branch master updated: * tp/Texinfo/XS/xspara.c (xspara__add_next): Check return value for u8_mbtouc as documented, by checking for 0xfffd (replacement character), instead of using the same as mbrtowc.

[Gavin D. Smith]

This is an automated email from the git hooks/post-receive script.

gavin pushed a commit to branch master
in repository texinfo.

The following commit(s) were added to refs/heads/master by this push:
     new ad43617b34 * tp/Texinfo/XS/xspara.c (xspara__add_next): Check return 
value for u8_mbtouc as documented, by checking for 0xfffd (replacement 
character), instead of using the same as mbrtowc.
ad43617b34 is described below

commit ad43617b34e52323e45ae3480f6b141f357c9b7f
Author: Gavin Smith <gavinsmith0123@gmail.com>
AuthorDate: Sat Nov 18 21:35:45 2023 +0000

    * tp/Texinfo/XS/xspara.c (xspara__add_next): Check return value
    for u8_mbtouc as documented, by checking for 0xfffd (replacement
    character), instead of using the same as mbrtowc.
---
 ChangeLog              |  6 ++++++
 tp/Texinfo/XS/xspara.c | 27 +++++++++------------------
 2 files changed, 15 insertions(+), 18 deletions(-)

diff --git a/ChangeLog b/ChangeLog
index 98d9c335a1..44b1007e61 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,9 @@
+2023-11-18  Gavin Smith <gavinsmith0123@gmail.com>
+
+       * tp/Texinfo/XS/xspara.c (xspara__add_next): Check return value
+       for u8_mbtouc as documented, by checking for 0xfffd (replacement
+       character), instead of using the same as mbrtowc. 
+
 2023-11-18  Gavin Smith <gavinsmith0123@gmail.com>
 
        * tp/Texinfo/XS/parsetexi/context_stack.c (pop_context):
diff --git a/tp/Texinfo/XS/xspara.c b/tp/Texinfo/XS/xspara.c
index e458aa4eae..ed0bea930d 100644
--- a/tp/Texinfo/XS/xspara.c
+++ b/tp/Texinfo/XS/xspara.c
@@ -548,24 +548,17 @@ xspara__add_next (TEXT *result, char *word, int word_len,
                 }
 
               char_len = u8_mbtouc (&w, (uint8_t *) p, left);
-              if (char_len == (size_t) -2) {
-                /* unfinished multibyte character */
-                char_len = left;
-              } else if (char_len == (size_t) -1) {
-                /* invalid character */
-                char_len = 1;
-              } else if (char_len == 0) {
-                /* not sure what this means but we must avoid an infinite loop.
-                   Possibly only happens with invalid strings */
-                char_len = 1;
-              }
+              if (w == 0xfffd) /* bug - invalid string */
+                {
+                  if (char_len <= 0)
+                    char_len = 1; /* avoid an infinte loop */
+                }
               left -= char_len;
+              p += char_len;
 
               columns = uc_width (w, "UTF-8");
               if (columns > 0)
                 len += columns;
-
-              p += char_len;
             }
           state.word_counter += len;
        }
@@ -736,12 +729,10 @@ xspara_add_text (char *text, int len)
               char32_t wc;
               next_len = u8_mbtouc (&wc, (uint8_t *) q, len);
 
-              if ((long) next_len == 0)
-                break; /* Null character. Shouldn't happen. */
-              else if ((long) next_len < 0)
+              if (wc == 0xfffd) /* bug - invalid string */
                 {
-                  q++; len--; /* Invalid.  Just try to keep going. */
-                  continue;
+                  if (next_len <= 0)
+                    next_len = 1; /* avoid an infinte loop */
                 }
 
              /* Note: width == 0 includes accent characters. */