Introduce the -v flag which outputs the actual index of each code

[sgt/charset] / iso2022.c

diff --git a/iso2022.c b/iso2022.c
index 24eb811..a570860 100644 (file)
--- a/iso2022.c
+++ b/iso2022.c
@@ -35,6 +35,15 @@
 
 enum {S4, S6, M4, M6};
 
+static long int emacs_big5_1_to_unicode(int, int);
+static long int emacs_big5_2_to_unicode(int, int);
+static long int cns11643_1_to_unicode(int, int);
+static long int cns11643_2_to_unicode(int, int);
+static long int cns11643_3_to_unicode(int, int);
+static long int cns11643_4_to_unicode(int, int);
+static long int cns11643_5_to_unicode(int, int);
+static long int cns11643_6_to_unicode(int, int);
+static long int cns11643_7_to_unicode(int, int);
 static long int null_dbcs_to_unicode(int, int);
 
 const struct iso2022_subcharset {
@@ -69,10 +78,19 @@ const struct iso2022_subcharset {
 #if 0
     { M4, 0, '@' }, /* JIS C 6226-1978 */
 #endif
+    { M4, 0, '0', -0x21, 0, &emacs_big5_1_to_unicode },
+    { M4, 0, '1', -0x21, 0, &emacs_big5_2_to_unicode },
     { M4, 0, 'A', -0x21, 0, &gb2312_to_unicode },
     { M4, 0, 'B', -0x21, 0, &jisx0208_to_unicode },
     { M4, 0, 'C', -0x21, 0, &ksx1001_to_unicode },
     { M4, 0, 'D', -0x21, 0, &jisx0212_to_unicode },
+    { M4, 0, 'G', -0x21, 0, &cns11643_1_to_unicode },
+    { M4, 0, 'H', -0x21, 0, &cns11643_2_to_unicode },
+    { M4, 0, 'I', -0x21, 0, &cns11643_3_to_unicode },
+    { M4, 0, 'J', -0x21, 0, &cns11643_4_to_unicode },
+    { M4, 0, 'K', -0x21, 0, &cns11643_5_to_unicode },
+    { M4, 0, 'L', -0x21, 0, &cns11643_6_to_unicode },
+    { M4, 0, 'M', -0x21, 0, &cns11643_7_to_unicode },
     { M4, 0, '~', 0, 0, &null_dbcs_to_unicode }, /* empty 94^n-set */
     { M6, 0, '~', 0, 0, &null_dbcs_to_unicode }, /* empty 96^n-set */
 };
@@ -82,6 +100,63 @@ static long int null_dbcs_to_unicode(int r, int c)
     return ERROR;
 }
 
+/*
+ * Emacs encodes Big5 in COMPOUND_TEXT as two 94x94 character sets.
+ * We treat Big5 as a 94x191 character set with a bunch of undefined
+ * columns in the middle, so we have to mess around a bit to make
+ * things fit.
+ */
+
+static long int emacs_big5_1_to_unicode(int r, int c)
+{
+    unsigned long s;
+    s = r * 94 + c;
+    r = s / 157;
+    c = s % 157;
+    if (c >= 64) c += 34; /* Skip over the gap */
+    return big5_to_unicode(r, c);
+}
+
+static long int emacs_big5_2_to_unicode(int r, int c)
+{
+    unsigned long s;
+    s = r * 94 + c;
+    r = s / 157 + 40;
+    c = s % 157;
+    if (c >= 64) c += 34; /* Skip over the gap */
+    return big5_to_unicode(r, c);
+}
+
+/* Wrappers for cns11643_to_unicode() */
+static long int cns11643_1_to_unicode(int r, int c)
+{
+    return cns11643_to_unicode(0, r, c);
+}
+static long int cns11643_2_to_unicode(int r, int c)
+{
+    return cns11643_to_unicode(1, r, c);
+}
+static long int cns11643_3_to_unicode(int r, int c)
+{
+    return cns11643_to_unicode(2, r, c);
+}
+static long int cns11643_4_to_unicode(int r, int c)
+{
+    return cns11643_to_unicode(3, r, c);
+}
+static long int cns11643_5_to_unicode(int r, int c)
+{
+    return cns11643_to_unicode(4, r, c);
+}
+static long int cns11643_6_to_unicode(int r, int c)
+{
+    return cns11643_to_unicode(5, r, c);
+}
+static long int cns11643_7_to_unicode(int r, int c)
+{
+    return cns11643_to_unicode(6, r, c);
+}
+
 /* States, or "what we're currently accumulating". */
 enum {
     IDLE,      /* None of the below */
@@ -730,6 +805,13 @@ int main(void)
     iso2022_read_test(TESTSTR("\x1b-A\x1b%/2\x80\x89"
                              "big5-0\2\xa1\x40\xa1\x40"),
                      0x3000, 0xa1, 0x40, 0, -1);
+    /* Emacs Big5-in-ISO-2022 mapping */
+    iso2022_read_test(TESTSTR("\x1b$(0&x86\x1b(B  \x1b$(0DeBv"),
+                     0x5143, 0x6c23, ' ', ' ', 0x958b, 0x767c, 0, -1);
+    /* Test from RFC 1922 (ISO-2022-CN) */
+    iso2022_read_test(TESTSTR("\x1b$)A\x0e=;;;\x1b$)GG(_P\x0f"),
+                     0x4EA4, 0x6362, 0x4EA4, 0x63db, 0, -1);
+    
     printf("read tests completed\n");
     printf("total: %d errors\n", total_errs);
     return (total_errs != 0);