Fix tracing span, using official mwbot crate

Cargo.toml

@@ -1,6 +1,6 @@
 [package]
-name = "dancing-bot-teachers"
-version = "0.1.0"
+name = "teachertracker-rs"
+version = "0.1.3"
 edition = "2024"
 authors = ["Lukas Wölfer <coding@thasky.one>"]
 description = "A MediaWiki bot that updates score information of teachers"
@@ -13,7 +13,8 @@ categories = ["web-programming", "api-bindings", "automation"]
 [dependencies]
 chrono = "0.4.41"
 futures = "0.3.31"
-mwbot = { git = "https://gitlab.wikimedia.org/repos/mwbot-rs/mwbot.git", rev = "05cbb12188f18e2da710de158d89a9a4f1b42689", default-features = false, features = ["generators", "mwbot_derive"] }
+# mwbot = { git = "https://gitlab.wikimedia.org/repos/mwbot-rs/mwbot.git", rev = "05cbb12188f18e2da710de158d89a9a4f1b42689", default-features = false, features = ["generators", "mwbot_derive"] }
+mwbot = { version = "0.7.0", default-features = false, features = ["generators", "mwbot_derive"] }
 rand = "0.9.2"
 reqwest = "0.12.22"
 serde = { version = "1.0.219", features = ["derive"] }

src/dance_info.rs

@@ -51,6 +51,8 @@ pub struct CompState {
     pub rank: DanceRank,
     pub points: u16,
 }
+
+#[derive(Debug)]
 pub struct DanceInfo {
     pub firstname: String,
     pub lastname: String,

src/main.rs

@@ -13,11 +13,13 @@
     reason = "Disable this for most of the time, enable this for cleanup later"
 )]
 
+#![feature(never_type)]
+
 use mwbot::{
-    Bot,
+    Bot, ConfigError,
     generators::{Generator, SortDirection, categories::CategoryMemberSort},
 };
-use std::{error::Error, path::Path};
+use std::path::Path;
 
 use crate::watchdog::watch_wanted;
 
@@ -41,30 +43,37 @@ fn list_teacher_pages(bot: &Bot) -> tokio::sync::mpsc::Receiver<Result<mwbot::Pa
     pages.generate(bot)
 }
 
-fn main() {
+#[must_use]
+pub fn app_signature() -> String {
+    format!("{} [{}]", env!("CARGO_PKG_NAME"), env!("CARGO_PKG_VERSION"))
+}
+
+#[derive(thiserror::Error, Debug)]
+pub enum AppError {
+    #[error("Runtime error: {0}")]
+    RuntimeError(#[from] std::io::Error),
+    #[error("Bot initialization error: {0}")]
+    BotError(#[from] ConfigError),
+}
+
+fn main() -> Result<(), AppError> {
     tracing_subscriber::fmt()
         .with_level(true)
         .with_max_level(tracing::Level::INFO)
         .init();
+    tracing::info!("Starting {}", app_signature());
 
-    let rt = match tokio::runtime::Builder::new_current_thread()
+    let rt = tokio::runtime::Builder::new_current_thread()
         .enable_all()
-        .build()
-    {
-        Ok(o) => o,
-        Err(e) => {
-            tracing::error!("Could not start runtime: {e}");
-            return;
-        }
-    };
-    rt.block_on(async {
-        let bot = match Bot::from_path(Path::new("./mwbot.toml")).await {
-            Ok(x) => x,
-            Err(e) => {
-                dbg!(e);
-                return;
-            }
-        };
-        futures::join!(watch_wanted(bot.clone()), updater::update_wsdc(bot));
-    });
+        .build()?;
+
+    let bot = rt.block_on(Bot::from_path(Path::new("./mwbot.toml")))?;
+
+    #[allow(
+        unreachable_code,
+        reason = "This is a false positive I think, I just want to loop infinitely on two futures"
+    )]
+    rt.block_on(async { futures::join!(watch_wanted(bot.clone()), updater::update_wsdc(bot)) });
+
+    Ok(())
 }

src/updater.rs

@@ -2,23 +2,38 @@ use std::time::Duration;
 
 use mwbot::Bot;
 use rand::seq::SliceRandom as _;
+use tokio::time::sleep;
 
 use crate::{watchdog::generate_page, wikiinfo::index_wsdc_ids};
 
 pub async fn update_wsdc(bot: Bot) -> ! {
     loop {
-        let mut l = index_wsdc_ids(&bot).await;
-        l.shuffle(&mut rand::rng());
-        tracing::info!("We have to update {} pages", l.len());
-        let wait_duration = Duration::from_secs(6 * 3600);
-        for (index, page) in l {
-            tracing::info!("Next up: {index}");
-            tokio::time::sleep(wait_duration).await;
-            if generate_page(index, page).await {
-                tracing::info!("Updated {index}");
-            } else {
-                tracing::error!("Error updating {index}");
-            }
+        update_all_teachers(&bot).await;
+    }
+}
+
+/// Updates all teachers once
+async fn update_all_teachers(bot: &Bot) {
+    let mut l = index_wsdc_ids(bot).await;
+    l.shuffle(&mut rand::rng());
+    tracing::info!("We have to update {} pages", l.len());
+    let wait_duration = Duration::from_hours(6);
+
+    for (index, page) in l {
+        process_page(wait_duration, index, page).await;
+    }
+    tracing::info!("Updates all pages");
+}
+
+#[tracing::instrument(skip(page, wait_duration))]
+async fn process_page(wait_duration: Duration, index: u32, page: mwbot::Page) {
+    tracing::info!("Next up");
+    sleep(wait_duration).await;
+
+    match generate_page(index, page).await {
+        Ok(()) => (),
+        Err(err) => {
+            tracing::error!("Error updating: {err}");
         }
     }
 }

src/watchdog.rs

@@ -1,65 +1,92 @@
 use std::time::Duration;
 
+use crate::app_signature;
+use crate::wikipage::InfoCompileError;
+use crate::worldsdc::DanceInfoError;
 use crate::{wikiinfo::wanted_ids, wikipage::page_from_info, worldsdc::fetch_wsdc_info};
-use mwbot::Bot;
 use mwbot::SaveOptions;
-use tracing::Level;
+use mwbot::{Bot, Page};
 
-pub async fn watch_wanted(bot: Bot) -> ! {
-    let span = tracing::span!(Level::INFO, "wanted_watchdog");
-    let _enter = span.enter();
-
-    let mut ignored_ids = vec![];
-    loop {
-        if ignored_ids.is_empty() {
-            tracing::info!("Watchdog check...");
-        } else {
-            tracing::info!("Watchdog check [{} failed ids]...", ignored_ids.len());
-        }
-        let wanted = wanted_ids(bot.clone()).await;
-        let mut new_ignored = vec![];
-        for (id, page) in wanted.into_iter().filter(|(x, _)| ignored_ids.contains(x)) {
-            if !generate_page(id, page).await {
-                new_ignored.push(id);
-            }
-        }
-        if !new_ignored.is_empty() {
-            ignored_ids.extend(new_ignored);
-        }
-        tokio::time::sleep(Duration::from_secs(30)).await;
-    }
+pub struct Ticker {
+    count: usize,
+    max: usize,
 }
 
-pub async fn generate_page(id: u32, page: mwbot::Page) -> bool {
-    tracing::info!("Generating page for {id}");
-    let info = match fetch_wsdc_info(id).await {
-        Ok(o) => o,
-        Err(e) => {
-            tracing::error!("Error fetching wsdc info for {id}: {e}");
-            return false;
-        }
-    };
-    let code = match page_from_info(info) {
-        Ok(o) => o,
-        Err(e) => {
-            tracing::error!("Creating wikicode for {id}: {e}");
-            return false;
-        }
-    };
+impl Ticker {
+    pub const fn new(max: usize) -> Self {
+        Self { count: 0, max }
+    }
 
-    match page
-        .save(
-            code,
-            &SaveOptions::summary("Created WSDC info from worldsdc.com")
-                .mark_as_bot(true)
-                .mark_as_minor(false),
-        )
-        .await
-    {
-        Ok(_) => true,
-        Err(e) => {
-            tracing::error!("Could not save page for {id}: {e}");
+    /// Returns `true` if the ticker has "ticked" (i.e., reached `max` and reset).
+    pub const fn tick(&mut self) -> bool {
+        self.count += 1;
+        if self.count >= self.max {
+            self.count = 0;
+            true
+        } else {
             false
         }
     }
 }
+
+/// Continuously monitors teacher IDs that do not have a corresponding teacher WSDC page, ignoring those that fail processing.
+#[tracing::instrument(skip_all)]
+pub async fn watch_wanted(bot: Bot) -> ! {
+    let mut ignored_ids = vec![];
+    let mut heartbeat_ticker = Ticker::new(120);
+    loop {
+        if heartbeat_ticker.tick() {
+            tracing::info!(failed_id_count = ignored_ids.len(), "Watchdog check...");
+        }
+        let wanted = wanted_ids(bot.clone()).await;
+        let new_ignored = update_wanted_ids(&wanted, &ignored_ids).await;
+        ignored_ids.extend(new_ignored);
+        tokio::time::sleep(Duration::from_secs(30)).await;
+    }
+}
+
+async fn update_wanted_ids(wanted: &[(u32, Page)], ignored_ids: &[u32]) -> Vec<u32> {
+    let mut new_ignored = vec![];
+
+    for (id, page) in wanted.iter().filter(|(x, _)| ignored_ids.contains(x)) {
+        let span = tracing::info_span!("update", id);
+        let _enter = span.enter();
+        if let Err(e) = generate_page(*id, page.clone()).await {
+            tracing::error!("{e}");
+            new_ignored.push(*id);
+        }
+    }
+    new_ignored
+}
+
+use thiserror::Error;
+
+#[derive(Error, Debug)]
+pub enum GeneratePageError {
+    #[error("Error fetching WSDC info for {0}")]
+    Fetch(#[from] DanceInfoError),
+    #[error("Error creating wikicode for {0}")]
+    Wikicode(#[from] InfoCompileError),
+    #[error("Error saving page for {0}")]
+    Save(#[from] mwbot::Error),
+}
+
+pub async fn generate_page(id: u32, page: mwbot::Page) -> Result<(), GeneratePageError> {
+    tracing::info!("Generating page for {id}");
+    let info = fetch_wsdc_info(id).await?;
+
+    let code = page_from_info(info)?;
+
+    page.save(
+        code,
+        &SaveOptions::summary(&format!(
+            "Created WSDC info from worldsdc.com -- {}",
+            app_signature()
+        ))
+        .mark_as_bot(true)
+        .mark_as_minor(false),
+    )
+    .await?;
+
+    Ok(())
+}

src/worldsdc/mod.rs

@@ -2,23 +2,34 @@ use std::collections::HashMap;
 
 use reqwest::ClientBuilder;
 
-use crate::dance_info::{CompState, DanceInfo, DanceRank, DanceRole};
+use crate::{
+    app_signature,
+    dance_info::{CompState, DanceInfo, DanceRank, DanceRole},
+};
 
 // mod caching;
 pub async fn fetch_wsdc_info(id: u32) -> Result<DanceInfo, DanceInfoError> {
     let client = ClientBuilder::new()
+        .user_agent(app_signature())
         .build()
         .map_err(DanceInfoError::ClientBuild)?;
 
     let mut params = HashMap::new();
-    params.insert("q", id.to_string());
-    let response = client
-        .request(
-            reqwest::Method::POST,
-            "https://points.worldsdc.com/lookup2020/find",
-        )
+
+    let url = if cfg!(test) {
+        // "https://o5grQU3Y.free.beeceptor.com/lookup2020/find"
+        "http://localhost:8000"
+    } else {
+        "https://points.worldsdc.com/lookup2020/find"
+    };
+    params.insert("num", id.to_string());
+    let request = client
+        .request(reqwest::Method::POST, url)
         .form(&params)
-        .send()
+        .build()
+        .map_err(DanceInfoError::RequestBuild)?;
+    let response = client
+        .execute(request)
         .await
         .map_err(DanceInfoError::Request)?;
 
@@ -26,10 +37,36 @@ pub async fn fetch_wsdc_info(id: u32) -> Result<DanceInfo, DanceInfoError> {
     Ok(x.into())
 }
 
+#[cfg(test)]
+mod tests {
+    #![allow(clippy::unwrap_used, reason = "Allow unwrap in tests")]
+    use crate::worldsdc::fetch_wsdc_info;
+
+    #[test]
+    #[ignore = "Only run when the mock api is setup"]
+    fn test_fetch_wsdc() {
+        let rt = match tokio::runtime::Builder::new_current_thread()
+            .enable_all()
+            .build()
+        {
+            Ok(o) => o,
+            Err(e) => {
+                tracing::error!("Could not start runtime: {e}");
+                return;
+            }
+        };
+        let x = rt.block_on(fetch_wsdc_info(7));
+        dbg!(&x);
+        x.unwrap();
+    }
+}
+
 #[derive(thiserror::Error, Debug)]
 pub enum DanceInfoError {
     #[error("Failed to build client: {0}")]
     ClientBuild(reqwest::Error),
+    #[error("Failed to build request: {0}")]
+    RequestBuild(reqwest::Error),
     #[error("Request error: {0}")]
     Request(reqwest::Error),
     #[error("Failed to parse response: {0}")]