/// <summary>
/// Parse the Data inline image.
/// </summary>
private HtmlImageInfo ReadDataUri(string src)
{
if (DataUri.TryCreate(src, out DataUri dataUri))
{
Size size;
knownContentType.TryGetValue(dataUri.Mime, out ImagePartType type);
var ipart = mainPart.AddImagePart(type);
using (var outputStream = ipart.GetStream(FileMode.Create))
{
outputStream.Write(dataUri.Data, 0, dataUri.Data.Length);
outputStream.Seek(0L, SeekOrigin.Begin);
size = GetImageSize(outputStream);
}
return(new HtmlImageInfo()
{
Source = src,
ImagePartId = mainPart.GetIdOfPart(ipart),
Size = size
});
}
return(null);
}
//____________________________________________________________________
//
// Public Functionality
/// <summary>
/// Download the remote or local image located at the specified url.
/// </summary>
public HtmlImageInfo Download(string imageUri)
{
if (prefetchedImages.Contains(imageUri))
{
return(prefetchedImages[imageUri]);
}
if (DataUri.IsWellFormed(imageUri)) // data inline, encoded in base64
{
return(ReadDataUri(imageUri));
}
return(DownloadRemoteImage(imageUri));
}
/// <summary>
/// Parse an instance of the Data URI scheme.
/// </summary>
/// <param name="uri">The inline Data URI to parse.</param>
/// <param name="result">When this method returns, contains a DataUri constructed
/// from <paramref name="uri"/>. This parameter is passed uninitialized.</param>
/// <returns>A <see cref="Boolean"/> value that is true if the DataUri was
/// successfully created; otherwise, false.</returns>
public static bool TryCreate(string uri, out DataUri result)
{
// expected format: data:[<MIME-type>][;charset=<encoding>][;base64],<data>
// The encoding is indicated by ;base64. If it's present the data is encoded as base64. Without it the data (as a sequence of octets)
// is represented using ASCII encoding for octets inside the range of safe URL characters and using the standard %xx hex encoding
// of URLs for octets outside that range. If <MIME-type> is omitted, it defaults to text/plain;charset=US-ASCII.
// (As a shorthand, the type can be omitted but the charset parameter supplied.)
// Some browsers (Chrome, Opera, Safari, Firefox) accept a non-standard ordering if both ;base64 and ;charset are supplied,
// while Internet Explorer requires that the charset's specification must precede the base64 token.
// http://en.wikipedia.org/wiki/Data_URI_scheme
// We will stick for IE compliance for the moment...
Match match = dataUriRegex.Match(uri);
result = null;
if (!match.Success)
{
return(false);
}
byte[] rawData;
string mime;
Encoding charSet = Encoding.ASCII;
// if mime is omitted, set default value as it stands in the norm
if (match.Groups["mime"].Length == 0)
{
mime = "text/plain";
}
else
{
mime = match.Groups["mime"].Value;
}
if (match.Groups["charset"].Length > 0)
{
try
{
charSet = Encoding.GetEncoding(match.Groups["charset"].Value);
}
catch (ArgumentException)
{
// charSet was not recognized
return(false);
}
}
// is it encoded in base64?
if (match.Groups["base64"].Length > 0)
{
// be careful that the raw data is encoded for url (standard %xx hex encoding)
string base64 = HttpUtility.HtmlDecode(match.Groups["data"].Value);
try
{
rawData = Convert.FromBase64String(base64);
}
catch (FormatException)
{
// Base64 data is invalid
return(false);
}
}
else
{
// the <data> represents some text (like html snippet) and must be decoded.
string raw = HttpUtility.UrlDecode(match.Groups["data"].Value);
try
{
// we convert back to UTF-8 for easier processing later and to have a "referential" encoding
rawData = Encoding.Convert(charSet, Encoding.UTF8, charSet.GetBytes(raw));
}
catch (ArgumentException)
{
// UTF-8 Encoded data is invalid
return(false);
}
}
result = new DataUri(mime, rawData);
return(true);
}
