1 files changed, 88 insertions, 0 deletions
diff --git a/utf8.c b/utf8.c
new file mode 100644
index 0000000..6954f96
--- /dev/null
+++ b/utf8.c
@@ -0,0 +1,88 @@
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include "utf8.h"
+
+/* defines outUTF8(), fromUTF8()
+   to translate Unicode chars to, from UTF-8 byte sequences - DAT 12.4.2009 */
+
+#define out(u,fil) putc((int)(u),(fil))
+
+/*
+FILE *f;
+int nextch()
+{ return getc(f); }
+*/
+
+static char errs[24];
+char *p;
+
+void utf8report()
+{ fprintf(stderr,"protocol error - %s sequence:%s\n",
+          strstr(errs,"EOF")?"incomplete":"invalid",errs);
+  exit(1);
+}
+
+#define foo(a) if(a==EOF)strcat(p," EOF"),p+=4;else sprintf(p," %#x",a),p+=strlen(p)
+#define err(a) {*errs='\0';p=errs;foo(a); utf8report(); }
+#define err2(a,b)  {*errs='\0';p=errs;foo(a);foo(b); utf8report(); }
+#define err3(a,b,c)  {*errs='\0';p=errs;foo(a);foo(b);foo(c); utf8report(); }
+#define err4(a,b,c,d)  {*errs='\0';p=errs;foo(a);foo(b);foo(c);foo(d); utf8report(); }
+
+#define nextch(x) ((x)=getc(fil))
+
+/*
+#define nextch(x) ((x)=*(*p)++)
+unicode scanUTF8(char **p)
+*/
+
+unicode fromUTF8(FILE *fil)
+/* returns a unicode value or EOF for end of input */
+{ unsigned c0,c1,c2,c3;
+  if((nextch(c0))==EOF)return(EOF);
+  if(c0<=0x7f) /* ascii */
+    return(c0);
+  if((c0&0xe0)==0xc0)
+    { /* 2 bytes */
+      if((nextch(c1))==EOF)err2(c0,c1); 
+      if((c1&0xc0)!=0x80)err2(c0,c1);
+      return((c0&0x1f)<<6|c1&0x3f);
+    }
+  if((c0&0xf0)==0xe0)
+    { /* 3 bytes */
+      if((nextch(c1))==EOF)err2(c0,c1);
+      if((c1&0xc0)!=0x80)err2(c0,c1);
+      if((nextch(c2))==EOF)err3(c0,c1,c2);
+      if((c2&0xc0)!=0x80)err3(c0,c1,c2);
+      return((c0&0xf)<<12|(c1&0x3f)<<6|c2&0x3f);
+    }
+  if((c0&0xf8)==0xf0)
+    { /* 4 bytes */
+      if((nextch(c1))==EOF)err2(c0,c1);
+      if((c1&0xc0)!=0x80)err2(c0,c1);
+      if((nextch(c2))==EOF)err3(c0,c1,c2);
+      if((c2&0xc0)!=0x80)err3(c0,c1,c2);
+      if((nextch(c3))==EOF)err4(c0,c1,c2,c3);
+      if((c3&0xc0)!=0x80)err4(c0,c1,c2,c3);
+      return((c0&7)<<18|(c1&0x3f)<<12|(c2&0x3f)<<6|c3&0x3f);
+    }
+  err(c0);
+}
+
+void outUTF8(unicode u, FILE *fil)
+{ if(u<=0x7f)
+  /* ascii */
+    out(u,fil); else
+  if(u<=0x7ff)
+  /* latin1 and other chars requiring 2 octets */
+    out(0xc0|(u&0x7c0)>>6,fil),out(0x80|u&0x3f,fil); else
+  if(u<=0xffff)
+  /* to here is basic multilingual plane */
+    out(0xe0|(u&0xf000)>>12,fil),out(0x80|(u&0xfc0)>>6,fil),out(0x80|u&0x3f,fil); else
+  if(u<=0x10ffff)
+  /* other planes - rarely used - 4 octets */
+    out(0xf0|(u&0x1c0000)>>18,fil),out(0x80|(u&0x3f000)>>12,fil),out(0x80|(u&0xfc0)>>6,fil),
+        out(0x80|u&0x3f,fil); else
+  /* codes above 0x10ffff not valid */
+  fprintf(stderr,"char 0x%lx out of unicode range\n",u),exit(1);
+}