simplify substitute_path_rule_matches using filename_ncmp

At the time this function was written, there was no filename_ncmp,
only FILENAME_CMP. So, in order to do an n-cmp, we had to make a local
copy of the first n characters of our string and use that to perform
the comparison. This patch simplifies the function's implementation,
now that we have filename_ncmp.

gdb/ChangeLog:

        * source.c (substitute_path_rule_matches): Simplify using
        filename_ncmp instead of FILENAME_CMP.

Tested on x86_64-linux.

1 files changed, 2 insertions, 9 deletions

diff --git a/gdb/source.c b/gdb/source.c
index cb05939..c985a1b 100644
--- a/gdb/source.c
+++ b/gdb/source.c
@@ -932,21 +932,14 @@ substitute_path_rule_matches (const struct substitute_path_rule *rule,
 {
   const int from_len = strlen (rule->from);
   const int path_len = strlen (path);
-  char *path_start;
 
   if (path_len < from_len)
     return 0;
 
   /* The substitution rules are anchored at the start of the path,
-     so the path should start with rule->from.  There is no filename
-     comparison routine, so we need to extract the first FROM_LEN
-     characters from PATH first and use that to do the comparison.  */
+     so the path should start with rule->from.  */
 
-  path_start = alloca (from_len + 1);
-  strncpy (path_start, path, from_len);
-  path_start[from_len] = '\0';
-
-  if (FILENAME_CMP (path_start, rule->from) != 0)
+  if (filename_ncmp (path, rule->from, from_len) != 0)
     return 0;
 
   /* Make sure that the region in the path that matches the substitution