/// <summary> /// Returns up to maxEncodings codepages that are assumed to be apropriate /// </summary> /// <param name="input">array containing the raw data</param> /// <param name="maxEncodings">maxiumum number of encodings to detect</param> /// <returns>an array of Encoding with assumed encodings</returns> public static Encoding[] DetectInputCodepages(byte[] input, int maxEncodings) { if (maxEncodings < 1) throw new ArgumentOutOfRangeException("maxEncodings", "at least one encoding must be returned"); if (input == null) throw new ArgumentNullException("input"); // empty strings can always be encoded as ASCII if (input.Length == 0) return new Encoding[] { Encoding.ASCII }; // expand the string to be at least 256 bytes if (input.Length < 256) { byte[] newInput = new byte[256]; int steps = 256 / input.Length; for (int i = 0; i < steps; i++) Array.Copy(input, 0, newInput, input.Length * i, input.Length); int rest = 256 % input.Length; if (rest > 0) Array.Copy(input, 0, newInput, steps * input.Length, rest); input = newInput; } List<Encoding> result = new List<Encoding>(); // get the IMultiLanguage" interface IMultiLanguage2 multilang2 = new CMultiLanguageClass(); if (multilang2 == null) throw new COMException("Failed to get IMultilang2"); try { DetectEncodingInfo[] detectedEncdings = new DetectEncodingInfo[maxEncodings]; int scores = detectedEncdings.Length; int srcLen = input.Length; // setup options (none) const MLDETECTCP options = MLDETECTCP.MLDETECTCP_NONE; // finally... call to DetectInputCodepage multilang2.DetectInputCodepage(options, 0, ref input[0], ref srcLen, ref detectedEncdings[0], ref scores); // get result if (scores > 0) { for (int i = 0; i < scores; i++) { // add the result result.Add(Encoding.GetEncoding((int)detectedEncdings[i].nCodePage)); } } } finally { Marshal.FinalReleaseComObject(multilang2); } // nothing found return result.ToArray(); }
public static Encoding[] GetEncodings(byte[] input, int maxEncodings) { if (input.IsNullOrEmpty()) { return(new[] { Default }); } if (maxEncodings < 1) { maxEncodings = 1; } // expand the string to be at least 256 bytes if (input.Length < 256) { byte[] newInput = new byte[256]; int steps = 256 / input.Length; for (int i = 0; i < steps; i++) { Array.Copy(input, 0, newInput, input.Length * i, input.Length); } int rest = 256 % input.Length; if (rest > 0) { Array.Copy(input, 0, newInput, steps * input.Length, rest); } input = newInput; } List <Encoding> result = new List <Encoding>(); // get the IMultiLanguage" interface IMultiLanguage2 multiLang2 = new CMultiLanguageClass(); if (multiLang2 == null) { throw new COMException("Failed to get " + nameof(IMultiLanguage2)); } try { DetectEncodingInfo[] detectedEncodings = new DetectEncodingInfo[maxEncodings]; int scores = detectedEncodings.Length; int srcLen = input.Length; // finally... call to DetectInputCodepage multiLang2.DetectInputCodepage(MLDETECTCP.MLDETECTCP_NONE, 0, ref input[0], ref srcLen, ref detectedEncodings[0], ref scores); // get result if (scores > 0) { for (int i = 0; i < scores; i++) { result.Add(Encoding.GetEncoding((int)detectedEncodings[i].nCodePage)); } } } finally { Marshal.FinalReleaseComObject(multiLang2); } return(result.ToArray()); }
/// <summary> /// Rerurns up to maxEncodings codpages that are assumed to be apropriate /// </summary> /// <param name="input">array containing the raw data</param> /// <param name="maxEncodings">maxiumum number of encodings to detect</param> /// <returns>an array of Encoding with assumed encodings</returns> public static Encoding[] DetectInputCodepages(byte[] input, int maxEncodings) { if (maxEncodings < 1) { throw new ArgumentOutOfRangeException("maxEncodings", "at least one encoding must be returned"); } if (input == null) { throw new ArgumentNullException("input"); } // empty strings can always be encoded as ASCII if (input.Length == 0) { return new Encoding[] { Encoding.ASCII }; } // expand the string to be at least 256 bytes if (input.Length < 256) { byte[] newInput = new byte[256]; int steps = 256 / input.Length; for (int i = 0; i < steps; i++) { Array.Copy(input, 0, newInput, input.Length * i, input.Length); } int rest = 256 % input.Length; if (rest > 0) { Array.Copy(input, 0, newInput, steps * input.Length, rest); } input = newInput; } List<Encoding> result = new List<Encoding>(); // get the IMultiLanguage" interface IMultiLanguage2 multilang2 = new CMultiLanguageClass(); if (multilang2 == null) { throw new COMException("Failed to get IMultilang2"); } try { DetectEncodingInfo[] detectedEncdings = new DetectEncodingInfo[maxEncodings]; int scores = detectedEncdings.Length; int srcLen = input.Length; // setup options (none) const Mldetectcp options = Mldetectcp.MldetectcpNone; // finally... call to DetectInputCodepage multilang2.DetectInputCodepage(options, 0, ref input[0], ref srcLen, ref detectedEncdings[0], ref scores); // get result if (scores > 0) { for (int i = 0; i < scores; i++) { // add the result result.Add(Encoding.GetEncoding((int)detectedEncdings[i].nCodePage)); } } } finally { Marshal.FinalReleaseComObject(multilang2); } // nothing found return result.ToArray(); }
public static Encoding DetectEncoding(byte[] data, int dataLength, DetectOption flags) { Encoding encoding2; if (data == null) { throw new ArgumentNullException("data"); } if (data.Length < dataLength) { throw new ArgumentException("data length is less than dataLength"); } if (dataLength == 0) { return null; } if ((flags & DetectOption.TrySimpleDetectFirst) > DetectOption.Default) { Encoding encoding = SimpleDetectEncoding(data, dataLength); if (encoding != null) { return encoding; } } IMultiLanguage2 o = new CMultiLanguageClass() as IMultiLanguage2; if (o == null) { return (((flags & DetectOption.TrySimpleDetectFirst) > DetectOption.Default) ? null : SimpleDetectEncoding(data, dataLength)); } try { DetectEncodingInfo[] lpEncoding = new DetectEncodingInfo[1]; int length = lpEncoding.Length; switch (o.DetectInputCodepage(((MLDETECTCP) flags) & MLDETECTCP.MLDETECTCP_MASK, 0, data, ref dataLength, lpEncoding, ref length)) { case 0: case 1: if (length <= 0) { break; } return Encoding.GetEncoding((int) lpEncoding[0].nCodePage); } encoding2 = null; } finally { Marshal.FinalReleaseComObject(o); } return encoding2; }