0340541002 2025-04-24 1: use crate::{
0340541002 2025-04-24 2: command,
0340541002 2025-04-24 3: sql::Db,
0340541002 2025-04-24 4: };
1c444d34ff 2024-08-28 5:
9171c791eb 2021-11-13 6: use std::{
c1e27b74ed 2021-11-13 7: borrow::Cow,
9171c791eb 2021-11-13 8: collections::{
9171c791eb 2021-11-13 9: BTreeMap,
9171c791eb 2021-11-13 10: HashSet,
9171c791eb 2021-11-13 11: },
dc7c43b010 2026-01-01 12: sync::Arc,
44575a91d3 2025-07-09 13: };
44575a91d3 2025-07-09 14:
dc7c43b010 2026-01-01 15: use async_compat::Compat;
acb0a4ac54 2025-09-28 16: use chrono::{
acb0a4ac54 2025-09-28 17: DateTime,
acb0a4ac54 2025-09-28 18: Local,
44575a91d3 2025-07-09 19: };
5a4aab7687 2025-06-29 20: use lazy_static::lazy_static;
5a4aab7687 2025-06-29 21: use regex::Regex;
acb0a4ac54 2025-09-28 22: use reqwest::header::{
acb0a4ac54 2025-09-28 23: CACHE_CONTROL,
acb0a4ac54 2025-09-28 24: EXPIRES,
acb0a4ac54 2025-09-28 25: LAST_MODIFIED
dc7c43b010 2026-01-01 26: };
dc7c43b010 2026-01-01 27: use smol::{
dc7c43b010 2026-01-01 28: Timer,
dc7c43b010 2026-01-01 29: lock::Mutex,
acb0a4ac54 2025-09-28 30: };
bb89b6fab8 2025-06-15 31: use tgbot::{
bb89b6fab8 2025-06-15 32: api::Client,
bb89b6fab8 2025-06-15 33: handler::UpdateHandler,
bb89b6fab8 2025-06-15 34: types::{
bb89b6fab8 2025-06-15 35: Bot,
bb89b6fab8 2025-06-15 36: ChatPeerId,
bb89b6fab8 2025-06-15 37: Command,
bb89b6fab8 2025-06-15 38: GetBot,
bb89b6fab8 2025-06-15 39: Message,
bb89b6fab8 2025-06-15 40: ParseMode,
bb89b6fab8 2025-06-15 41: SendMessage,
bb89b6fab8 2025-06-15 42: Update,
bb89b6fab8 2025-06-15 43: UpdateType,
bb89b6fab8 2025-06-15 44: UserPeerId,
1bd041d00f 2025-05-20 45: },
1bd041d00f 2025-05-20 46: };
44575a91d3 2025-07-09 47: use stacked_errors::{
44575a91d3 2025-07-09 48: Result,
44575a91d3 2025-07-09 49: StackableErr,
44575a91d3 2025-07-09 50: anyhow,
44575a91d3 2025-07-09 51: bail,
44575a91d3 2025-07-09 52: };
5a4aab7687 2025-06-29 53:
5a4aab7687 2025-06-29 54: lazy_static!{
5a4aab7687 2025-06-29 55: pub static ref RE_SPECIAL: Regex = Regex::new(r"([\-_*\[\]()~`>#+|{}\.!])").unwrap();
5a4aab7687 2025-06-29 56: }
5a4aab7687 2025-06-29 57:
5a4aab7687 2025-06-29 58: /// Encodes special HTML entities to prevent them interfering with Telegram HTML
5a4aab7687 2025-06-29 59: pub fn encode (text: &str) -> Cow<'_, str> {
5a4aab7687 2025-06-29 60: RE_SPECIAL.replace_all(text, "\\$1")
5a4aab7687 2025-06-29 61: }
5a4aab7687 2025-06-29 62:
d7fbc271ac 2026-01-06 63: // This one does nothing except making sure only one token exists for each id
d7fbc271ac 2026-01-06 64: pub struct Token {
d7fbc271ac 2026-01-06 65: running: Arc<Mutex<HashSet<i32>>>,
d7fbc271ac 2026-01-06 66: my_id: i32,
d7fbc271ac 2026-01-06 67: }
d7fbc271ac 2026-01-06 68:
d7fbc271ac 2026-01-06 69: impl Token {
d7fbc271ac 2026-01-06 70: fn new (running: &Arc<Mutex<HashSet<i32>>>, my_id: i32) -> Option<Token> {
d7fbc271ac 2026-01-06 71: let running = running.clone();
d7fbc271ac 2026-01-06 72: smol::block_on(async {
d7fbc271ac 2026-01-06 73: let mut set = running.lock_arc().await;
d7fbc271ac 2026-01-06 74: if set.contains(&my_id) {
d7fbc271ac 2026-01-06 75: None
d7fbc271ac 2026-01-06 76: } else {
d7fbc271ac 2026-01-06 77: set.insert(my_id);
d7fbc271ac 2026-01-06 78: Some(Token {
d7fbc271ac 2026-01-06 79: running,
d7fbc271ac 2026-01-06 80: my_id,
d7fbc271ac 2026-01-06 81: })
d7fbc271ac 2026-01-06 82: }
d7fbc271ac 2026-01-06 83: })
d7fbc271ac 2026-01-06 84: }
d7fbc271ac 2026-01-06 85: }
d7fbc271ac 2026-01-06 86:
d7fbc271ac 2026-01-06 87: impl Drop for Token {
d7fbc271ac 2026-01-06 88: fn drop (&mut self) {
d7fbc271ac 2026-01-06 89: smol::block_on(async {
d7fbc271ac 2026-01-06 90: let mut set = self.running.lock_arc().await;
d7fbc271ac 2026-01-06 91: set.remove(&self.my_id);
d7fbc271ac 2026-01-06 92: })
d7fbc271ac 2026-01-06 93: }
d7fbc271ac 2026-01-06 94: }
d7fbc271ac 2026-01-06 95:
9171c791eb 2021-11-13 96: #[derive(Clone)]
9171c791eb 2021-11-13 97: pub struct Core {
bb89b6fab8 2025-06-15 98: owner_chat: ChatPeerId,
bb89b6fab8 2025-06-15 99: // max_delay: u16,
bb89b6fab8 2025-06-15 100: pub tg: Client,
bb89b6fab8 2025-06-15 101: pub me: Bot,
0340541002 2025-04-24 102: pub db: Db,
d7fbc271ac 2026-01-06 103: running: Arc<Mutex<HashSet<i32>>>,
45e34762e4 2023-05-28 104: http_client: reqwest::Client,
dc7c43b010 2026-01-01 105: }
dc7c43b010 2026-01-01 106:
dc7c43b010 2026-01-01 107: pub struct Post {
dc7c43b010 2026-01-01 108: uri: String,
dc7c43b010 2026-01-01 109: title: String,
dc7c43b010 2026-01-01 110: authors: String,
dc7c43b010 2026-01-01 111: summary: String,
0340541002 2025-04-24 112: }
0340541002 2025-04-24 113:
9171c791eb 2021-11-13 114: impl Core {
0340541002 2025-04-24 115: pub async fn new(settings: config::Config) -> Result<Core> {
44575a91d3 2025-07-09 116: let owner_chat = ChatPeerId::from(settings.get_int("owner").stack()?);
44575a91d3 2025-07-09 117: let api_key = settings.get_string("api_key").stack()?;
42b29b744b 2026-01-01 118: let tg = Client::new(&api_key).stack()?
42b29b744b 2026-01-01 119: .with_host(settings.get_string("api_gateway").stack()?);
45e34762e4 2023-05-28 120:
45e34762e4 2023-05-28 121: let mut client = reqwest::Client::builder();
9d8a6738fd 2023-07-30 122: if let Ok(proxy) = settings.get_string("proxy") {
44575a91d3 2025-07-09 123: let proxy = reqwest::Proxy::all(proxy).stack()?;
45e34762e4 2023-05-28 124: client = client.proxy(proxy);
45e34762e4 2023-05-28 125: }
44575a91d3 2025-07-09 126: let http_client = client.build().stack()?;
44575a91d3 2025-07-09 127: let me = tg.execute(GetBot).await.stack()?;
0340541002 2025-04-24 128: let core = Core {
f988dfd28f 2022-02-13 129: tg,
e624ef9d66 2025-04-20 130: me,
e624ef9d66 2025-04-20 131: owner_chat,
44575a91d3 2025-07-09 132: db: Db::new(&settings.get_string("pg").stack()?)?,
d7fbc271ac 2026-01-06 133: running: Arc::new(Mutex::new(HashSet::new())),
45e34762e4 2023-05-28 134: http_client,
bb89b6fab8 2025-06-15 135: // max_delay: 60,
0340541002 2025-04-24 136: };
bb89b6fab8 2025-06-15 137: let clone = core.clone();
dc7c43b010 2026-01-01 138: smol::spawn(Compat::new(async move {
26339860ce 2022-02-13 139: loop {
26339860ce 2022-02-13 140: let delay = match &clone.autofetch().await {
26339860ce 2022-02-13 141: Err(err) => {
07b34bcad6 2025-08-05 142: if let Err(err) = clone.send(format!("🛑 {err}"), None, None).await {
e624ef9d66 2025-04-20 143: eprintln!("Autofetch error: {err:?}");
26339860ce 2022-02-13 144: };
cb86e770f9 2022-03-15 145: std::time::Duration::from_secs(60)
26339860ce 2022-02-13 146: },
26339860ce 2022-02-13 147: Ok(time) => *time,
9171c791eb 2021-11-13 148: };
dc7c43b010 2026-01-01 149: Timer::after(delay).await;
9171c791eb 2021-11-13 150: }
dc7c43b010 2026-01-01 151: })).detach();
9171c791eb 2021-11-13 152: Ok(core)
9171c791eb 2021-11-13 153: }
9171c791eb 2021-11-13 154:
bb89b6fab8 2025-06-15 155: pub async fn send <S>(&self, msg: S, target: Option<ChatPeerId>, mode: Option<ParseMode>) -> Result<Message>
e624ef9d66 2025-04-20 156: where S: Into<String> {
e624ef9d66 2025-04-20 157: let msg = msg.into();
e624ef9d66 2025-04-20 158:
e624ef9d66 2025-04-20 159: let mode = mode.unwrap_or(ParseMode::Html);
e624ef9d66 2025-04-20 160: let target = target.unwrap_or(self.owner_chat);
44575a91d3 2025-07-09 161: self.tg.execute(
44575a91d3 2025-07-09 162: SendMessage::new(target, msg)
44575a91d3 2025-07-09 163: .with_parse_mode(mode)
44575a91d3 2025-07-09 164: ).await.stack()
44575a91d3 2025-07-09 165: }
44575a91d3 2025-07-09 166:
acb0a4ac54 2025-09-28 167: pub async fn check (&self, id: i32, real: bool, last_scrape: Option<DateTime<Local>>) -> Result<String> {
44575a91d3 2025-07-09 168: let mut posted: i32 = 0;
44575a91d3 2025-07-09 169: let mut conn = self.db.begin().await.stack()?;
44575a91d3 2025-07-09 170:
d7fbc271ac 2026-01-06 171: let token = Token::new(&self.running, id);
d7fbc271ac 2026-01-06 172: if token.is_none() {
d7fbc271ac 2026-01-06 173: bail!("check is already running");
d7fbc271ac 2026-01-06 174: }
d7fbc271ac 2026-01-06 175: let source = conn.get_source(id, self.owner_chat).await.stack()?;
d7fbc271ac 2026-01-06 176: conn.set_scrape(id).await.stack()?;
d7fbc271ac 2026-01-06 177: let destination = ChatPeerId::from(match real {
d7fbc271ac 2026-01-06 178: true => source.channel_id,
d7fbc271ac 2026-01-06 179: false => source.owner,
d7fbc271ac 2026-01-06 180: });
d7fbc271ac 2026-01-06 181: let mut this_fetch: Option<DateTime<chrono::FixedOffset>> = None;
d7fbc271ac 2026-01-06 182: let mut posts: BTreeMap<DateTime<chrono::FixedOffset>, Post> = BTreeMap::new();
d7fbc271ac 2026-01-06 183:
d7fbc271ac 2026-01-06 184: let mut builder = self.http_client.get(&source.url);
d7fbc271ac 2026-01-06 185: if let Some(last_scrape) = last_scrape {
d7fbc271ac 2026-01-06 186: builder = builder.header(LAST_MODIFIED, last_scrape.to_rfc2822());
d7fbc271ac 2026-01-06 187: };
d7fbc271ac 2026-01-06 188: let response = builder.send().await.stack()?;
d7fbc271ac 2026-01-06 189: {
d7fbc271ac 2026-01-06 190: let headers = response.headers();
d7fbc271ac 2026-01-06 191: let expires = headers.get(EXPIRES);
d7fbc271ac 2026-01-06 192: let cache = headers.get(CACHE_CONTROL);
d7fbc271ac 2026-01-06 193: if expires.is_some() || cache.is_some() {
d7fbc271ac 2026-01-06 194: println!("{} {} {:?} {:?} {:?}", Local::now().to_rfc2822(), &source.url, last_scrape, expires, cache);
d7fbc271ac 2026-01-06 195: }
d7fbc271ac 2026-01-06 196: }
d7fbc271ac 2026-01-06 197: let status = response.status();
d7fbc271ac 2026-01-06 198: let content = response.bytes().await.stack()?;
d7fbc271ac 2026-01-06 199: match rss::Channel::read_from(&content[..]) {
d7fbc271ac 2026-01-06 200: Ok(feed) => {
d7fbc271ac 2026-01-06 201: for item in feed.items() {
d7fbc271ac 2026-01-06 202: if let Some(link) = item.link() {
d7fbc271ac 2026-01-06 203: let date = match item.pub_date() {
d7fbc271ac 2026-01-06 204: Some(feed_date) => DateTime::parse_from_rfc2822(feed_date),
d7fbc271ac 2026-01-06 205: None => DateTime::parse_from_rfc3339(match item.dublin_core_ext() {
d7fbc271ac 2026-01-06 206: Some(dates) => &dates.dates()[0],
d7fbc271ac 2026-01-06 207: None => bail!("Feed item misses posting date."),
d7fbc271ac 2026-01-06 208: }),
d7fbc271ac 2026-01-06 209: }.stack()?;
d7fbc271ac 2026-01-06 210: let uri = link.to_string();
d7fbc271ac 2026-01-06 211: let title = item.title().unwrap_or("").to_string();
d7fbc271ac 2026-01-06 212: let authors = item.author().unwrap_or("").to_string();
d7fbc271ac 2026-01-06 213: let summary = item.content().unwrap_or("").to_string();
d7fbc271ac 2026-01-06 214: posts.insert(date, Post{
d7fbc271ac 2026-01-06 215: uri,
d7fbc271ac 2026-01-06 216: title,
d7fbc271ac 2026-01-06 217: authors,
d7fbc271ac 2026-01-06 218: summary,
d7fbc271ac 2026-01-06 219: });
d7fbc271ac 2026-01-06 220: }
d7fbc271ac 2026-01-06 221: };
d7fbc271ac 2026-01-06 222: },
d7fbc271ac 2026-01-06 223: Err(err) => match err {
d7fbc271ac 2026-01-06 224: rss::Error::InvalidStartTag => {
d7fbc271ac 2026-01-06 225: match atom_syndication::Feed::read_from(&content[..]) {
d7fbc271ac 2026-01-06 226: Ok(feed) => {
d7fbc271ac 2026-01-06 227: for item in feed.entries() {
d7fbc271ac 2026-01-06 228: let date = item.published().unwrap();
d7fbc271ac 2026-01-06 229: let uri = item.links()[0].href().to_string();
d7fbc271ac 2026-01-06 230: let title = item.title().to_string();
d7fbc271ac 2026-01-06 231: let authors = item.authors().iter().map(|x| format!("{} <{:?}>", x.name(), x.email())).collect::<Vec<String>>().join(", ");
d7fbc271ac 2026-01-06 232: let summary = if let Some(sum) = item.summary() { sum.value.clone() } else { String::new() };
d7fbc271ac 2026-01-06 233: posts.insert(*date, Post{
d7fbc271ac 2026-01-06 234: uri,
d7fbc271ac 2026-01-06 235: title,
d7fbc271ac 2026-01-06 236: authors,
d7fbc271ac 2026-01-06 237: summary,
d7fbc271ac 2026-01-06 238: });
d7fbc271ac 2026-01-06 239: };
d7fbc271ac 2026-01-06 240: },
d7fbc271ac 2026-01-06 241: Err(err) => {
d7fbc271ac 2026-01-06 242: bail!("Unsupported or mangled content:\n{:?}\n{err}\n{status:#?}\n", &source.url)
d7fbc271ac 2026-01-06 243: },
d7fbc271ac 2026-01-06 244: }
d7fbc271ac 2026-01-06 245: },
d7fbc271ac 2026-01-06 246: rss::Error::Eof => (),
d7fbc271ac 2026-01-06 247: _ => bail!("Unsupported or mangled content:\n{:?}\n{err}\n{status:#?}\n", &source.url)
d7fbc271ac 2026-01-06 248: }
d7fbc271ac 2026-01-06 249: };
d7fbc271ac 2026-01-06 250: for (date, post) in posts.iter() {
d7fbc271ac 2026-01-06 251: let post_url: Cow<str> = match source.url_re {
d7fbc271ac 2026-01-06 252: Some(ref x) => sedregex::ReplaceCommand::new(x).stack()?.execute(&post.uri),
d7fbc271ac 2026-01-06 253: None => post.uri.clone().into(),
d7fbc271ac 2026-01-06 254: };
d7fbc271ac 2026-01-06 255: if let Some(exists) = conn.exists(&post_url, id).await.stack()? {
d7fbc271ac 2026-01-06 256: if ! exists {
d7fbc271ac 2026-01-06 257: if this_fetch.is_none() || *date > this_fetch.unwrap() {
d7fbc271ac 2026-01-06 258: this_fetch = Some(*date);
d7fbc271ac 2026-01-06 259: };
d7fbc271ac 2026-01-06 260: self.send( match &source.iv_hash {
d7fbc271ac 2026-01-06 261: Some(hash) => format!("<a href=\"https://t.me/iv?url={post_url}&rhash={hash}\"> </a>{post_url}"),
d7fbc271ac 2026-01-06 262: None => format!("{post_url}"),
d7fbc271ac 2026-01-06 263: }, Some(destination), Some(ParseMode::Html)).await.stack()?;
d7fbc271ac 2026-01-06 264: conn.add_post(id, date, &post_url).await.stack()?;
d7fbc271ac 2026-01-06 265: };
d7fbc271ac 2026-01-06 266: };
d7fbc271ac 2026-01-06 267: posted += 1;
d7fbc271ac 2026-01-06 268: };
d7fbc271ac 2026-01-06 269: posts.clear();
0340541002 2025-04-24 270: Ok(format!("Posted: {posted}"))
26339860ce 2022-02-13 271: }
26339860ce 2022-02-13 272:
bb89b6fab8 2025-06-15 273: async fn autofetch(&self) -> Result<std::time::Duration> {
26339860ce 2022-02-13 274: let mut delay = chrono::Duration::minutes(1);
9910c2209c 2023-08-04 275: let now = chrono::Local::now();
bb89b6fab8 2025-06-15 276: let queue = {
44575a91d3 2025-07-09 277: let mut conn = self.db.begin().await.stack()?;
44575a91d3 2025-07-09 278: conn.get_queue().await.stack()?
bb89b6fab8 2025-06-15 279: };
bb89b6fab8 2025-06-15 280: for row in queue {
9910c2209c 2023-08-04 281: if let Some(next_fetch) = row.next_fetch {
9910c2209c 2023-08-04 282: if next_fetch < now {
acb0a4ac54 2025-09-28 283: if let (Some(owner), Some(source_id), last_scrape) = (row.owner, row.source_id, row.last_scrape) {
bb89b6fab8 2025-06-15 284: let clone = Core {
bb89b6fab8 2025-06-15 285: owner_chat: ChatPeerId::from(owner),
9910c2209c 2023-08-04 286: ..self.clone()
9910c2209c 2023-08-04 287: };
b4af85e31b 2025-06-29 288: let source = {
44575a91d3 2025-07-09 289: let mut conn = self.db.begin().await.stack()?;
b4af85e31b 2025-06-29 290: match conn.get_one(owner, source_id).await {
b4af85e31b 2025-06-29 291: Ok(Some(source)) => source.to_string(),
4c144972c0 2026-01-02 292: Ok(None) => "Source not found in database?".to_string(),
b4af85e31b 2025-06-29 293: Err(err) => format!("Failed to fetch source data:\n{err}"),
b4af85e31b 2025-06-29 294: }
b4af85e31b 2025-06-29 295: };
dc7c43b010 2026-01-01 296: smol::spawn(Compat::new(async move {
acb0a4ac54 2025-09-28 297: if let Err(err) = clone.check(source_id, true, Some(last_scrape)).await {
4c144972c0 2026-01-02 298: if let Err(err) = clone.send(&format!("🛑 {source}\n{}", encode(&err.to_string())), None, Some(ParseMode::MarkdownV2)).await {
ec71df21e1 2025-11-07 299: eprintln!("Check error: {err}");
79c91a5357 2024-08-02 300: // clone.disable(&source_id, owner).await.unwrap();
9910c2209c 2023-08-04 301: };
9910c2209c 2023-08-04 302: };
dc7c43b010 2026-01-01 303: })).detach();
9910c2209c 2023-08-04 304: }
9910c2209c 2023-08-04 305: } else if next_fetch - now < delay {
9910c2209c 2023-08-04 306: delay = next_fetch - now;
9910c2209c 2023-08-04 307: }
9910c2209c 2023-08-04 308: }
9910c2209c 2023-08-04 309: };
44575a91d3 2025-07-09 310: delay.to_std().stack()
9910c2209c 2023-08-04 311: }
9910c2209c 2023-08-04 312:
bb89b6fab8 2025-06-15 313: pub async fn list (&self, owner: UserPeerId) -> Result<String> {
b4af85e31b 2025-06-29 314: let mut reply: Vec<String> = vec![];
9910c2209c 2023-08-04 315: reply.push("Channels:".into());
44575a91d3 2025-07-09 316: let mut conn = self.db.begin().await.stack()?;
44575a91d3 2025-07-09 317: for row in conn.get_list(owner).await.stack()? {
b4af85e31b 2025-06-29 318: reply.push(row.to_string());
e624ef9d66 2025-04-20 319: };
b4af85e31b 2025-06-29 320: Ok(reply.join("\n\n"))
bb89b6fab8 2025-06-15 321: }
bb89b6fab8 2025-06-15 322: }
bb89b6fab8 2025-06-15 323:
bb89b6fab8 2025-06-15 324: impl UpdateHandler for Core {
bb89b6fab8 2025-06-15 325: async fn handle (&self, update: Update) {
bb89b6fab8 2025-06-15 326: if let UpdateType::Message(msg) = update.update_type {
bb89b6fab8 2025-06-15 327: if let Ok(cmd) = Command::try_from(msg) {
bb89b6fab8 2025-06-15 328: let msg = cmd.get_message();
bb89b6fab8 2025-06-15 329: let words = cmd.get_args();
fae13a0e55 2025-06-28 330: let command = cmd.get_name();
fae13a0e55 2025-06-28 331: let res = match command {
fae13a0e55 2025-06-28 332: "/check" | "/clean" | "/enable" | "/delete" | "/disable" => command::command(self, command, msg, words).await,
bb89b6fab8 2025-06-15 333: "/start" => command::start(self, msg).await,
bb89b6fab8 2025-06-15 334: "/list" => command::list(self, msg).await,
fae13a0e55 2025-06-28 335: "/add" | "/update" => command::update(self, command, msg, words).await,
bb89b6fab8 2025-06-15 336: any => Err(anyhow!("Unknown command: {any}")),
bb89b6fab8 2025-06-15 337: };
bb89b6fab8 2025-06-15 338: if let Err(err) = res {
07b34bcad6 2025-08-05 339: if let Err(err2) = self.send(format!("\\#error\n```\n{err}\n```"),
bb89b6fab8 2025-06-15 340: Some(msg.chat.get_id()),
bb89b6fab8 2025-06-15 341: Some(ParseMode::MarkdownV2)
bb89b6fab8 2025-06-15 342: ).await{
bb89b6fab8 2025-06-15 343: dbg!(err2);
bb89b6fab8 2025-06-15 344: };
bb89b6fab8 2025-06-15 345: }
bb89b6fab8 2025-06-15 346: };
bb89b6fab8 2025-06-15 347: };
e624ef9d66 2025-04-20 348: }
9171c791eb 2021-11-13 349: }