Lines of
src/core.rs
from check-in fae13a0e55
that are changed by the sequence of edits moving toward
check-in b4af85e31b:
1: use crate::{
2: command,
3: sql::Db,
4: };
5:
6: use std::{
7: borrow::Cow,
8: collections::{
9: BTreeMap,
10: HashSet,
11: },
12: sync::{
13: Arc,
14: Mutex
15: },
16: };
17:
18: use anyhow::{
19: anyhow,
20: bail,
21: Result,
22: };
23: use async_std::task;
24: use chrono::DateTime;
25: use tgbot::{
26: api::Client,
27: handler::UpdateHandler,
28: types::{
29: Bot,
30: ChatPeerId,
31: Command,
32: GetBot,
33: Message,
34: ParseMode,
35: SendMessage,
36: Update,
37: UpdateType,
38: UserPeerId,
39: },
40: };
41:
42: #[derive(Clone)]
43: pub struct Core {
44: owner_chat: ChatPeerId,
45: // max_delay: u16,
46: pub tg: Client,
47: pub me: Bot,
48: pub db: Db,
49: sources: Arc<Mutex<HashSet<Arc<i32>>>>,
50: http_client: reqwest::Client,
51: }
52:
53: impl Core {
54: pub async fn new(settings: config::Config) -> Result<Core> {
55: let owner_chat = ChatPeerId::from(settings.get_int("owner")?);
56: let api_key = settings.get_string("api_key")?;
57: let tg = Client::new(&api_key)?;
58:
59: let mut client = reqwest::Client::builder();
60: if let Ok(proxy) = settings.get_string("proxy") {
61: let proxy = reqwest::Proxy::all(proxy)?;
62: client = client.proxy(proxy);
63: }
64: let http_client = client.build()?;
65: let me = tg.execute(GetBot).await?;
66: let core = Core {
67: tg,
68: me,
69: owner_chat,
70: db: Db::new(&settings.get_string("pg")?)?,
71: sources: Arc::new(Mutex::new(HashSet::new())),
72: http_client,
73: // max_delay: 60,
74: };
75: let clone = core.clone();
76: task::spawn(async move {
77: loop {
78: let delay = match &clone.autofetch().await {
79: Err(err) => {
80: if let Err(err) = clone.send(format!("š {err:?}"), None, None).await {
81: eprintln!("Autofetch error: {err:?}");
82: };
83: std::time::Duration::from_secs(60)
84: },
85: Ok(time) => *time,
86: };
87: task::sleep(delay).await;
88: }
89: });
90: Ok(core)
91: }
92:
93: pub async fn send <S>(&self, msg: S, target: Option<ChatPeerId>, mode: Option<ParseMode>) -> Result<Message>
94: where S: Into<String> {
95: let msg = msg.into();
96:
97: let mode = mode.unwrap_or(ParseMode::Html);
98: let target = target.unwrap_or(self.owner_chat);
99: Ok(self.tg.execute(
100: SendMessage::new(target, msg)
101: .with_parse_mode(mode)
102: ).await?)
103: }
104:
105: pub async fn check (&self, id: i32, real: bool) -> Result<String> {
106: let mut posted: i32 = 0;
107: let mut conn = self.db.begin().await?;
108:
109: let id = {
110: let mut set = self.sources.lock().unwrap();
111: match set.get(&id) {
112: Some(id) => id.clone(),
113: None => {
114: let id = Arc::new(id);
115: set.insert(id.clone());
116: id.clone()
117: },
118: }
119: };
120: let count = Arc::strong_count(&id);
121: if count == 2 {
122: let source = conn.get_source(*id, self.owner_chat).await?;
123: conn.set_scrape(*id).await?;
124: let destination = ChatPeerId::from(match real {
125: true => source.channel_id,
126: false => source.owner,
127: });
128: let mut this_fetch: Option<DateTime<chrono::FixedOffset>> = None;
129: let mut posts: BTreeMap<DateTime<chrono::FixedOffset>, String> = BTreeMap::new();
130:
131: let response = self.http_client.get(&source.url).send().await?;
132: let status = response.status();
133: let content = response.bytes().await?;
134: match rss::Channel::read_from(&content[..]) {
135: Ok(feed) => {
136: for item in feed.items() {
137: if let Some(link) = item.link() {
138: let date = match item.pub_date() {
139: Some(feed_date) => DateTime::parse_from_rfc2822(feed_date),
140: None => DateTime::parse_from_rfc3339(&item.dublin_core_ext().unwrap().dates()[0]),
141: }?;
142: let url = link;
143: posts.insert(date, url.to_string());
144: }
145: };
146: },
147: Err(err) => match err {
148: rss::Error::InvalidStartTag => {
149: match atom_syndication::Feed::read_from(&content[..]) {
150: Ok(feed) => {
151: for item in feed.entries() {
152: let date = item.published().unwrap();
153: let url = item.links()[0].href();
154: posts.insert(*date, url.to_string());
155: };
156: },
157: Err(err) => {
158: bail!("Unsupported or mangled content:\n{:?}\n{err:#?}\n{status:#?}\n", &source.url)
159: },
160: }
161: },
162: rss::Error::Eof => (),
163: _ => bail!("Unsupported or mangled content:\n{:?}\n{err:#?}\n{status:#?}\n", &source.url)
164: }
165: };
166: for (date, url) in posts.iter() {
167: let post_url: Cow<str> = match source.url_re {
168: Some(ref x) => sedregex::ReplaceCommand::new(x)?.execute(url),
169: None => url.into(),
170: };
171: if let Some(exists) = conn.exists(&post_url, *id).await? {
172: if ! exists {
173: if this_fetch.is_none() || *date > this_fetch.unwrap() {
174: this_fetch = Some(*date);
175: };
176: self.send( match &source.iv_hash {
177: Some(hash) => format!("<a href=\"https://t.me/iv?url={post_url}&rhash={hash}\"> </a>{post_url}"),
178: None => format!("{post_url}"),
179: }, Some(destination), Some(ParseMode::Html)).await?;
180: conn.add_post(*id, date, &post_url).await?;
181: };
182: };
183: posted += 1;
184: };
185: posts.clear();
186: };
187: Ok(format!("Posted: {posted}"))
188: }
189:
190: async fn autofetch(&self) -> Result<std::time::Duration> {
191: let mut delay = chrono::Duration::minutes(1);
192: let now = chrono::Local::now();
193: let queue = {
194: let mut conn = self.db.begin().await?;
195: conn.get_queue().await?
196: };
197: for row in queue {
198: if let Some(next_fetch) = row.next_fetch {
199: if next_fetch < now {
200: if let (Some(owner), Some(source_id)) = (row.owner, row.source_id) {
201: let clone = Core {
202: owner_chat: ChatPeerId::from(owner),
203: ..self.clone()
204: };
205: task::spawn(async move {
206: if let Err(err) = clone.check(source_id, true).await {
fae13a0e55 2025-06-28 207: if let Err(err) = clone.send(&format!("š {err:?}"), None, None).await {
208: eprintln!("Check error: {err:?}");
209: // clone.disable(&source_id, owner).await.unwrap();
210: };
211: };
212: });
213: }
214: } else if next_fetch - now < delay {
215: delay = next_fetch - now;
216: }
217: }
218: };
219: Ok(delay.to_std()?)
220: }
221:
222: pub async fn list (&self, owner: UserPeerId) -> Result<String> {
fae13a0e55 2025-06-28 223: let mut reply: Vec<Cow<str>> = vec![];
224: reply.push("Channels:".into());
225: let mut conn = self.db.begin().await?;
226: for row in conn.get_list(owner).await? {
fae13a0e55 2025-06-28 227: reply.push(format!("\n\\#ļøā£ {} \\*ļøā£ `{}` {}\nš `{}`", row.source_id, row.channel,
fae13a0e55 2025-06-28 228: match row.enabled {
fae13a0e55 2025-06-28 229: true => "š enabled",
fae13a0e55 2025-06-28 230: false => "ā disabled",
fae13a0e55 2025-06-28 231: }, row.url).into());
fae13a0e55 2025-06-28 232: if let Some(hash) = &row.iv_hash {
fae13a0e55 2025-06-28 233: reply.push(format!("IV: `{hash}`").into());
fae13a0e55 2025-06-28 234: }
fae13a0e55 2025-06-28 235: if let Some(re) = &row.url_re {
fae13a0e55 2025-06-28 236: reply.push(format!("RE: `{re}`").into());
fae13a0e55 2025-06-28 237: }
238: };
fae13a0e55 2025-06-28 239: Ok(reply.join("\n"))
240: }
241: }
242:
243: impl UpdateHandler for Core {
244: async fn handle (&self, update: Update) {
245: if let UpdateType::Message(msg) = update.update_type {
246: if let Ok(cmd) = Command::try_from(msg) {
247: let msg = cmd.get_message();
248: let words = cmd.get_args();
249: let command = cmd.get_name();
250: let res = match command {
251: "/check" | "/clean" | "/enable" | "/delete" | "/disable" => command::command(self, command, msg, words).await,
252: "/start" => command::start(self, msg).await,
253: "/list" => command::list(self, msg).await,
254: "/add" | "/update" => command::update(self, command, msg, words).await,
255: any => Err(anyhow!("Unknown command: {any}")),
256: };
257: if let Err(err) = res {
258: if let Err(err2) = self.send(format!("\\#error\n```\n{err:?}\n```"),
259: Some(msg.chat.get_id()),
260: Some(ParseMode::MarkdownV2)
261: ).await{
262: dbg!(err2);
263: };
264: }
265: };
266: };
267: }
268: }