X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FEml.pm;h=2ccbb6597de406680376ad9a02a419ee53d1015b;hb=1dde99e0badebfaf509cee4f15bbfa284996c8fc;hp=1adaff04f7be871d1ab9343c4bb737436b603d09;hpb=d35a49e114579a4df7bcb23d013ef0101cbb3ec2;p=public-inbox.git

diff --git a/lib/PublicInbox/Eml.pm b/lib/PublicInbox/Eml.pm
index 1adaff04..2ccbb659 100644
--- a/lib/PublicInbox/Eml.pm
+++ b/lib/PublicInbox/Eml.pm
@@ -41,6 +41,7 @@ $PublicInbox::EmlContentFoo::STRICT_PARAMS = 0;
 our $MAXPARTS = 1000; # same as SpamAssassin
 our $MAXDEPTH = 20; # seems enough, Perl sucks, here
 our $MAXBOUNDLEN = 2048; # same as postfix
+our $header_size_limit = 102400; # same as postfix
 
 my %MIME_ENC = (qp => \&enc_qp, base64 => \&encode_base64);
 my %MIME_DEC = (qp => \&dec_qp, base64 => \&decode_base64);
@@ -68,19 +69,47 @@ sub re_memo ($) {
 					/ismx
 }
 
+sub hdr_truncate ($) {
+	my $len = length($_[0]);
+	substr($_[0], $header_size_limit, $len) = '';
+	my $end = rindex($_[0], "\n");
+	if ($end >= 0) {
+		++$end;
+		substr($_[0], $end, $len) = '';
+		warn "header of $len bytes truncated to $end bytes\n";
+	} else {
+		$_[0] = '';
+		warn <<EOF
+header of $len bytes without `\\n' within $header_size_limit ignored
+EOF
+	}
+}
+
 # compatible with our uses of Email::MIME
 sub new {
 	my $ref = ref($_[1]) ? $_[1] : \(my $cpy = $_[1]);
-	if ($$ref =~ /(?:\r?\n(\r?\n))/gs) { # likely
-		# This can modify $$ref in-place and to avoid memcpy/memmove
-		# on a potentially large $$ref.  It does need to make a
-		# copy for $hdr, though.  Idea stolen from Email::Simple
-		my $hdr = substr($$ref, 0, pos($$ref), ''); # sv_chop on $$ref
+	# substr() can modify the first arg in-place and to avoid
+	# memcpy/memmove on a potentially large scalar.  It does need
+	# to make a copy for $hdr, though.  Idea stolen from Email::Simple.
+
+	# We also prefer index() on common LFLF emails since it's faster
+	# and re scan can bump RSS by length($$ref) on big strings
+	if (index($$ref, "\r\n") < 0 && (my $pos = index($$ref, "\n\n")) >= 0) {
+		# likely on *nix
+		my $hdr = substr($$ref, 0, $pos + 2, ''); # sv_chop on $$ref
+		chop($hdr); # lower SvCUR
+		hdr_truncate($hdr) if length($hdr) > $header_size_limit;
+		bless { hdr => \$hdr, crlf => "\n", bdy => $ref }, __PACKAGE__;
+	} elsif ($$ref =~ /\r?\n(\r?\n)/s) {
+		my $hdr = substr($$ref, 0, $+[0], ''); # sv_chop on $$ref
 		substr($hdr, -(length($1))) = ''; # lower SvCUR
+		hdr_truncate($hdr) if length($hdr) > $header_size_limit;
 		bless { hdr => \$hdr, crlf => $1, bdy => $ref }, __PACKAGE__;
 	} elsif ($$ref =~ /^[a-z0-9-]+[ \t]*:/ims && $$ref =~ /(\r?\n)\z/s) {
 		# body is optional :P
-		bless { hdr => \($$ref), crlf => $1 }, __PACKAGE__;
+		my $hdr = substr($$ref, 0, $header_size_limit + 1);
+		hdr_truncate($hdr) if length($hdr) > $header_size_limit;
+		bless { hdr => \$hdr, crlf => $1 }, __PACKAGE__;
 	} else { # nothing useful
 		my $hdr = $$ref = '';
 		bless { hdr => \$hdr, crlf => "\n" }, __PACKAGE__;
@@ -90,8 +119,8 @@ sub new {
 sub new_sub {
 	my (undef, $ref) = @_;
 	# special case for messages like <85k5su9k59.fsf_-_@lola.goethe.zz>
-	$$ref =~ /\A(?:(\r?\n))/gs or goto &new;
-	my $hdr = substr($$ref, 0, pos($$ref), ''); # sv_chop on $$ref
+	$$ref =~ /\A(\r?\n)/s or goto &new;
+	my $hdr = substr($$ref, 0, $+[0], ''); # sv_chop on $$ref
 	bless { hdr => \$hdr, crlf => $1, bdy => $ref }, __PACKAGE__;
 }
 
@@ -131,8 +160,12 @@ sub mp_descend ($$) {
 	# Cut at the the first epilogue, not subsequent ones.
 	# *sigh* just the regexp match alone seems to bump RSS by
 	# length($$bdy) on a ~30M string:
-	$$bdy =~ /((?:\r?\n)?^--$bnd--[ \t]*\r?$)/gsm and
-		substr($$bdy, pos($$bdy) - length($1)) = '';
+	my $epilogue_missing;
+	if ($$bdy =~ /(?:\r?\n)?^--$bnd--[ \t]*\r?$/sm) {
+		substr($$bdy, $-[0]) = '';
+	} else {
+		$epilogue_missing = 1;
+	}
 
 	# *Sigh* split() doesn't work in-place and return CoW strings
 	# because Perl wants to "\0"-terminate strings.  So split()
@@ -150,6 +183,10 @@ sub mp_descend ($$) {
 
 	if (@parts) { # the usual path if we got this far:
 		undef $bdy; # release memory ASAP if $nr > 0
+
+		# compatibility with Email::MIME
+		$parts[-1] =~ s/\n\r?\n\z/\n/s if $epilogue_missing;
+
 		@parts = grep /[^ \t\r\n]/s, @parts; # ignore empty parts
 
 		# Keep "From: someone..." from preamble in old,