Python ZipFile.debug方法代码示例

# 需要导入模块: from zipfile import ZipFile [as 别名]
# 或者: from zipfile.ZipFile import debug [as 别名]
    def generate_plugin(self):
        """Generate ZIP file with specified stylesheets."""
        self._preprocess()

        __output = self.out

        __temp = StringIO.StringIO()
        __failed = False
        try:
            __zip = ZipFile(__temp, "w")
            __zip.debug = 3
            try:
                # integrator
                self._run_generation(__zip, self.__generate_integrator,
                                    "%s/integrator.xml" % (self.plugin_name))
                # plugin
                self._run_generation(__zip, self.__generate_plugin_file,
                                    "%s/plugin.xml" % (self.plugin_name))
                # catalog
                self._run_generation(__zip, self.__generate_catalog,
                                    "%s/cfg/catalog.xml" % (self.plugin_name))
                # font-mappings
#                self._run_generation(__zip, self.__generate_font_mappings,
#                                    "%s/cfg/fo/font-mappins.xml" % (self.plugin_name))
                # custom XSLT
                self._run_generation(__zip, self.__generate_custom,
                                    "%s/cfg/fo/xsl/custom.xsl" % (self.plugin_name))
                # custom XSLT attribute sets
                self._run_generation(__zip, self.__generate_custom_attr,
                                    "%s/cfg/fo/attrs/custom.xsl" % (self.plugin_name))
                # shell XSLT
                if self.override_shell:
                    self._run_generation(__zip, self.__generate_shell,
                                        "%s/xsl/fo/topic2fo_shell_%s.xsl" % (self.plugin_name, self.formatter))
#                if not self.link_pagenumber or self.table_continued:
                for lang in self.variable_languages:
                    self._run_generation(__zip, lambda: self.__generate_vars(lang),
                                         "%s/cfg/common/vars/%s.xml" % (self.plugin_name, lang))
#                if self.generate_shell:
#                    # shell XSLT
#                    self._run_generation(__zip, self.__generate_shell,
#                                        "%s/xsl/fo/.xsl" % (self.plugin_name))
            except:
                __failed = True
                raise Exception("Failed to write plugin", sys.exc_info()[1]), None, sys.exc_info()[2]
            finally:
                if __zip != None:
                    __zip.close()
            if not __failed:
                __output.write(__temp.getvalue())
        except:
            __failed = True
            raise Exception("Failed to write ZIP file to output", sys.exc_info()[1]), None, sys.exc_info()[2]
        finally:
            __temp.close()

# 需要导入模块: from zipfile import ZipFile [as 别名]
# 或者: from zipfile.ZipFile import debug [as 别名]
    def writeStoryImpl(self, out):

        ## Python 2.5 ZipFile is rather more primative than later
        ## versions.  It can operate on a file, or on a StringIO, but
        ## not on an open stream.  OTOH, I suspect we would have had
        ## problems with closing and opening again to change the
        ## compression type anyway.
        zipio = StringIO.StringIO()

        ## mimetype must be first file and uncompressed.  Python 2.5
        ## ZipFile can't change compression type file-by-file, so we
        ## have to close and re-open
        outputepub = ZipFile(zipio, 'w', compression=ZIP_STORED)
        outputepub.debug=3
        outputepub.writestr('mimetype','application/epub+zip')
        outputepub.close()

        ## Re-open file for content.
        outputepub = ZipFile(zipio, 'a', compression=ZIP_DEFLATED)
        outputepub.debug=3
        
        ## Create META-INF/container.xml file.  The only thing it does is
        ## point to content.opf
        containerdom = getDOMImplementation().createDocument(None, "container", None)
        containertop = containerdom.documentElement
        containertop.setAttribute("version","1.0")
        containertop.setAttribute("xmlns","urn:oasis:names:tc:opendocument:xmlns:container")
        rootfiles = containerdom.createElement("rootfiles")
        containertop.appendChild(rootfiles)
        rootfiles.appendChild(newTag(containerdom,"rootfile",{"full-path":"content.opf",
                                                              "media-type":"application/oebps-package+xml"}))
        outputepub.writestr("META-INF/container.xml",containerdom.toxml(encoding='utf-8'))
        containerdom.unlink()
        del containerdom

        ## Epub has two metadata files with real data.  We're putting
        ## them in content.opf (pointed to by META-INF/container.xml)
        ## and toc.ncx (pointed to by content.opf)

        ## content.opf contains metadata, a 'manifest' list of all
        ## other included files, and another 'spine' list of the items in the
        ## file

        uniqueid= 'fanficfare-uid:%s-u%s-s%s' % (
            self.getMetadata('site'),
            self.story.getList('authorId')[0],
            self.getMetadata('storyId'))
        
        contentdom = getDOMImplementation().createDocument(None, "package", None)
        package = contentdom.documentElement
        package.setAttribute("version","2.0")
        package.setAttribute("xmlns","http://www.idpf.org/2007/opf")
        package.setAttribute("unique-identifier","fanficfare-uid")
        metadata=newTag(contentdom,"metadata",
                        attrs={"xmlns:dc":"http://purl.org/dc/elements/1.1/",
                               "xmlns:opf":"http://www.idpf.org/2007/opf"})
        package.appendChild(metadata)

        metadata.appendChild(newTag(contentdom,"dc:identifier",
                                    text=uniqueid,
                                    attrs={"id":"fanficfare-uid"}))

        if self.getMetadata('title'):
            metadata.appendChild(newTag(contentdom,"dc:title",text=self.getMetadata('title')))

        if self.getMetadata('author'):
            if self.story.isList('author'):
                for auth in self.story.getList('author'):
                    metadata.appendChild(newTag(contentdom,"dc:creator",
                                                attrs={"opf:role":"aut"},
                                                text=auth))
            else:
                metadata.appendChild(newTag(contentdom,"dc:creator",
                                            attrs={"opf:role":"aut"},
                                            text=self.getMetadata('author')))

        metadata.appendChild(newTag(contentdom,"dc:contributor",text="FanFicFare [https://github.com/JimmXinu/FanFicFare]",attrs={"opf:role":"bkp"}))
        metadata.appendChild(newTag(contentdom,"dc:rights",text=""))
        if self.story.getMetadata('langcode'):
            metadata.appendChild(newTag(contentdom,"dc:language",text=self.story.getMetadata('langcode')))
        else:
            metadata.appendChild(newTag(contentdom,"dc:language",text='en'))

        #  published, created, updated, calibre
        #  Leave calling self.story.getMetadataRaw directly in case date format changes.
        if self.story.getMetadataRaw('datePublished'):
            metadata.appendChild(newTag(contentdom,"dc:date",
                                        attrs={"opf:event":"publication"},
                                        text=self.story.getMetadataRaw('datePublished').strftime("%Y-%m-%d")))
        
        if self.story.getMetadataRaw('dateCreated'):
            metadata.appendChild(newTag(contentdom,"dc:date",
                                        attrs={"opf:event":"creation"},
                                        text=self.story.getMetadataRaw('dateCreated').strftime("%Y-%m-%d")))
        
        if self.story.getMetadataRaw('dateUpdated'):
            metadata.appendChild(newTag(contentdom,"dc:date",
                                        attrs={"opf:event":"modification"},
                                        text=self.story.getMetadataRaw('dateUpdated').strftime("%Y-%m-%d")))
            metadata.appendChild(newTag(contentdom,"meta",
#.........这里部分代码省略.........

# 需要导入模块: from zipfile import ZipFile [as 别名]
# 或者: from zipfile.ZipFile import debug [as 别名]
def reset_orig_chapters_epub(inputio,outfile):
    inputepub = ZipFile(inputio, 'r') # works equally well with a path or a blob

    ## build zip in memory in case updating in place(CLI).
    zipio = StringIO()

    ## Write mimetype file, must be first and uncompressed.
    ## Older versions of python(2.4/5) don't allow you to specify
    ## compression by individual file.
    ## Overwrite if existing output file.
    outputepub = ZipFile(zipio, 'w', compression=ZIP_STORED)
    outputepub.debug = 3
    outputepub.writestr("mimetype", "application/epub+zip")
    outputepub.close()

    ## Re-open file for content.
    outputepub = ZipFile(zipio, "a", compression=ZIP_DEFLATED)
    outputepub.debug = 3

    changed = False

    unmerge_tocncxdoms = {}
    ## spin through file contents, saving any unmerge toc.ncx files.
    for zf in inputepub.namelist():
        ## logger.debug("zf:%s"%zf)
        if zf.endswith('/toc.ncx'):
            ## logger.debug("toc.ncx zf:%s"%zf)
            unmerge_tocncxdoms[zf] = parseString(inputepub.read(zf))

    tocncxdom = parseString(inputepub.read('toc.ncx'))
    ## spin through file contents.
    for zf in inputepub.namelist():
        if zf not in ['mimetype','toc.ncx'] and not zf.endswith('/toc.ncx'):
            entrychanged = False
            data = inputepub.read(zf)
            # if isinstance(data,unicode):
            #     logger.debug("\n\n\ndata is unicode\n\n\n")
            if re.match(r'.*/file\d+\.xhtml',zf):
                #logger.debug("zf:%s"%zf)
                data = data.decode('utf-8')
                soup = make_soup(data)

                chapterorigtitle = None
                tag = soup.find('meta',{'name':'chapterorigtitle'})
                if tag:
                    chapterorigtitle = tag['content']

                # toctitle is separate for add_chapter_numbers:toconly users.
                chaptertoctitle = None
                tag = soup.find('meta',{'name':'chaptertoctitle'})
                if tag:
                    chaptertoctitle = tag['content']
                    chaptertoctitle = chapterorigtitle

                chaptertitle = None
                tag = soup.find('meta',{'name':'chaptertitle'})
                if tag:
                    chaptertitle = tag['content']
                    chaptertitle_tag = tag

                #logger.debug("chaptertitle:(%s) chapterorigtitle:(%s)"%(chaptertitle, chapterorigtitle))
                if chaptertitle and chapterorigtitle and chapterorigtitle != chaptertitle:
                    origdata = data
                    # data = data.replace(u'<meta name="chaptertitle" content="'+chaptertitle+u'"></meta>',
                    #                     u'<meta name="chaptertitle" content="'+chapterorigtitle+u'"></meta>')
                    # data = data.replace(u'<title>'+chaptertitle+u'</title>',u'<title>'+chapterorigtitle+u'</title>')
                    # data = data.replace(u'<h3>'+chaptertitle+u'</h3>',u'<h3>'+chapterorigtitle+u'</h3>')
                    chaptertitle_tag['content'] = chapterorigtitle
                    title_tag = soup.find('title')
                    if title_tag and title_tag.string == chaptertitle:
                        title_tag.string.replace_with(chapterorigtitle)

                    h3_tag = soup.find('h3')
                    if h3_tag and h3_tag.string == chaptertitle:
                        h3_tag.string.replace_with(chapterorigtitle)

                    data = unicode(soup)

                    entrychanged = ( origdata != data )
                    changed = changed or entrychanged

                    if entrychanged:
                        logger.debug("\nentrychanged:%s\n"%zf)
                        _replace_tocncx(tocncxdom,zf,chaptertoctitle)
                        ## Also look for and update individual
                        ## book toc.ncx files for anthology in case
                        ## it's unmerged.
                        zf_toc = zf[:zf.rfind('/OEBPS/')]+'/toc.ncx'
                        mergedprefix_len = len(zf[:zf.rfind('/OEBPS/')])+1

                        if zf_toc in unmerge_tocncxdoms:
                            _replace_tocncx(unmerge_tocncxdoms[zf_toc],zf[mergedprefix_len:],chaptertoctitle)

                outputepub.writestr(zf,data.encode('utf-8'))
            else:
                # possibly binary data, thus no .encode().
                outputepub.writestr(zf,data)

    for tocnm, tocdom in unmerge_tocncxdoms.items():
        outputepub.writestr(tocnm,tocdom.toxml(encoding='utf-8'))
#.........这里部分代码省略.........

# 需要导入模块: from zipfile import ZipFile [as 别名]
# 或者: from zipfile.ZipFile import debug [as 别名]
    def write_split_epub(self,
                         outputio,
                         linenums,
                         changedtocs={},
                         authoropts=[],
                         titleopt=None,
                         descopt=None,
                         tags=[],
                         languages=['en'],
                         coverjpgpath=None):

        files = self.get_split_files(linenums)

        ## Write mimetype file, must be first and uncompressed.
        ## Older versions of python(2.4/5) don't allow you to specify
        ## compression by individual file.
        ## Overwrite if existing output file.
        outputepub = ZipFile(outputio, "w", compression=ZIP_STORED)
        outputepub.debug = 3
        outputepub.writestr("mimetype", "application/epub+zip")
        outputepub.close()

        ## Re-open file for content.
        outputepub = ZipFile(outputio, "a", compression=ZIP_DEFLATED)
        outputepub.debug = 3

        ## Create META-INF/container.xml file.  The only thing it does is
        ## point to content.opf
        containerdom = getDOMImplementation().createDocument(None, "container", None)
        containertop = containerdom.documentElement
        containertop.setAttribute("version","1.0")
        containertop.setAttribute("xmlns","urn:oasis:names:tc:opendocument:xmlns:container")
        rootfiles = containerdom.createElement("rootfiles")
        containertop.appendChild(rootfiles)
        rootfiles.appendChild(newTag(containerdom,"rootfile",{"full-path":"content.opf",
                                                              "media-type":"application/oebps-package+xml"}))
        outputepub.writestr("META-INF/container.xml",containerdom.toprettyxml(indent='   ',encoding='utf-8'))


####    ## create content.opf file.
        uniqueid="epubsplit-uid-%d" % time() # real sophisticated uid scheme.
        contentdom = getDOMImplementation().createDocument(None, "package", None)
        package = contentdom.documentElement

        package.setAttribute("version","2.0")
        package.setAttribute("xmlns","http://www.idpf.org/2007/opf")
        package.setAttribute("unique-identifier","epubsplit-id")
        metadata=newTag(contentdom,"metadata",
                        attrs={"xmlns:dc":"http://purl.org/dc/elements/1.1/",
                               "xmlns:opf":"http://www.idpf.org/2007/opf"})
        metadata.appendChild(newTag(contentdom,"dc:identifier",text=uniqueid,attrs={"id":"epubsplit-id"}))
        if( titleopt is None ):
            titleopt = self.origtitle+" Split"
        metadata.appendChild(newTag(contentdom,"dc:title",text=titleopt))

        if( authoropts and len(authoropts) > 0  ):
            useauthors=authoropts
        else:
            useauthors=self.origauthors

        usedauthors=dict()
        for author in useauthors:
            if( not usedauthors.has_key(author) ):
                usedauthors[author]=author
                metadata.appendChild(newTag(contentdom,"dc:creator",
                                            attrs={"opf:role":"aut"},
                                            text=author))

        metadata.appendChild(newTag(contentdom,"dc:contributor",text="epubsplit",attrs={"opf:role":"bkp"}))
        metadata.appendChild(newTag(contentdom,"dc:rights",text="Copyrights as per source stories"))

        if languages:
            for l in languages:
                metadata.appendChild(newTag(contentdom,"dc:language",text=l))
        else:
            metadata.appendChild(newTag(contentdom,"dc:language",text="en"))

        if not descopt:
            # created now, but not filled in until TOC generation to save loops.
            description = newTag(contentdom,"dc:description",text="Split from %s by %s."%(self.origtitle,", ".join(self.origauthors)))
        else:
            description = newTag(contentdom,"dc:description",text=descopt)
        metadata.appendChild(description)

        for tag in tags:
            metadata.appendChild(newTag(contentdom,"dc:subject",text=tag))

        package.appendChild(metadata)

        manifest = contentdom.createElement("manifest")
        package.appendChild(manifest)
        spine = newTag(contentdom,"spine",attrs={"toc":"ncx"})
        package.appendChild(spine)

        manifest.appendChild(newTag(contentdom,"item",
                                    attrs={'id':'ncx',
                                           'href':'toc.ncx',
                                           'media-type':'application/x-dtbncx+xml'}))

        if coverjpgpath:
#.........这里部分代码省略.........

# 需要导入模块: from zipfile import ZipFile [as 别名]
# 或者: from zipfile.ZipFile import debug [as 别名]
def doUnMerge(inputio,outdir=None):
    epub = ZipFile(inputio, 'r') # works equally well with inputio as a path or a blob
    outputios = []

    ## Find the .opf file.
    container = epub.read("META-INF/container.xml")
    containerdom = parseString(container)
    rootfilenodelist = containerdom.getElementsByTagName("rootfile")
    rootfilename = rootfilenodelist[0].getAttribute("full-path")

    contentdom = parseString(epub.read(rootfilename))

    ## Save the path to the .opf file--hrefs inside it are relative to it.
    relpath = get_path_part(rootfilename)
    #print("relpath:%s"%relpath)
            
    # spin through the manifest--only place there are item tags.
    # Correction--only place there *should* be item tags.  But
    # somebody found one that did.
    manifesttag=contentdom.getElementsByTagNameNS("*","manifest")[0]
    for item in manifesttag.getElementsByTagNameNS("*","item"):
        # look for our fake media-type for original rootfiles.
        if( item.getAttribute("media-type") == "origrootfile/xml" ):
            # found one, assume the dir containing it is a complete
            # original epub, do initial setup of epub.
            itemhref = relpath+unquote(item.getAttribute("href"))
            #print("Found origrootfile:%s"%itemhref)
            curepubpath = re.sub(r'([^\d/]+/)+$','',get_path_part(itemhref))
            savehref = itemhref[len(curepubpath):]
            #print("curepubpath:%s"%curepubpath)
            
            outputio = StringIO()
            outputepub = ZipFile(outputio, "w", compression=ZIP_STORED)
            outputepub.debug = 3
            outputepub.writestr("mimetype", "application/epub+zip")
            outputepub.close()
        
            ## Re-open file for content.
            outputepub = ZipFile(outputio, "a", compression=ZIP_DEFLATED)
            outputepub.debug = 3
            ## Create META-INF/container.xml file.  The only thing it does is
            ## point to content.opf
            containerdom = getDOMImplementation().createDocument(None, "container", None)
            containertop = containerdom.documentElement
            containertop.setAttribute("version","1.0")
            containertop.setAttribute("xmlns","urn:oasis:names:tc:opendocument:xmlns:container")
            rootfiles = containerdom.createElement("rootfiles")
            containertop.appendChild(rootfiles)
            rootfiles.appendChild(newTag(containerdom,"rootfile",{"full-path":savehref,
                                                                  "media-type":"application/oebps-package+xml"}))
            outputepub.writestr("META-INF/container.xml",containerdom.toprettyxml(indent='   ',encoding='utf-8'))

            outputepub.writestr(savehref,epub.read(itemhref))
            
            for item2 in contentdom.getElementsByTagName("item"):
                item2href = relpath+unquote(item2.getAttribute("href"))
                if item2href.startswith(curepubpath) and item2href != itemhref:
                    save2href = item2href[len(curepubpath):]
                    #print("Found %s -> %s"%(item2href,save2href))
                    outputepub.writestr(save2href,epub.read(item2href))

            # declares all the files created by Windows.  otherwise, when
            # it runs in appengine, windows unzips the files as 000 perms.
            for zf in outputepub.filelist:
                zf.create_system = 0
            outputepub.close()
            
            outputios.append(outputio)

    if outdir:
        outfilenames=[]
        for count,epubIO in enumerate(outputios):
            filename="%s/%d.epub"%(outdir,count)
            print("write %s"%filename)
            outstream = open(filename,"wb")
            outstream.write(epubIO.getvalue())
            outstream.close()
            outfilenames.append(filename)
        return outfilenames
    else:
        return outputios

# 需要导入模块: from zipfile import ZipFile [as 别名]
# 或者: from zipfile.ZipFile import debug [as 别名]
def doMerge(outputio,
            files,
            authoropts=[],
            titleopt=None,
            descopt=None,
            tags=[],
            languages=['en'],
            titlenavpoints=True,
            flattentoc=False,
            printtimes=False,
            coverjpgpath=None,
            keepmetadatafiles=False,
            source=None):
    '''
    outputio = output file name or StringIO.
    files = list of input file names or StringIOs.
    authoropts = list of authors to use, otherwise add from all input
    titleopt = title, otherwise '<first title> Anthology'
    descopt = description, otherwise '<title> by <author>' list for all input
    tags = dc:subject tags to include, otherwise none.
    languages = dc:language tags to include
    titlenavpoints if true, put in a new TOC entry for each epub, nesting each epub's chapters under it
    flattentoc if true, flatten TOC down to one level only.
    coverjpgpath, Path to a jpg to use as cover image.
    '''

    printt = partial(cond_print,printtimes)
    
    ## Python 2.5 ZipFile is rather more primative than later
    ## versions.  It can operate on a file, or on a StringIO, but
    ## not on an open stream.  OTOH, I suspect we would have had
    ## problems with closing and opening again to change the
    ## compression type anyway.

    filecount=0
    t = time()
    
    ## Write mimetype file, must be first and uncompressed.
    ## Older versions of python(2.4/5) don't allow you to specify
    ## compression by individual file.
    ## Overwrite if existing output file.
    outputepub = ZipFile(outputio, "w", compression=ZIP_STORED)
    outputepub.debug = 3
    outputepub.writestr("mimetype", "application/epub+zip")
    outputepub.close()

    ## Re-open file for content.
    outputepub = ZipFile(outputio, "a", compression=ZIP_DEFLATED)
    outputepub.debug = 3

    ## Create META-INF/container.xml file.  The only thing it does is
    ## point to content.opf
    containerdom = getDOMImplementation().createDocument(None, "container", None)
    containertop = containerdom.documentElement
    containertop.setAttribute("version","1.0")
    containertop.setAttribute("xmlns","urn:oasis:names:tc:opendocument:xmlns:container")
    rootfiles = containerdom.createElement("rootfiles")
    containertop.appendChild(rootfiles)
    rootfiles.appendChild(newTag(containerdom,"rootfile",{"full-path":"content.opf",
                                                          "media-type":"application/oebps-package+xml"}))
    outputepub.writestr("META-INF/container.xml",containerdom.toprettyxml(indent='   ',encoding='utf-8'))    

    ## Process input epubs.
    
    items = [] # list of (id, href, type) tuples(all strings) -- From .opfs' manifests
    items.append(("ncx","toc.ncx","application/x-dtbncx+xml")) ## we'll generate the toc.ncx file,
                                                               ## but it needs to be in the items manifest.
    itemrefs = [] # list of strings -- idrefs from .opfs' spines
    navmaps = [] # list of navMap DOM elements -- TOC data for each from toc.ncx files
    is_ffdl_epub = [] # list of t/f

    itemhrefs = {} # hash of item[id]s to itemref[href]s -- to find true start of book(s).
    firstitemhrefs = []

    booktitles = [] # list of strings -- Each book's title
    allauthors = [] # list of lists of strings -- Each book's list of authors.

    filelist = []
    
    printt("prep output:%s"%(time()-t))
    t = time()
    
    booknum=1
    firstmetadom = None
    for file in files:
        if file == None : continue
        
        book = "%d" % booknum
        bookdir = "%d/" % booknum
        bookid = "a%d" % booknum
        #print "book %d" % booknum
        
        epub = ZipFile(file, 'r')

        ## Find the .opf file.
        container = epub.read("META-INF/container.xml")
        containerdom = parseString(container)
        rootfilenodelist = containerdom.getElementsByTagNameNS("*","rootfile")
        rootfilename = rootfilenodelist[0].getAttribute("full-path")

#.........这里部分代码省略.........

# 需要导入模块: from zipfile import ZipFile [as 别名]
# 或者: from zipfile.ZipFile import debug [as 别名]
def reset_orig_chapters_epub(inputio,outfile):
    inputepub = ZipFile(inputio, 'r') # works equally well with a path or a blob

    ## build zip in memory in case updating in place(CLI).
    zipio = StringIO()
    
    ## Write mimetype file, must be first and uncompressed.
    ## Older versions of python(2.4/5) don't allow you to specify
    ## compression by individual file.
    ## Overwrite if existing output file.
    outputepub = ZipFile(zipio, 'w', compression=ZIP_STORED)
    outputepub.debug = 3
    outputepub.writestr("mimetype", "application/epub+zip")
    outputepub.close()

    ## Re-open file for content.
    outputepub = ZipFile(zipio, "a", compression=ZIP_DEFLATED)
    outputepub.debug = 3

    changed = False
    
    tocncxdom = parseString(inputepub.read('toc.ncx'))
    ## spin through file contents.
    for zf in inputepub.namelist():
        if zf not in ['mimetype','toc.ncx'] :
            entrychanged = False
            data = inputepub.read(zf)
            # if isinstance(data,unicode):
            #     logger.debug("\n\n\ndata is unicode\n\n\n")
            if re.match(r'.*/file\d+\.xhtml',zf):
                data = data.decode('utf-8')
                soup = bs.BeautifulSoup(data,"html5lib")
                
                chapterorigtitle = None
                tag = soup.find('meta',{'name':'chapterorigtitle'})
                if tag:
                    chapterorigtitle = tag['content']

                # toctitle is separate for add_chapter_numbers:toconly users.
                chaptertoctitle = None
                tag = soup.find('meta',{'name':'chaptertoctitle'})
                if tag:
                    chaptertoctitle = tag['content']
                elif chapterorigtitle:
                    chaptertoctitle = chapterorigtitle
                    
                chaptertitle = None
                tag = soup.find('meta',{'name':'chaptertitle'})
                if tag:
                    chaptertitle = tag['content']

                if chaptertitle and chapterorigtitle and chapterorigtitle != chaptertitle:
                    origdata = data
                    # print("\n%s\n%s\n"%(chapterorigtitle,chaptertitle))
                    data = data.replace(u'<meta name="chaptertitle" content="'+chaptertitle+u'"></meta>',
                                        u'<meta name="chaptertitle" content="'+chapterorigtitle+u'"></meta>')
                    data = data.replace(u'<title>'+chaptertitle+u'</title>',u'<title>'+chapterorigtitle+u'</title>')
                    data = data.replace(u'<h3>'+chaptertitle+u'</h3>',u'<h3>'+chapterorigtitle+u'</h3>')

                    entrychanged = ( origdata != data )
                    changed = changed or entrychanged
                    
                    if entrychanged:
                        ## go after the TOC entry, too.
                        # <navPoint id="file0005" playOrder="6">
                        #   <navLabel>
                        #     <text>5. (new) Chapter 4</text>
                        #   </navLabel>
                        #   <content src="OEBPS/file0005.xhtml"/>
                        # </navPoint>
                        for contenttag in tocncxdom.getElementsByTagName("content"):
                            if contenttag.getAttribute('src') == zf:
                                texttag = contenttag.parentNode.getElementsByTagName('navLabel')[0].getElementsByTagName('text')[0]
                                texttag.childNodes[0].replaceWholeText(chaptertoctitle)
                                # logger.debug("text label:%s"%texttag.toxml())
                                continue
                    
                outputepub.writestr(zf,data.encode('utf-8'))
            else:
                # possibly binary data, thus no .encode().
                outputepub.writestr(zf,data)

    outputepub.writestr('toc.ncx',tocncxdom.toxml(encoding='utf-8'))
    outputepub.close()
    # declares all the files created by Windows.  otherwise, when
    # it runs in appengine, windows unzips the files as 000 perms.
    for zf in outputepub.filelist:
        zf.create_system = 0
        
    # only *actually* write if changed.
    if changed:
        if isinstance(outfile,basestring):
            with open(outfile,"wb") as outputio:
                outputio.write(zipio.getvalue())
        else:
            outfile.write(zipio.getvalue())

    inputepub.close()
    zipio.close()
            
#.........这里部分代码省略.........

# 需要导入模块: from zipfile import ZipFile [as 别名]
# 或者: from zipfile.ZipFile import debug [as 别名]
def reset_orig_chapters_epub(inputio, outfile):
    inputepub = ZipFile(inputio, "r")  # works equally well with a path or a blob

    ## build zip in memory in case updating in place(CLI).
    zipio = StringIO()

    ## Write mimetype file, must be first and uncompressed.
    ## Older versions of python(2.4/5) don't allow you to specify
    ## compression by individual file.
    ## Overwrite if existing output file.
    outputepub = ZipFile(zipio, "w", compression=ZIP_STORED)
    outputepub.debug = 3
    outputepub.writestr("mimetype", "application/epub+zip")
    outputepub.close()

    ## Re-open file for content.
    outputepub = ZipFile(zipio, "a", compression=ZIP_DEFLATED)
    outputepub.debug = 3

    changed = False

    unmerge_tocncxdoms = {}
    ## spin through file contents, saving any unmerge toc.ncx files.
    for zf in inputepub.namelist():
        ## logger.debug("zf:%s"%zf)
        if zf.endswith("/toc.ncx"):
            ## logger.debug("toc.ncx zf:%s"%zf)
            unmerge_tocncxdoms[zf] = parseString(inputepub.read(zf))

    tocncxdom = parseString(inputepub.read("toc.ncx"))
    ## spin through file contents.
    for zf in inputepub.namelist():
        if zf not in ["mimetype", "toc.ncx"] and not zf.endswith("/toc.ncx"):
            entrychanged = False
            data = inputepub.read(zf)
            # if isinstance(data,unicode):
            #     logger.debug("\n\n\ndata is unicode\n\n\n")
            if re.match(r".*/file\d+\.xhtml", zf):
                # logger.debug("zf:%s"%zf)
                data = data.decode("utf-8")
                soup = bs.BeautifulSoup(data, "html5lib")

                chapterorigtitle = None
                tag = soup.find("meta", {"name": "chapterorigtitle"})
                if tag:
                    chapterorigtitle = tag["content"]

                # toctitle is separate for add_chapter_numbers:toconly users.
                chaptertoctitle = None
                tag = soup.find("meta", {"name": "chaptertoctitle"})
                if tag:
                    chaptertoctitle = tag["content"]
                elif chapterorigtitle:
                    chaptertoctitle = chapterorigtitle

                chaptertitle = None
                tag = soup.find("meta", {"name": "chaptertitle"})
                if tag:
                    chaptertitle = tag["content"]

                if chaptertitle and chapterorigtitle and chapterorigtitle != chaptertitle:
                    origdata = data
                    # print("\n%s\n%s\n"%(chapterorigtitle,chaptertitle))
                    data = data.replace(
                        u'<meta name="chaptertitle" content="' + chaptertitle + u'"></meta>',
                        u'<meta name="chaptertitle" content="' + chapterorigtitle + u'"></meta>',
                    )
                    data = data.replace(
                        u"<title>" + chaptertitle + u"</title>", u"<title>" + chapterorigtitle + u"</title>"
                    )
                    data = data.replace(u"<h3>" + chaptertitle + u"</h3>", u"<h3>" + chapterorigtitle + u"</h3>")

                    entrychanged = origdata != data
                    changed = changed or entrychanged

                    if entrychanged:
                        _replace_tocncx(tocncxdom, zf, chaptertoctitle)
                        ## Also look for and update individual
                        ## book toc.ncx files for anthology in case
                        ## it's unmerged.
                        zf_toc = zf[: zf.rfind("/OEBPS/")] + "/toc.ncx"
                        mergedprefix_len = len(zf[: zf.rfind("/OEBPS/")]) + 1

                        if zf_toc in unmerge_tocncxdoms:
                            _replace_tocncx(unmerge_tocncxdoms[zf_toc], zf[mergedprefix_len:], chaptertoctitle)

                outputepub.writestr(zf, data.encode("utf-8"))
            else:
                # possibly binary data, thus no .encode().
                outputepub.writestr(zf, data)

    for tocnm, tocdom in unmerge_tocncxdoms.items():
        outputepub.writestr(tocnm, tocdom.toxml(encoding="utf-8"))

    outputepub.writestr("toc.ncx", tocncxdom.toxml(encoding="utf-8"))
    outputepub.close()
    # declares all the files created by Windows.  otherwise, when
    # it runs in appengine, windows unzips the files as 000 perms.
    for zf in outputepub.filelist:
        zf.create_system = 0
#.........这里部分代码省略.........

# 需要导入模块: from zipfile import ZipFile [as 别名]
# 或者: from zipfile.ZipFile import debug [as 别名]
def doMerge(outputio,
            files,
            authoropts=[],
            titleopt=None,
            descopt=None,
            tags=[],
            languages=['en'],
            titlenavpoints=True,
            originalnavpoints=True,
            flattentoc=False,
            printtimes=False,
            coverjpgpath=None,
            keepmetadatafiles=False,
            source=None):
    '''
    outputio = output file name or StringIO.
    files = list of input file names or StringIOs.
    authoropts = list of authors to use, otherwise add from all input
    titleopt = title, otherwise '<first title> Anthology'
    descopt = description, otherwise '<title> by <author>' list for all input
    tags = dc:subject tags to include, otherwise none.
    languages = dc:language tags to include
    titlenavpoints if true, put in a new TOC entry for each epub, nesting each epub's chapters under it
    originalnavpoints if true, include the original TOCs from each epub
    flattentoc if true, flatten TOC down to one level only.
    coverjpgpath, Path to a jpg to use as cover image.
    '''

    printt = partial(cond_print,printtimes)

    ## Python 2.5 ZipFile is rather more primative than later
    ## versions.  It can operate on a file, or on a StringIO, but
    ## not on an open stream.  OTOH, I suspect we would have had
    ## problems with closing and opening again to change the
    ## compression type anyway.

    filecount=0
    t = time()

    ## Write mimetype file, must be first and uncompressed.
    ## Older versions of python(2.4/5) don't allow you to specify
    ## compression by individual file.
    ## Overwrite if existing output file.
    outputepub = ZipFile(outputio, "w", compression=ZIP_STORED, allowZip64=True)
    outputepub.debug = 3
    outputepub.writestr("mimetype", "application/epub+zip")
    outputepub.close()

    ## Re-open file for content.
    outputepub = ZipFile(outputio, "a", compression=ZIP_DEFLATED, allowZip64=True)
    outputepub.debug = 3

    ## Create META-INF/container.xml file.  The only thing it does is
    ## point to content.opf
    containerdom = getDOMImplementation().createDocument(None, "container", None)
    containertop = containerdom.documentElement
    containertop.setAttribute("version","1.0")
    containertop.setAttribute("xmlns","urn:oasis:names:tc:opendocument:xmlns:container")
    rootfiles = containerdom.createElement("rootfiles")
    containertop.appendChild(rootfiles)
    rootfiles.appendChild(newTag(containerdom,"rootfile",{"full-path":"content.opf",
                                                          "media-type":"application/oebps-package+xml"}))
    outputepub.writestr("META-INF/container.xml",containerdom.toprettyxml(indent='   ',encoding='utf-8'))

    ## Process input epubs.

    items = [] # list of (id, href, type) tuples(all strings) -- From .opfs' manifests
    items.append(("ncx","toc.ncx","application/x-dtbncx+xml")) ## we'll generate the toc.ncx file,
                                                               ## but it needs to be in the items manifest.
    itemrefs = [] # list of strings -- idrefs from .opfs' spines
    navmaps = [] # list of navMap DOM elements -- TOC data for each from toc.ncx files
    is_ffdl_epub = [] # list of t/f

    itemhrefs = {} # hash of item[id]s to itemref[href]s -- to find true start of book(s).
    firstitemhrefs = []

    booktitles = [] # list of strings -- Each book's title
    allauthors = [] # list of lists of strings -- Each book's list of authors.

    filelist = []

    printt("prep output:%s"%(time()-t))
    t = time()

    booknum=1
    firstmetadom = None
    for file in files:
        if file == None : continue

        book = "%d" % booknum
        bookdir = "%d/" % booknum
        bookid = "a%d" % booknum

        epub = ZipFile(file, 'r')

        ## Find the .opf file.
        container = epub.read("META-INF/container.xml")
        containerdom = parseString(container)
        rootfilenodelist = containerdom.getElementsByTagNameNS("*","rootfile")
        rootfilename = rootfilenodelist[0].getAttribute("full-path")
#.........这里部分代码省略.........