From a88949f234bff68baefa380d1291ff12878c274c Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Rapha=C3=ABl=20Bournhonesque?= <raphael@bournhonesque.eu>
Date: Fri, 28 Jul 2023 12:01:47 +0200
Subject: [PATCH] fix: don't let bot index most facet pages

Crawling bots can't visit all page and crawl OFF continuously.
We want to limit crawlers on interesting pages, so we return a noindex
html page on most facet pages (except most interesting ones such as
brand, category,...)
---
 lib/ProductOpener/Routing.pm           |   6 ++
 tests/integration/facet_page_crawler.t | 122 +++++++++++++++++++++++++
 2 files changed, 128 insertions(+)
 create mode 100644 tests/integration/facet_page_crawler.t

diff --git a/lib/ProductOpener/Routing.pm b/lib/ProductOpener/Routing.pm
index 089dee67ae63d..81a01cc72906f 100644
--- a/lib/ProductOpener/Routing.pm
+++ b/lib/ProductOpener/Routing.pm
@@ -429,6 +429,12 @@ sub analyze_request ($request_ref) {
 
 			my $tagtype = $request_ref->{tagtype};
 
+			if (    ($tagtype !~ /^brands|categories|labels|ingredients|additives$/)
+				and ($request_ref->{is_crawl_bot} eq 1))
+			{
+				$request_ref->{no_index} = 1;
+			}
+
 			if (($#components >= 0)) {
 				$request_ref->{tag} = shift @components;
 
diff --git a/tests/integration/facet_page_crawler.t b/tests/integration/facet_page_crawler.t
new file mode 100644
index 0000000000000..1fa89631d385c
--- /dev/null
+++ b/tests/integration/facet_page_crawler.t
@@ -0,0 +1,122 @@
+#!/usr/bin/perl -w
+
+use ProductOpener::PerlStandards;
+
+use Test::More;
+use ProductOpener::APITest qw/:all/;
+use ProductOpener::Test qw/:all/;
+use ProductOpener::TestDefaults qw/:all/;
+
+use File::Basename "dirname";
+
+use Storable qw(dclone);
+
+remove_all_users();
+
+remove_all_products();
+
+wait_application_ready();
+
+my $ua = new_client();
+
+my $CRAWLING_BOT_USER_AGENT
+	= 'Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko; compatible; bingbot/2.0; +http://www.bing.com/bingbot.htm) Chrome/';
+my $NORMAL_USER_USER_AGENT
+	= 'Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:109.0) Gecko/20100101 Firefox/114.0Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:109.0) Gecko/20100101 Firefox/114.0';
+
+my %product_form = (
+	%{dclone(\%default_product_form)},
+	(
+		code => '0200000000235',
+		product_name => "Only-Product",
+	)
+);
+
+edit_product($ua, \%product_form);
+
+my $tests_ref = [
+	# Normal user should have access to product page
+	{
+		test_case => 'normal-user-access-product-page',
+		method => 'GET',
+		path => '/product/0200000000235/only-product',
+		headers_in => {'User-Agent' => $CRAWLING_BOT_USER_AGENT},
+		expected_status_code => 200,
+		expected_type => 'html',
+		response_content_must_match => '<title>Only-Product - 100 g</title>'
+	},
+	# Crawling bot should have access to product page
+	{
+		test_case => 'crawler-access-product-page',
+		method => 'GET',
+		path => '/product/0200000000235/only-product',
+		headers_in => {'User-Agent' => $NORMAL_USER_USER_AGENT},
+		expected_status_code => 200,
+		expected_type => 'html',
+		response_content_must_match => '<title>Only-Product - 100 g</title>'
+	},
+	# Crawling bot should receive a noindex page for nested facets
+	{
+		test_case => 'crawler-access-nested-facet-page',
+		method => 'GET',
+		path => '/category/hazelnut-spreads/brand/nutella',
+		headers_in => {'User-Agent' => $CRAWLING_BOT_USER_AGENT},
+		expected_status_code => 200,
+		expected_type => 'html',
+		response_content_must_match => '<h1>NOINDEX</h1>'
+	},
+	# Normal user should have access to nested facets
+	{
+		test_case => 'normal-user-access-nested-facet-page',
+		method => 'GET',
+		path => '/category/hazelnut-spreads/brand/nutella',
+		headers_in => {'User-Agent' => $NORMAL_USER_USER_AGENT},
+		expected_status_code => 200,
+		expected_type => 'html',
+		response_content_must_not_match => '<h1>NOINDEX</h1>'
+	},
+	# Crawling bot should have access to specific facet pages (such as category)
+	{
+		test_case => 'crawler-access-category-facet-page',
+		method => 'GET',
+		path => '/category/hazelnut-spreads',
+		headers_in => {'User-Agent' => $CRAWLING_BOT_USER_AGENT},
+		expected_status_code => 200,
+		expected_type => 'html',
+		response_content_must_not_match => '<h1>NOINDEX</h1>'
+	},
+	# Normal user should have access to facet pages
+	{
+		test_case => 'normal-user-access-category-facet-page',
+		method => 'GET',
+		path => '/category/hazelnut-spreads',
+		headers_in => {'User-Agent' => $NORMAL_USER_USER_AGENT},
+		expected_status_code => 200,
+		expected_type => 'html',
+		response_content_must_not_match => '<h1>NOINDEX</h1>'
+	},
+	# Crawling bot should receive a noindex page for most facet pages (including editor facet)
+	{
+		test_case => 'crawler-access-editor-facet-page',
+		method => 'GET',
+		path => '/editor/unknown-user',
+		headers_in => {'User-Agent' => $CRAWLING_BOT_USER_AGENT},
+		expected_status_code => 200,
+		expected_type => 'html',
+		response_content_must_match => '<h1>NOINDEX</h1>'
+	},
+	# Normal user should have access to editor facet
+	{
+		test_case => 'normal-user-access-editor-facet-page',
+		method => 'GET',
+		path => '/editor/unknown-user',
+		headers_in => {'User-Agent' => $NORMAL_USER_USER_AGENT},
+		expected_status_code => 200,
+		expected_type => 'html',
+		response_content_must_match => 'Unknown user.'
+	},
+];
+
+execute_api_tests(__FILE__, $tests_ref);
+
+done_testing();