Fix (most) UniCaseTest failures (fix #27)

Upper- or lowercasing Unicode strings is not possible by mapping one
character to one character in each case, in contradiction of what the
GLib functions g_unichar_tolower() and g_unichar_toupper() would seem
to suggest. The casing operations now operate on the whole string.

Unfortunately, this introduces a new bug, #28, since there is no
title-casing function in GLib that allows for the title-case of a
character to be longer than one character.

diff --git a/libchimara/case.c b/libchimara/case.c
index 5b1a5d755edadc1eddc6beeda73a880ac7860bc7..3176a9a1adf2d60dd8aa7a204175db92772d1bce 100644 (file)
--- a/libchimara/case.c
+++ b/libchimara/case.c
@@ -95,15 +95,25 @@ glk_buffer_to_lower_case_uni(glui32 *buf, glui32 len, glui32 numchars)
 {
        g_return_val_if_fail(buf != NULL && (len > 0 || numchars > 0), 0);
        g_return_val_if_fail(numchars <= len, 0);
+
+       long outchars;
+
+       /* Lowercase the string */
+       char *utf8 = convert_ucs4_to_utf8(buf, numchars);
+       if(!utf8)
+               return numchars;
+       char *lowered = g_utf8_strdown(utf8, -1);
+       g_free(utf8);
+       gunichar *outbuf = convert_utf8_to_ucs4(lowered, &outchars);
+       g_free(lowered);
+       if(!outbuf)
+               return numchars;
        
-       /* GLib has a function that converts _one_ UCS-4 character to _one_
-       lowercase UCS-4 character; so apparently we don't have to worry about the
-       string length changing... */
-       glui32 *ptr;
-       for(ptr = buf; ptr < buf + numchars; ptr++)
-               *ptr = g_unichar_tolower(*ptr);
+       /* Copy the output buffer to the original buffer */
+       memcpy(buf, outbuf, MIN(outchars, len) * 4);
+       g_free(outbuf);
        
-       return numchars;
+       return outchars;
 }
 
 /**
@@ -123,14 +133,24 @@ glk_buffer_to_upper_case_uni(glui32 *buf, glui32 len, glui32 numchars)
        g_return_val_if_fail(buf != NULL && (len > 0 || numchars > 0), 0);
        g_return_val_if_fail(numchars <= len, 0);
        
-       /* GLib has a function that converts _one_ UCS-4 character to _one_
-       uppercase UCS-4 character; so apparently we don't have to worry about the
-       string length changing... */
-       glui32 *ptr;
-       for(ptr = buf; ptr < buf + numchars; ptr++)
-               *ptr = g_unichar_toupper(*ptr);
+       long outchars;
        
-       return numchars;
+       /* Uppercase the string */
+       char *utf8 = convert_ucs4_to_utf8(buf, numchars);
+       if(!utf8)
+               return numchars;
+       char *uppered = g_utf8_strup(utf8, -1);
+       g_free(utf8);
+       gunichar *outbuf = convert_utf8_to_ucs4(uppered, &outchars);
+       g_free(uppered);
+       if(!outbuf)
+               return numchars;
+
+       /* Copy the output buffer to the original buffer */
+       memcpy(buf, outbuf, MIN(outchars, len) * 4);
+       g_free(outbuf);
+
+       return outchars;
 }
 
 /**
@@ -162,9 +182,14 @@ glk_buffer_to_title_case_uni(glui32 *buf, glui32 len, glui32 numchars, glui32 lo
        g_return_val_if_fail(buf != NULL && (len > 0 || numchars > 0), 0);
        g_return_val_if_fail(numchars <= len, 0);
        
-       /* GLib has a function that converts _one_ UCS-4 character to _one_
-       titlecase UCS-4 character; so apparently we don't have to worry about the
-       string length changing... */
+       /* FIXME: This is wrong. g_unichar_totitle() which returns the titlecase of
+        one Unicode code point, but that only returns the correct result if the
+        titlecase character is also one code point.
+        For example, the one-character 'ffi' ligature should be title-cased to the
+        three-character string 'Ffi'. This code leaves it as the 'ffi' ligature,
+        which is incorrect.
+        However, nothing much can be done about it unless GLib gets a
+        g_utf8_strtitle() function, or we roll our own. */
        *buf = g_unichar_totitle(*buf);
        /* Call lowercase on the rest of the string */
        if(lowerrest)