Python sigil_bs4.BeautifulSoup类代码示例

def performPageMapUpdates(data, currentdir, 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 = {}
    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('#')
                    url = parts[0]
                    fragment = ""
                    if len(parts) > 1:
                        fragment = parts[1]
                    bookrelpath = os.path.join(currentdir, unquoteurl(url))
                    bookrelpath = os.path.normpath(bookrelpath)
                    bookrelpath = bookrelpath.replace(os.sep, "/")
                    if bookrelpath in updates:
                        attribute_value = updates[bookrelpath]
                        if fragment != "":
                            attribute_value = attribute_value + "#" + fragment
                        attribute_value = quoteurl(attribute_value)
                        tag[att] = attribute_value
    newdata = soup.decodexml(indent_level=0, formatter='minimal', indent_chars="  ")
    return newdata

def performNCXSourceUpdates(data, currentdir, keylist, valuelist):
    # rebuild serialized lookup dictionary
    updates = {}
    for i in range(0, len(keylist)):
        updates[ keylist[i] ] = valuelist[i]
    xmlbuilder = LXMLTreeBuilderForXML(parser=None, empty_element_tags=ebook_xml_empty_tags)
    soup = BeautifulSoup(data, features=None, builder=xmlbuilder)
    for tag in soup.find_all("content"):
        if "src" in tag.attrs:
            src = tag["src"]
            if src.find(":") == -1:
                parts = src.split('#')
                url = parts[0]
                fragment = ""
                if len(parts) > 1:
                    fragment = parts[1]
                bookrelpath = os.path.join(currentdir, unquoteurl(url))
                bookrelpath = os.path.normpath(bookrelpath)
                bookrelpath = bookrelpath.replace(os.sep, "/")
                if bookrelpath in updates:
                    attribute_value = updates[bookrelpath]
                    if fragment != "":
                        attribute_value = attribute_value + "#" + fragment
                    attribute_value = quoteurl(attribute_value)
                    tag["src"] = attribute_value
    newdata = soup.decodexml(indent_level=0, formatter='minimal', indent_chars="  ")
    return newdata

def diagnose(data):
    """Diagnostic suite for isolating common problems."""
    print("Diagnostic running on Beautiful Soup %s" % __version__)
    print("Python version %s" % sys.version)

    basic_parsers = ["html.parser", "html5lib", "lxml"]
    for name in basic_parsers:
        for builder in builder_registry.builders:
            if name in builder.features:
                break
        else:
            basic_parsers.remove(name)
            print((
                "I noticed that %s is not installed. Installing it may help." %
                name))

    if 'lxml' in basic_parsers:
        basic_parsers.append(["lxml", "xml"])
        try:
            from lxml import etree
            print("Found lxml version %s" % ".".join(map(str,etree.LXML_VERSION)))
        except ImportError as e:
            print (
                "lxml is not installed or couldn't be imported.")


    if 'html5lib' in basic_parsers:
        try:
            import html5lib
            print("Found html5lib version %s" % html5lib.__version__)
        except ImportError as e:
            print (
                "html5lib is not installed or couldn't be imported.")

    if hasattr(data, 'read'):
        data = data.read()
    elif os.path.exists(data):
        print('"%s" looks like a filename. Reading data from the file.' % data)
        data = open(data).read()
    elif data.startswith("http:") or data.startswith("https:"):
        print('"%s" looks like a URL. Beautiful Soup is not an HTTP client.' % data)
        print("You need to use some other library to get the document behind the URL, and feed that document to Beautiful Soup.")
        return
    print()

    for parser in basic_parsers:
        print("Trying to parse your markup with %s" % parser)
        success = False
        try:
            soup = BeautifulSoup(data, parser)
            success = True
        except Exception as e:
            print("%s could not parse the markup." % parser)
            traceback.print_exc()
        if success:
            print("Here's what %s did with the markup:" % parser)
            print(soup.prettify())

        print("-" * 80)

def repairXML(data, self_closing_tags=ebook_xml_empty_tags, indent_chars="  "):
    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')
    xmlbuilder = LXMLTreeBuilderForXML(parser=None, empty_element_tags=self_closing_tags)
    soup = BeautifulSoup(data, features=None, from_encoding="utf-8", builder=xmlbuilder)
    newdata = soup.decodexml(indent_level=0, formatter='minimal', indent_chars=indent_chars)
    return newdata

def repairXML(data, mtype="", indent_chars="  "):
    data = _remove_xml_header(data)
    data = _make_it_sane(data)
    voidtags = get_void_tags(mtype)
    # 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=voidtags)
    soup = BeautifulSoup(data, features=None, from_encoding="utf-8", builder=xmlbuilder)
    newdata = soup.decodexml(indent_level=0, formatter='minimal', indent_chars=indent_chars)
    return newdata

def repairXML(data, mtype="", indent_chars="  "):
    newdata = _remove_xml_header(data)
    # if well-formed - don't mess with it
    if _well_formed(newdata):
        return data
    newdata = _make_it_sane(newdata)
    if not _well_formed(newdata):
        newdata = _reformat(newdata)
        if mtype == "application/oebps-package+xml":
            newdata = newdata.decode('utf-8')
            newdata = Opf_Parser(newdata).rebuild_opfxml()
    # lxml requires utf-8 on Mac, won't work with unicode
    if isinstance(newdata, str):
        newdata = newdata.encode('utf-8')
    voidtags = get_void_tags(mtype)
    xmlbuilder = LXMLTreeBuilderForXML(parser=None, empty_element_tags=voidtags)
    soup = BeautifulSoup(newdata, features=None, from_encoding="utf-8", builder=xmlbuilder)
    newdata = soup.decodexml(indent_level=0, formatter='minimal', indent_chars=indent_chars)
    return newdata

def anchorNCXUpdates(data, originating_filename, keylist, valuelist):
    # rebuild serialized lookup dictionary
    id_dict = {}
    for i in range(0, len(keylist)):
        id_dict[ keylist[i] ] = valuelist[i]
    xmlbuilder = LXMLTreeBuilderForXML(parser=None, empty_element_tags=ebook_xml_empty_tags)
    soup = BeautifulSoup(data, features=None, builder=xmlbuilder)
    original_filename_with_relative_path = TEXT_FOLDER_NAME  + "/" + originating_filename
    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) and (len(parts) > 1) and (parts[0] == original_filename_with_relative_path) and (parts[1] != ""):
                    fragment_id = parts[1]
                    if fragment_id in id_dict:
                        attribute_value = TEXT_FOLDER_NAME + "/" + quoteurl(id_dict[fragment_id]) + "#" + fragment_id
                        tag["src"] = attribute_value
    newdata = soup.decodexml(indent_level=0, formatter='minimal', indent_chars="  ")
    return newdata

def repairXML(data, self_closing_tags=ebook_xml_empty_tags, indent_chars="  "):
    xmlbuilder = LXMLTreeBuilderForXML(parser=None, empty_element_tags=self_closing_tags)
    soup = BeautifulSoup(data, features=None, builder=xmlbuilder)
    newdata = soup.decodexml(indent_level=0, formatter='minimal', indent_chars=indent_chars)
    return newdata