summaryrefslogtreecommitdiff
path: root/lib/utf8.pm
diff options
context:
space:
mode:
authorJarkko Hietaniemi <jhi@iki.fi>2002-04-16 12:34:36 +0000
committerJarkko Hietaniemi <jhi@iki.fi>2002-04-16 12:34:36 +0000
commitc20e2abd87bda89a16a4bc4b0eaecf500fbede5c (patch)
tree2178be0864ad475b7f483634a5ec7614792d6fba /lib/utf8.pm
parentddbf912790641017a0ae28c039dec8489579cc0c (diff)
downloadperl-c20e2abd87bda89a16a4bc4b0eaecf500fbede5c.tar.gz
Let's not promise too much: use utf8 only works on identifier
names, not package or subroutine names (admittedly limited [1], but that's what the Camel says, and that's what we are going to stick to for 5.8.0). Also document that use vars does not do utf8. [1] The obvious problem in both is that package and subroutine names need to mappable to the filesystem. p4raw-id: //depot/perl@15947
Diffstat (limited to 'lib/utf8.pm')
-rw-r--r--lib/utf8.pm6
1 files changed, 3 insertions, 3 deletions
diff --git a/lib/utf8.pm b/lib/utf8.pm
index c748a4993f..e0c4ac1966 100644
--- a/lib/utf8.pm
+++ b/lib/utf8.pm
@@ -56,9 +56,9 @@ Enabling the C<utf8> pragma has the following effect:
Bytes in the source text that have their high-bit set will be treated
as being part of a literal UTF-8 character. This includes most
-literals such as identifiers, string constants, constant regular
-expression patterns and package names. On EBCDIC platforms characters
-in the Latin 1 character set are treated as being part of a literal
+literals such as identifier names, string constants, and constant
+regular expression patterns. On EBCDIC platforms characters in
+the Latin 1 character set are treated as being part of a literal
UTF-EBCDIC character.
=back