diff options
author | Eric Wong <e@80x24.org> | 2014-09-15 02:34:33 +0000 |
---|---|---|
committer | Eric Wong <e@80x24.org> | 2014-09-15 02:44:10 +0000 |
commit | 9713697ffa1e811e378f4233baf4fe2fb133214a (patch) | |
tree | 8034b0438f561e5e2c530ae138c643c80cd1e713 /lib/PublicInbox/Hval.pm | |
parent | b16868d658478c3fa07c44f802d37b266e5ee822 (diff) | |
download | public-inbox-9713697ffa1e811e378f4233baf4fe2fb133214a.tar.gz |
Some Message-IDs are crazy long, so support SHA-1s for them instead. This allows shorter URLs to be generated and are less likely However, we'll still favor short Message-IDs whenever possible.
Diffstat (limited to 'lib/PublicInbox/Hval.pm')
-rw-r--r-- | lib/PublicInbox/Hval.pm | 22 |
1 files changed, 15 insertions, 7 deletions
diff --git a/lib/PublicInbox/Hval.pm b/lib/PublicInbox/Hval.pm index a93e8c24..9be163aa 100644 --- a/lib/PublicInbox/Hval.pm +++ b/lib/PublicInbox/Hval.pm @@ -5,28 +5,36 @@ package PublicInbox::Hval; use strict; use warnings; -use fields qw(raw); +use fields qw(raw href); use Encode qw(find_encoding); use URI::Escape qw(uri_escape_utf8); my $enc_ascii = find_encoding('us-ascii'); sub new { - my ($class, $raw) = @_; + my ($class, $raw, $href) = @_; my $self = fields::new($class); # we never care about leading/trailing whitespace $raw =~ s/\A\s*//; $raw =~ s/\s*\z//; $self->{raw} = $raw; + $self->{href} = defined $href ? $href : $raw; $self; } sub new_msgid { - my ($class, $raw) = @_; - $raw =~ s/\A<//; - $raw =~ s/>\z//; - $class->new($raw); + my ($class, $msgid) = @_; + $msgid =~ s/\A\s*<?//; + $msgid =~ s/>?\s*\z//; + + if (length($msgid) <= 40) { + $class->new($msgid); + } else { + require Digest::SHA; + my $hex = Digest::SHA::sha1_hex($msgid); + $class->new($msgid, $hex); + } } sub new_oneline { @@ -52,7 +60,7 @@ sub ascii_html { } sub as_html { ascii_html($_[0]->{raw}) } -sub as_href { ascii_html(uri_escape_utf8($_[0]->{raw})) } +sub as_href { ascii_html(uri_escape_utf8($_[0]->{href})) } sub raw { if (defined $_[1]) { |