diff options
Diffstat (limited to 'Emby.Server.Implementations/TextEncoding/UniversalDetector/Core/JapaneseContextAnalyser.cs')
| -rw-r--r-- | Emby.Server.Implementations/TextEncoding/UniversalDetector/Core/JapaneseContextAnalyser.cs | 50 |
1 files changed, 31 insertions, 19 deletions
diff --git a/Emby.Server.Implementations/TextEncoding/UniversalDetector/Core/JapaneseContextAnalyser.cs b/Emby.Server.Implementations/TextEncoding/UniversalDetector/Core/JapaneseContextAnalyser.cs index 7d28224c5..a2bf04ba0 100644 --- a/Emby.Server.Implementations/TextEncoding/UniversalDetector/Core/JapaneseContextAnalyser.cs +++ b/Emby.Server.Implementations/TextEncoding/UniversalDetector/Core/JapaneseContextAnalyser.cs @@ -160,7 +160,7 @@ namespace UniversalDetector.Core { // This is just one way to calculate confidence. It works well for me. if (totalRel > MINIMUM_DATA_THRESHOLD) - return ((float)(totalRel - relSample[0]))/totalRel; + return ((float)(totalRel - relSample[0])) / totalRel; else return DONT_KNOW; } @@ -181,22 +181,28 @@ namespace UniversalDetector.Core // to record those bytes as well and analyse the character once it // is complete, but since a character will not make much difference, // 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); i += charLen; - if (i > max) { + if (i > max) + { needToSkipCharNum = i - max; lastCharOrder = -1; - } else { - if (order != -1 && lastCharOrder != -1) { - totalRel ++; - if (totalRel > MAX_REL_THRESHOLD) { + } + else + { + if (order != -1 && lastCharOrder != -1) + { + totalRel++; + if (totalRel > MAX_REL_THRESHOLD) + { done = true; break; } relSample[jp2CharContext[lastCharOrder, order]]++; - } - lastCharOrder = order; + } + lastCharOrder = order; } } } @@ -210,7 +216,8 @@ namespace UniversalDetector.Core // Only 2-bytes characters are of our interest int order = (charLen == 2) ? GetOrder(buf, offset) : -1; - if (order != -1 && lastCharOrder != -1) { + if (order != -1 && lastCharOrder != -1) + { totalRel++; // count this sequence to its category counter relSample[jp2CharContext[lastCharOrder, order]]++; @@ -221,7 +228,8 @@ namespace UniversalDetector.Core public void Reset() { totalRel = 0; - for (int i = 0; i < CATEGORIES_NUM; i++) { + for (int i = 0; i < CATEGORIES_NUM; i++) + { relSample[i] = 0; needToSkipCharNum = 0; lastCharOrder = -1; @@ -254,8 +262,9 @@ namespace UniversalDetector.Core charLen = 1; // return its order if it is hiragana - if (buf[offset] == HIRAGANA_FIRST_BYTE) { - byte low = buf[offset+1]; + if (buf[offset] == HIRAGANA_FIRST_BYTE) + { + byte low = buf[offset + 1]; if (low >= 0x9F && low <= 0xF1) return low - 0x9F; } @@ -265,8 +274,9 @@ namespace UniversalDetector.Core protected override int GetOrder(byte[] buf, int offset) { // We are only interested in Hiragana - if (buf[offset] == HIRAGANA_FIRST_BYTE) { - byte low = buf[offset+1]; + if (buf[offset] == HIRAGANA_FIRST_BYTE) + { + byte low = buf[offset + 1]; if (low >= 0x9F && low <= 0xF1) return low - 0x9F; } @@ -292,8 +302,9 @@ namespace UniversalDetector.Core charLen = 1; // return its order if it is hiragana - if (high == HIRAGANA_FIRST_BYTE) { - byte low = buf[offset+1]; + if (high == HIRAGANA_FIRST_BYTE) + { + byte low = buf[offset + 1]; if (low >= 0xA1 && low <= 0xF3) return low - 0xA1; } @@ -303,8 +314,9 @@ namespace UniversalDetector.Core protected override int GetOrder(byte[] buf, int offset) { // We are only interested in Hiragana - if (buf[offset] == HIRAGANA_FIRST_BYTE) { - byte low = buf[offset+1]; + if (buf[offset] == HIRAGANA_FIRST_BYTE) + { + byte low = buf[offset + 1]; if (low >= 0xA1 && low <= 0xF3) return low - 0xA1; } |