_count[group] = countlst self.manifest_id_to_properties[id] = properties self.manifest_id_to_fallback[id] = fallback self.manifest_id_to_overlay[id] = overlay continue # spine if tname == "spine": if tattr is not None: self.spine_ppd = tattr.get("page-progression-direction", None) continue if tname == "itemref" and "spine" in prefix: idref = tattr.pop("idref", "") linear = tattr.pop("linear", None) properties = tattr.pop("properties", None) self.spine.append((idref, linear, properties)) continue # guide if tname == "reference" and "guide" in prefix: type = tattr.pop("type", '') title = tattr.pop("title", '') href = unquoteurl(tattr.pop("href", '')) self.guide.append((type, title, href)) continue # bindings (stored but ignored for now) if tname in ["mediaType", "mediatype"] and "bindings" in prefix: mtype = tattr.pop("media-type", "") handler = tattr.pop("handler", "") self.bindings.append((mtype, handler)) continue # determine unique ShortPathName for each bookpath # start with filename and work back up the folders # spn = {} # dupset = set() # nameset = {} # lvl = 1 # for bkpath in self.bookpaths: # aname = build_short_name(bkpath, lvl) # spn[bkpath] = aname # if aname in nameset: # dupset.add(aname) # nameset[aname].append(bkpath) # else: # nameset[aname]=[bkpath] # # now work just through any to-do list of duplicates # until all duplicates are gone # # todolst = list(dupset) # while(todolst): # dupset = set() # lvl += 1 # for aname in todolst: # bklst = nameset[aname] # del nameset[aname] # for bkpath in bklst: # newname = build_short_name(bkpath, lvl) # spn[bkpath] = newname # if newname in nameset: # dupset.add(newname) # nameset[newname].append(bkpath) # else: # nameset[newname] = [bkpath] # todolst = list(dupset) # finally sort by number of files in dir to find default folders for each group dirlst = [] use_lower_case = False for group in self.group_folder.keys(): folders = self.group_folder[group] cnts = self.group_count[group] folders = [x for _, x in sorted(zip(cnts, folders), reverse=True)] self.group_folder[group] = folders if group in [ "Text", "Styles", "Images", "Audio", "Fonts", "Video", "Misc" ]: afolder = folders[0] if afolder.find(group.lower()) > -1: use_lower_case = True dirlst.append(folders[0]) # now back fill any missing values # commonbase will end with a / commonbase = longestCommonPath(dirlst) if commonbase == "/": commonbase = "" for group in ["Styles", "Images", "Audio", "Fonts", "Video", "Misc"]: folders = self.group_folder.get(group, []) gname = group if use_lower_case: gname = gname.lower() if not folders: folders = [commonbase + gname] self.group_folder[group] = folders
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": href = tattr.get("href", "") href = unquoteurl(href) if href.find(":") == -1: # first strip off any fragment fragment = "" if href.find("#") != -1: href, fragment = href.split("#") # find destination bookpath 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) 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 map_href_to_id(self, href, ow): href = unicode_str(href) href = unquoteurl(href) return self.href_to_id.get(href, ow)
def build_bookpath(self, href, starting_dir): href = unicode_str(href) href = unquoteurl(href) starting_dir = unicode_str(starting_dir) return buildBookPath(href, starting_dir)