Code to support Cyrillic, Greek, Arabic, Baltic symbol sets, see the

reference bug #690706 for details.  Thanks to Josef Hinteregger for
writing and testing the code to support the new sets.

git-svn-id: http://svn.ghostscript.com/ghostscript/trunk@12278 a1074d23-0009-0410-80fe-cf8c14f379e6

1 files changed, 15 insertions, 20 deletions

diff --git a/pcl/pctpm.h b/pcl/pctpm.h
index 37257b58b..cdab7c349 100644
--- a/pcl/pctpm.h
+++ b/pcl/pctpm.h
@@ -18,28 +18,23 @@
 
 #include "gx.h"
 
-typedef struct pcl_text_parsing_method_s {
-    byte    min1, max1;
-    byte    min2, max2;
+typedef enum  {
+/* http://lprng.sourceforge.net/DISTRIB/RESOURCES/DOCS/pcl5comp.pdf */
+    tpm_0_SBCS    = 0,
+    tpm_21_DBCS7  = 21,   /* e.g. GB_2312-80 (China), JIS_X0208-1983 (Japan), KSC_5601 (Korea) */
+    tpm_31_sjis   = 31,
+    tpm_38_DBCS8  = 38,	  
+    tpm_83_utf8   = 83,	  /* http://docs.hp.com/en/5991-7956/5991-7956.pdf */
+                          /* HP-UX 11i v3 International Printing Features  */
+    tpm_1008_utf8 = 1008  /* http://www.lexmark.com/publications/pdfs/TRef_3Q01.pdf */
 } pcl_text_parsing_method_t;
 
-#define pcl_char_is_2_byte(ch, tpm)                     \
-    ( ((ch) >= (tpm)->min1) &&                          \
-      ((ch) <= (tpm)->max2) &&                          \
-      (((ch) <= (tpm)->max1) || ((ch) >= (tpm)->min2)) )
+int pcl_char_bytelen(byte ch, pcl_text_parsing_method_t tpm);
+gs_char pcl_char_get_char(pcl_text_parsing_method_t tpm, const byte ** psrc, int len);
 
-#define pcl_tpm_is_single_byte(tpm) ((tpm)->max1 == 0)
-
-/* Single-byte only */
-#define pcl_tpm_0_data  { 0xff, 0, 0xff, 0 }
-
-/* 0x21-0xff are double-byte */
-#define pcl_tpm_21_data { 0x21, 0xff, 0x21, 0xff }
-
-/* 0x81-0x9f, 0xe0-0xfc are double-byte */
-#define pcl_tpm_31_data { 0x81, 0x9f, 0xe0, 0xfc }
-
-/* 0x80-0xff are double-byte */
-#define pcl_tpm_38_data { 0x80, 0xff, 0x80, 0xff }
+/*
+ *  was 0xfffd  
+ */
+#define INVALID_UC 0xffff
 
 #endif			/* pctpm_INCLUDED */