From cb75a7aeb9fd530b1816b760c8c7f6cfb428c8e3 Mon Sep 17 00:00:00 2001
From: Eric Wong <e@80x24.org>
Date: Tue, 1 Dec 2020 01:35:46 +0000
Subject: nntpd: move {newsgroup} name check to config

With 50K newsgroups in the config file, this doesn't slow down
`PublicInbox::Config->new->fill_all' any measurable amount on
my busy old workstation.

This should prevent invalid newsgroup names from getting into
into extindex and catch user errors sooner, rather than later.

v2:
  - delete {newsgroup} if invalid to avoid ->nntp_url link
  - simplify -imapd and explain remaining check
---
 lib/PublicInbox/Config.pm | 21 +++++++++++++++++----
 lib/PublicInbox/IMAPD.pm  | 13 +++++++------
 lib/PublicInbox/NNTPD.pm  | 16 ++--------------
 3 files changed, 26 insertions(+), 24 deletions(-)

(limited to 'lib')

diff --git a/lib/PublicInbox/Config.pm b/lib/PublicInbox/Config.pm
index e7aea99b..9b9d5c19 100644
--- a/lib/PublicInbox/Config.pm
+++ b/lib/PublicInbox/Config.pm
@@ -437,10 +437,23 @@ EOF
 			$self->{-by_list_id}->{lc($list_id)} = $ibx;
 		}
 	}
-	if (my $ng = $ibx->{newsgroup}) {
-		# PublicInbox::NNTPD does stricter (and more expensive checks),
-		# keep this lean for startup speed
-		$self->{-by_newsgroup}->{$ng} = $ibx unless ref($ng);
+	if (my $ngname = $ibx->{newsgroup}) {
+		if (ref($ngname)) {
+			delete $ibx->{newsgroup};
+			warn 'multiple newsgroups not supported: '.
+				join(', ', @$ngname). "\n";
+		# Newsgroup name needs to be compatible with RFC 3977
+		# wildmat-exact and RFC 3501 (IMAP) ATOM-CHAR.
+		# Leave out a few chars likely to cause problems or conflicts:
+		# '|', '<', '>', ';', '#', '$', '&',
+		} elsif ($ngname =~ m![^A-Za-z0-9/_\.\-\~\@\+\=:]!) {
+			delete $ibx->{newsgroup};
+			warn "newsgroup name invalid: `$ngname'\n";
+		} else {
+			# PublicInbox::NNTPD does stricter ->nntp_usable
+			# checks, keep this lean for startup speed
+			$self->{-by_newsgroup}->{$ngname} = $ibx;
+		}
 	}
 	$self->{-by_name}->{$name} = $ibx;
 	if ($ibx->{obfuscate}) {
diff --git a/lib/PublicInbox/IMAPD.pm b/lib/PublicInbox/IMAPD.pm
index 366b6922..4a37734e 100644
--- a/lib/PublicInbox/IMAPD.pm
+++ b/lib/PublicInbox/IMAPD.pm
@@ -27,12 +27,13 @@ sub new {
 sub imapd_refresh_ibx { # pi_config->each_inbox cb
 	my ($ibx, $imapd) = @_;
 	my $ngname = $ibx->{newsgroup} or return;
-	if (ref $ngname) {
-		warn 'multiple newsgroups not supported: '.
-			join(', ', @$ngname). "\n";
-		return;
-	} elsif ($ngname =~ m![^a-z0-9/_\.\-\~\@\+\=:]! ||
-		 $ngname =~ /\.[0-9]+\z/) {
+
+	# We require lower-case since IMAP mailbox names are
+	# case-insensitive (but -nntpd matches INN in being
+	# case-sensitive
+	if ($ngname =~ m![^a-z0-9/_\.\-\~\@\+\=:]! ||
+			# don't confuse with 50K slices
+			$ngname =~ /\.[0-9]+\z/) {
 		warn "mailbox name invalid: newsgroup=`$ngname'\n";
 		return;
 	}
diff --git a/lib/PublicInbox/NNTPD.pm b/lib/PublicInbox/NNTPD.pm
index 5e287857..967850e9 100644
--- a/lib/PublicInbox/NNTPD.pm
+++ b/lib/PublicInbox/NNTPD.pm
@@ -38,20 +38,8 @@ sub refresh_groups {
 	my $groups = $pi_config->{-by_newsgroup}; # filled during each_inbox
 	$pi_config->each_inbox(sub {
 		my ($ibx) = @_;
-		my $ngname = $ibx->{newsgroup} or return;
-		if (ref $ngname) {
-			warn 'multiple newsgroups not supported: '.
-				join(', ', @$ngname). "\n";
-		# Newsgroup name needs to be compatible with RFC 3977
-		# wildmat-exact and RFC 3501 (IMAP) ATOM-CHAR.
-		# Leave out a few chars likely to cause problems or conflicts:
-		# '|', '<', '>', ';', '#', '$', '&',
-		} elsif ($ngname =~ m![^A-Za-z0-9/_\.\-\~\@\+\=:]!) {
-			warn "newsgroup name invalid: `$ngname'\n";
-			delete $groups->{$ngname};
-		} elsif ($ibx->nntp_usable) {
-			# Only valid if msgmap and search works
-
+		my $ngname = $ibx->{newsgroup} // return;
+		if ($ibx->nntp_usable) { # only valid if msgmap and over works
 			# preload to avoid fragmentation:
 			$ibx->description;
 			$ibx->base_url;
-- 
cgit v1.2.3-24-ge0c7