Test that S_grok_bslash_N() copes if S_reg() restarts the sizing parse.

S_reg() can discover midway through parsing the pattern to determine its
size, that the pattern will actually need to be encoded as UTF-8. If
calculations so far have been done in terms of bytes, then the macro
REQUIRE_UTF8 is used to restart the parse, so that sizes can be calculated
correctly for UTF-8.

It is possible to trigger this restart when processing multi-character
charnames interpolated into the pattern using \N{}. Test that this is
handled correctly.

diff --git a/t/re/pat_advanced.t b/t/re/pat_advanced.t
index fc13c6f..faa8859 100644 (file)
--- a/t/re/pat_advanced.t
+++ b/t/re/pat_advanced.t
@@ -1082,6 +1082,14 @@ sub run_tests {
         eval "q(W) =~ /\\N{$name}/";
         ok ! $w, 'Verify that latin1 letter in name doesnt give warning';
 
+        # This tests the code path that restarts the parse when the recursive
+        # call to S_reg() from within S_grok_bslash_N() discovers that the
+        # pattern needs to be recalculated as UTF-8.  use eval to avoid
+        # needing literal Unicode in this source file:
+        my $r = eval "qr/\\N{\x{100}\x{100}}/";
+        isnt $r, undef, "Generated regex for multi-char UTF-8 charname"
+           or diag($@);
+        ok "\x{100}\x{100}" =~ $r, "which matches";
     }
 
     {