ICU-491.11.1.tar.gz

[apple/icu.git] / icuSources / i18n / csrutf8.cpp

diff --git a/icuSources/i18n/csrutf8.cpp b/icuSources/i18n/csrutf8.cpp
index 8af5307bdc39faf458e00b779767c23e51e6d47a..b87c277fbf948447f034c5800a55c5a5d0bef263 100644 (file)
--- a/icuSources/i18n/csrutf8.cpp
+++ b/icuSources/i18n/csrutf8.cpp
@@ -1,6 +1,6 @@
 /*
  **********************************************************************
 /*
  **********************************************************************

- *   Copyright (C) 2005-2006, International Business Machines
+ *   Copyright (C) 2005-2008, International Business Machines

  *   Corporation and others.  All Rights Reserved.
  **********************************************************************
  */
  *   Corporation and others.  All Rights Reserved.
  **********************************************************************
  */


@@ -35,76 +35,76 @@ int32_t CharsetRecog_UTF8::match(InputText* det) {
     if (det->fRawLength >= 3 && 
         input[0] == 0xEF && input[1] == 0xBB && input[2] == 0xBF) {
             hasBOM = TRUE;
     if (det->fRawLength >= 3 && 
         input[0] == 0xEF && input[1] == 0xBB && input[2] == 0xBF) {
             hasBOM = TRUE;

-        }
-
-        // Scan for multi-byte sequences
-        for (i=0; i < det->fRawLength; i += 1) {
-            int32_t b = input[i];
-
-            if ((b & 0x80) == 0) {
-                continue;   // ASCII
-            }
+    }

 
 

-            // Hi bit on char found.  Figure out how long the sequence should be
-            if ((b & 0x0E0) == 0x0C0) {
-                trailBytes = 1;                
-            } else if ((b & 0x0F0) == 0x0E0) {
-                trailBytes = 2;
-            } else if ((b & 0x0F8) == 0xF0) {
-                trailBytes = 3;
-            } else {
-                numInvalid += 1;
+    // Scan for multi-byte sequences
+    for (i=0; i < det->fRawLength; i += 1) {
+        int32_t b = input[i];

 
 

-                if (numInvalid > 5) {
-                    break;
-                }
+        if ((b & 0x80) == 0) {
+            continue;   // ASCII
+        }

 
 

-                trailBytes = 0;
+        // Hi bit on char found.  Figure out how long the sequence should be
+        if ((b & 0x0E0) == 0x0C0) {
+            trailBytes = 1;
+        } else if ((b & 0x0F0) == 0x0E0) {
+            trailBytes = 2;
+        } else if ((b & 0x0F8) == 0xF0) {
+            trailBytes = 3;
+        } else {
+            numInvalid += 1;
+
+            if (numInvalid > 5) {
+                break;

             }
 
             }
 

-            // Verify that we've got the right number of trail bytes in the sequence
-            for (;;) {
-                i += 1;
+            trailBytes = 0;
+        }

 
 

-                if (i >= det->fRawLength) {
-                    break;
-                }
+        // Verify that we've got the right number of trail bytes in the sequence
+        for (;;) {
+            i += 1;

 
 

-                b = input[i];
+            if (i >= det->fRawLength) {
+                break;
+            }

 
 

-                if ((b & 0xC0) != 0x080) {
-                    numInvalid += 1;
-                    break;
-                }
+            b = input[i];

 
 

-                if (--trailBytes == 0) {
-                    numValid += 1;
-                    break;
-                }
+            if ((b & 0xC0) != 0x080) {
+                numInvalid += 1;
+                break;

             }
 
             }
 

+            if (--trailBytes == 0) {
+                numValid += 1;
+                break;
+            }

         }
 
         }
 

-        // Cook up some sort of confidence score, based on presense of a BOM
-        //    and the existence of valid and/or invalid multi-byte sequences.
-        confidence = 0;
-        if (hasBOM && numInvalid == 0) {
-            confidence = 100;
-        } else if (hasBOM && numValid > numInvalid*10) {
-            confidence = 80;
-        } else if (numValid > 3 && numInvalid == 0) {
-            confidence = 100;            
-        } else if (numValid > 0 && numInvalid == 0) {
-            confidence = 80;
-        } else if (numValid == 0 && numInvalid == 0) {
-            // Plain ASCII.  
-            confidence = 10;            
-        } else if (numValid > numInvalid*10) {
-            // Probably corruput utf-8 data.  Valid sequences aren't likely by chance.
-            confidence = 25;
-        }
-
-        return confidence;
+    }
+
+    // Cook up some sort of confidence score, based on presense of a BOM
+    //    and the existence of valid and/or invalid multi-byte sequences.
+    confidence = 0;
+    if (hasBOM && numInvalid == 0) {
+        confidence = 100;
+    } else if (hasBOM && numValid > numInvalid*10) {
+        confidence = 80;
+    } else if (numValid > 3 && numInvalid == 0) {
+        confidence = 100;
+    } else if (numValid > 0 && numInvalid == 0) {
+        confidence = 80;
+    } else if (numValid == 0 && numInvalid == 0) {
+        // Plain ASCII.
+        confidence = 10;
+    } else if (numValid > numInvalid*10) {
+        // Probably corruput utf-8 data.  Valid sequences aren't likely by chance.
+        confidence = 25;
+    }
+
+    return confidence;

 }
 
 U_NAMESPACE_END
 }
 
 U_NAMESPACE_END