[PATCH 0/5] view*: minor cleanups and fixes

[PATCH 0/5] view*: minor cleanups and fixes

[Eric Wong]

Just a few odds and ends which bother me when reading
the code or render HTML result.

Eric Wong (5):
  view: note we assume UTF-8 on unknown encodings
  view: use defined-or operator to simplify checks
  view: dedupe_subject: allow "0" is a valid Subject
  viewdiff: reduce sub parameter count
  view: inline flush_quote sub

 lib/PublicInbox/View.pm     | 37 +++++++++++++++----------------------
 lib/PublicInbox/ViewDiff.pm | 21 ++++++++++++---------
 2 files changed, 27 insertions(+), 31 deletions(-)

[PATCH 1/5] view: note we assume UTF-8 on unknown encodings

[Eric Wong]

Clarify that we're assuming the text is UTF-8, since users
may have no idea how it's mangled.
---
 lib/PublicInbox/View.pm | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/lib/PublicInbox/View.pm b/lib/PublicInbox/View.pm
index 70c10604..d897aeec 100644
--- a/lib/PublicInbox/View.pm
+++ b/lib/PublicInbox/View.pm
@@ -518,7 +518,9 @@ sub attach_link ($$$$;$) {
 	my $rv = $ctx->{obuf};
 	$$rv .= qq($nl<a\nhref="$ctx->{mhref}$idx-$sfn">);
 	if ($err) {
-		$$rv .= "[-- Warning: decoded text below may be mangled --]\n";
+		$$rv .= <<EOF;
+[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
+EOF
 	}
 	$$rv .= "[-- Attachment #$idx: ";
 	my $ts = "Type: $ct, Size: $size bytes";

[PATCH 2/5] view: use defined-or operator to simplify checks

[Eric Wong]

We depend on Perl 5.10 features in other places.  Shorten the
lifetime of the `$desc' scalar while we're at it.
---
 lib/PublicInbox/View.pm | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/lib/PublicInbox/View.pm b/lib/PublicInbox/View.pm
index d897aeec..9ef1f68a 100644
--- a/lib/PublicInbox/View.pm
+++ b/lib/PublicInbox/View.pm
@@ -504,9 +504,6 @@ sub attach_link ($$$$;$) {
 	# spotting MUA problems:
 	$ct =~ s/;.*// unless $err;
 	$ct = ascii_html($ct);
-	my $desc = $part->header('Content-Description');
-	$desc = $fn unless defined $desc;
-	$desc = '' unless defined $desc;
 	my $sfn;
 	if (defined $fn && $fn =~ /\A$PublicInbox::Hval::FN\z/o) {
 		$sfn = $fn;
@@ -524,6 +521,7 @@ EOF
 	}
 	$$rv .= "[-- Attachment #$idx: ";
 	my $ts = "Type: $ct, Size: $size bytes";
+	my $desc = $part->header('Content-Description') // $fn // '';
 	$desc = ascii_html($desc);
 	$$rv .= ($desc eq '') ? "$ts --]" : "$desc --]\n[-- $ts --]";
 	$$rv .= "</a>\n";

[PATCH 3/5] view: dedupe_subject: allow "0" is a valid Subject

[Eric Wong]

While rare in practice (even by spammers), A single "0" could
theoretically be the entire contents of a Subject line.  So
use the Perl 5.10+ defined-or operator to improve correctness
of subject deduplication.
---
 lib/PublicInbox/View.pm | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/lib/PublicInbox/View.pm b/lib/PublicInbox/View.pm
index 9ef1f68a..f0584cb7 100644
--- a/lib/PublicInbox/View.pm
+++ b/lib/PublicInbox/View.pm
@@ -888,20 +888,20 @@ sub missing_thread {
 sub dedupe_subject {
 	my ($prev_subj, $subj, $val) = @_;
 
-	my $omit = ''; # '"' denotes identical text omitted
+	my $omit; # '"' denotes identical text omitted
 	my (@prev_pop, @curr_pop);
 	while (@$prev_subj && @$subj && $subj->[-1] eq $prev_subj->[-1]) {
 		push(@prev_pop, pop(@$prev_subj));
 		push(@curr_pop, pop(@$subj));
-		$omit ||= $val;
+		$omit //= $val;
 	}
 	pop @$subj if @$subj && $subj->[-1] =~ /^re:\s*/i;
 	if (scalar(@curr_pop) == 1) {
-		$omit = '';
+		$omit = undef;
 		push @$prev_subj, @prev_pop;
 		push @$subj, @curr_pop;
 	}
-	$omit;
+	$omit // '';
 }
 
 sub skel_dump { # walk_thread callback

Re: [PATCH 3/5] view: dedupe_subject: allow "0" is a valid Subject

[Eric Wong]

$subject =~ s/is/as/

That should be:

Subject: [PATCH 3/5] view: dedupe_subject: allow "0" as a valid Subject

[PATCH 4/5] viewdiff: reduce sub parameter count

[Eric Wong]

We're slowly moving towards doing all of our output buffering
into a single buffer, so passing that around on the stack as
a dedicated parameter is confusing.
---
 lib/PublicInbox/View.pm     |  2 +-
 lib/PublicInbox/ViewDiff.pm | 21 ++++++++++++---------
 2 files changed, 13 insertions(+), 10 deletions(-)

diff --git a/lib/PublicInbox/View.pm b/lib/PublicInbox/View.pm
index f0584cb7..12ef6431 100644
--- a/lib/PublicInbox/View.pm
+++ b/lib/PublicInbox/View.pm
@@ -589,7 +589,7 @@ sub add_text_body { # callback for msg_iter
 		if ($cur =~ /\A>/) {
 			flush_quote($rv, $l, \$cur);
 		} elsif ($diff) {
-			flush_diff($rv, $ctx, \$cur);
+			flush_diff($ctx, \$cur);
 		} else {
 			# regular lines, OK
 			$$rv .= $l->to_html($cur);
diff --git a/lib/PublicInbox/ViewDiff.pm b/lib/PublicInbox/ViewDiff.pm
index 5d391a13..f7422712 100644
--- a/lib/PublicInbox/ViewDiff.pm
+++ b/lib/PublicInbox/ViewDiff.pm
@@ -120,8 +120,8 @@ sub anchor1 ($$) {
 	$ok ? "<a\nhref=#i$attr\nid=$attr>diff</a> --git" : undef
 }
 
-sub diff_header ($$$$) {
-	my ($dst, $x, $ctx, $top) = @_;
+sub diff_header ($$$) {
+	my ($x, $ctx, $top) = @_;
 	my (undef, undef, $pa, $pb) = splice(@$top, 0, 4); # ignore oid_{a,b}
 	my $spfx = $ctx->{-spfx};
 	my $dctx = { spfx => $spfx };
@@ -159,15 +159,17 @@ sub diff_header ($$$$) {
 		warn "BUG? <$$x> had no ^index line";
 	}
 	$$x =~ s!^diff --git!anchor1($ctx, $pb) // 'diff --git'!ems;
+	my $dst = $ctx->{obuf};
 	$$dst .= qq(<span\nclass="head">);
 	$$dst .= $$x;
 	$$dst .= '</span>';
 	$dctx;
 }
 
-sub diff_before_or_after ($$$) {
-	my ($dst, $ctx, $x) = @_;
+sub diff_before_or_after ($$) {
+	my ($ctx, $x) = @_;
 	my $linkify = $ctx->{-linkify};
+	my $dst = $ctx->{obuf};
 	for my $y (split(/(^---\n)/sm, $$x)) {
 		if ($y =~ /\A---\n\z/s) {
 			$$dst .= "---\n"; # all HTML is "\r\n" => "\n"
@@ -186,20 +188,21 @@ sub diff_before_or_after ($$$) {
 }
 
 # callers must do CRLF => LF conversion before calling this
-sub flush_diff ($$$) {
-	my ($dst, $ctx, $cur) = @_;
+sub flush_diff ($$) {
+	my ($ctx, $cur) = @_;
 
 	my @top = split($EXTRACT_DIFFS, $$cur);
 	$$cur = undef;
 
 	my $linkify = $ctx->{-linkify};
+	my $dst = $ctx->{obuf};
 	my $dctx; # {}, keys: Q, oid_a, oid_b
 
 	while (defined(my $x = shift @top)) {
 		if (scalar(@top) >= 4 &&
 				$top[1] =~ $IS_OID &&
 				$top[0] =~ $IS_OID) {
-			$dctx = diff_header($dst, \$x, $ctx, \@top);
+			$dctx = diff_header(\$x, $ctx, \@top);
 		} elsif ($dctx) {
 			my $after = '';
 
@@ -238,9 +241,9 @@ sub flush_diff ($$$) {
 					$$dst .= $linkify->to_html($s);
 				}
 			}
-			diff_before_or_after($dst, $ctx, \$after) unless $dctx;
+			diff_before_or_after($ctx, \$after) unless $dctx;
 		} else {
-			diff_before_or_after($dst, $ctx, \$x);
+			diff_before_or_after($ctx, \$x);
 		}
 	}
 }

[PATCH 5/5] view: inline flush_quote sub

[Eric Wong]

No point in having an extra sub for a short, commonly
called function in the same file.
---
 lib/PublicInbox/View.pm | 19 ++++++-------------
 1 file changed, 6 insertions(+), 13 deletions(-)

diff --git a/lib/PublicInbox/View.pm b/lib/PublicInbox/View.pm
index 12ef6431..1e53d8dc 100644
--- a/lib/PublicInbox/View.pm
+++ b/lib/PublicInbox/View.pm
@@ -482,17 +482,6 @@ sub multipart_text_as_html {
 	msg_iter($_[0], \&add_text_body, $_[1], 1);
 }
 
-sub flush_quote {
-	my ($s, $l, $quot) = @_;
-
-	my $rv = $l->to_html($$quot);
-
-	# we use a <span> here to allow users to specify their own
-	# color for quoted text
-	$$quot = undef;
-	$$s .= qq(<span\nclass="q">) . $rv . '</span>'
-}
-
 sub attach_link ($$$$;$) {
 	my ($ctx, $ct, $p, $fn, $err) = @_;
 	my ($part, $depth, @idx) = @$p;
@@ -587,14 +576,18 @@ sub add_text_body { # callback for msg_iter
 	my $l = $ctx->{-linkify} //= PublicInbox::Linkify->new;
 	foreach my $cur (@sections) {
 		if ($cur =~ /\A>/) {
-			flush_quote($rv, $l, \$cur);
+			# we use a <span> here to allow users to specify
+			# their own color for quoted text
+			$$rv .= qq(<span\nclass="q">);
+			$$rv .= $l->to_html($cur);
+			$$rv .= '</span>';
 		} elsif ($diff) {
 			flush_diff($ctx, \$cur);
 		} else {
 			# regular lines, OK
 			$$rv .= $l->to_html($cur);
-			$cur = undef;
 		}
+		$cur = undef;
 	}
 
 	obfuscate_addrs($ibx, $$rv) if $ibx->{obfuscate};