51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
|
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
|
+
+
+
-
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
-
+
+
+
+
+
+
+
+
+
+
+
+
+
-
+
|
bail,
};
lazy_static!{
pub static ref RE_SPECIAL: Regex = Regex::new(r"([\-_*\[\]()~`>#+|{}\.!])").unwrap();
}
/// Escape characters that are special in Telegram MarkdownV2 by prefixing them with a backslash.
///
/// This ensures the returned string can be used as MarkdownV2-formatted Telegram message content
/// Encodes special HTML entities to prevent them interfering with Telegram HTML
/// without special characters being interpreted as MarkdownV2 markup.
pub fn encode (text: &str) -> Cow<'_, str> {
RE_SPECIAL.replace_all(text, "\\$1")
}
// This one does nothing except making sure only one token exists for each id
pub struct Token {
running: Arc<Mutex<HashSet<i32>>>,
my_id: i32,
}
impl Token {
/// Attempts to acquire a per-id token by inserting `my_id` into the shared `running` set.
///
/// If the id was not already present, the function inserts it and returns `Some(Token)`.
/// When the returned `Token` is dropped, the id will be removed from the `running` set,
/// allowing subsequent acquisitions for the same id.
///
/// # Parameters
///
/// - `running`: Shared set tracking active ids.
/// - `my_id`: Identifier to acquire a token for.
///
/// # Returns
///
/// `Ok(Token)` if the id was successfully acquired, `Error` if a token for the id is already active.
async fn new (running: &Arc<Mutex<HashSet<i32>>>, my_id: i32) -> Result<Token> {
let running = running.clone();
let mut set = running.lock_arc().await;
if set.contains(&my_id) {
bail!("Token already taken");
} else {
set.insert(my_id);
Ok(Token {
running,
my_id,
})
}
}
}
impl Drop for Token {
/// Releases this token's claim on the shared running-set when the token is dropped.
///
/// The token's identifier is removed from the shared `running` set so that future
/// operations for the same id may proceed.
///
/// TODO: is using block_on inside block_on safe? Currently tested and working fine.
fn drop (&mut self) {
smol::block_on(async {
let mut set = self.running.lock_arc().await;
set.remove(&self.my_id);
})
}
}
#[derive(Clone)]
pub struct Core {
owner_chat: ChatPeerId,
// max_delay: u16,
pub tg: Client,
pub me: Bot,
pub db: Db,
sources: Arc<Mutex<HashSet<Arc<i32>>>>,
running: Arc<Mutex<HashSet<i32>>>,
http_client: reqwest::Client,
}
pub struct Post {
uri: String,
title: String,
authors: String,
summary: String,
}
impl Core {
/// Create a Core instance from configuration and start its background autofetch loop.
///
/// The provided `settings` must include:
/// - `owner` (integer): chat id to use as the default destination,
/// - `api_key` (string): Telegram bot API key,
/// - `api_gateway` (string): Telegram API gateway host,
/// - `pg` (string): PostgreSQL connection string,
/// - optional `proxy` (string): proxy URL for the HTTP client.
///
/// On success returns an initialized `Core` with Telegram and HTTP clients, database connection,
/// an empty running set for per-id tokens, and a spawned background task that periodically runs
/// `autofetch`. If any required setting is missing or initialization fails, an error is returned.
pub async fn new(settings: config::Config) -> Result<Core> {
let owner_chat = ChatPeerId::from(settings.get_int("owner").stack()?);
let api_key = settings.get_string("api_key").stack()?;
let tg = Client::new(&api_key).stack()?
.with_host(settings.get_string("api_gateway").stack()?);
let mut client = reqwest::Client::builder();
if let Ok(proxy) = settings.get_string("proxy") {
let proxy = reqwest::Proxy::all(proxy).stack()?;
client = client.proxy(proxy);
}
let http_client = client.build().stack()?;
let me = tg.execute(GetBot).await.stack()?;
let core = Core {
tg,
me,
owner_chat,
db: Db::new(&settings.get_string("pg").stack()?)?,
sources: Arc::new(Mutex::new(HashSet::new())),
running: Arc::new(Mutex::new(HashSet::new())),
http_client,
// max_delay: 60,
};
let clone = core.clone();
smol::spawn(Compat::new(async move {
loop {
let delay = match &clone.autofetch().await {
|
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
|
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
|
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
-
-
-
-
-
-
+
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
+
+
+
+
+
+
+
+
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
-
-
-
-
-
-
-
-
-
+
+
+
+
+
+
+
+
-
-
-
+
|
let target = target.unwrap_or(self.owner_chat);
self.tg.execute(
SendMessage::new(target, msg)
.with_parse_mode(mode)
).await.stack()
}
/// Fetches the feed for a source, sends any newly discovered posts to the appropriate chat, and records them in the database.
///
/// This acquires a per-source guard to prevent concurrent checks for the same `id`. If a check is already running for
/// the given `id`, the function returns an error. If `last_scrape` is provided, it is sent as the `If-Modified-Since`
/// header to the feed request. The function parses RSS or Atom feeds, sends unseen post URLs to either the source's
/// channel (when `real` is true) or the source owner (when `real` is false), and persists posted entries so they are
/// not reposted later.
///
/// Parameters:
/// - `id`: Identifier of the source to check.
/// - `real`: When `true`, send posts to the source's channel; when `false`, send to the source owner.
/// - `last_scrape`: Optional timestamp used to set the `If-Modified-Since` header for the HTTP request.
///
/// # Returns
///
/// `Posted: N` where `N` is the number of posts processed and sent.
pub async fn check (&self, id: i32, real: bool, last_scrape: Option<DateTime<Local>>) -> Result<String> {
let mut posted: i32 = 0;
let mut conn = self.db.begin().await.stack()?;
let id = {
let mut set = self.sources.lock_arc().await;
match set.get(&id) {
Some(id) => id.clone(),
None => {
let id = Arc::new(id);
let _token = Token::new(&self.running, id).await.stack()?;
set.insert(id.clone());
id.clone()
},
}
};
let count = Arc::strong_count(&id);
if count == 2 {
let source = conn.get_source(*id, self.owner_chat).await.stack()?;
conn.set_scrape(*id).await.stack()?;
let destination = ChatPeerId::from(match real {
true => source.channel_id,
false => source.owner,
});
let mut this_fetch: Option<DateTime<chrono::FixedOffset>> = None;
let mut posts: BTreeMap<DateTime<chrono::FixedOffset>, Post> = BTreeMap::new();
let source = conn.get_source(id, self.owner_chat).await.stack()?;
conn.set_scrape(id).await.stack()?;
let destination = ChatPeerId::from(match real {
true => source.channel_id,
false => source.owner,
});
let mut this_fetch: Option<DateTime<chrono::FixedOffset>> = None;
let mut posts: BTreeMap<DateTime<chrono::FixedOffset>, Post> = BTreeMap::new();
let mut builder = self.http_client.get(&source.url);
if let Some(last_scrape) = last_scrape {
builder = builder.header(LAST_MODIFIED, last_scrape.to_rfc2822());
};
let response = builder.send().await.stack()?;
{
let headers = response.headers();
let expires = headers.get(EXPIRES);
let cache = headers.get(CACHE_CONTROL);
if expires.is_some() || cache.is_some() {
println!("{} {} {:?} {:?} {:?}", Local::now().to_rfc2822(), &source.url, last_scrape, expires, cache);
}
}
let status = response.status();
let content = response.bytes().await.stack()?;
match rss::Channel::read_from(&content[..]) {
Ok(feed) => {
for item in feed.items() {
if let Some(link) = item.link() {
let date = match item.pub_date() {
Some(feed_date) => DateTime::parse_from_rfc2822(feed_date),
None => DateTime::parse_from_rfc3339(&item.dublin_core_ext().unwrap().dates()[0]),
}.stack()?;
let uri = link.to_string();
let title = item.title().unwrap_or("").to_string();
let authors = item.author().unwrap_or("").to_string();
let summary = item.content().unwrap_or("").to_string();
posts.insert(date, Post{
uri,
title,
authors,
summary,
});
}
};
},
Err(err) => match err {
rss::Error::InvalidStartTag => {
match atom_syndication::Feed::read_from(&content[..]) {
Ok(feed) => {
for item in feed.entries() {
let date = item.published().unwrap();
let uri = item.links()[0].href().to_string();
let title = item.title().to_string();
let authors = item.authors().iter().map(|x| format!("{} <{:?}>", x.name(), x.email())).collect::<Vec<String>>().join(", ");
let summary = if let Some(sum) = item.summary() { sum.value.clone() } else { String::new() };
posts.insert(*date, Post{
uri,
title,
authors,
summary,
});
};
},
Err(err) => {
bail!("Unsupported or mangled content:\n{:?}\n{err}\n{status:#?}\n", &source.url)
},
}
},
rss::Error::Eof => (),
_ => bail!("Unsupported or mangled content:\n{:?}\n{err}\n{status:#?}\n", &source.url)
}
};
for (date, post) in posts.iter() {
let post_url: Cow<str> = match source.url_re {
Some(ref x) => sedregex::ReplaceCommand::new(x).stack()?.execute(&post.uri),
None => post.uri.clone().into(),
};
if let Some(exists) = conn.exists(&post_url, *id).await.stack()? {
let mut builder = self.http_client.get(&source.url);
if let Some(last_scrape) = last_scrape {
builder = builder.header(LAST_MODIFIED, last_scrape.to_rfc2822());
};
let response = builder.send().await.stack()?;
#[cfg(debug_assertions)]
{
let headers = response.headers();
let expires = headers.get(EXPIRES);
let cache = headers.get(CACHE_CONTROL);
if expires.is_some() || cache.is_some() {
println!("{} {} {:?} {:?} {:?}", Local::now().to_rfc2822(), &source.url, last_scrape, expires, cache);
}
}
let status = response.status();
let content = response.bytes().await.stack()?;
match rss::Channel::read_from(&content[..]) {
Ok(feed) => {
for item in feed.items() {
if let Some(link) = item.link() {
let date = match item.pub_date() {
Some(feed_date) => DateTime::parse_from_rfc2822(feed_date),
None => DateTime::parse_from_rfc3339(match item.dublin_core_ext() {
Some(ext) => {
let dates = ext.dates();
if dates.is_empty() {
bail!("Feed item has Dublin Core extension but no dates.")
} else {
&dates[0]
}
},
None => bail!("Feed item misses posting date."),
}),
}.stack()?;
let uri = link.to_string();
let title = item.title().unwrap_or("").to_string();
let authors = item.author().unwrap_or("").to_string();
let summary = item.content().unwrap_or("").to_string();
posts.insert(date, Post{
uri,
title,
authors,
summary,
});
}
};
},
Err(err) => match err {
rss::Error::InvalidStartTag => {
match atom_syndication::Feed::read_from(&content[..]) {
Ok(feed) => {
for item in feed.entries() {
let date = item.published()
.stack_err("Feed item missing publishing date.")?;
let uri = {
let links = item.links();
if links.is_empty() {
bail!("Feed item missing post links.");
} else {
links[0].href().to_string()
}
};
let title = item.title().to_string();
let authors = item.authors().iter().map(|x| format!("{} <{:?}>", x.name(), x.email())).collect::<Vec<String>>().join(", ");
let summary = if let Some(sum) = item.summary() { sum.value.clone() } else { String::new() };
posts.insert(*date, Post{
uri,
title,
authors,
summary,
});
};
},
Err(err) => {
bail!("Unsupported or mangled content:\n{:?}\n{err}\n{status:#?}\n", &source.url)
},
}
},
rss::Error::Eof => (),
_ => bail!("Unsupported or mangled content:\n{:?}\n{err}\n{status:#?}\n", &source.url)
}
};
for (date, post) in posts.iter() {
let post_url: Cow<str> = match source.url_re {
Some(ref x) => sedregex::ReplaceCommand::new(x).stack()?.execute(&post.uri),
None => post.uri.clone().into(),
};
if ! conn.exists(&post_url, id).await.stack()? {
if ! exists {
if this_fetch.is_none() || *date > this_fetch.unwrap() {
this_fetch = Some(*date);
};
self.send( match &source.iv_hash {
Some(hash) => format!("<a href=\"https://t.me/iv?url={post_url}&rhash={hash}\"> </a>{post_url}"),
None => format!("{post_url}"),
}, Some(destination), Some(ParseMode::Html)).await.stack()?;
conn.add_post(*id, date, &post_url).await.stack()?;
if this_fetch.is_none() || *date > this_fetch.unwrap() {
this_fetch = Some(*date);
};
self.send( match &source.iv_hash {
Some(hash) => format!("<a href=\"https://t.me/iv?url={post_url}&rhash={hash}\"> </a>{post_url}"),
None => format!("{post_url}"),
}, Some(destination), Some(ParseMode::Html)).await.stack()?;
conn.add_post(id, date, &post_url).await.stack()?;
};
};
posted += 1;
};
posts.clear();
};
posts.clear();
Ok(format!("Posted: {posted}"))
}
async fn autofetch(&self) -> Result<std::time::Duration> {
let mut delay = chrono::Duration::minutes(1);
let now = chrono::Local::now();
let queue = {
|
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
|
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
|
-
+
-
+
|
owner_chat: ChatPeerId::from(owner),
..self.clone()
};
let source = {
let mut conn = self.db.begin().await.stack()?;
match conn.get_one(owner, source_id).await {
Ok(Some(source)) => source.to_string(),
Ok(None) => "Source not found in database.stack()?".to_string(),
Ok(None) => "Source not found in database?".to_string(),
Err(err) => format!("Failed to fetch source data:\n{err}"),
}
};
smol::spawn(Compat::new(async move {
if let Err(err) = clone.check(source_id, true, Some(last_scrape)).await {
if let Err(err) = clone.send(&format!("{source}\n\nš {}", encode(&err.to_string())), None, Some(ParseMode::MarkdownV2)).await {
if let Err(err) = clone.send(&format!("š {source}\n{}", encode(&err.to_string())), None, Some(ParseMode::MarkdownV2)).await {
eprintln!("Check error: {err}");
// clone.disable(&source_id, owner).await.unwrap();
};
};
})).detach();
}
} else if next_fetch - now < delay {
|