treewide: consolidate "From " line removal

Aside from our prior import bugs (fixed in a0c07cba0e5d8b6a
(mda: drop leading "From " lines again, 2016-06-26)), we'll
always have to be dealing with mutt piping messages to us and
`git format-patch' output.  So just share the regexp so we
can use it everywhere.

In may be desirable to allow importing messages with a leading
"From " line for FUSE, even.

Additionally, some instances of this regexp needlessly added
optional `\r?' (CR) checks ahead of the `\n' (LF) element; but
they're pointless anyways since [^\n]* is enough to exclude all
non-LF bytes.

9 files changed, 22 insertions, 25 deletions

diff --git a/lib/PublicInbox/Eml.pm b/lib/PublicInbox/Eml.pm
index 8b999e1a..24060ec8 100644
--- a/lib/PublicInbox/Eml.pm
+++ b/lib/PublicInbox/Eml.pm
@@ -528,4 +528,10 @@ sub willneed { re_memo($_) for @_ }
 willneed(qw(From To Cc Date Subject Content-Type In-Reply-To References
                 Message-ID X-Alt-Message-ID));
 
+# This fixes an old bug from import (pre-a0c07cba0e5d8b6a)
+# mutt also pipes single RFC822 messages with a "From " line,
+# but no Content-Length or "From " escaping.
+# "git format-patch" also generates such files by default.
+sub strip_from { $_[0] =~ s/\A[\r\n]*From [^\n]*\n//s }
+
 1;
diff --git a/lib/PublicInbox/IMAP.pm b/lib/PublicInbox/IMAP.pm
index 3c64cefa..e4a9e304 100644
--- a/lib/PublicInbox/IMAP.pm
+++ b/lib/PublicInbox/IMAP.pm
@@ -664,7 +664,7 @@ sub op_eml_new { $_[4] = PublicInbox::Eml->new($_[3]) }
 # s/From / fixes old bug from import (pre-a0c07cba0e5d8b6a)
 sub to_crlf_full {
         ${$_[0]} =~ s/(?<!\r)\n/\r\n/sg;
-        ${$_[0]} =~ s/\A[\r\n]*From [^\r\n]*\r\n//s;
+        PublicInbox::Eml::strip_from(${$_[0]});
 }
 
 sub op_crlf_bref { to_crlf_full($_[3]) }
diff --git a/lib/PublicInbox/Import.pm b/lib/PublicInbox/Import.pm
index 7175884c..cd03da05 100644
--- a/lib/PublicInbox/Import.pm
+++ b/lib/PublicInbox/Import.pm
@@ -118,9 +118,6 @@ sub _cat_blob ($$$) {
         $n == $len or croak "cat-blob: short read: $n < $len";
         my $lf = chop $buf;
         croak "bad read on final byte: <$lf>" if $lf ne "\n";
-
-        # fixup some bugginess in old versions:
-        $buf =~ s/\A[\r\n]*From [^\r\n]*\r?\n//s;
         \$buf;
 }
 
@@ -136,8 +133,9 @@ sub check_remove_v1 {
         my $info = _check_path($r, $w, $tip, $path) or return ('MISSING',undef);
         $info =~ m!\A100644 blob ([a-f0-9]{40,})\t!s or die "not blob: $info";
         my $oid = $1;
-        my $msg = _cat_blob($r, $w, $oid) or die "BUG: cat-blob $1 failed";
-        my $cur = PublicInbox::Eml->new($msg);
+        my $bref = _cat_blob($r, $w, $oid) or die "BUG: cat-blob $1 failed";
+        PublicInbox::Eml::strip_from($$bref);
+        my $cur = PublicInbox::Eml->new($bref);
         my $cur_s = $cur->header('Subject') // '';
         my $cur_m = $mime->header('Subject') // '';
         if ($cur_s ne $cur_m || norm_body($cur) ne norm_body($mime)) {
diff --git a/lib/PublicInbox/LeiInput.pm b/lib/PublicInbox/LeiInput.pm
index 93f8b6b8..28b73ca9 100644
--- a/lib/PublicInbox/LeiInput.pm
+++ b/lib/PublicInbox/LeiInput.pm
@@ -84,10 +84,7 @@ sub input_fh {
                         return $self->{lei}->child_error(0, <<"");
 error reading $name: $!
 
-                # mutt pipes single RFC822 messages with a "From " line,
-                # but no Content-Length or "From " escaping.
-                # "git format-patch" also generates such files by default.
-                $buf =~ s/\A[\r\n]*From [^\r\n]*\r?\n//s;
+                PublicInbox::Eml::strip_from($buf);
 
                 # a user may feed just a body: git diff | lei rediff -U9
                 if ($self->{-force_eml}) {
diff --git a/lib/PublicInbox/LeiInspect.pm b/lib/PublicInbox/LeiInspect.pm
index f801610f..65c64cf2 100644
--- a/lib/PublicInbox/LeiInspect.pm
+++ b/lib/PublicInbox/LeiInspect.pm
@@ -254,7 +254,7 @@ sub inspect_start ($$) {
 sub do_inspect { # lei->do_env cb
         my ($lei) = @_;
         my $str = delete $lei->{istr};
-        $str =~ s/\A[\r\n]*From [^\r\n]*\r?\n//s;
+        PublicInbox::Eml::strip_from($str);
         my $eml = PublicInbox::Eml->new(\$str);
         inspect_start($lei, [ 'blob:'.$lei->git_oid($eml)->hexdigest,
                         map { "mid:$_" } @{mids($eml)} ]);
diff --git a/lib/PublicInbox/LeiToMail.pm b/lib/PublicInbox/LeiToMail.pm
index 8771592d..ead60b38 100644
--- a/lib/PublicInbox/LeiToMail.pm
+++ b/lib/PublicInbox/LeiToMail.pm
@@ -53,8 +53,7 @@ sub _mbox_hdr_buf ($$$) {
         }
         my $buf = delete $eml->{hdr};
 
-        # fixup old bug from import (pre-a0c07cba0e5d8b6a)
-        $$buf =~ s/\A[\r\n]*From [^\r\n]*\r?\n//s;
+        PublicInbox::Eml::strip_from($$buf);
         my $ident = $smsg->{blob} // 'lei';
         if (defined(my $pct = $smsg->{pct})) { $ident .= "=$pct" }
 
diff --git a/lib/PublicInbox/Mbox.pm b/lib/PublicInbox/Mbox.pm
index bf61bb0e..52f88ae3 100644
--- a/lib/PublicInbox/Mbox.pm
+++ b/lib/PublicInbox/Mbox.pm
@@ -89,17 +89,15 @@ sub emit_raw {
 
 sub msg_hdr ($$) {
         my ($ctx, $eml) = @_;
-        my $header_obj = $eml->header_obj;
 
-        # drop potentially confusing headers, ssoma already should've dropped
-        # Lines and Content-Length
-        foreach my $d (qw(Lines Bytes Content-Length Status)) {
-                $header_obj->header_set($d);
+        # drop potentially confusing headers, various importers should've
+        # already dropped these, but we can't trust stuff we've cloned
+        for my $d (qw(Lines Bytes Content-Length Status)) {
+                $eml->header_set($d);
         }
-        my $crlf = $header_obj->crlf;
-        my $buf = $header_obj->as_string;
-        # fixup old bug from import (pre-a0c07cba0e5d8b6a)
-        $buf =~ s/\A[\r\n]*From [^\r\n]*\r?\n//s;
+        my $crlf = $eml->crlf;
+        my $buf = $eml->header_obj->as_string;
+        PublicInbox::Eml::strip_from($buf);
         "From mboxrd\@z Thu Jan  1 00:00:00 1970" . $crlf . $buf . $crlf;
 }
 
diff --git a/lib/PublicInbox/MboxReader.pm b/lib/PublicInbox/MboxReader.pm
index e4209022..d67fb4eb 100644
--- a/lib/PublicInbox/MboxReader.pm
+++ b/lib/PublicInbox/MboxReader.pm
@@ -93,7 +93,7 @@ sub _mbox_cl ($$$;@) {
                         undef $mbfh;
                 }
                 while (my $hdr = _extract_hdr(\$buf)) {
-                        $$hdr =~ s/\A[\r\n]*From [^\n]*\n//s or
+                        PublicInbox::Eml::strip_from($$hdr) or
                                 die "E: no 'From ' line in:\n", Dumper($hdr);
                         my $eml = PublicInbox::Eml->new($hdr);
                         next unless $eml->raw_size;
diff --git a/lib/PublicInbox/NNTP.pm b/lib/PublicInbox/NNTP.pm
index 316b7775..603cf094 100644
--- a/lib/PublicInbox/NNTP.pm
+++ b/lib/PublicInbox/NNTP.pm
@@ -523,8 +523,7 @@ sub msg_hdr_write ($$) {
         set_nntp_headers($eml, $smsg);
 
         my $hdr = $eml->{hdr} // \(my $x = '');
-        # fixup old bug from import (pre-a0c07cba0e5d8b6a)
-        $$hdr =~ s/\A[\r\n]*From [^\r\n]*\r?\n//s;
+        PublicInbox::Eml::strip_from($$hdr);
         $$hdr =~ s/(?<!\r)\n/\r\n/sg; # Alpine barfs without this
 
         # for leafnode compatibility, we need to ensure Message-ID headers