Merge branch 'ossfuzz-11576-variant-text-unicode-unescape' into 'master'

gvariant: Fix error handling for parsing Unicode escapes

See merge request GNOME/glib!511

glib/gstrfuncs.c

@@ -1140,6 +1140,11 @@ g_parse_long_long (const gchar  *nptr,
  * changing the current locale, since that would not be
  * thread-safe.
  *
+ * Note that input with a leading minus sign (`-`) is accepted, and will return
+ * the negation of the parsed number, unless that would overflow a #guint64.
+ * Critically, this means you cannot assume that a short fixed length input will
+ * never result in a low return value, as the input could have a leading `-`.
+ *
  * This function is typically used when reading configuration
  * files or other non-user input that should be locale independent.
  * To handle input from the user you should normally use the
@@ -3304,7 +3309,8 @@ g_ascii_string_to_signed (const gchar  *str,
  * @base that is within inclusive bounds limited by @min and @max. If
  * this is true, then the converted number is stored in @out_num. An
  * empty string is not a valid input. A string with leading or
- * trailing whitespace is also an invalid input.
+ * trailing whitespace is also an invalid input. A string with a leading sign
+ * (`-` or `+`) is not a valid input for the unsigned parser.
  *
  * @base can be between 2 and 36 inclusive. Hexadecimal numbers must
  * not be prefixed with "0x" or "0X". Such a problem does not exist

glib/gvariant-parser.c

@@ -1528,18 +1528,21 @@ string_free (AST *ast)
   g_slice_free (String, string);
 }
 
+/* Accepts exactly @length hexadecimal digits. No leading sign or `0x`/`0X` prefix allowed.
+ * No leading/trailing space allowed. */
 static gboolean
 unicode_unescape (const gchar  *src,
                   gint         *src_ofs,
                   gchar        *dest,
                   gint         *dest_ofs,
-                  gint          length,
+                  gsize         length,
                   SourceRef    *ref,
                   GError      **error)
 {
   gchar buffer[9];
-  guint64 value;
+  guint64 value = 0;
   gchar *end;
+  gsize n_valid_chars;
 
   (*src_ofs)++;
 
@@ -1547,13 +1550,24 @@ unicode_unescape (const gchar  *src,
   strncpy (buffer, src + *src_ofs, length);
   buffer[length] = '\0';
 
-  value = g_ascii_strtoull (buffer, &end, 0x10);
+  for (n_valid_chars = 0; n_valid_chars < length; n_valid_chars++)
+    if (!g_ascii_isxdigit (buffer[n_valid_chars]))
+      break;
+
+  if (n_valid_chars == length)
+    value = g_ascii_strtoull (buffer, &end, 0x10);
 
   if (value == 0 || end != buffer + length)
     {
-      parser_set_error (error, ref, NULL,
+      SourceRef escape_ref;
+
+      escape_ref = *ref;
+      escape_ref.start += *src_ofs;
+      escape_ref.end = escape_ref.start + n_valid_chars;
+
+      parser_set_error (error, &escape_ref, NULL,
                         G_VARIANT_PARSE_ERROR_INVALID_CHARACTER,
-                        "invalid %d-character unicode escape", length);
+                        "invalid %" G_GSIZE_FORMAT "-character unicode escape", length);
       return FALSE;
     }
 

glib/tests/gvariant.c

@@ -4039,6 +4039,22 @@ test_parse_failures (void)
     "b\"\\\"",                  "0-4:",            "unterminated string constant",
     "b'\\'a",                   "0-5:",            "unterminated string constant",
     "b\"\\\"a",                 "0-5:",            "unterminated string constant",
+    "'\\u-ff4'",                "3:",              "invalid 4-character unicode escape",
+    "'\\u+ff4'",                "3:",              "invalid 4-character unicode escape",
+    "'\\u'",                    "3:",              "invalid 4-character unicode escape",
+    "'\\u0'",                   "3-4:",            "invalid 4-character unicode escape",
+    "'\\uHELLO'",               "3:",              "invalid 4-character unicode escape",
+    "'\\u ff4'",                "3:",              "invalid 4-character unicode escape",
+    "'\\u012'",                 "3-6:",            "invalid 4-character unicode escape",
+    "'\\u0xff4'",               "3-4:",            "invalid 4-character unicode escape",
+    "'\\U-ff4'",                "3:",              "invalid 8-character unicode escape",
+    "'\\U+ff4'",                "3:",              "invalid 8-character unicode escape",
+    "'\\U'",                    "3:",              "invalid 8-character unicode escape",
+    "'\\U0'",                   "3-4:",            "invalid 8-character unicode escape",
+    "'\\UHELLO'",               "3:",              "invalid 8-character unicode escape",
+    "'\\U ff4'",                "3:",              "invalid 8-character unicode escape",
+    "'\\U0123456'",             "3-10:",           "invalid 8-character unicode escape",
+    "'\\U0xff4'",               "3-4:",            "invalid 8-character unicode escape",
   };
   guint i;
 
@@ -4674,8 +4690,8 @@ test_print_context (void)
     { NULL, "[1, 2, 3, 'str']", " ^        ^^^^^" },
     { G_VARIANT_TYPE_UINT16, "{ 'abc':'def' }", "  ^^^^^^^^^^^^^^^" },
     { NULL, "<5", "    ^" },
-    { NULL, "'ab\\ux'", "  ^^^^^^^" },
-    { NULL, "'ab\\U00efx'", "  ^^^^^^^^^^^" }
+    { NULL, "'ab\\ux'", "       ^ " },
+    { NULL, "'ab\\U00efx'", "       ^^^^  " }
   };
   GVariant *v;
   gchar *s;

glib/tests/strfuncs.c

@@ -1098,6 +1098,7 @@ test_strtoll (void)
   check_uint64 ("18446744073709551616", "", 10, G_MAXUINT64, ERANGE);
   check_uint64 ("20xyz", "xyz", 10, 20, 0);
   check_uint64 ("-1", "", 10, G_MAXUINT64, 0);
+  check_uint64 ("-FF4", "", 16, -((guint64) 0xFF4), 0);
 
   check_int64 ("0", "", 10, 0, 0);
   check_int64 ("9223372036854775807", "", 10, G_MAXINT64, 0);