relay_filter/
web_crawlers.rs

1//! Filters events coming from user agents known to be web crawlers.
2
3use once_cell::sync::Lazy;
4use regex::Regex;
5
6use crate::{FilterConfig, FilterStatKey, Filterable};
7
8static WEB_CRAWLERS: Lazy<Regex> = Lazy::new(|| {
9    Regex::new(
10        r"(?ix)
11        Mediapartners-Google|
12        AdsBot-Google|
13        Googlebot|
14        FeedFetcher-Google|
15        BingBot|                    # Bing search
16        BingPreview|
17        Baiduspider|                # Baidu search
18        Slurp|                      # Yahoo
19        Sogou|                      # Sogou
20        facebook|                   # facebook
21        meta-|                      # meta/facebook
22        ia_archiver|                # Alexa
23        bots?[/\s\);]|              # Generic bot
24        spider[/\s\);]|             # Generic spider
25        Slack|                      # Slack - see https://api.slack.com/robots
26        Calypso\sAppCrawler|        # Google indexing bot
27        pingdom|                    # Pingdom
28        lyticsbot|                  # Lytics
29        AWS\sSecurity\sScanner|     # AWS Security Scanner causing DisallowedHost errors in Django, see
30                                    # https://forums.aws.amazon.com/thread.jspa?messageID=932404
31                                    # and https://github.com/getsentry/sentry-python/issues/641
32        HubSpot\sCrawler|           # HubSpot web crawler (web-crawlers@hubspot.com)
33        Bytespider|                 # Bytedance
34        Better\sUptime|             # Better Uptime
35        Cloudflare-Healthchecks|    # Cloudflare Health Checks
36        GTmetrix|                   # GTmetrix
37        BrightEdgeOnCrawl|          # BrightEdge - see https://www.brightedge.com/news/press-releases/brightedge-acquires-oncrawl-future-proof-web-30-strategies
38        ELB-HealthChecker|          # AWS Elastic Load Balancing Health Checks
39        naver.me/spd|               # Yeti/1.1 - naver.me
40        ClaudeBot|                  # Anthropic - see https://support.anthropic.com/en/articles/8896518-does-anthropic-crawl-data-from-the-web-and-how-can-site-owners-block-the-crawler
41        CCBot|                      # CCBot - see https://commoncrawl.org/ccbot
42        OAI-SearchBot|              # OpenAI - see https://platform.openai.com/docs/bots
43        GPTBot|                     # OpenAI - see https://platform.openai.com/docs/bots
44        PerplexityBot|              # Perplexity - see https://docs.perplexity.ai/guides/bots
45        Applebot|                   # Apple - see https://support.apple.com/en-us/119829
46        DuckDuckBot                 # DuckDuckGo - see https://duckduckgo.com/duckduckgo-help-pages/results/duckduckbot
47    "
48    )
49    .expect("Invalid web crawlers filter Regex")
50});
51
52static ALLOWED_WEB_CRAWLERS: Lazy<Regex> = Lazy::new(|| {
53    Regex::new(
54        r"(?ix)
55        Slackbot\s1\.\d+|            # Slack - see https://api.slack.com/robots
56        SentryUptimeBot              # Uptime Checker https://docs.sentry.io/product/alerts/uptime-monitoring/
57    ",
58    )
59    .expect("Invalid allowed web crawlers filter Regex")
60});
61
62/// Checks if the event originates from a known web crawler.
63fn matches(user_agent: Option<&str>) -> bool {
64    if let Some(user_agent) = user_agent {
65        WEB_CRAWLERS.is_match(user_agent) && !ALLOWED_WEB_CRAWLERS.is_match(user_agent)
66    } else {
67        false
68    }
69}
70
71/// Filters events originating from a known web crawler.
72pub fn should_filter<F: Filterable>(item: &F, config: &FilterConfig) -> Result<(), FilterStatKey> {
73    if !config.is_enabled {
74        return Ok(());
75    }
76
77    if matches(item.user_agent()) {
78        return Err(FilterStatKey::WebCrawlers);
79    }
80
81    Ok(())
82}
83
84#[cfg(test)]
85mod tests {
86    use super::*;
87    use crate::testutils;
88
89    #[test]
90    fn test_filter_when_disabled() {
91        let evt = testutils::get_event_with_user_agent("Googlebot");
92        let filter_result = should_filter(&evt, &FilterConfig { is_enabled: false });
93        assert_eq!(
94            filter_result,
95            Ok(()),
96            "Event filtered although filter should have been disabled"
97        )
98    }
99
100    #[test]
101    fn test_filter_banned_user_agents() {
102        let user_agents = [
103            "Mediapartners-Google",
104            "AdsBot-Google",
105            "Googlebot",
106            "FeedFetcher-Google",
107            "BingBot",
108            "BingPreview",
109            "Baiduspider",
110            "Slurp",
111            "Sogou",
112            "facebook",
113            "facebookexternalhit/1.1 (+http://www.facebook.com/externalhit_uatext.php)",
114            "facebookcatalog/1.0",
115            "meta-externalagent/1.1 (+https://developers.facebook.com/docs/sharing/webmasters/crawler)",
116            "meta-externalfetcher/1.1",
117            "ia_archiver",
118            "bots ",
119            "bots;",
120            "bots)",
121            "spider ",
122            "spider;",
123            "spider)",
124            "Calypso AppCrawler",
125            "pingdom",
126            "lyticsbot",
127            "AWS Security Scanner",
128            "Mozilla/5.0 (Linux; Android 6.0.1; Calypso AppCrawler Build/MMB30Y; wv) AppleWebKit/537.36 (KHTML, like Gecko) Version/4.0 Chrome/53.0.2785.124 Mobile Safari/537.36",
129            "Slackbot-LinkExpanding 1.0 (+https://api.slack.com/robots)",
130            "Slack-ImgProxy 0.19 (+https://api.slack.com/robots)",
131            "Twitterbot/1.0",
132            "FeedFetcher-Google; (+http://www.google.com/feedfetcher.html)",
133            "Mozilla/5.0 (compatible; Googlebot/2.1; +http://www.google.com/bot.html)",
134            "AdsBot-Google (+http://www.google.com/adsbot.html)",
135            "Mozilla/5.0 (compatible; HubSpot Crawler; web-crawlers@hubspot.com)",
136            "Mozilla/5.0 (Linux; Android 5.0) AppleWebKit/537.36 (KHTML, like Gecko) Mobile Safari/537.36 (compatible; Bytespider; spider-feedback@bytedance.com)",
137            "Better Uptime Bot Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/74.0.3729.169 Safari/537.36",
138            "Mozilla/5.0 (compatible;Cloudflare-Healthchecks/1.0;+https://www.cloudflare.com/; healthcheck-id: 0d1ca23e292c8c14)",
139            "Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/117.0.0.0 Safari/537.36 GTmetrix",
140            "Mozilla/5.0 (compatible; BrightEdgeOnCrawl/1.0; +http://www.oncrawl.com)",
141            "ELB-HealthChecker/2.0",
142            "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko; compatible; Yeti/1.1; +https://naver.me/spd) Chrome/127.0.0.0 Safari/537.36",
143            "Mozilla/5.0; ClaudeBot",
144            "Mozilla/5.0; CCBot",
145            "; OAI-SearchBot/1.0; +https://openai.com/searchbot",
146            "Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko); compatible; GPTBot/1.1; +https://openai.com/gptbot",
147            "Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko; compatible; PerplexityBot/1.0; +https://perplexity.ai/perplexitybot)",
148            "Mozilla/5.0 (Device; OS_version) AppleWebKit/WebKit_version (KHTML, like Gecko)Version/Safari_version [Mobile/Mobile_version] Safari/WebKit_version (Applebot/Applebot_version; +http://www.apple.com/go/applebot)",
149            "DuckDuckBot/1.1; (+http://duckduckgo.com/duckduckbot.html)",
150        ];
151
152        for banned_user_agent in &user_agents {
153            let event = testutils::get_event_with_user_agent(banned_user_agent);
154            let filter_result = should_filter(&event, &FilterConfig { is_enabled: true });
155            assert_ne!(
156                filter_result,
157                Ok(()),
158                "Failed to filter web crawler with user agent '{banned_user_agent}'"
159            );
160        }
161    }
162
163    #[test]
164    fn test_dont_filter_normal_user_agents() {
165        let normal_user_agents = [
166            "some user agent",
167            "IE",
168            "ie",
169            "opera",
170            "safari",
171            "APIs-Google (+https://developers.google.com/webmasters/APIs-Google.html)",
172            "Mozilla/5.0 (Windows NT 6.1) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/41.0.2228.0 Safari/537.36",
173            "Slackbot 1.0(+https://api.slack.com/robots)",
174            "SentryUptimeBot/1.0 (+http://docs.sentry.io/product/alerts/uptime-monitoring/)",
175            "Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko); compatible; ChatGPT-User/1.0; +https://openai.com/bot",
176        ];
177        for user_agent in &normal_user_agents {
178            let event = testutils::get_event_with_user_agent(user_agent);
179            let filter_result = should_filter(&event, &FilterConfig { is_enabled: true });
180            assert_eq!(
181                filter_result,
182                Ok(()),
183                "Failed benign user agent '{user_agent}'"
184            );
185        }
186    }
187}