Making sure an invalid starting byte of a multibyte sequence is properly

terminated, to prevent the displaying of ghost characters.

git-svn-id: svn://svn.savannah.gnu.org/nano/trunk/nano@5206 35c25a1d-7b9e-4130-9fde-d3aeb78583b8

diff --git a/ChangeLog b/ChangeLog
index bdc85d08b4ffaa3c4405f3d12101a549372f5258..1ef7b1767f96880d7ff3edddd3b29f1898197e83 100644 (file)
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,6 +1,14 @@
 2015-04-18  Benno Schulenberg  <bensberg@justemail.net>
        * src/global.c, src/nano.c, doc/man/nanorc.5, doc/texinfo/nano.texi:
        Make the descriptions of the multibuffer feature more accurate.
+       * src/winio.c (display_string): Make sure an invalid starting byte
+       of a multibyte sequence is properly terminated, so that it doesn't
+       pick up lingering bytes of any previous content.  This prevents the
+       displaying of ghosts -- characters that aren't really there -- when a
+       file contains valid ánd invalid UTF-8 sequences.  For an example see:
+       https://lists.gnu.org/archive/html/nano-devel/2015-04/msg00052.html.
+       Also make two comments more accurate: an invalid multibyte sequence
+       will never be categorized as a control character or anything else.
 
 2015-04-18  Mark Oteiza  <mvoteiza@udel.edu>
        * doc/syntax/{python,ruby,sh,tex}.nanorc: Add a linter definition.

diff --git a/src/winio.c b/src/winio.c
index 9d6587a42d3ecf1660b191539cbbc7850a9c167c..917f7d6e975b86f765c53c9545463ba4915b2bfd 100644 (file)
--- a/src/winio.c
+++ b/src/winio.c
@@ -2004,9 +2004,7 @@ char *display_string(const char *buf, size_t start_col, size_t len, bool
                converted[index++] = ' ';
                start_col++;
            }
-       /* If buf contains a control character, interpret it.  If buf
-        * contains an invalid multibyte control character, display it
-        * as such. */
+       /* If buf contains a control character, interpret it. */
        } else if (is_cntrl_mbchar(buf_mb)) {
            char *ctrl_buf_mb = charalloc(mb_cur_max());
            int ctrl_buf_mb_len, i;
@@ -2036,13 +2034,17 @@ char *display_string(const char *buf, size_t start_col, size_t len, bool
 #endif
                converted[index++] = ' ';
            start_col++;
-       /* If buf contains a non-control character, interpret it.  If
-        * buf contains an invalid multibyte non-control character,
-        * display it as such. */
+       /* If buf contains a non-control character, interpret it.  If buf
+        * contains an invalid multibyte sequence, display it as such. */
        } else {
            char *nctrl_buf_mb = charalloc(mb_cur_max());
            int nctrl_buf_mb_len, i;
 
+           /* Make sure an invalid sequence-starter byte is properly
+            * terminated, so that it doesn't pick up lingering bytes
+            * of any previous content. */
+           null_at(&buf_mb, buf_mb_len);
+
            nctrl_buf_mb = mbrep(buf_mb, nctrl_buf_mb,
                &nctrl_buf_mb_len);