From: Martin Mares <mj@ucw.cz>
Date: Thu, 1 Mar 2001 16:57:58 +0000 (+0000)
Subject: Defined a GET_TAGGED_CHAR macro to read our internal representation
X-Git-Tag: holmes-import~1540
X-Git-Url: http://mj.ucw.cz/gitweb/?a=commitdiff_plain;h=91373ee6489fbe08db12bef2cf6df704b31e1378;p=libucw.git

Defined a GET_TAGGED_CHAR macro to read our internal representation
of tagged text, mapping the tags to character codes >= 0x80000000.
---

diff --git a/lib/index.h b/lib/index.h
index 6394b65c..70a8186d 100644
--- a/lib/index.h
+++ b/lib/index.h
@@ -62,3 +62,23 @@ struct fingerprint {
 };
 
 void fingerprint(byte *string, struct fingerprint *fp);
+
+/* Reading of tagged text (Unicode values, tags mapped to 0x80000000 and higher) */
+
+#define GET_TAGGED_CHAR(p,u) do {				\
+  u = *p;							\
+  if (u >= 0xc0)						\
+    GET_UTF8(p,u);						\
+  else if (u >= 0x80)						\
+    {								\
+      p++;							\
+      if (u >= 0xb0)						\
+	u += 0x80020000;					\
+      else if (u >= 0xa0)					\
+	u = 0x80010000 + ((u & 0x0f) << 6) + (*p++ & 0x3f);	\
+      else							\
+	u += 0x80000000;					\
+    }								\
+  else								\
+    p++;							\
+} while (0)