1 use crate::{settings::structs::Settings, ApiError, IpAddr};
2 use actix_web::dev::ConnectionInfo;
3 use chrono::{DateTime, FixedOffset, NaiveDateTime};
4 use itertools::Itertools;
5 use rand::{distributions::Alphanumeric, thread_rng, Rng};
6 use regex::{Regex, RegexBuilder};
10 static ref EMAIL_REGEX: Regex = Regex::new(r"^[a-zA-Z0-9.!#$%&’*+/=?^_`{|}~-]+@[a-zA-Z0-9-]+(?:\.[a-zA-Z0-9-]+)*$").expect("compile regex");
11 static ref SLUR_REGEX: Regex = {
12 let mut slurs = r"(fag(g|got|tard)?\b|cock\s?sucker(s|ing)?|ni((g{2,}|q)+|[gq]{2,})[e3r]+(s|z)?|mudslime?s?|kikes?|\bspi(c|k)s?\b|\bchinks?|gooks?|bitch(es|ing|y)?|whor(es?|ing)|\btr(a|@)nn?(y|ies?)|\b(b|re|r)tard(ed)?s?)".to_string();
13 if let Some(additional_slurs) = Settings::get().additional_slurs {
15 slurs.push_str(&additional_slurs);
17 RegexBuilder::new(&slurs).case_insensitive(true).build().expect("compile regex")
21 static ref USERNAME_MATCHES_REGEX: Regex = Regex::new(r"/u/[a-zA-Z][0-9a-zA-Z_]*").expect("compile regex");
22 // TODO keep this old one, it didn't work with port well tho
23 // static ref MENTIONS_REGEX: Regex = Regex::new(r"@(?P<name>[\w.]+)@(?P<domain>[a-zA-Z0-9._-]+\.[a-zA-Z0-9_-]+)").expect("compile regex");
24 static ref MENTIONS_REGEX: Regex = Regex::new(r"@(?P<name>[\w.]+)@(?P<domain>[a-zA-Z0-9._:-]+)").expect("compile regex");
25 static ref VALID_ACTOR_NAME_REGEX: Regex = Regex::new(r"^[a-zA-Z0-9_]{3,}$").expect("compile regex");
26 static ref VALID_POST_TITLE_REGEX: Regex = Regex::new(r".*\S.*").expect("compile regex");
27 static ref VALID_MATRIX_ID_REGEX: Regex = Regex::new(r"^@[A-Za-z0-9._=-]+:[A-Za-z0-9.-]+\.[A-Za-z]{2,}$").expect("compile regex");
28 // taken from https://en.wikipedia.org/wiki/UTM_parameters
29 static ref CLEAN_URL_PARAMS_REGEX: Regex = Regex::new(r"^utm_source|utm_medium|utm_campaign|utm_term|utm_content|gclid|gclsrc|dclid|fbclid$").expect("compile regex");
32 pub fn naive_from_unix(time: i64) -> NaiveDateTime {
33 NaiveDateTime::from_timestamp(time, 0)
36 pub fn convert_datetime(datetime: NaiveDateTime) -> DateTime<FixedOffset> {
37 DateTime::<FixedOffset>::from_utc(datetime, FixedOffset::east(0))
40 pub fn remove_slurs(test: &str) -> String {
41 SLUR_REGEX.replace_all(test, "*removed*").to_string()
44 pub(crate) fn slur_check(test: &str) -> Result<(), Vec<&str>> {
45 let mut matches: Vec<&str> = SLUR_REGEX.find_iter(test).map(|mat| mat.as_str()).collect();
48 matches.sort_unstable();
51 if matches.is_empty() {
58 pub fn check_slurs(text: &str) -> Result<(), ApiError> {
59 if let Err(slurs) = slur_check(text) {
60 Err(ApiError::err(&slurs_vec_to_str(slurs)))
66 pub fn check_slurs_opt(text: &Option<String>) -> Result<(), ApiError> {
68 Some(t) => check_slurs(t),
73 pub(crate) fn slurs_vec_to_str(slurs: Vec<&str>) -> String {
74 let start = "No slurs - ";
75 let combined = &slurs.join(", ");
76 [start, combined].concat()
79 pub fn generate_random_string() -> String {
81 .sample_iter(&Alphanumeric)
87 pub fn markdown_to_html(text: &str) -> String {
88 comrak::markdown_to_html(text, &comrak::ComrakOptions::default())
91 // TODO nothing is done with community / group webfingers yet, so just ignore those for now
92 #[derive(Clone, PartialEq, Eq, Hash)]
93 pub struct MentionData {
99 pub fn is_local(&self) -> bool {
100 Settings::get().hostname.eq(&self.domain)
102 pub fn full_name(&self) -> String {
103 format!("@{}@{}", &self.name, &self.domain)
107 pub fn scrape_text_for_mentions(text: &str) -> Vec<MentionData> {
108 let mut out: Vec<MentionData> = Vec::new();
109 for caps in MENTIONS_REGEX.captures_iter(text) {
110 out.push(MentionData {
111 name: caps["name"].to_string(),
112 domain: caps["domain"].to_string(),
115 out.into_iter().unique().collect()
118 pub fn is_valid_actor_name(name: &str) -> bool {
119 name.chars().count() <= Settings::get().actor_name_max_length
120 && VALID_ACTOR_NAME_REGEX.is_match(name)
123 // Can't do a regex here, reverse lookarounds not supported
124 pub fn is_valid_display_name(name: &str) -> bool {
125 !name.starts_with('@')
126 && !name.starts_with('\u{200b}')
127 && name.chars().count() >= 3
128 && name.chars().count() <= Settings::get().actor_name_max_length
131 pub fn is_valid_matrix_id(matrix_id: &str) -> bool {
132 VALID_MATRIX_ID_REGEX.is_match(matrix_id)
135 pub fn is_valid_post_title(title: &str) -> bool {
136 VALID_POST_TITLE_REGEX.is_match(title)
139 pub fn get_ip(conn_info: &ConnectionInfo) -> IpAddr {
142 .realip_remote_addr()
143 .unwrap_or("127.0.0.1:12345")
146 .unwrap_or("127.0.0.1")
151 pub fn clean_url_params(mut url: Url) -> Url {
152 if url.query().is_some() {
155 .filter(|q| !CLEAN_URL_PARAMS_REGEX.is_match(&q.0))
156 .map(|q| format!("{}={}", q.0, q.1))
158 url.set_query(Some(&new_query));
165 use crate::utils::clean_url_params;
169 fn test_clean_url_params() {
170 let url = Url::parse("https://example.com/path/123?utm_content=buffercf3b2&utm_medium=social&username=randomuser&id=123").unwrap();
171 let cleaned = clean_url_params(url);
172 let expected = Url::parse("https://example.com/path/123?username=randomuser&id=123").unwrap();
173 assert_eq!(expected.to_string(), cleaned.to_string());
175 let url = Url::parse("https://example.com/path/123").unwrap();
176 let cleaned = clean_url_params(url.clone());
177 assert_eq!(url.to_string(), cleaned.to_string());