]> Untitled Git - lemmy.git/blobdiff - server/src/lib.rs
Merge branch 'master' of https://github.com/makigi-io/makigi into makigi-io-master
[lemmy.git] / server / src / lib.rs
index b5c26762f7aaa620a51e694293a1b00454169f4e..ebfe17d9c9ef803c3863a2ab52ca83ca4e45f937 100644 (file)
@@ -11,108 +11,50 @@ pub extern crate actix;
 pub extern crate actix_web;
 pub extern crate bcrypt;
 pub extern crate chrono;
-pub extern crate crypto;
+pub extern crate comrak;
 pub extern crate dotenv;
 pub extern crate jsonwebtoken;
 pub extern crate lettre;
 pub extern crate lettre_email;
 pub extern crate rand;
 pub extern crate regex;
+pub extern crate rss;
 pub extern crate serde;
 pub extern crate serde_json;
+pub extern crate sha2;
 pub extern crate strum;
 
 pub mod api;
 pub mod apub;
 pub mod db;
+pub mod rate_limit;
+pub mod routes;
 pub mod schema;
-pub mod websocket;
-pub mod nodeinfo;
+pub mod settings;
 pub mod version;
+pub mod websocket;
 
+use actix_web::dev::ConnectionInfo;
 use chrono::{DateTime, NaiveDateTime, Utc};
-use dotenv::dotenv;
 use lettre::smtp::authentication::{Credentials, Mechanism};
 use lettre::smtp::extension::ClientId;
 use lettre::smtp::ConnectionReuseParameters;
-use lettre::{SmtpClient, Transport};
+use lettre::{ClientSecurity, SmtpClient, Transport};
 use lettre_email::Email;
+use log::error;
+use percent_encoding::{utf8_percent_encode, NON_ALPHANUMERIC};
 use rand::distributions::Alphanumeric;
 use rand::{thread_rng, Rng};
-use regex::Regex;
-use std::env;
-
-pub struct Settings {
-  db_url: String,
-  hostname: String,
-  jwt_secret: String,
-  rate_limit_message: i32,
-  rate_limit_message_per_second: i32,
-  rate_limit_post: i32,
-  rate_limit_post_per_second: i32,
-  rate_limit_register: i32,
-  rate_limit_register_per_second: i32,
-  email_config: Option<EmailConfig>,
-}
-
-pub struct EmailConfig {
-  smtp_server: String,
-  smtp_login: String,
-  smtp_password: String,
-  smtp_from_address: String,
-}
-
-impl Settings {
-  fn get() -> Self {
-    dotenv().ok();
-
-    let email_config =
-      if env::var("SMTP_SERVER").is_ok() && !env::var("SMTP_SERVER").unwrap().eq("") {
-        Some(EmailConfig {
-          smtp_server: env::var("SMTP_SERVER").expect("SMTP_SERVER must be set"),
-          smtp_login: env::var("SMTP_LOGIN").expect("SMTP_LOGIN must be set"),
-          smtp_password: env::var("SMTP_PASSWORD").expect("SMTP_PASSWORD must be set"),
-          smtp_from_address: env::var("SMTP_FROM_ADDRESS").expect("SMTP_FROM_ADDRESS must be set"),
-        })
-      } else {
-        None
-      };
+use regex::{Regex, RegexBuilder};
+use serde::Deserialize;
 
-    Settings {
-      db_url: env::var("DATABASE_URL").expect("DATABASE_URL must be set"),
-      hostname: env::var("HOSTNAME").unwrap_or("rrr".to_string()),
-      jwt_secret: env::var("JWT_SECRET").unwrap_or("changeme".to_string()),
-      rate_limit_message: env::var("RATE_LIMIT_MESSAGE")
-        .unwrap_or("30".to_string())
-        .parse()
-        .unwrap(),
-      rate_limit_message_per_second: env::var("RATE_LIMIT_MESSAGE_PER_SECOND")
-        .unwrap_or("60".to_string())
-        .parse()
-        .unwrap(),
-      rate_limit_post: env::var("RATE_LIMIT_POST")
-        .unwrap_or("3".to_string())
-        .parse()
-        .unwrap(),
-      rate_limit_post_per_second: env::var("RATE_LIMIT_POST_PER_SECOND")
-        .unwrap_or("600".to_string())
-        .parse()
-        .unwrap(),
-      rate_limit_register: env::var("RATE_LIMIT_REGISTER")
-        .unwrap_or("1".to_string())
-        .parse()
-        .unwrap(),
-      rate_limit_register_per_second: env::var("RATE_LIMIT_REGISTER_PER_SECOND")
-        .unwrap_or("3600".to_string())
-        .parse()
-        .unwrap(),
-      email_config: email_config,
-    }
-  }
-  fn api_endpoint(&self) -> String {
-    format!("{}/api/v1", self.hostname)
-  }
-}
+use crate::settings::Settings;
+
+pub type ConnectionId = usize;
+pub type PostId = i32;
+pub type CommunityId = i32;
+pub type UserId = i32;
+pub type IPAddr = String;
 
 pub fn to_datetime_utc(ndt: NaiveDateTime) -> DateTime<Utc> {
   DateTime::<Utc>::from_utc(ndt, Utc)
@@ -130,12 +72,43 @@ pub fn is_email_regex(test: &str) -> bool {
   EMAIL_REGEX.is_match(test)
 }
 
+pub fn is_image_content_type(test: &str) -> Result<(), failure::Error> {
+  if attohttpc::get(test)
+    .send()?
+    .headers()
+    .get("Content-Type")
+    .ok_or_else(|| format_err!("No Content-Type header"))?
+    .to_str()?
+    .starts_with("image/")
+  {
+    Ok(())
+  } else {
+    Err(format_err!("Not an image type."))
+  }
+}
+
 pub fn remove_slurs(test: &str) -> String {
   SLUR_REGEX.replace_all(test, "*removed*").to_string()
 }
 
-pub fn has_slurs(test: &str) -> bool {
-  SLUR_REGEX.is_match(test)
+pub fn slur_check(test: &str) -> Result<(), Vec<&str>> {
+  let mut matches: Vec<&str> = SLUR_REGEX.find_iter(test).map(|mat| mat.as_str()).collect();
+
+  // Unique
+  matches.sort_unstable();
+  matches.dedup();
+
+  if matches.is_empty() {
+    Ok(())
+  } else {
+    Err(matches)
+  }
+}
+
+pub fn slurs_vec_to_str(slurs: Vec<&str>) -> String {
+  let start = "No slurs - ";
+  let combined = &slurs.join(", ");
+  [start, combined].concat()
 }
 
 pub fn extract_usernames(test: &str) -> Vec<&str> {
@@ -162,45 +135,172 @@ pub fn send_email(
   to_username: &str,
   html: &str,
 ) -> Result<(), String> {
-  let email_config = Settings::get().email_config.ok_or("no_email_setup")?;
+  let email_config = Settings::get().email.ok_or("no_email_setup")?;
 
   let email = Email::builder()
     .to((to_email, to_username))
-    .from((
-      email_config.smtp_login.to_owned(),
-      email_config.smtp_from_address,
-    ))
+    .from(email_config.smtp_from_address.to_owned())
     .subject(subject)
     .html(html)
     .build()
     .unwrap();
 
-  let mut mailer = SmtpClient::new_simple(&email_config.smtp_server)
-    .unwrap()
-    .hello_name(ClientId::Domain("localhost".to_string()))
-    .credentials(Credentials::new(
-      email_config.smtp_login.to_owned(),
-      email_config.smtp_password.to_owned(),
-    ))
-    .smtp_utf8(true)
-    .authentication_mechanism(Mechanism::Plain)
-    .connection_reuse(ConnectionReuseParameters::ReuseUnlimited)
-    .transport();
+  let mailer = if email_config.use_tls {
+    SmtpClient::new_simple(&email_config.smtp_server).unwrap()
+  } else {
+    SmtpClient::new(&email_config.smtp_server, ClientSecurity::None).unwrap()
+  }
+  .hello_name(ClientId::Domain(Settings::get().hostname))
+  .smtp_utf8(true)
+  .authentication_mechanism(Mechanism::Plain)
+  .connection_reuse(ConnectionReuseParameters::ReuseUnlimited);
+  let mailer = if let (Some(login), Some(password)) =
+    (&email_config.smtp_login, &email_config.smtp_password)
+  {
+    mailer.credentials(Credentials::new(login.to_owned(), password.to_owned()))
+  } else {
+    mailer
+  };
 
-  let result = mailer.send(email.into());
+  let mut transport = mailer.transport();
+  let result = transport.send(email.into());
+  transport.close();
 
   match result {
     Ok(_) => Ok(()),
-    Err(_) => Err("no_email_setup".to_string()),
+    Err(e) => Err(e.to_string()),
+  }
+}
+
+#[derive(Deserialize, Debug)]
+pub struct IframelyResponse {
+  title: Option<String>,
+  description: Option<String>,
+  thumbnail_url: Option<String>,
+  html: Option<String>,
+}
+
+pub fn fetch_iframely(url: &str) -> Result<IframelyResponse, failure::Error> {
+  let fetch_url = format!("http://iframely/oembed?url={}", url);
+  let text: String = attohttpc::get(&fetch_url).send()?.text()?;
+  let res: IframelyResponse = serde_json::from_str(&text)?;
+  Ok(res)
+}
+
+#[derive(Deserialize, Debug, Clone)]
+pub struct PictrsResponse {
+  files: Vec<PictrsFile>,
+  msg: String,
+}
+
+#[derive(Deserialize, Debug, Clone)]
+pub struct PictrsFile {
+  file: String,
+  delete_token: String,
+}
+
+pub fn fetch_pictrs(image_url: &str) -> Result<PictrsResponse, failure::Error> {
+  is_image_content_type(image_url)?;
+
+  let fetch_url = format!(
+    "http://pictrs:8080/image/download?url={}",
+    utf8_percent_encode(image_url, NON_ALPHANUMERIC) // TODO this might not be needed
+  );
+  let text = attohttpc::get(&fetch_url).send()?.text()?;
+  let res: PictrsResponse = serde_json::from_str(&text)?;
+  if res.msg == "ok" {
+    Ok(res)
+  } else {
+    Err(format_err!("{}", &res.msg))
   }
 }
 
+fn fetch_iframely_and_pictrs_data(
+  url: Option<String>,
+) -> (
+  Option<String>,
+  Option<String>,
+  Option<String>,
+  Option<String>,
+) {
+  match &url {
+    Some(url) => {
+      // Fetch iframely data
+      let (iframely_title, iframely_description, iframely_thumbnail_url, iframely_html) =
+        match fetch_iframely(url) {
+          Ok(res) => (res.title, res.description, res.thumbnail_url, res.html),
+          Err(e) => {
+            error!("iframely err: {}", e);
+            (None, None, None, None)
+          }
+        };
+
+      // Fetch pictrs thumbnail
+      let pictrs_thumbnail = match iframely_thumbnail_url {
+        Some(iframely_thumbnail_url) => match fetch_pictrs(&iframely_thumbnail_url) {
+          Ok(res) => Some(res.files[0].file.to_owned()),
+          Err(e) => {
+            error!("pictrs err: {}", e);
+            None
+          }
+        },
+        // Try to generate a small thumbnail if iframely is not supported
+        None => match fetch_pictrs(&url) {
+          Ok(res) => Some(res.files[0].file.to_owned()),
+          Err(e) => {
+            error!("pictrs err: {}", e);
+            None
+          }
+        },
+      };
+
+      (
+        iframely_title,
+        iframely_description,
+        iframely_html,
+        pictrs_thumbnail,
+      )
+    }
+    None => (None, None, None, None),
+  }
+}
+
+pub fn markdown_to_html(text: &str) -> String {
+  comrak::markdown_to_html(text, &comrak::ComrakOptions::default())
+}
+
+pub fn get_ip(conn_info: &ConnectionInfo) -> String {
+  conn_info
+    .remote()
+    .unwrap_or("127.0.0.1:12345")
+    .split(':')
+    .next()
+    .unwrap_or("127.0.0.1")
+    .to_string()
+}
+
+pub fn is_valid_username(name: &str) -> bool {
+  VALID_USERNAME_REGEX.is_match(name)
+}
+
+pub fn is_valid_community_name(name: &str) -> bool {
+  VALID_COMMUNITY_NAME_REGEX.is_match(name)
+}
+
 #[cfg(test)]
 mod tests {
-  use crate::{extract_usernames, has_slurs, is_email_regex, remove_slurs, Settings};
+  use crate::{
+    extract_usernames, is_email_regex, is_image_content_type, is_valid_community_name,
+    is_valid_username, remove_slurs, slur_check, slurs_vec_to_str,
+  };
+
   #[test]
-  fn test_api() {
-    assert_eq!(Settings::get().api_endpoint(), "rrr/api/v1");
+  fn test_image() {
+    assert!(is_image_content_type("https://1734811051.rsc.cdn77.org/data/images/full/365645/as-virus-kills-navajos-in-their-homes-tribal-women-provide-lifeline.jpg?w=600?w=650").is_ok());
+    assert!(is_image_content_type(
+      "https://twitter.com/BenjaminNorton/status/1259922424272957440?s=20"
+    )
+    .is_err());
   }
 
   #[test]
@@ -209,18 +309,50 @@ mod tests {
     assert!(!is_email_regex("nada_neutho"));
   }
 
+  #[test]
+  fn test_valid_register_username() {
+    assert!(is_valid_username("Hello_98"));
+    assert!(is_valid_username("ten"));
+    assert!(!is_valid_username("Hello-98"));
+    assert!(!is_valid_username("a"));
+    assert!(!is_valid_username(""));
+  }
+
+  #[test]
+  fn test_valid_community_name() {
+    assert!(is_valid_community_name("example"));
+    assert!(is_valid_community_name("example_community"));
+    assert!(!is_valid_community_name("Example"));
+    assert!(!is_valid_community_name("Ex"));
+    assert!(!is_valid_community_name(""));
+  }
+
   #[test]
   fn test_slur_filter() {
     let test =
-      "coons test dindu ladyboy tranny retardeds. This is a bunch of other safe text.".to_string();
+      "coons test dindu ladyboy tranny retardeds. Capitalized Niggerz. This is a bunch of other safe text.";
     let slur_free = "No slurs here";
     assert_eq!(
       remove_slurs(&test),
-      "*removed* test *removed* *removed* *removed* *removed*. This is a bunch of other safe text."
+      "*removed* test *removed* *removed* *removed* *removed*. Capitalized *removed*. This is a bunch of other safe text."
         .to_string()
     );
-    assert!(has_slurs(&test));
-    assert!(!has_slurs(slur_free));
+
+    let has_slurs_vec = vec![
+      "Niggerz",
+      "coons",
+      "dindu",
+      "ladyboy",
+      "retardeds",
+      "tranny",
+    ];
+    let has_slurs_err_str = "No slurs - Niggerz, coons, dindu, ladyboy, retardeds, tranny";
+
+    assert_eq!(slur_check(test), Err(has_slurs_vec));
+    assert_eq!(slur_check(slur_free), Ok(()));
+    if let Err(slur_vec) = slur_check(test) {
+      assert_eq!(&slurs_vec_to_str(slur_vec), has_slurs_err_str);
+    }
   }
 
   #[test]
@@ -230,6 +362,21 @@ mod tests {
     assert_eq!(usernames, expected);
   }
 
+  // These helped with testing
+  // #[test]
+  // fn test_iframely() {
+  //   let res = fetch_iframely("https://www.redspark.nu/?p=15341");
+  //   assert!(res.is_ok());
+  // }
+
+  // #[test]
+  // fn test_pictshare() {
+  //   let res = fetch_pictshare("https://upload.wikimedia.org/wikipedia/en/2/27/The_Mandalorian_logo.jpg");
+  //   assert!(res.is_ok());
+  //   let res_other = fetch_pictshare("https://upload.wikimedia.org/wikipedia/en/2/27/The_Mandalorian_logo.jpgaoeu");
+  //   assert!(res_other.is_err());
+  // }
+
   // #[test]
   // fn test_send_email() {
   //  let result =  send_email("not a subject", "test_email@gmail.com", "ur user", "<h1>HI there</h1>");
@@ -239,6 +386,8 @@ mod tests {
 
 lazy_static! {
   static ref EMAIL_REGEX: Regex = Regex::new(r"^[a-zA-Z0-9.!#$%&’*+/=?^_`{|}~-]+@[a-zA-Z0-9-]+(?:\.[a-zA-Z0-9-]+)*$").unwrap();
-  static ref SLUR_REGEX: Regex = Regex::new(r"(fag(g|got|tard)?|maricos?|cock\s?sucker(s|ing)?|nig(\b|g?(a|er)?s?)\b|dindu(s?)|mudslime?s?|kikes?|mongoloids?|towel\s*heads?|\bspi(c|k)s?\b|\bchinks?|niglets?|beaners?|\bnips?\b|\bcoons?\b|jungle\s*bunn(y|ies?)|jigg?aboo?s?|\bpakis?\b|rag\s*heads?|gooks?|cunts?|bitch(es|ing|y)?|puss(y|ies?)|twats?|feminazis?|whor(es?|ing)|\bslut(s|t?y)?|\btrann?(y|ies?)|ladyboy(s?)|\b(b|re|r)tard(ed)?s?)").unwrap();
+  static ref SLUR_REGEX: Regex = RegexBuilder::new(r"(fag(g|got|tard)?|maricos?|cock\s?sucker(s|ing)?|nig(\b|g?(a|er)?(s|z)?)\b|dindu(s?)|mudslime?s?|kikes?|mongoloids?|towel\s*heads?|\bspi(c|k)s?\b|\bchinks?|niglets?|beaners?|\bnips?\b|\bcoons?\b|jungle\s*bunn(y|ies?)|jigg?aboo?s?|\bpakis?\b|rag\s*heads?|gooks?|cunts?|bitch(es|ing|y)?|puss(y|ies?)|twats?|feminazis?|whor(es?|ing)|\bslut(s|t?y)?|\btrann?(y|ies?)|ladyboy(s?)|\b(b|re|r)tard(ed)?s?)").case_insensitive(true).build().unwrap();
   static ref USERNAME_MATCHES_REGEX: Regex = Regex::new(r"/u/[a-zA-Z][0-9a-zA-Z_]*").unwrap();
+  static ref VALID_USERNAME_REGEX: Regex = Regex::new(r"^[a-zA-Z0-9_]{3,20}$").unwrap();
+  static ref VALID_COMMUNITY_NAME_REGEX: Regex = Regex::new(r"^[a-z0-9_]{3,20}$").unwrap();
 }