Visual Studio Reformat: Emby.Server.Implementations Part T-T

This commit is contained in:
Erwin de Haan 2019-01-13 20:22:56 +01:00
parent 0efc699e3d
commit 25f0315e91
39 changed files with 1054 additions and 892 deletions

View file

@ -1,14 +1,14 @@
using System;
using System.Collections.Generic;
using System.Linq;
using MediaBrowser.Controller.Configuration;
using MediaBrowser.Controller.Dto;
using MediaBrowser.Controller.Entities; using MediaBrowser.Controller.Entities;
using MediaBrowser.Controller.Entities.TV; using MediaBrowser.Controller.Entities.TV;
using MediaBrowser.Controller.Library; using MediaBrowser.Controller.Library;
using MediaBrowser.Controller.TV; using MediaBrowser.Controller.TV;
using MediaBrowser.Model.Entities; using MediaBrowser.Model.Entities;
using MediaBrowser.Model.Querying; using MediaBrowser.Model.Querying;
using System;
using System.Collections.Generic;
using System.Linq;
using MediaBrowser.Controller.Configuration;
using MediaBrowser.Controller.Dto;
namespace Emby.Server.Implementations.TV namespace Emby.Server.Implementations.TV
{ {

View file

@ -1,10 +1,8 @@
using System; using System;
using System.Collections.Generic; using System.Collections.Generic;
using System.IO;
using System.IO.Compression;
using NLangDetect.Core.Utils;
using MediaBrowser.Model.Serialization;
using System.Linq; using System.Linq;
using MediaBrowser.Model.Serialization;
using NLangDetect.Core.Utils;
namespace NLangDetect.Core namespace NLangDetect.Core
{ {

View file

@ -1,8 +1,8 @@
using System; using System;
using System.IO;
using System.IO.Compression; using System.IO.Compression;
using System.Xml; using System.Xml;
using NLangDetect.Core.Utils; using NLangDetect.Core.Utils;
using System.IO;
namespace NLangDetect.Core namespace NLangDetect.Core
{ {

View file

@ -1,10 +1,9 @@
using System;
using System.Collections.Generic; using System.Collections.Generic;
using System.Globalization; using System.Globalization;
using System.IO; using System.IO;
using System.Reflection;
using System.Text.RegularExpressions;
using System.Linq; using System.Linq;
using System; using System.Text.RegularExpressions;
namespace NLangDetect.Core.Utils namespace NLangDetect.Core.Utils
{ {

View file

@ -1,9 +1,9 @@
using System; using System;
using System.Text; using System.Text;
using MediaBrowser.Model.IO; using MediaBrowser.Model.IO;
using Microsoft.Extensions.Logging;
using MediaBrowser.Model.Serialization; using MediaBrowser.Model.Serialization;
using MediaBrowser.Model.Text; using MediaBrowser.Model.Text;
using Microsoft.Extensions.Logging;
using NLangDetect.Core; using NLangDetect.Core;
using UniversalDetector; using UniversalDetector;

View file

@ -57,22 +57,29 @@ namespace UniversalDetector.Core
int codingState = 0; int codingState = 0;
int max = offset + len; int max = offset + len;
for (int i = offset; i < max; i++) { for (int i = offset; i < max; i++)
{
codingState = codingSM.NextState(buf[i]); codingState = codingSM.NextState(buf[i]);
if (codingState == SMModel.ERROR) { if (codingState == SMModel.ERROR)
{
state = ProbingState.NotMe; state = ProbingState.NotMe;
break; break;
} }
if (codingState == SMModel.ITSME) { if (codingState == SMModel.ITSME)
{
state = ProbingState.FoundIt; state = ProbingState.FoundIt;
break; break;
} }
if (codingState == SMModel.START) { if (codingState == SMModel.START)
{
int charLen = codingSM.CurrentCharLen; int charLen = codingSM.CurrentCharLen;
if (i == offset) { if (i == offset)
{
lastChar[1] = buf[offset]; lastChar[1] = buf[offset];
distributionAnalyser.HandleOneChar(lastChar, 0, charLen); distributionAnalyser.HandleOneChar(lastChar, 0, charLen);
} else { }
else
{
distributionAnalyser.HandleOneChar(buf, i - 1, charLen); distributionAnalyser.HandleOneChar(buf, i - 1, charLen);
} }
} }

View file

@ -97,9 +97,11 @@ namespace UniversalDetector.Core
{ {
//we only care about 2-bytes character in our distribution analysis //we only care about 2-bytes character in our distribution analysis
int order = (charLen == 2) ? GetOrder(buf, offset) : -1; int order = (charLen == 2) ? GetOrder(buf, offset) : -1;
if (order >= 0) { if (order >= 0)
{
totalChars++; totalChars++;
if (order < tableSize) { // order is valid if (order < tableSize)
{ // order is valid
if (512 > charToFreqOrder[order]) if (512 > charToFreqOrder[order])
freqChars++; freqChars++;
} }
@ -124,7 +126,8 @@ namespace UniversalDetector.Core
// negative answer // negative answer
if (totalChars <= 0 || freqChars <= MINIMUM_DATA_THRESHOLD) if (totalChars <= 0 || freqChars <= MINIMUM_DATA_THRESHOLD)
return SURE_NO; return SURE_NO;
if (totalChars != freqChars) { if (totalChars != freqChars)
{
float r = freqChars / ((totalChars - freqChars) * typicalDistributionRatio); float r = freqChars / ((totalChars - freqChars) * typicalDistributionRatio);
if (r < SURE_YES) if (r < SURE_YES)
return r; return r;
@ -2559,12 +2562,15 @@ namespace UniversalDetector.Core
/// </summary> /// </summary>
public override int GetOrder(byte[] buf, int offset) public override int GetOrder(byte[] buf, int offset)
{ {
if (buf[offset] >= 0xA4) { if (buf[offset] >= 0xA4)
{
if (buf[offset + 1] >= 0xA1) if (buf[offset + 1] >= 0xA1)
return 157 * (buf[offset] - 0xA4) + buf[offset + 1] - 0xA1 + 63; return 157 * (buf[offset] - 0xA4) + buf[offset + 1] - 0xA1 + 63;
else else
return 157 * (buf[offset] - 0xA4) + buf[offset + 1] - 0x40; return 157 * (buf[offset] - 0xA4) + buf[offset + 1] - 0x40;
} else { }
else
{
return -1; return -1;
} }
} }

View file

@ -40,7 +40,8 @@ using System.IO;
namespace UniversalDetector.Core namespace UniversalDetector.Core
{ {
public enum ProbingState { public enum ProbingState
{
Detecting = 0, // no sure answer yet, but caller can ask for confidence Detecting = 0, // no sure answer yet, but caller can ask for confidence
FoundIt = 1, // positive answer FoundIt = 1, // positive answer
NotMe = 2 // negative answer NotMe = 2 // negative answer
@ -107,21 +108,27 @@ namespace UniversalDetector.Core
{ {
byte[] result = null; byte[] result = null;
using (MemoryStream ms = new MemoryStream(buf.Length)) { using (MemoryStream ms = new MemoryStream(buf.Length))
{
bool meetMSB = false; bool meetMSB = false;
int max = offset + len; int max = offset + len;
int prev = offset; int prev = offset;
int cur = offset; int cur = offset;
while (cur < max) { while (cur < max)
{
byte b = buf[cur]; byte b = buf[cur];
if ((b & 0x80) != 0) { if ((b & 0x80) != 0)
{
meetMSB = true; meetMSB = true;
} else if (b < CAPITAL_A || (b > CAPITAL_Z && b < SMALL_A) }
|| b > SMALL_Z) { else if (b < CAPITAL_A || (b > CAPITAL_Z && b < SMALL_A)
if (meetMSB && cur > prev) { || b > SMALL_Z)
{
if (meetMSB && cur > prev)
{
ms.Write(buf, prev, cur - prev); ms.Write(buf, prev, cur - prev);
ms.WriteByte(SPACE); ms.WriteByte(SPACE);
meetMSB = false; meetMSB = false;
@ -149,14 +156,16 @@ namespace UniversalDetector.Core
{ {
byte[] result = null; byte[] result = null;
using (MemoryStream ms = new MemoryStream(buf.Length)) { using (MemoryStream ms = new MemoryStream(buf.Length))
{
bool inTag = false; bool inTag = false;
int max = offset + len; int max = offset + len;
int prev = offset; int prev = offset;
int cur = offset; int cur = offset;
while (cur < max) { while (cur < max)
{
byte b = buf[cur]; byte b = buf[cur];
@ -167,8 +176,10 @@ namespace UniversalDetector.Core
// it's ascii, but it's not a letter // it's ascii, but it's not a letter
if ((b & 0x80) == 0 && (b < CAPITAL_A || b > SMALL_Z if ((b & 0x80) == 0 && (b < CAPITAL_A || b > SMALL_Z
|| (b > CAPITAL_Z && b < SMALL_A))) { || (b > CAPITAL_Z && b < SMALL_A)))
if (cur > prev && !inTag) { {
if (cur > prev && !inTag)
{
ms.Write(buf, prev, cur - prev); ms.Write(buf, prev, cur - prev);
ms.WriteByte(SPACE); ms.WriteByte(SPACE);
} }

View file

@ -60,7 +60,8 @@ namespace UniversalDetector.Core
// for each byte we get its class, if it is first byte, // for each byte we get its class, if it is first byte,
// we also get byte length // we also get byte length
int byteCls = model.GetClass(b); int byteCls = model.GetClass(b);
if (currentState == SMModel.START) { if (currentState == SMModel.START)
{
currentBytePos = 0; currentBytePos = 0;
currentCharLen = model.charLenTable[byteCls]; currentCharLen = model.charLenTable[byteCls];
} }

View file

@ -62,23 +62,30 @@ namespace UniversalDetector.Core
int codingState; int codingState;
int max = offset + len; int max = offset + len;
for (int i = offset; i < max; i++) { for (int i = offset; i < max; i++)
{
codingState = codingSM.NextState(buf[i]); codingState = codingSM.NextState(buf[i]);
if (codingState == SMModel.ERROR) { if (codingState == SMModel.ERROR)
{
state = ProbingState.NotMe; state = ProbingState.NotMe;
break; break;
} }
if (codingState == SMModel.ITSME) { if (codingState == SMModel.ITSME)
{
state = ProbingState.FoundIt; state = ProbingState.FoundIt;
break; break;
} }
if (codingState == SMModel.START) { if (codingState == SMModel.START)
{
int charLen = codingSM.CurrentCharLen; int charLen = codingSM.CurrentCharLen;
if (i == offset) { if (i == offset)
{
lastChar[1] = buf[offset]; lastChar[1] = buf[offset];
contextAnalyser.HandleOneChar(lastChar, 0, charLen); contextAnalyser.HandleOneChar(lastChar, 0, charLen);
distributionAnalyser.HandleOneChar(lastChar, 0, charLen); distributionAnalyser.HandleOneChar(lastChar, 0, charLen);
} else { }
else
{
contextAnalyser.HandleOneChar(buf, i - 1, charLen); contextAnalyser.HandleOneChar(buf, i - 1, charLen);
distributionAnalyser.HandleOneChar(buf, i - 1, charLen); distributionAnalyser.HandleOneChar(buf, i - 1, charLen);
} }

View file

@ -60,22 +60,29 @@ namespace UniversalDetector.Core
int codingState; int codingState;
int max = offset + len; int max = offset + len;
for (int i = offset; i < max; i++) { for (int i = offset; i < max; i++)
{
codingState = codingSM.NextState(buf[i]); codingState = codingSM.NextState(buf[i]);
if (codingState == SMModel.ERROR) { if (codingState == SMModel.ERROR)
{
state = ProbingState.NotMe; state = ProbingState.NotMe;
break; break;
} }
if (codingState == SMModel.ITSME) { if (codingState == SMModel.ITSME)
{
state = ProbingState.FoundIt; state = ProbingState.FoundIt;
break; break;
} }
if (codingState == SMModel.START) { if (codingState == SMModel.START)
{
int charLen = codingSM.CurrentCharLen; int charLen = codingSM.CurrentCharLen;
if (i == offset) { if (i == offset)
{
lastChar[1] = buf[offset]; lastChar[1] = buf[offset];
distributionAnalyser.HandleOneChar(lastChar, 0, charLen); distributionAnalyser.HandleOneChar(lastChar, 0, charLen);
} else { }
else
{
distributionAnalyser.HandleOneChar(buf, i - 1, charLen); distributionAnalyser.HandleOneChar(buf, i - 1, charLen);
} }
} }

View file

@ -56,22 +56,29 @@ namespace UniversalDetector.Core
int codingState; int codingState;
int max = offset + len; int max = offset + len;
for (int i = 0; i < max; i++) { for (int i = 0; i < max; i++)
{
codingState = codingSM.NextState(buf[i]); codingState = codingSM.NextState(buf[i]);
if (codingState == SMModel.ERROR) { if (codingState == SMModel.ERROR)
{
state = ProbingState.NotMe; state = ProbingState.NotMe;
break; break;
} }
if (codingState == SMModel.ITSME) { if (codingState == SMModel.ITSME)
{
state = ProbingState.FoundIt; state = ProbingState.FoundIt;
break; break;
} }
if (codingState == SMModel.START) { if (codingState == SMModel.START)
{
int charLen = codingSM.CurrentCharLen; int charLen = codingSM.CurrentCharLen;
if (i == offset) { if (i == offset)
{
lastChar[1] = buf[offset]; lastChar[1] = buf[offset];
distributionAnalyser.HandleOneChar(lastChar, 0, charLen); distributionAnalyser.HandleOneChar(lastChar, 0, charLen);
} else { }
else
{
distributionAnalyser.HandleOneChar(buf, i - 1, charLen); distributionAnalyser.HandleOneChar(buf, i - 1, charLen);
} }
} }

View file

@ -67,22 +67,30 @@ namespace UniversalDetector.Core
{ {
int max = offset + len; int max = offset + len;
for (int i = offset; i < max && state == ProbingState.Detecting; i++) { for (int i = offset; i < max && state == ProbingState.Detecting; i++)
for (int j = activeSM - 1; j >= 0; j--) { {
for (int j = activeSM - 1; j >= 0; j--)
{
// byte is feed to all active state machine // byte is feed to all active state machine
int codingState = codingSM[j].NextState(buf[i]); int codingState = codingSM[j].NextState(buf[i]);
if (codingState == SMModel.ERROR) { if (codingState == SMModel.ERROR)
{
// got negative answer for this state machine, make it inactive // got negative answer for this state machine, make it inactive
activeSM--; activeSM--;
if (activeSM == 0) { if (activeSM == 0)
{
state = ProbingState.NotMe; state = ProbingState.NotMe;
return state; return state;
} else if (j != activeSM) { }
else if (j != activeSM)
{
CodingStateMachine t = codingSM[activeSM]; CodingStateMachine t = codingSM[activeSM];
codingSM[activeSM] = codingSM[j]; codingSM[activeSM] = codingSM[j];
codingSM[j] = t; codingSM[j] = t;
} }
} else if (codingState == SMModel.ITSME) { }
else if (codingState == SMModel.ITSME)
{
state = ProbingState.FoundIt; state = ProbingState.FoundIt;
detectedCharset = codingSM[j].ModelName; detectedCharset = codingSM[j].ModelName;
return state; return state;

View file

@ -64,22 +64,29 @@ namespace UniversalDetector.Core
int codingState = SMModel.START; int codingState = SMModel.START;
int max = offset + len; int max = offset + len;
for (int i = offset; i < max; i++) { for (int i = offset; i < max; i++)
{
codingState = codingSM.NextState(buf[i]); codingState = codingSM.NextState(buf[i]);
if (codingState == SMModel.ERROR) { if (codingState == SMModel.ERROR)
{
state = ProbingState.NotMe; state = ProbingState.NotMe;
break; break;
} }
if (codingState == SMModel.ITSME) { if (codingState == SMModel.ITSME)
{
state = ProbingState.FoundIt; state = ProbingState.FoundIt;
break; break;
} }
if (codingState == SMModel.START) { if (codingState == SMModel.START)
{
int charLen = codingSM.CurrentCharLen; int charLen = codingSM.CurrentCharLen;
if (i == offset) { if (i == offset)
{
lastChar[1] = buf[offset]; lastChar[1] = buf[offset];
analyser.HandleOneChar(lastChar, 0, charLen); analyser.HandleOneChar(lastChar, 0, charLen);
} else { }
else
{
analyser.HandleOneChar(buf, i - 1, charLen); analyser.HandleOneChar(buf, i - 1, charLen);
} }
} }
@ -87,7 +94,8 @@ namespace UniversalDetector.Core
lastChar[0] = buf[max - 1]; lastChar[0] = buf[max - 1];
if (state == ProbingState.Detecting) { if (state == ProbingState.Detecting)
{
if (analyser.GotEnoughData() && GetConfidence() > SHORTCUT_THRESHOLD) if (analyser.GotEnoughData() && GetConfidence() > SHORTCUT_THRESHOLD)
state = ProbingState.FoundIt; state = ProbingState.FoundIt;
} }

View file

@ -36,7 +36,6 @@
* *
* ***** END LICENSE BLOCK ***** */ * ***** END LICENSE BLOCK ***** */
using System;
/** /**
* General ideas of the Hebrew charset recognition * General ideas of the Hebrew charset recognition
@ -217,14 +216,17 @@ namespace UniversalDetector.Core
int max = offset + len; int max = offset + len;
for (int i = offset; i < max; i++) { for (int i = offset; i < max; i++)
{
byte b = buf[i]; byte b = buf[i];
// a word just ended // a word just ended
if (b == 0x20) { if (b == 0x20)
{
// *(curPtr-2) was not a space so prev is not a 1 letter word // *(curPtr-2) was not a space so prev is not a 1 letter word
if (beforePrev != 0x20) { if (beforePrev != 0x20)
{
// case (1) [-2:not space][-1:final letter][cur:space] // case (1) [-2:not space][-1:final letter][cur:space]
if (IsFinal(prev)) if (IsFinal(prev))
finalCharLogicalScore++; finalCharLogicalScore++;
@ -233,7 +235,9 @@ namespace UniversalDetector.Core
finalCharVisualScore++; finalCharVisualScore++;
} }
} else { }
else
{
// case (3) [-2:space][-1:final letter][cur:not space] // case (3) [-2:space][-1:final letter][cur:not space]
if ((beforePrev == 0x20) && (IsFinal(prev)) && (b != ' ')) if ((beforePrev == 0x20) && (IsFinal(prev)) && (b != ' '))
++finalCharVisualScore; ++finalCharVisualScore;

View file

@ -181,16 +181,22 @@ namespace UniversalDetector.Core
// to record those bytes as well and analyse the character once it // to record those bytes as well and analyse the character once it
// is complete, but since a character will not make much difference, // is complete, but since a character will not make much difference,
// skipping it will simplify our logic and improve performance. // skipping it will simplify our logic and improve performance.
for (int i = needToSkipCharNum+offset; i < max; ) { for (int i = needToSkipCharNum + offset; i < max;)
{
int order = GetOrder(buf, i, out charLen); int order = GetOrder(buf, i, out charLen);
i += charLen; i += charLen;
if (i > max) { if (i > max)
{
needToSkipCharNum = i - max; needToSkipCharNum = i - max;
lastCharOrder = -1; lastCharOrder = -1;
} else { }
if (order != -1 && lastCharOrder != -1) { else
{
if (order != -1 && lastCharOrder != -1)
{
totalRel++; totalRel++;
if (totalRel > MAX_REL_THRESHOLD) { if (totalRel > MAX_REL_THRESHOLD)
{
done = true; done = true;
break; break;
} }
@ -210,7 +216,8 @@ namespace UniversalDetector.Core
// Only 2-bytes characters are of our interest // Only 2-bytes characters are of our interest
int order = (charLen == 2) ? GetOrder(buf, offset) : -1; int order = (charLen == 2) ? GetOrder(buf, offset) : -1;
if (order != -1 && lastCharOrder != -1) { if (order != -1 && lastCharOrder != -1)
{
totalRel++; totalRel++;
// count this sequence to its category counter // count this sequence to its category counter
relSample[jp2CharContext[lastCharOrder, order]]++; relSample[jp2CharContext[lastCharOrder, order]]++;
@ -221,7 +228,8 @@ namespace UniversalDetector.Core
public void Reset() public void Reset()
{ {
totalRel = 0; totalRel = 0;
for (int i = 0; i < CATEGORIES_NUM; i++) { for (int i = 0; i < CATEGORIES_NUM; i++)
{
relSample[i] = 0; relSample[i] = 0;
needToSkipCharNum = 0; needToSkipCharNum = 0;
lastCharOrder = -1; lastCharOrder = -1;
@ -254,7 +262,8 @@ namespace UniversalDetector.Core
charLen = 1; charLen = 1;
// return its order if it is hiragana // return its order if it is hiragana
if (buf[offset] == HIRAGANA_FIRST_BYTE) { if (buf[offset] == HIRAGANA_FIRST_BYTE)
{
byte low = buf[offset + 1]; byte low = buf[offset + 1];
if (low >= 0x9F && low <= 0xF1) if (low >= 0x9F && low <= 0xF1)
return low - 0x9F; return low - 0x9F;
@ -265,7 +274,8 @@ namespace UniversalDetector.Core
protected override int GetOrder(byte[] buf, int offset) protected override int GetOrder(byte[] buf, int offset)
{ {
// We are only interested in Hiragana // We are only interested in Hiragana
if (buf[offset] == HIRAGANA_FIRST_BYTE) { if (buf[offset] == HIRAGANA_FIRST_BYTE)
{
byte low = buf[offset + 1]; byte low = buf[offset + 1];
if (low >= 0x9F && low <= 0xF1) if (low >= 0x9F && low <= 0xF1)
return low - 0x9F; return low - 0x9F;
@ -292,7 +302,8 @@ namespace UniversalDetector.Core
charLen = 1; charLen = 1;
// return its order if it is hiragana // return its order if it is hiragana
if (high == HIRAGANA_FIRST_BYTE) { if (high == HIRAGANA_FIRST_BYTE)
{
byte low = buf[offset + 1]; byte low = buf[offset + 1];
if (low >= 0xA1 && low <= 0xF3) if (low >= 0xA1 && low <= 0xF3)
return low - 0xA1; return low - 0xA1;
@ -303,7 +314,8 @@ namespace UniversalDetector.Core
protected override int GetOrder(byte[] buf, int offset) protected override int GetOrder(byte[] buf, int offset)
{ {
// We are only interested in Hiragana // We are only interested in Hiragana
if (buf[offset] == HIRAGANA_FIRST_BYTE) { if (buf[offset] == HIRAGANA_FIRST_BYTE)
{
byte low = buf[offset + 1]; byte low = buf[offset + 1];
if (low >= 0xA1 && low <= 0xF3) if (low >= 0xA1 && low <= 0xF3)
return low - 0xA1; return low - 0xA1;

View file

@ -36,7 +36,6 @@
* *
* ***** END LICENSE BLOCK ***** */ * ***** END LICENSE BLOCK ***** */
using System;
namespace UniversalDetector.Core namespace UniversalDetector.Core
{ {
@ -135,10 +134,12 @@ namespace UniversalDetector.Core
byte[] newbuf = FilterWithEnglishLetters(buf, offset, len); byte[] newbuf = FilterWithEnglishLetters(buf, offset, len);
byte charClass, freq; byte charClass, freq;
for (int i = 0; i < newbuf.Length; i++) { for (int i = 0; i < newbuf.Length; i++)
{
charClass = Latin1_CharToClass[newbuf[i]]; charClass = Latin1_CharToClass[newbuf[i]];
freq = Latin1ClassModel[lastCharClass * CLASS_NUM + charClass]; freq = Latin1ClassModel[lastCharClass * CLASS_NUM + charClass];
if (freq == 0) { if (freq == 0)
{
state = ProbingState.NotMe; state = ProbingState.NotMe;
break; break;
} }
@ -155,13 +156,17 @@ namespace UniversalDetector.Core
float confidence = 0.0f; float confidence = 0.0f;
int total = 0; int total = 0;
for (int i = 0; i < FREQ_CAT_NUM; i++) { for (int i = 0; i < FREQ_CAT_NUM; i++)
{
total += freqCounter[i]; total += freqCounter[i];
} }
if (total <= 0) { if (total <= 0)
{
confidence = 0.0f; confidence = 0.0f;
} else { }
else
{
confidence = freqCounter[3] * 1.0f / total; confidence = freqCounter[3] * 1.0f / total;
confidence -= freqCounter[1] * 20.0f / total; confidence -= freqCounter[1] * 20.0f / total;
} }

View file

@ -36,7 +36,6 @@
* *
* ***** END LICENSE BLOCK ***** */ * ***** END LICENSE BLOCK ***** */
using System;
namespace UniversalDetector.Core namespace UniversalDetector.Core
{ {
@ -67,7 +66,8 @@ namespace UniversalDetector.Core
public override string GetCharsetName() public override string GetCharsetName()
{ {
if (bestGuess == -1) { if (bestGuess == -1)
{
GetConfidence(); GetConfidence();
if (bestGuess == -1) if (bestGuess == -1)
bestGuess = 0; bestGuess = 0;
@ -78,12 +78,16 @@ namespace UniversalDetector.Core
public override void Reset() public override void Reset()
{ {
activeNum = 0; activeNum = 0;
for (int i = 0; i < probers.Length; i++) { for (int i = 0; i < probers.Length; i++)
if (probers[i] != null) { {
if (probers[i] != null)
{
probers[i].Reset(); probers[i].Reset();
isActive[i] = true; isActive[i] = true;
++activeNum; ++activeNum;
} else { }
else
{
isActive[i] = false; isActive[i] = false;
} }
} }
@ -100,13 +104,18 @@ namespace UniversalDetector.Core
bool keepNext = true; bool keepNext = true;
int max = offset + len; int max = offset + len;
for (int i = offset; i < max; i++) { for (int i = offset; i < max; i++)
if ((buf[i] & 0x80) != 0) { {
if ((buf[i] & 0x80) != 0)
{
highbyteBuf[hptr++] = buf[i]; highbyteBuf[hptr++] = buf[i];
keepNext = true; keepNext = true;
} else { }
else
{
//if previous is highbyte, keep this even it is a ASCII //if previous is highbyte, keep this even it is a ASCII
if (keepNext) { if (keepNext)
{
highbyteBuf[hptr++] = buf[i]; highbyteBuf[hptr++] = buf[i];
keepNext = false; keepNext = false;
} }
@ -115,18 +124,23 @@ namespace UniversalDetector.Core
ProbingState st = ProbingState.NotMe; ProbingState st = ProbingState.NotMe;
for (int i = 0; i < probers.Length; i++) { for (int i = 0; i < probers.Length; i++)
{
if (!isActive[i]) if (!isActive[i])
continue; continue;
st = probers[i].HandleData(highbyteBuf, 0, hptr); st = probers[i].HandleData(highbyteBuf, 0, hptr);
if (st == ProbingState.FoundIt) { if (st == ProbingState.FoundIt)
{
bestGuess = i; bestGuess = i;
state = ProbingState.FoundIt; state = ProbingState.FoundIt;
break; break;
} else if (st == ProbingState.NotMe) { }
else if (st == ProbingState.NotMe)
{
isActive[i] = false; isActive[i] = false;
activeNum--; activeNum--;
if (activeNum <= 0) { if (activeNum <= 0)
{
state = ProbingState.NotMe; state = ProbingState.NotMe;
break; break;
} }
@ -140,16 +154,23 @@ namespace UniversalDetector.Core
float bestConf = 0.0f; float bestConf = 0.0f;
float cf = 0.0f; float cf = 0.0f;
if (state == ProbingState.FoundIt) { if (state == ProbingState.FoundIt)
{
return 0.99f; return 0.99f;
} else if (state == ProbingState.NotMe) { }
else if (state == ProbingState.NotMe)
{
return 0.01f; return 0.01f;
} else { }
for (int i = 0; i < PROBERS_NUM; i++) { else
{
for (int i = 0; i < PROBERS_NUM; i++)
{
if (!isActive[i]) if (!isActive[i])
continue; continue;
cf = probers[i].GetConfidence(); cf = probers[i].GetConfidence();
if (bestConf < cf) { if (bestConf < cf)
{
bestConf = cf; bestConf = cf;
bestGuess = i; bestGuess = i;
} }
@ -162,10 +183,14 @@ namespace UniversalDetector.Core
{ {
float cf; float cf;
GetConfidence(); GetConfidence();
for (int i = 0; i < PROBERS_NUM; i++) { for (int i = 0; i < PROBERS_NUM; i++)
if (!isActive[i]) { {
if (!isActive[i])
{
//Console.WriteLine(" MBCS inactive: {0} (confidence is too low).", ProberName[i]); //Console.WriteLine(" MBCS inactive: {0} (confidence is too low).", ProberName[i]);
} else { }
else
{
cf = probers[i].GetConfidence(); cf = probers[i].GetConfidence();
//Console.WriteLine(" MBCS {0}: [{1}]", cf, ProberName[i]); //Console.WriteLine(" MBCS {0}: [{1}]", cf, ProberName[i]);
} }

View file

@ -36,7 +36,6 @@
* *
* ***** END LICENSE BLOCK ***** */ * ***** END LICENSE BLOCK ***** */
using System;
namespace UniversalDetector.Core namespace UniversalDetector.Core
{ {
@ -88,19 +87,24 @@ namespace UniversalDetector.Core
if (newBuf.Length == 0) if (newBuf.Length == 0)
return state; // Nothing to see here, move on. return state; // Nothing to see here, move on.
for (int i = 0; i < PROBERS_NUM; i++) { for (int i = 0; i < PROBERS_NUM; i++)
{
if (!isActive[i]) if (!isActive[i])
continue; continue;
st = probers[i].HandleData(newBuf, 0, newBuf.Length); st = probers[i].HandleData(newBuf, 0, newBuf.Length);
if (st == ProbingState.FoundIt) { if (st == ProbingState.FoundIt)
{
bestGuess = i; bestGuess = i;
state = ProbingState.FoundIt; state = ProbingState.FoundIt;
break; break;
} else if (st == ProbingState.NotMe) { }
else if (st == ProbingState.NotMe)
{
isActive[i] = false; isActive[i] = false;
activeNum--; activeNum--;
if (activeNum <= 0) { if (activeNum <= 0)
{
state = ProbingState.NotMe; state = ProbingState.NotMe;
break; break;
} }
@ -112,7 +116,8 @@ namespace UniversalDetector.Core
public override float GetConfidence() public override float GetConfidence()
{ {
float bestConf = 0.0f, cf; float bestConf = 0.0f, cf;
switch (state) { switch (state)
{
case ProbingState.FoundIt: case ProbingState.FoundIt:
return 0.99f; //sure yes return 0.99f; //sure yes
case ProbingState.NotMe: case ProbingState.NotMe:
@ -138,7 +143,8 @@ namespace UniversalDetector.Core
{ {
float cf = GetConfidence(); float cf = GetConfidence();
// Console.WriteLine(" SBCS Group Prober --------begin status"); // Console.WriteLine(" SBCS Group Prober --------begin status");
for (int i = 0; i < PROBERS_NUM; i++) { for (int i = 0; i < PROBERS_NUM; i++)
{
if (isActive[i]) if (isActive[i])
probers[i].DumpStatus(); probers[i].DumpStatus();
//else //else
@ -151,12 +157,16 @@ namespace UniversalDetector.Core
public override void Reset() public override void Reset()
{ {
int activeNum = 0; int activeNum = 0;
for (int i = 0; i < PROBERS_NUM; i++) { for (int i = 0; i < PROBERS_NUM; i++)
if (probers[i] != null) { {
if (probers[i] != null)
{
probers[i].Reset(); probers[i].Reset();
isActive[i] = true; isActive[i] = true;
activeNum++; activeNum++;
} else { }
else
{
isActive[i] = false; isActive[i] = false;
} }
} }
@ -167,7 +177,8 @@ namespace UniversalDetector.Core
public override string GetCharsetName() public override string GetCharsetName()
{ {
//if we have no answer yet //if we have no answer yet
if (bestGuess == -1) { if (bestGuess == -1)
{
GetConfidence(); GetConfidence();
//no charset seems positive //no charset seems positive
if (bestGuess == -1) if (bestGuess == -1)

View file

@ -36,7 +36,6 @@
* *
* ***** END LICENSE BLOCK ***** */ * ***** END LICENSE BLOCK ***** */
using System;
namespace UniversalDetector.Core namespace UniversalDetector.Core
{ {
@ -89,16 +88,19 @@ namespace UniversalDetector.Core
{ {
int max = offset + len; int max = offset + len;
for (int i = offset; i < max; i++) { for (int i = offset; i < max; i++)
{
byte order = model.GetOrder(buf[i]); byte order = model.GetOrder(buf[i]);
if (order < SYMBOL_CAT_ORDER) if (order < SYMBOL_CAT_ORDER)
totalChar++; totalChar++;
if (order < SAMPLE_SIZE) { if (order < SAMPLE_SIZE)
{
freqChar++; freqChar++;
if (lastOrder < SAMPLE_SIZE) { if (lastOrder < SAMPLE_SIZE)
{
totalSeqs++; totalSeqs++;
if (!reversed) if (!reversed)
++(seqCounters[model.GetPrecedence(lastOrder * SAMPLE_SIZE + order)]); ++(seqCounters[model.GetPrecedence(lastOrder * SAMPLE_SIZE + order)]);
@ -109,8 +111,10 @@ namespace UniversalDetector.Core
lastOrder = order; lastOrder = order;
} }
if (state == ProbingState.Detecting) { if (state == ProbingState.Detecting)
if (totalSeqs > SB_ENOUGH_REL_THRESHOLD) { {
if (totalSeqs > SB_ENOUGH_REL_THRESHOLD)
{
float cf = GetConfidence(); float cf = GetConfidence();
if (cf > POSITIVE_SHORTCUT_THRESHOLD) if (cf > POSITIVE_SHORTCUT_THRESHOLD)
state = ProbingState.FoundIt; state = ProbingState.FoundIt;
@ -139,7 +143,8 @@ namespace UniversalDetector.Core
// POSITIVE_APPROACH // POSITIVE_APPROACH
float r = 0.0f; float r = 0.0f;
if (totalSeqs > 0) { if (totalSeqs > 0)
{
r = 1.0f * seqCounters[POSITIVE_CAT] / totalSeqs / model.TypicalPositiveRatio; r = 1.0f * seqCounters[POSITIVE_CAT] / totalSeqs / model.TypicalPositiveRatio;
r = r * freqChar / totalChar; r = r * freqChar / totalChar;
if (r >= 1.0f) if (r >= 1.0f)

View file

@ -69,23 +69,30 @@ namespace UniversalDetector.Core
int codingState; int codingState;
int max = offset + len; int max = offset + len;
for (int i = offset; i < max; i++) { for (int i = offset; i < max; i++)
{
codingState = codingSM.NextState(buf[i]); codingState = codingSM.NextState(buf[i]);
if (codingState == SMModel.ERROR) { if (codingState == SMModel.ERROR)
{
state = ProbingState.NotMe; state = ProbingState.NotMe;
break; break;
} }
if (codingState == SMModel.ITSME) { if (codingState == SMModel.ITSME)
{
state = ProbingState.FoundIt; state = ProbingState.FoundIt;
break; break;
} }
if (codingState == SMModel.START) { if (codingState == SMModel.START)
{
int charLen = codingSM.CurrentCharLen; int charLen = codingSM.CurrentCharLen;
if (i == offset) { if (i == offset)
{
lastChar[1] = buf[offset]; lastChar[1] = buf[offset];
contextAnalyser.HandleOneChar(lastChar, 2 - charLen, charLen); contextAnalyser.HandleOneChar(lastChar, 2 - charLen, charLen);
distributionAnalyser.HandleOneChar(lastChar, 0, charLen); distributionAnalyser.HandleOneChar(lastChar, 0, charLen);
} else { }
else
{
contextAnalyser.HandleOneChar(buf, i + 1 - charLen, charLen); contextAnalyser.HandleOneChar(buf, i + 1 - charLen, charLen);
distributionAnalyser.HandleOneChar(buf, i - 1, charLen); distributionAnalyser.HandleOneChar(buf, i - 1, charLen);
} }

View file

@ -36,7 +36,6 @@
* *
* ***** END LICENSE BLOCK ***** */ * ***** END LICENSE BLOCK ***** */
using System;
namespace UniversalDetector.Core namespace UniversalDetector.Core
{ {

View file

@ -36,7 +36,6 @@
* *
* ***** END LICENSE BLOCK ***** */ * ***** END LICENSE BLOCK ***** */
using System;
namespace UniversalDetector.Core namespace UniversalDetector.Core
{ {

View file

@ -51,7 +51,8 @@ namespace UniversalDetector.Core
Reset(); Reset();
} }
public override string GetCharsetName() { public override string GetCharsetName()
{
return "UTF-8"; return "UTF-8";
} }
@ -67,21 +68,25 @@ namespace UniversalDetector.Core
int codingState = SMModel.START; int codingState = SMModel.START;
int max = offset + len; int max = offset + len;
for (int i = offset; i < max; i++) { for (int i = offset; i < max; i++)
{
codingState = codingSM.NextState(buf[i]); codingState = codingSM.NextState(buf[i]);
if (codingState == SMModel.ERROR) { if (codingState == SMModel.ERROR)
{
state = ProbingState.NotMe; state = ProbingState.NotMe;
break; break;
} }
if (codingState == SMModel.ITSME) { if (codingState == SMModel.ITSME)
{
state = ProbingState.FoundIt; state = ProbingState.FoundIt;
break; break;
} }
if (codingState == SMModel.START) { if (codingState == SMModel.START)
{
if (codingSM.CurrentCharLen >= 2) if (codingSM.CurrentCharLen >= 2)
numOfMBChar++; numOfMBChar++;
} }
@ -98,11 +103,14 @@ namespace UniversalDetector.Core
float unlike = 0.99f; float unlike = 0.99f;
float confidence = 0.0f; float confidence = 0.0f;
if (numOfMBChar < 6) { if (numOfMBChar < 6)
{
for (int i = 0; i < numOfMBChar; i++) for (int i = 0; i < numOfMBChar; i++)
unlike *= ONE_CHAR_PROB; unlike *= ONE_CHAR_PROB;
confidence = 1.0f - unlike; confidence = 1.0f - unlike;
} else { }
else
{
confidence = 0.99f; confidence = 0.99f;
} }
return confidence; return confidence;

View file

@ -70,7 +70,8 @@ namespace UniversalDetector.Core
protected CharsetProber escCharsetProber; protected CharsetProber escCharsetProber;
protected string detectedCharset; protected string detectedCharset;
public UniversalDetector(int languageFilter) { public UniversalDetector(int languageFilter)
{
this.start = true; this.start = true;
this.inputState = InputState.PureASCII; this.inputState = InputState.PureASCII;
this.lastChar = 0x00; this.lastChar = 0x00;
@ -80,7 +81,8 @@ namespace UniversalDetector.Core
public virtual void Feed(byte[] buf, int offset, int len) public virtual void Feed(byte[] buf, int offset, int len)
{ {
if (done) { if (done)
{
return; return;
} }
@ -88,10 +90,13 @@ namespace UniversalDetector.Core
gotData = true; gotData = true;
// If the data starts with BOM, we know it is UTF // If the data starts with BOM, we know it is UTF
if (start) { if (start)
{
start = false; start = false;
if (len > 3) { if (len > 3)
switch (buf[0]) { {
switch (buf[0])
{
case 0xEF: case 0xEF:
if (0xBB == buf[1] && 0xBF == buf[2]) if (0xBB == buf[1] && 0xBF == buf[2])
detectedCharset = "UTF-8"; detectedCharset = "UTF-8";
@ -118,22 +123,27 @@ namespace UniversalDetector.Core
break; break;
} // switch } // switch
} }
if (detectedCharset != null) { if (detectedCharset != null)
{
done = true; done = true;
return; return;
} }
} }
for (int i = 0; i < len; i++) { for (int i = 0; i < len; i++)
{
// other than 0xa0, if every other character is ascii, the page is ascii // other than 0xa0, if every other character is ascii, the page is ascii
if ((buf[i] & 0x80) != 0 && buf[i] != 0xA0) { if ((buf[i] & 0x80) != 0 && buf[i] != 0xA0)
{
// we got a non-ascii byte (high-byte) // we got a non-ascii byte (high-byte)
if (inputState != InputState.Highbyte) { if (inputState != InputState.Highbyte)
{
inputState = InputState.Highbyte; inputState = InputState.Highbyte;
// kill EscCharsetProber if it is active // kill EscCharsetProber if it is active
if (escCharsetProber != null) { if (escCharsetProber != null)
{
escCharsetProber = null; escCharsetProber = null;
} }
@ -145,9 +155,12 @@ namespace UniversalDetector.Core
if (charsetProbers[2] == null) if (charsetProbers[2] == null)
charsetProbers[2] = new Latin1Prober(); charsetProbers[2] = new Latin1Prober();
} }
} else { }
else
{
if (inputState == InputState.PureASCII && if (inputState == InputState.PureASCII &&
(buf[i] == 0x33 || (buf[i] == 0x7B && lastChar == 0x7E))) { (buf[i] == 0x33 || (buf[i] == 0x7B && lastChar == 0x7E)))
{
// found escape character or HZ "~{" // found escape character or HZ "~{"
inputState = InputState.EscASCII; inputState = InputState.EscASCII;
} }
@ -157,25 +170,31 @@ namespace UniversalDetector.Core
ProbingState st = ProbingState.NotMe; ProbingState st = ProbingState.NotMe;
switch (inputState) { switch (inputState)
{
case InputState.EscASCII: case InputState.EscASCII:
if (escCharsetProber == null) { if (escCharsetProber == null)
{
escCharsetProber = new EscCharsetProber(); escCharsetProber = new EscCharsetProber();
} }
st = escCharsetProber.HandleData(buf, offset, len); st = escCharsetProber.HandleData(buf, offset, len);
if (st == ProbingState.FoundIt) { if (st == ProbingState.FoundIt)
{
done = true; done = true;
detectedCharset = escCharsetProber.GetCharsetName(); detectedCharset = escCharsetProber.GetCharsetName();
} }
break; break;
case InputState.Highbyte: case InputState.Highbyte:
for (int i = 0; i < PROBERS_NUM; i++) { for (int i = 0; i < PROBERS_NUM; i++)
if (charsetProbers[i] != null) { {
if (charsetProbers[i] != null)
{
st = charsetProbers[i].HandleData(buf, offset, len); st = charsetProbers[i].HandleData(buf, offset, len);
#if DEBUG #if DEBUG
charsetProbers[i].DumpStatus(); charsetProbers[i].DumpStatus();
#endif #endif
if (st == ProbingState.FoundIt) { if (st == ProbingState.FoundIt)
{
done = true; done = true;
detectedCharset = charsetProbers[i].GetCharsetName(); detectedCharset = charsetProbers[i].GetCharsetName();
return; return;
@ -195,38 +214,47 @@ namespace UniversalDetector.Core
/// </summary> /// </summary>
public virtual void DataEnd() public virtual void DataEnd()
{ {
if (!gotData) { if (!gotData)
{
// we haven't got any data yet, return immediately // we haven't got any data yet, return immediately
// caller program sometimes call DataEnd before anything has // caller program sometimes call DataEnd before anything has
// been sent to detector // been sent to detector
return; return;
} }
if (detectedCharset != null) { if (detectedCharset != null)
{
done = true; done = true;
Report(detectedCharset, 1.0f); Report(detectedCharset, 1.0f);
return; return;
} }
if (inputState == InputState.Highbyte) { if (inputState == InputState.Highbyte)
{
float proberConfidence = 0.0f; float proberConfidence = 0.0f;
float maxProberConfidence = 0.0f; float maxProberConfidence = 0.0f;
int maxProber = 0; int maxProber = 0;
for (int i = 0; i < PROBERS_NUM; i++) { for (int i = 0; i < PROBERS_NUM; i++)
if (charsetProbers[i] != null) { {
if (charsetProbers[i] != null)
{
proberConfidence = charsetProbers[i].GetConfidence(); proberConfidence = charsetProbers[i].GetConfidence();
if (proberConfidence > maxProberConfidence) { if (proberConfidence > maxProberConfidence)
{
maxProberConfidence = proberConfidence; maxProberConfidence = proberConfidence;
maxProber = i; maxProber = i;
} }
} }
} }
if (maxProberConfidence > MINIMUM_THRESHOLD) { if (maxProberConfidence > MINIMUM_THRESHOLD)
{
Report(charsetProbers[maxProber].GetCharsetName(), maxProberConfidence); Report(charsetProbers[maxProber].GetCharsetName(), maxProberConfidence);
} }
} else if (inputState == InputState.PureASCII) { }
else if (inputState == InputState.PureASCII)
{
Report("ASCII", 1.0f); Report("ASCII", 1.0f);
} }
} }