X-Git-Url: http://www.chiark.greenend.org.uk/ucgi/~ianmdlvl/git?a=blobdiff_plain;f=src%2Fshared%2Futf8.c;h=c559c13678d62b7cd275f05721febe73b5205924;hb=d9c67ea112724e271c39553d966eae612e272e34;hp=c3d97cc783f06e2805a4a2394a162e87a3b8b0e1;hpb=8f6ce71fe79d897b67157d92869db87ee2042af6;p=elogind.git

diff --git a/src/shared/utf8.c b/src/shared/utf8.c
index c3d97cc78..c559c1367 100644
--- a/src/shared/utf8.c
+++ b/src/shared/utf8.c
@@ -3,6 +3,7 @@
 /***
   This file is part of systemd.
 
+  Copyright 2008-2011 Kay Sievers
   Copyright 2012 Lennart Poettering
 
   systemd is free software; you can redistribute it and/or modify it
@@ -19,7 +20,7 @@
   along with systemd; If not, see <http://www.gnu.org/licenses/>.
 ***/
 
-/* This file is based on the GLIB utf8 validation functions. The
+/* Parts of this file are based on the GLIB utf8 validation functions. The
  * original license text follows. */
 
 /* gutf8.c - Operations on UTF-8 strings.
@@ -97,7 +98,7 @@ static int utf8_encoded_expected_len(const char *str) {
 }
 
 /* decode one unicode char */
-static int utf8_encoded_to_unichar(const char *str) {
+int utf8_encoded_to_unichar(const char *str) {
         int unichar;
         int len;
         int i;
@@ -135,19 +136,21 @@ static int utf8_encoded_to_unichar(const char *str) {
         return unichar;
 }
 
-bool utf8_is_printable(const char* str, size_t length) {
+bool utf8_is_printable_newline(const char* str, size_t length, bool newline) {
         const uint8_t *p;
 
         assert(str);
 
-        for (p = (const uint8_t*) str; length; p++) {
+        for (p = (const uint8_t*) str; length;) {
                 int encoded_len = utf8_encoded_valid_unichar((const char *)p);
-                int32_t val = utf8_encoded_to_unichar((const char*)p);
+                int val = utf8_encoded_to_unichar((const char*)p);
 
-                if (encoded_len < 0 || val < 0 || is_unicode_control(val))
+                if (encoded_len < 0 || val < 0 || is_unicode_control(val) ||
+                    (!newline && val == '\n'))
                         return false;
 
                 length -= encoded_len;
+                p += encoded_len;
         }
 
         return true;
@@ -159,7 +162,9 @@ const char *utf8_is_valid(const char *str) {
         assert(str);
 
         for (p = (const uint8_t*) str; *p; ) {
-                int len = utf8_encoded_valid_unichar((const char *)p);
+                int len;
+
+                len = utf8_encoded_valid_unichar((const char *)p);
 
                 if (len < 0)
                         return NULL;
@@ -170,37 +175,42 @@ const char *utf8_is_valid(const char *str) {
         return str;
 }
 
-char *ascii_is_valid(const char *str) {
-        const char *p;
+char *utf8_escape_invalid(const char *str) {
+        char *p, *s;
 
         assert(str);
 
-        for (p = str; *p; p++)
-                if ((unsigned char) *p >= 128)
-                        return NULL;
+        p = s = malloc(strlen(str) * 4 + 1);
+        if (!p)
+                return NULL;
 
-        return (char*) str;
-}
+        while (*str) {
+                int len;
 
-char *ascii_filter(const char *str) {
-        const char *s;
-        char *r, *d;
-        size_t l;
+                len = utf8_encoded_valid_unichar(str);
+                if (len > 0) {
+                        s = mempcpy(s, str, len);
+                        str += len;
+                } else {
+                        s = mempcpy(s, UTF8_REPLACEMENT_CHARACTER, strlen(UTF8_REPLACEMENT_CHARACTER));
+                        str += 1;
+                }
+        }
+        *s = '\0';
 
-        assert(str);
+        return p;
+}
 
-        l = strlen(str);
-        r = malloc(l + 1);
-        if (!r)
-                return NULL;
+char *ascii_is_valid(const char *str) {
+        const char *p;
 
-        for (s = str, d = r; *s; s++)
-                if ((unsigned char) *s < 128)
-                        *(d++) = *s;
+        assert(str);
 
-        *d = 0;
+        for (p = str; *p; p++)
+                if ((unsigned char) *p >= 128)
+                        return NULL;
 
-        return r;
+        return (char*) str;
 }
 
 char *utf16_to_utf8(const void *s, size_t length) {