aboutsummaryrefslogtreecommitdiffstats
path: root/util
diff options
context:
space:
mode:
Diffstat (limited to '')
-rw-r--r--util/ChangeLog11
-rw-r--r--util/miscutil.c25
-rw-r--r--util/strgutil.c204
-rw-r--r--util/ttyio.c23
4 files changed, 202 insertions, 61 deletions
diff --git a/util/ChangeLog b/util/ChangeLog
index 95b06019a..a451e4a9a 100644
--- a/util/ChangeLog
+++ b/util/ChangeLog
@@ -1,3 +1,14 @@
+Tue Aug 31 17:20:44 CEST 1999 Werner Koch <[email protected]>
+
+
+ * strgutil (utf8_to_native): Implemented.
+ (check_utf8_string): Removed.
+
+ * miscutil.c (make_printable_string): Fixed possible buffer overflow.
+ (print_utf8_string): New.
+
+ * ttyio.c (tty_print_utf8_string): New.
+
Mon Aug 30 20:38:33 CEST 1999 Werner Koch <[email protected]>
diff --git a/util/miscutil.c b/util/miscutil.c
index 2b95d97d6..eb72415bb 100644
--- a/util/miscutil.c
+++ b/util/miscutil.c
@@ -194,6 +194,29 @@ print_string( FILE *fp, const byte *p, size_t n, int delim )
}
/****************
+ * Print an UTF8 string to FP and filter all control characters out.
+ */
+void
+print_utf8_string( FILE *fp, const byte *p, size_t n )
+{
+ size_t i;
+ char *buf;
+
+ /* we can handle plain ascii simpler, so check for it first */
+ for(i=0; i < n; i++ ) {
+ if( p[i] & 0x80 )
+ break;
+ }
+ if( i < n ) {
+ buf = utf8_to_native( p, n );
+ fputs( buf, fp );
+ m_free( buf );
+ }
+ else
+ print_string( fp, p, n, 0 );
+}
+
+/****************
* This function returns a string which is suitable for printing
* Caller must release it with m_free()
*/
@@ -211,7 +234,7 @@ make_printable_string( const byte *p, size_t n, int delim )
|| *p=='\v' || *p=='\b' || !*p )
buflen += 2;
else
- buflen += 3;
+ buflen += 4;
}
else
buflen++;
diff --git a/util/strgutil.c b/util/strgutil.c
index 87eaad423..9ab63a047 100644
--- a/util/strgutil.c
+++ b/util/strgutil.c
@@ -376,77 +376,161 @@ native_to_utf8( const char *string )
/****************
- * Convert string, which is in UTF8 to native encoding. Replace
- * illegal encodings by some "\xnn".
+ * Convert string, which is in UTF8 to native encoding.
+ * illegal encodings by some "\xnn" and quote all control characters
*/
char *
-utf8_to_native( const char *string )
+utf8_to_native( const char *string, size_t length )
{
- #if 0
+ int nleft;
+ int i;
+ byte encbuf[7];
+ int encidx;
const byte *s;
size_t n;
- byte *buffer, *p;
-
- /* quick check whether we actually have characters with bit 8 set */
- for( s=string; *s; s++ )
- if( *s & 0x80 )
- break;
- if( !*s ) /* that is easy */
- return m_strdup(string);
-
- /* count the extended utf-8 characters */
- 110x xxxx
- 1110 xxxx
- 1111 0xxx
- for( n=1, s=string; *s; s++ ) {
- if( !(*s & 0x80) )
- n++;
- else if( (*s & 0xe0) == 0xc0 )
- n += 2;
- else if( (*s & 0xf0) == 0xe0 )
- n += 3;
- else if( (*s & 0xf8) == 0xf0 )
- n += 4;
- else
- n++; /* invalid encoding */
- }
+ byte *buffer = NULL, *p = NULL;
+ unsigned long val = 0;
+ size_t slen;
+ int resync = 0;
+
+ /* 1. pass (p==NULL): count the extended utf-8 characters */
+ /* 2. pass (p!=NULL): create string */
+ for( ;; ) {
+ for( slen=length, nleft=encidx=0, n=0, s=string; slen; s++, slen-- ) {
+ if( resync ) {
+ if( !(*s < 128 || (*s >= 0xc0 && *s <= 0xfd)) ) {
+ /* still invalid */
+ if( p ) {
+ sprintf(p, "\\x%02x", *s );
+ p += 4;
+ }
+ n += 4;
+ continue;
+ }
+ resync = 0;
+ }
+ if( !nleft ) {
+ if( !(*s & 0x80) ) { /* plain ascii */
+ if( iscntrl( *s ) ) {
+ n++;
+ if( p )
+ *p++ = '\\';
+ switch( *s ) {
+ case '\n': n++; if( p ) *p++ = 'n'; break;
+ case '\r': n++; if( p ) *p++ = 'r'; break;
+ case '\f': n++; if( p ) *p++ = 'f'; break;
+ case '\v': n++; if( p ) *p++ = 'v'; break;
+ case '\b': n++; if( p ) *p++ = 'b'; break;
+ case 0 : n++; if( p ) *p++ = '0'; break;
+ default: n += 3;
+ sprintf( p, "x%02x", *s );
+ p += 3;
+ break;
+ }
+ }
+ else {
+ if( p ) *p++ = *s;
+ n++;
+ }
+ }
+ else if( (*s & 0xe0) == 0xc0 ) { /* 110x xxxx */
+ val = *s & 0x1f;
+ nleft = 1;
+ encbuf[encidx=0] = *s;
+ }
+ else if( (*s & 0xf0) == 0xe0 ) { /* 1110 xxxx */
+ val = *s & 0x0f;
+ nleft = 2;
+ encbuf[encidx=0] = *s;
+ }
+ else if( (*s & 0xf8) == 0xf0 ) { /* 1111 0xxx */
+ val = *s & 0x07;
+ nleft = 3;
+ encbuf[encidx=0] = *s;
+ }
+ else if( (*s & 0xfc) == 0xf8 ) { /* 1111 10xx */
+ val = *s & 0x03;
+ nleft = 4;
+ encbuf[encidx=0] = *s;
+ }
+ else if( (*s & 0xfe) == 0xfc ) { /* 1111 110x */
+ val = *s & 0x01;
+ nleft = 5;
+ encbuf[encidx=0] = *s;
+ }
+ else { /* invalid encoding: print as \xnn */
+ if( p ) {
+ sprintf(p, "\\x%02x", *s );
+ p += 4;
+ }
+ n += 4;
+ resync = 1;
+ }
+ }
+ else if( *s < 0x80 || *s >= 0xc0 ) { /* invalid */
+ if( p ) {
+ sprintf(p, "\\x%02x", *s );
+ p += 4;
+ }
+ n += 4;
+ nleft = 0;
+ resync = 1;
+ }
+ else {
+ encbuf[++encidx] = *s;
+ val <<= 6;
+ val |= *s & 0x3f;
+ if( !--nleft ) { /* ready */
+ if( active_charset ) { /* table lookup */
+ for(i=0; i < 128; i++ ) {
+ if( active_charset[i] == val )
+ break;
+ }
+ if( i < 128 ) { /* we can print this one */
+ if( p ) *p++ = i+128;
+ n++;
+ }
+ else { /* we do not have a translation: print utf8 */
+ if( p ) {
+ for(i=0; i < encidx; i++ ) {
+ sprintf(p, "\\x%02x", encbuf[i] );
+ p += 4;
+ }
+ }
+ n += encidx*4;
+ }
+ }
+ else { /* native set */
+ if( val >= 0x80 && val < 256 ) {
+ n++; /* we can simply print this character */
+ if( p ) *p++ = val;
+ }
+ else { /* we do not have a translation: print utf8 */
+ if( p ) {
+ for(i=0; i < encidx; i++ ) {
+ sprintf(p, "\\x%02x", encbuf[i] );
+ p += 4;
+ }
+ }
+ n += encidx*4;
+ }
+ }
- buffer = p = m_alloc( n );
- for( s=string; *s; ) {
- if( !(*s & 0x80) )
- *p++ = *s++;
- else if( (*s & 0xe0) == 0xc0 ) {
- u32 val;
- if( (s[1] & 0xc0) != 0x80 )
- ;
- val = (*s << 6) | (s[1] & 0x3f);
+ }
+
+ }
+ }
+ if( !buffer ) { /* allocate the buffer after the first pass */
+ buffer = p = m_alloc( n + 1 );
+ }
+ else {
+ *p = 0; /* make a string */
+ return buffer;
}
- else if( (*s & 0xf0) == 0xe0 )
- n += 3;
- else if( (*s & 0xf8) == 0xf0 )
- n += 4;
- else
- n++; /* invalid encoding */
}
- #endif
- return m_strdup(string);
-
}
-/****************
- * check whether string is a valid UTF8 string.
- * Returns 0 = Okay
- * 1 = Too short
- * 2 = invalid encoding
- */
-int
-check_utf8_string( const char *string )
-{
- /*fixme */
- return 0;
-}
-
/*********************************************
********** missing string functions *********
diff --git a/util/ttyio.c b/util/ttyio.c
index 3f5eb7900..74f6ce0f6 100644
--- a/util/ttyio.c
+++ b/util/ttyio.c
@@ -235,6 +235,29 @@ tty_print_string( byte *p, size_t n )
#endif
}
+void
+tty_print_utf8_string( byte *p, size_t n )
+{
+ size_t i;
+ char *buf;
+
+ if (no_terminal)
+ return;
+
+ /* we can handle plain ascii simpler, so check for it first */
+ for(i=0; i < n; i++ ) {
+ if( p[i] & 0x80 )
+ break;
+ }
+ if( i < n ) {
+ buf = utf8_to_native( p, n );
+ tty_printf("%s", buf );
+ m_free( buf );
+ }
+ else
+ tty_print_string( p, n );
+}
+
generated by cgit v1.2.3 (git 2.25.1) at 2025-05-07 06:25:33 +0000