Make g_utf8_make_valid public

Based on a patch by Simon van der Linden and rebased onto current GLib,
with improved documentation loosely based on Telepathy's
tp_utf8_make_valid().

Signed-off-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Bug: https://bugzilla.gnome.org/show_bug.cgi?id=591603
Bug: https://bugzilla.gnome.org/show_bug.cgi?id=610969
Reviewed-by: Colin Walters <walters@verbum.org>

6 files changed, 36 insertions, 19 deletions

diff --git a/docs/reference/glib/glib-sections.txt b/docs/reference/glib/glib-sections.txt
index d8f1643bd..4745446d0 100644
--- a/docs/reference/glib/glib-sections.txt
+++ b/docs/reference/glib/glib-sections.txt
@@ -2923,6 +2923,7 @@ g_utf8_strrchr
 g_utf8_strreverse
 g_utf8_substring
 g_utf8_validate
+g_utf8_make_valid
 
 <SUBSECTION>
 g_utf8_strup
diff --git a/glib/gconvert.c b/glib/gconvert.c
index 892337120..1de941edc 100644
--- a/glib/gconvert.c
+++ b/glib/gconvert.c
@@ -1939,7 +1939,7 @@ g_filename_display_name (const gchar *filename)
    * by a question mark
    */
   if (!display_name) 
-    display_name = _g_utf8_make_valid (filename);
+    display_name = g_utf8_make_valid (filename);
 
   return display_name;
 }
diff --git a/glib/gkeyfile.c b/glib/gkeyfile.c
index 4c444ab65..70d8fda3e 100644
--- a/glib/gkeyfile.c
+++ b/glib/gkeyfile.c
@@ -1198,7 +1198,7 @@ g_key_file_parse_line (GKeyFile     *key_file,
 				     &parse_error);
   else
     {
-      gchar *line_utf8 = _g_utf8_make_valid (line);
+      gchar *line_utf8 = g_utf8_make_valid (line);
       g_set_error (error, G_KEY_FILE_ERROR,
                    G_KEY_FILE_ERROR_PARSE,
                    _("Key file contains line “%s” which is not "
@@ -1330,7 +1330,7 @@ g_key_file_parse_key_value_pair (GKeyFile     *key_file,
     {
       if (g_ascii_strcasecmp (value, "UTF-8") != 0)
         {
-	  gchar *value_utf8 = _g_utf8_make_valid (value);
+	  gchar *value_utf8 = g_utf8_make_valid (value);
           g_set_error (error, G_KEY_FILE_ERROR,
                        G_KEY_FILE_ERROR_UNKNOWN_ENCODING,
                        _("Key file contains unsupported "
@@ -1863,7 +1863,7 @@ g_key_file_get_string (GKeyFile     *key_file,
 
   if (!g_utf8_validate (value, -1, NULL))
     {
-      gchar *value_utf8 = _g_utf8_make_valid (value);
+      gchar *value_utf8 = g_utf8_make_valid (value);
       g_set_error (error, G_KEY_FILE_ERROR,
                    G_KEY_FILE_ERROR_UNKNOWN_ENCODING,
                    _("Key file contains key “%s” with value “%s” "
@@ -1979,7 +1979,7 @@ g_key_file_get_string_list (GKeyFile     *key_file,
 
   if (!g_utf8_validate (value, -1, NULL))
     {
-      gchar *value_utf8 = _g_utf8_make_valid (value);
+      gchar *value_utf8 = g_utf8_make_valid (value);
       g_set_error (error, G_KEY_FILE_ERROR,
                    G_KEY_FILE_ERROR_UNKNOWN_ENCODING,
                    _("Key file contains key “%s” with value “%s” "
@@ -4291,7 +4291,7 @@ g_key_file_parse_value_as_integer (GKeyFile     *key_file,
 
   if (*value == '\0' || (*eof_int != '\0' && !g_ascii_isspace(*eof_int)))
     {
-      gchar *value_utf8 = _g_utf8_make_valid (value);
+      gchar *value_utf8 = g_utf8_make_valid (value);
       g_set_error (error, G_KEY_FILE_ERROR,
 		   G_KEY_FILE_ERROR_INVALID_VALUE,
 		   _("Value “%s” cannot be interpreted "
@@ -4304,7 +4304,7 @@ g_key_file_parse_value_as_integer (GKeyFile     *key_file,
   int_value = long_value;
   if (int_value != long_value || errno == ERANGE)
     {
-      gchar *value_utf8 = _g_utf8_make_valid (value);
+      gchar *value_utf8 = g_utf8_make_valid (value);
       g_set_error (error,
 		   G_KEY_FILE_ERROR, 
 		   G_KEY_FILE_ERROR_INVALID_VALUE,
@@ -4338,7 +4338,7 @@ g_key_file_parse_value_as_double  (GKeyFile     *key_file,
 
   if (*end_of_valid_d != '\0' || end_of_valid_d == value)
     {
-      gchar *value_utf8 = _g_utf8_make_valid (value);
+      gchar *value_utf8 = g_utf8_make_valid (value);
       g_set_error (error, G_KEY_FILE_ERROR,
 		   G_KEY_FILE_ERROR_INVALID_VALUE,
 		   _("Value “%s” cannot be interpreted "
@@ -4377,7 +4377,7 @@ g_key_file_parse_value_as_boolean (GKeyFile     *key_file,
   else if (strcmp_sized (value, length, "false") == 0 || strcmp_sized (value, length, "0") == 0)
     return FALSE;
 
-  value_utf8 = _g_utf8_make_valid (value);
+  value_utf8 = g_utf8_make_valid (value);
   g_set_error (error, G_KEY_FILE_ERROR,
                G_KEY_FILE_ERROR_INVALID_VALUE,
                _("Value “%s” cannot be interpreted "
diff --git a/glib/gmarkup.c b/glib/gmarkup.c
index 834eb9e2c..dc828d531 100644
--- a/glib/gmarkup.c
+++ b/glib/gmarkup.c
@@ -423,7 +423,7 @@ set_error (GMarkupParseContext  *context,
   /* Make sure that the GError message is valid UTF-8
    * even if it is complaining about invalid UTF-8 in the markup
    */
-  s_valid = _g_utf8_make_valid (s);
+  s_valid = g_utf8_make_valid (s);
   set_error_literal (context, error, code, s);
 
   g_free (s);
diff --git a/glib/gunicode.h b/glib/gunicode.h
index ce35f6b71..558b37254 100644
--- a/glib/gunicode.h
+++ b/glib/gunicode.h
@@ -885,9 +885,8 @@ GLIB_AVAILABLE_IN_ALL
 gchar *g_utf8_collate_key_for_filename (const gchar *str,
                                         gssize       len) G_GNUC_MALLOC;
 
-
-/* private */
-gchar *_g_utf8_make_valid (const gchar *name);
+GLIB_AVAILABLE_IN_2_52
+gchar *g_utf8_make_valid (const gchar *str);
 
 G_END_DECLS
 
diff --git a/glib/gutf8.c b/glib/gutf8.c
index eae10daa3..4c5caccfd 100644
--- a/glib/gutf8.c
+++ b/glib/gutf8.c
@@ -1735,19 +1735,36 @@ g_utf8_strreverse (const gchar *str,
   return result;
 }
 
-
+/**
+ * g_utf8_make_valid:
+ * @str: string to coerce into UTF-8
+ *
+ * If the provided string is valid UTF-8, return a copy of it. If not,
+ * return a copy in which bytes that could not be interpreted as valid Unicode
+ * are replaced with the Unicode replacement character (U+FFFD).
+ *
+ * For example, this is an appropriate function to use if you have received
+ * a string that was incorrectly declared to be UTF-8, and you need a valid
+ * UTF-8 version of it that can be logged or displayed to the user, with the
+ * assumption that it is close enough to ASCII or UTF-8 to be mostly
+ * readable as-is.
+ *
+ * Returns: (transfer full): a valid UTF-8 string whose content resembles @str
+ *
+ * Since: 2.52
+ */
 gchar *
-_g_utf8_make_valid (const gchar *name)
+g_utf8_make_valid (const gchar *str)
 {
   GString *string;
   const gchar *remainder, *invalid;
   gint remaining_bytes, valid_bytes;
 
-  g_return_val_if_fail (name != NULL, NULL);
+  g_return_val_if_fail (str != NULL, NULL);
 
   string = NULL;
-  remainder = name;
-  remaining_bytes = strlen (name);
+  remainder = str;
+  remaining_bytes = strlen (str);
 
   while (remaining_bytes != 0) 
     {
@@ -1767,7 +1784,7 @@ _g_utf8_make_valid (const gchar *name)
     }
   
   if (string == NULL)
-    return g_strdup (name);
+    return g_strdup (str);
   
   g_string_append (string, remainder);