ICU-307 lenient converter name matching

X-SVN-Rev: 1957

icu4c/source/common/ucnv_io.c

@@ -136,32 +136,48 @@ isAlias(const char *alias, UErrorCode *pErrorCode) {
     }
 }
 
-/* compare lowercase str1 with mixed-case str2, both being charset names */
-static int
-charsetNameCmp(const char *str1, const char *str2) {
+/**
+ * Do a fuzzy compare of a two converter/alias names.  The comparison
+ * is case-insensitive.  It also ignores the characters '-', '_', and
+ * ' ' (dash, underscore, and space).  Thus the strings "UTF-8",
+ * "utf_8", and "Utf 8" are exactly equivalent.
+ * 
+ * This is a symmetrical (commutative) operation; order of arguments
+ * is insignificant.  This is an important property for sorting the
+ * list (when the list is preprocessed into binary form) and for
+ * performing binary searches on it at run time.
+ * 
+ * @param name1 a converter name or alias, zero-terminated
+ * @param name2 a converter name or alias, zero-terminated
+ * @return 0 if the names match, or a negative value if the name1
+ * lexically precedes name2, or a positive value if the name1
+ * lexically follows name2.
+ */
+U_CFUNC int
+charsetNameCmp(const char *name1, const char *name2) {
     int rc;
     unsigned char c1, c2;
 
     for (;;) {
-        c1 = (unsigned char) *str1;
-        c2 = (unsigned char) *str2;
-        if (c1 == 0) {
-            if(c2 == 0) {
-                return 0;
-            } else {
-                return -1;
-            }
-        } else if (c2 == 0) {
-            return 1;
-        } else {
-            /* compare non-zero characters with lowercase */
-            rc = (int) c1 - (int) (unsigned char) uprv_tolower(c2);
-            if(rc != 0) {
-                return rc;
-            }
+        /* Ignore delimiters '-', '_', and ' ' */
+        while ((c1 = (unsigned char)*name1) == '-'
+               || c1 == '_' || c1 == ' ') ++name1;
+        while ((c2 = (unsigned char)*name2) == '-'
+               || c2 == '_' || c2 == ' ') ++name2;
+
+        /* If we reach the ends of both strings then they match */
+        if ((c1|c2)==0) {
+            return 0;
         }
-        ++str1;
-        ++str2;
+        
+        /* Case-insensitive comparison */
+        rc = (int)(unsigned char)uprv_tolower(c1) -
+             (int)(unsigned char)uprv_tolower(c2);
+        if (rc!=0) {
+            return rc;
+        }
+        ++name1;
+        ++name2;
     }
 }
 

icu4c/source/common/ucnv_io.h

@@ -99,4 +99,11 @@ ucnv_io_getDefaultConverterName(void);
 U_CFUNC void
 ucnv_io_setDefaultConverterName(const char *name);
 
+/**
+ * Lexically compare two converter names/aliases, ignoring case and
+ * the delimiters '-', '_', and ' '.
+ */
+U_CFUNC int
+charsetNameCmp(const char *name1, const char *name2);
+
 #endif /* _UCNV_IO */

icu4c/source/common/unicode/ucnv.h

@@ -122,7 +122,10 @@ U_CDECL_END
 
 /**
  * Creates a UConverter object with the names specified as a C string.
- * The actual name will be resolved with the alias file.
+ * The actual name will be resolved with the alias file
+ * using a case-insensitive string comparison that ignores
+ * the delimiters '-', '_', and ' ' (dash, underscore, and space).
+ * E.g., the names "UTF8", "utf-8", and "Utf 8" are all equivalent.
  * If <code>NULL</code> is passed for the converter name, it will create one with the
  * getDefaultName return value.
  *
@@ -151,8 +154,12 @@ UConverter* U_EXPORT2 ucnv_open   (const char *converterName, UErrorCode * err);
 
 /**
  * Creates a Unicode converter with the names specified as unicode string. The name should be limited to
- * the ASCII-7 alphanumerics range. The actual name will be resolved with the alias file.
- * if <TT>NULL</TT> is passed for the converter name, it will create one with the
+ * the ASCII-7 alphanumerics range.
+ * The actual name will be resolved with the alias file
+ * using a case-insensitive string comparison that ignores
+ * the delimiters '-', '_', and ' ' (dash, underscore, and space).
+ * E.g., the names "UTF8", "utf-8", and "Utf 8" are all equivalent.
+ * If <TT>NULL</TT> is passed for the converter name, it will create one with the
  * getDefaultName return value.
  * @param converterName : name of the uconv table in a zero terminated Unicode string
  * @param err outgoing error status <TT>U_MEMORY_ALLOCATION_ERROR, TABLE_NOT_FOUND</TT>
@@ -714,7 +721,9 @@ U_CAPI int32_t U_EXPORT2 ucnv_countAvailable (void);
 
 
 /**
- * Gives the number of aliases for given converter or alias name
+ * Gives the number of aliases for given converter or alias name.
+ * Note that additional aliases are recognized by ucnv_open().
+ * This method only enumerates the listed entries in the alias file.
  * @param alias alias name
  * @param pErrorCode result of operation
  * @return number of names on alias list
@@ -724,7 +733,9 @@ U_CAPI uint16_t
 ucnv_countAliases(const char *alias, UErrorCode *pErrorCode);
 
 /**
- * Gives the name of the alias at given index of alias list
+ * Gives the name of the alias at given index of alias list.
+ * Note that additional aliases are recognized by ucnv_open().
+ * This method only enumerates the listed entries in the alias file.
  * @param alias alias name
  * @param n index in alias list
  * @param pErrorCode result of operation
@@ -735,7 +746,9 @@ U_CAPI const char *
 ucnv_getAlias(const char *alias, uint16_t n, UErrorCode *pErrorCode);
 
 /**
- * Fill-up the list of alias names for the given alias
+ * Fill-up the list of alias names for the given alias.
+ * Note that additional aliases are recognized by ucnv_open().
+ * This method only enumerates the listed entries in the alias file.
  * @param alias alias name
  * @param aliases fill-in list, aliases is a pointer to an array of
  *        <code>ucnv_countAliases()</code> string-pointers

icu4c/source/tools/gencnval/gencnval.c

@@ -23,6 +23,7 @@
 #include "unicode/utypes.h"
 #include "cmemory.h"
 #include "cstring.h"
+#include "ucnv_io.h" /* charsetNameCmp */
 #include "filestrm.h"
 #include "unewdata.h"
 #include "uoptions.h"
@@ -326,7 +327,7 @@ allocString(uint32_t length) {
 
 static int
 compareAliases(const void *alias1, const void *alias2) {
-    return uprv_stricmp(((Alias *)alias1)->alias, ((Alias *)alias2)->alias);
+    return charsetNameCmp(((Alias*)alias1)->alias, ((Alias*)alias2)->alias);
 }
 
 /*

icu4c/source/tools/gencnval/gencnval.dsp

@@ -89,6 +89,10 @@ LINK32=link.exe
 
 SOURCE=.\gencnval.c
 # End Source File
+# Begin Source File
+
+SOURCE=..\..\common\ucnv_io.c
+# End Source File
 # End Group
 # Begin Group "Header Files"