/// <summary>
/// Initializes a new instance of the <see cref="MimeKit.IO.Filters.CharsetFilter"/> class.
/// </summary>
/// <remarks>
/// Creates a new <see cref="CharsetFilter"/> to convert text from the specified
/// source encoding into the target charset encoding.
/// </remarks>
/// <param name="sourceEncoding">Source encoding.</param>
/// <param name="targetEncoding">Target encoding.</param>
/// <exception cref="System.ArgumentNullException">
/// <para><paramref name="sourceEncoding"/> is <c>null</c>.</para>
/// <para>-or-</para>
/// <para><paramref name="targetEncoding"/> is <c>null</c>.</para>
/// </exception>
public CharsetFilter(Encoding sourceEncoding, Encoding targetEncoding)
{
if (sourceEncoding == null)
{
throw new ArgumentNullException(nameof(sourceEncoding));
}
if (targetEncoding == null)
{
throw new ArgumentNullException(nameof(targetEncoding));
}
SourceEncoding = sourceEncoding;
TargetEncoding = targetEncoding;
decoder = (Decoder)SourceEncoding.GetDecoder();
encoder = (Encoder)TargetEncoding.GetEncoder();
}
static IList<Word> GetRfc822Words (FormatOptions options, Encoding charset, string text, bool phrase)
{
var encoder = charset.GetEncoder ();
var words = new List<Word> ();
var chars = new char[2];
var saved = new Word ();
var word = new Word ();
int nchars, n, i = 0;
char c;
while (i < text.Length) {
c = text[i++];
if (c < 256 && IsBlank (c)) {
if (word.ByteCount > 0) {
words.Add (word);
word = new Word ();
}
word.StartIndex = i;
} else {
// save state in case adding this character exceeds the max line length
word.CopyTo (saved);
if (c < 127) {
if (IsCtrl (c)) {
word.Encoding = options.AllowMixedHeaderCharsets ? Math.Max (word.Encoding, 1) : 2;
word.Type = WordType.EncodedWord;
word.EncodeCount++;
} else if (phrase && !IsAtom (c)) {
// phrases can have quoted strings
if (word.Type == WordType.Atom)
word.Type = WordType.QuotedString;
}
if (c == '"' || c == '\\')
word.QuotedPairs++;
word.ByteCount++;
word.CharCount++;
nchars = 1;
} else if (c < 256) {
// iso-8859-1
word.Encoding = options.AllowMixedHeaderCharsets ? Math.Max (word.Encoding, 1) : 2;
word.Type = WordType.EncodedWord;
word.EncodeCount++;
word.ByteCount++;
word.CharCount++;
nchars = 1;
} else {
if (char.IsSurrogatePair (text, i - 1)) {
chars[1] = text[i++];
nchars = 2;
} else {
nchars = 1;
}
chars[0] = c;
try {
n = encoder.GetByteCount (chars, 0, nchars, true);
} catch {
n = 3;
}
word.Type = WordType.EncodedWord;
word.CharCount += nchars;
word.EncodeCount += n;
word.ByteCount += n;
word.Encoding = 2;
}
if (ExceedsMaxLineLength (options, charset, word)) {
// restore our previous state
saved.CopyTo (word);
i -= nchars;
// Note: if the word is longer than what we can fit on
// one line, then we need to encode it.
if (word.Type == WordType.Atom) {
word.Type = WordType.EncodedWord;
// in order to fit this long atom under MaxLineLength, we need to
// account for the added length of =?us-ascii?q?...?=
n = "us-ascii".Length + 7;
word.CharCount -= n;
word.ByteCount -= n;
i -= n;
}
words.Add (word);
saved.Type = word.Type;
word = new Word ();
// Note: the word-type needs to be preserved when breaking long words.
word.Type = saved.Type;
word.StartIndex = i;
}
}
}
if (word.ByteCount > 0)
words.Add (word);
return words;
}
static byte[] CharsetConvert (Encoding charset, char[] word, int length, out int converted)
{
var encoder = charset.GetEncoder ();
int count = encoder.GetByteCount (word, 0, length, true);
var encoded = new byte[count];
converted = encoder.GetBytes (word, 0, length, encoded, 0, true);
return encoded;
}
/// <summary>
/// Initializes a new instance of the <see cref="MimeKit.IO.Filters.CharsetFilter"/> class.
/// </summary>
/// <remarks>
/// Creates a new <see cref="CharsetFilter"/> to convert text from the specified
/// source encoding into the target charset encoding.
/// </remarks>
/// <param name="sourceEncoding">Source encoding.</param>
/// <param name="targetEncoding">Target encoding.</param>
/// <exception cref="System.ArgumentNullException">
/// <para><paramref name="sourceEncoding"/> is <c>null</c>.</para>
/// <para>-or-</para>
/// <para><paramref name="targetEncoding"/> is <c>null</c>.</para>
/// </exception>
public CharsetFilter (Encoding sourceEncoding, Encoding targetEncoding)
{
if (sourceEncoding == null)
throw new ArgumentNullException ("sourceEncoding");
if (targetEncoding == null)
throw new ArgumentNullException ("targetEncoding");
SourceEncoding = sourceEncoding;
TargetEncoding = targetEncoding;
decoder = (Decoder) SourceEncoding.GetDecoder ();
encoder = (Encoder) TargetEncoding.GetEncoder ();
}
