[PATCH 0/4] HTML micro-optimizations

[PATCH 0/4] HTML micro-optimizations

[Eric Wong]

Testing with a horriblely long 368-message thread on git@vger,
I've found some minor tweaks which give us some tiny speedups.
Unfortunately, these are minor and new features may end up
costing us more in the future.  Further optimizations are
needed.

Eric Wong (4):
      view: remove unused parameter
      view: stop chomping off whitespace at ends of messages
      view: do not modify array during iteration
      linkify: modify argument in place

 lib/PublicInbox/Linkify.pm | 17 +++++++----------
 lib/PublicInbox/View.pm    | 21 +++++++++------------
 2 files changed, 16 insertions(+), 22 deletions(-)

[PATCH 1/4] view: remove unused parameter

[Eric Wong]

And add a comment about it to remind our future selves.
---
 lib/PublicInbox/View.pm | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/lib/PublicInbox/View.pm b/lib/PublicInbox/View.pm
index b779665..cf40b55 100644
--- a/lib/PublicInbox/View.pm
+++ b/lib/PublicInbox/View.pm
@@ -441,7 +441,7 @@ sub attach_link ($$$$;$) {
 
 sub add_text_body {
 	my ($upfx, $p) = @_; # from msg_iter: [ Email::MIME, depth, @idx ]
-	my ($part, $depth, @idx) = @$p;
+	my ($part, $depth) = @$p; # attachment @idx is unused
 	my $ct = $part->content_type || 'text/plain';
 	my $fn = $part->filename;
 
-- 
EW

[PATCH 2/4] view: stop chomping off whitespace at ends of messages

[Eric Wong]

This allows a 3-4% speedup in $MESSAGE_ID/T/ page generation
speed for a 368+ message thread.  It also more faithfully
preserves the message as intended; even if the it makes the
sender look like a space-wasting slob :P
---
 lib/PublicInbox/View.pm | 12 +++++-------
 1 file changed, 5 insertions(+), 7 deletions(-)

diff --git a/lib/PublicInbox/View.pm b/lib/PublicInbox/View.pm
index cf40b55..97a8bcb 100644
--- a/lib/PublicInbox/View.pm
+++ b/lib/PublicInbox/View.pm
@@ -490,15 +490,13 @@ sub add_text_body {
 		}
 	}
 
-	my $end = "\n";
-	if (@quot) {
-		$end = '';
+	if (@quot) { # ugh, top posted
 		flush_quote(\$s, $l, \@quot);
+	} elsif ($s =~ /\n\z/s) { # common, last line ends with a newline
+		$s;
+	} else { # some editors don't do newlines...
+		$s .= "\n";
 	}
-	$s =~ s/[ \t]+$//sgm; # kill per-line trailing whitespace
-	$s =~ s/\A\n+//s; # kill leading blank lines
-	$s =~ s/\s+\z//s; # kill all trailing spaces
-	$s .= $end;
 }
 
 sub _msg_html_prepare {
-- 
EW

[PATCH 3/4] view: do not modify array during iteration

[Eric Wong]

This results in a half percent speedup or so doing
$MESSAGE_ID/T/ HTML generation for a 368 message thread.
---
 lib/PublicInbox/View.pm | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/lib/PublicInbox/View.pm b/lib/PublicInbox/View.pm
index 97a8bcb..39ca959 100644
--- a/lib/PublicInbox/View.pm
+++ b/lib/PublicInbox/View.pm
@@ -476,7 +476,7 @@ sub add_text_body {
 	}
 	my @quot;
 	my $l = PublicInbox::Linkify->new;
-	while (defined(my $cur = shift @lines)) {
+	foreach my $cur (@lines) {
 		if ($cur !~ /^>/) {
 			# show the previously buffered quote inline
 			flush_quote(\$s, $l, \@quot) if @quot;
-- 
EW

[PATCH 4/4] linkify: modify argument in place

[Eric Wong]

This results in over 1% speedup doing $MESSAGE_ID/T/ HTML
generation for a 368-message thread.
---
 lib/PublicInbox/Linkify.pm | 17 +++++++----------
 lib/PublicInbox/View.pm    |  5 ++---
 2 files changed, 9 insertions(+), 13 deletions(-)

diff --git a/lib/PublicInbox/Linkify.pm b/lib/PublicInbox/Linkify.pm
index acd2a47..8e1728c 100644
--- a/lib/PublicInbox/Linkify.pm
+++ b/lib/PublicInbox/Linkify.pm
@@ -22,11 +22,10 @@ my $LINK_RE = qr{(\()?\b((?:ftps?|https?|nntps?|gopher)://
 		 (?:\#[a-z0-9\-\._~!\$\&\';\(\)\*\+,;=:@/%\?]+)?
 		 )}xi;
 
-sub new { bless {}, shift }
+sub new { bless {}, $_[0] }
 
 sub linkify_1 {
-	my ($self, $s) = @_;
-	$s =~ s!$LINK_RE!
+	$_[1] =~ s!$LINK_RE!
 		my $beg = $1 || '';
 		my $url = $2;
 		my $end = '';
@@ -50,19 +49,17 @@ sub linkify_1 {
 
 		# only escape ampersands, others do not match LINK_RE
 		$url =~ s/&/&/g;
-		$self->{$key} = $url;
+		$_[0]->{$key} = $url;
 		$beg . 'PI-LINK-'. $key . $end;
 	!ge;
-	$s;
+	$_[1];
 }
 
 sub linkify_2 {
-	my ($self, $s) = @_;
-
 	# Added "PI-LINK-" prefix to avoid false-positives on git commits
-	$s =~ s!\bPI-LINK-([a-f0-9]{40})\b!
+	$_[1] =~ s!\bPI-LINK-([a-f0-9]{40})\b!
 		my $key = $1;
-		my $url = $self->{$key};
+		my $url = $_[0]->{$key};
 		if (defined $url) {
 			"<a\nhref=\"$url\">$url</a>";
 		} else {
@@ -70,7 +67,7 @@ sub linkify_2 {
 			$key;
 		}
 	!ge;
-	$s;
+	$_[1];
 }
 
 1;
diff --git a/lib/PublicInbox/View.pm b/lib/PublicInbox/View.pm
index 39ca959..e4e9d7d 100644
--- a/lib/PublicInbox/View.pm
+++ b/lib/PublicInbox/View.pm
@@ -482,9 +482,8 @@ sub add_text_body {
 			flush_quote(\$s, $l, \@quot) if @quot;
 
 			# regular line, OK
-			$cur = $l->linkify_1($cur);
-			$cur = ascii_html($cur);
-			$s .= $l->linkify_2($cur);
+			$l->linkify_1($cur);
+			$s .= $l->linkify_2(ascii_html($cur));
 		} else {
 			push @quot, $cur;
 		}
-- 
EW