fix copied code to be applicable to this codebase

re-write scrape code so it actually works
Fix message counting logic
2024-07-25 13:17:03 -06:00 · 2024-07-25 13:16:46 -06:00 · 2024-07-25 13:15:31 -06:00 · 2024-07-25 10:56:45 -06:00
4 changed files with 202 additions and 39 deletions
--- a/Cargo.lock
+++ b/Cargo.lock
@ -355,6 +355,8 @@ dependencies = [
 "once_cell",
 "poise",
 "tokio",
+ "tracing",
+ "tracing-subscriber",
 ]

 [[package]]
@ -731,6 +733,12 @@ dependencies = [
 "wasm-bindgen",
 ]

+[[package]]
+name = "lazy_static"
+version = "1.5.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "bbd2bcb4c963f2ddae06a2efc7e9f3591312473c50c6685e1f298068316e66fe"
+
 [[package]]
 name = "libc"
 version = "0.2.155"
@ -759,6 +767,15 @@ version = "0.4.22"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "a7a70ba024b9dc04c27ea2f0c0548feb474ec5c54bba33a7f72f873a39d07b24"

+[[package]]
+name = "matchers"
+version = "0.1.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "8263075bb86c5a1b1427b5ae862e8889656f126e9f77c484496e8b47cf5c5558"
+dependencies = [
+ "regex-automata 0.1.10",
+]
+
 [[package]]
 name = "memchr"
 version = "2.7.4"
@ -816,6 +833,16 @@ dependencies = [
 "windows-sys 0.48.0",
 ]

+[[package]]
+name = "nu-ansi-term"
+version = "0.46.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "77a8165726e8236064dbb45459242600304b42a5ea24ee2948e18e023bf7ba84"
+dependencies = [
+ "overload",
+ "winapi",
+]
+
 [[package]]
 name = "num-conv"
 version = "0.1.0"
@ -856,6 +883,12 @@ version = "1.19.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "3fdb12b2476b595f9358c5161aa467c2438859caa136dec86c26fdd2efe17b92"

+[[package]]
+name = "overload"
+version = "0.1.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "b15813163c1d831bf4a13c3610c05c0d03b39feb07f7e09fa234dac9b15aaf39"
+
 [[package]]
 name = "parking_lot"
 version = "0.12.3"
@ -1014,8 +1047,17 @@ checksum = "b91213439dad192326a0d7c6ee3955910425f441d7038e0d6933b0aec5c4517f"
 dependencies = [
 "aho-corasick",
 "memchr",
- "regex-automata",
- "regex-syntax",
+ "regex-automata 0.4.7",
+ "regex-syntax 0.8.4",
+]
+
+[[package]]
+name = "regex-automata"
+version = "0.1.10"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "6c230d73fb8d8c1b9c0b3135c5142a8acee3a0558fb8db5cf1cb65f8d7862132"
+dependencies = [
+ "regex-syntax 0.6.29",
 ]

 [[package]]
@ -1026,9 +1068,15 @@ checksum = "38caf58cc5ef2fed281f89292ef23f6365465ed9a41b7a7754eb4e26496c92df"
 dependencies = [
 "aho-corasick",
 "memchr",
- "regex-syntax",
+ "regex-syntax 0.8.4",
 ]

+[[package]]
+name = "regex-syntax"
+version = "0.6.29"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "f162c6dd7b008981e4d40210aca20b4bd0f9b60ca9271061b07f78537722f2e1"
+
 [[package]]
 name = "regex-syntax"
 version = "0.8.4"
@ -1321,6 +1369,15 @@ dependencies = [
 "digest",
 ]

+[[package]]
+name = "sharded-slab"
+version = "0.1.7"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "f40ca3c46823713e0d4209592e8d6e826aa57e928f09752619fc696c499637f6"
+dependencies = [
+ "lazy_static",
+]
+
 [[package]]
 name = "skeptic"
 version = "0.13.7"
@ -1466,6 +1523,16 @@ dependencies = [
 "syn 2.0.72",
 ]

+[[package]]
+name = "thread_local"
+version = "1.1.8"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "8b9ef9bad013ada3808854ceac7b46812a6465ba368859a37e2100283d2d719c"
+dependencies = [
+ "cfg-if",
+ "once_cell",
+]
+
 [[package]]
 name = "time"
 version = "0.3.36"
@ -1626,6 +1693,36 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "c06d3da6113f116aaee68e4d601191614c9053067f9ab7f6edbcb161237daa54"
 dependencies = [
 "once_cell",
+ "valuable",
+]
+
+[[package]]
+name = "tracing-log"
+version = "0.2.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "ee855f1f400bd0e5c02d150ae5de3840039a3f54b025156404e34c23c03f47c3"
+dependencies = [
+ "log",
+ "once_cell",
+ "tracing-core",
+]
+
+[[package]]
+name = "tracing-subscriber"
+version = "0.3.18"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "ad0f048c97dbd9faa9b7df56362b8ebcaa52adb06b498c050d2f4e32f90a7a8b"
+dependencies = [
+ "matchers",
+ "nu-ansi-term",
+ "once_cell",
+ "regex",
+ "sharded-slab",
+ "smallvec",
+ "thread_local",
+ "tracing",
+ "tracing-core",
+ "tracing-log",
 ]

 [[package]]
@ -1756,6 +1853,12 @@ version = "0.7.6"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "09cc8ee72d2a9becf2f2febe0205bbed8fc6615b7cb429ad062dc7b7ddd036a9"

+[[package]]
+name = "valuable"
+version = "0.1.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "830b7e5d4d90034032940e4ace0d9a9a057e7a45cd94e6c007832e39edb82f6d"
+
 [[package]]
 name = "version_check"
 version = "0.9.4"
@ -1891,6 +1994,22 @@ dependencies = [
 "rustls-pki-types",
 ]

+[[package]]
+name = "winapi"
+version = "0.3.9"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "5c839a674fcd7a98952e593242ea400abe93992746761e38641405d28b00f419"
+dependencies = [
+ "winapi-i686-pc-windows-gnu",
+ "winapi-x86_64-pc-windows-gnu",
+]
+
+[[package]]
+name = "winapi-i686-pc-windows-gnu"
+version = "0.4.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "ac3b87c63620426dd9b991e5ce0329eff545bccbbb34f3be09ff6fb6ab51b7b6"
+
 [[package]]
 name = "winapi-util"
 version = "0.1.8"
@ -1900,6 +2019,12 @@ dependencies = [
 "windows-sys 0.52.0",
 ]

+[[package]]
+name = "winapi-x86_64-pc-windows-gnu"
+version = "0.4.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "712e227841d057c1ee1cd2fb22fa7e5a5461ae8e48fa2ca79ec42cfc1931183f"
+
 [[package]]
 name = "windows-core"
 version = "0.52.0"
--- a/Cargo.toml
+++ b/Cargo.toml
@ -15,3 +15,5 @@ poise = { version = "0.6", features = ["cache"] }
 dotenv = "0.15.0"
 anyhow = "1.0.75"
 once_cell = "1.19.0"
+tracing = "0.1.40"
+tracing-subscriber = { version = "0.3.18", features = ["env-filter"] }
--- a/src/command.rs
+++ b/src/command.rs
@ -2,7 +2,8 @@ use std::{collections::HashMap, fmt::Display, sync::Arc};

 use crate::Context;
 use anyhow::Error;
-use poise::serenity_prelude::{Cache, CacheHttp, ChannelId, ChannelType, GetMessages, GuildChannel, Http, Message};
+use poise::{serenity_prelude::{Cache, CacheHttp, ChannelId, ChannelType, GetMessages, GuildChannel, Http, Message}, CreateReply};
+use tracing::{debug, error, trace};

 struct Server {
    channels: Vec<Channel>,
@ -80,7 +81,6 @@ impl Server {
        }
    }
    
-    // TODO this might be broken
    fn search_by_id<'a>(target: &'a mut Vec<Channel>, find: &ChannelId) -> Option<&'a mut Channel> {
        for child in target {
            if child.this.id == *find {
@ -136,31 +136,47 @@ impl Server {

    /// Scrapes messages for all the channels in `self`.
    async fn scrape_all(&mut self) {
-        let settings = GetMessages::default().limit(5);
        let cache: (&Arc<Cache>, &Http) = (&Arc::new(Cache::new()), &Http::new(&crate::ENV.token));
-        walk_channels(&mut self.channels, cache, settings).await;
+        walk_channels(&mut self.channels, cache).await;

-        // recursive walk thru the channels
-        async fn walk_channels(all: &mut Vec<Channel>, cache: impl CacheHttp + Clone, settings: GetMessages) {
+        /// Recursive walk thru the channels
+        async fn walk_channels(all: &mut Vec<Channel>, cache: impl CacheHttp + Clone) {
+            let settings = GetMessages::default().limit(5);
            for channel in all {
-                // get the messages
-                match Box::pin(channel.this.messages(cache.clone(), settings)).await {
-                    Ok(mesgs) => {
-                        // store messages in our server object
-                        channel.messages = mesgs;
-                        if channel.messages.is_empty() {
-                            eprintln!("{} was empty - (Or incorrect permissions)", channel.this.name);
-                        }
-                    },
-                    Err(e) => {
-                        eprintln!("{}", e);
-                    },
-                }
                // Clone *should* be cheap - it's Arc under the hood
-                Box::pin(walk_channels(&mut channel.children, cache.clone(), settings)).await;
+                // get the messages
+                get_messages(channel, cache.clone(), settings).await;
+            }           
+
+            /// Get all messages for 1 channel and children
+            async fn get_messages(channel: &mut Channel, cache: impl CacheHttp + Clone, settings: GetMessages) {
+                // Loop thru all the messages in the channel in batches.
+                // Adding each batch to the current channel's messages the whole time.
+                let mut last_id = channel.this.last_message_id;
+                while let Some(last) = last_id {
+                    match channel.this.messages(cache.clone(), settings.before(last)).await {
+                        Ok(mut ok) => {
+                            if ok.is_empty() {
+                                debug!("Reached the beginning of {}", channel.this.name);
+                                // Stop the loop if there are no more messages.
+                                last_id = None;  
+                            } else {
+                                trace!("Adding {} messages to \"{}\"", ok.len(), channel.this.name);
+                                channel.messages.append(&mut ok);
+                                // extract message id
+                                if let Some(l) = channel.messages.last() {
+                                    last_id = Some(l.id);
+                                }
+                            }                        
+                        },
+                        Err(e) => error!("Error while trying to get messages - {e}"),
+                    }
+                }
+                // Then recurse into children channels
+                Box::pin(walk_channels(&mut channel.children, cache.clone())).await;
            }
        }
-   }
+    }

    /// Walk thru all the channels and count the saved messages. Will only give relevant data if
    /// done after `scrape_all()`.
@ -168,6 +184,7 @@ impl Server {
        fn walk(this: &Vec<Channel>) -> usize {
            let mut total = 0;
            for channel in this {
+                total += channel.messages.len();
                total += walk(&channel.children);
            };
            total
@ -181,10 +198,14 @@ pub async fn scrape_all(ctx: Context<'_>) -> Result<(), Error> {
    let guild = ctx.guild_id().unwrap().to_partial_guild(ctx.serenity_context()).await.unwrap();
    if let Ok(map) = guild.channels(ctx.http()).await {
        let mut server = index(map).await;
-        server.scrape_all().await;
-
-        let _ = ctx.reply(&format!("Scraped {} messages", server.message_count())).await;
-    }
+        match ctx.reply("Starting scrape...").await {
+            Ok(ok) => {
+                server.scrape_all().await;
+                let _ = ok.edit(ctx, CreateReply::default().content(&format!("Scraped {} messages", server.message_count()))).await;
+            },
+            Err(e) => error!("{e} - While trying to reply to scrape command"),
+        }
+   }
    Ok(())
 }

@ -195,8 +216,8 @@ async fn index(map: HashMap<ChannelId, GuildChannel>) -> Server {
    map.into_iter().for_each(|(_id, current)| {
        // println!("{} {} {:?}", current.name, current.id, current.parent_id);
        server.add(current);
-        // TODO take note of position
-        // Take node of vc user limit
+        // TODO Take note of position
+        // TODO Take node of vc user limit
    });
    server.clean();
    server
--- a/src/main.rs
+++ b/src/main.rs
@ -1,5 +1,7 @@
 use once_cell::sync::Lazy;
 use poise::serenity_prelude::{self as serenity, GatewayIntents};
+use tracing::{debug, error, info, warn, Level};
+use tracing_subscriber::EnvFilter;
 mod command;

 pub struct Data {} // User data, which is stored and accessible in all command invocations
@ -12,16 +14,31 @@ static ENV: Lazy<BotEnv> = Lazy::new(|| {
 #[tokio::main]
 async fn main() {

+    // Start the tracing subscriber
+    let filter = EnvFilter::builder()
+            .parse("discord_egress=trace,tokio=warn")
+            .expect("Could not create env filter.")
+            ;
+
+    tracing_subscriber::fmt::fmt()
+        .with_max_level(Level::TRACE)
+        .with_target(true)
+        .with_env_filter(filter)
+        .with_thread_ids(false)
+        .with_file(false)
+        .without_time()
+        .init();
+
    // Generate sick text like this: 
    // http://www.patorjk.com/software/taag/#p=testall&f=Graffiti&t=hello%20world
-    println!(r#"
+    info!(r#"
    Invite this bot with:
    "#);
-    println!("https://discord.com/api/oauth2/authorize?client_id={}&permissions={}&scope=bot",
+    info!("https://discord.com/api/oauth2/authorize?client_id={}&permissions={}&scope=bot",
        ENV.id,
        ENV.intents.bits(),
    );
-    print!("\n");
+    info!("\n");

    // Setup framework
    let framework = poise::Framework::builder()
@ -65,9 +82,7 @@ fn read_env() -> BotEnv {
    // ==================== ID ===========================
    let id: String = std::env::var(DISCORD_ID)
        .unwrap_or_else(|_| {
-            println!(r#"
-WARN: Missing {DISCORD_ID}
-> This isn't really that problematic, just that the generated invite link won't work."#);
+            warn!("Missing {DISCORD_ID} This isn't really that problematic, just that the generated invite link won't work.");
            String::from("")
        });
    // ==================== Token ========================
@ -76,8 +91,8 @@ WARN: Missing {DISCORD_ID}
    // ==================== Intents ======================
    let intents_env: String = std::env::var(DISCORD_INTENTS)
        .unwrap_or_else(|msg| {
-            println!("DEBUG: what is: {msg}");
-            println!("ERROR: Missing {DISCORD_INTENTS}");
+            debug!("What is: {msg}");
+            error!("Missing {DISCORD_INTENTS}");
            "0".to_string()
        });
    // ==================== Parse Intents =================
@ -85,7 +100,7 @@ WARN: Missing {DISCORD_ID}
    let intents_u64 = intents_env.parse::<u64>().unwrap_or(0u64);
    let intents_truncated: GatewayIntents = GatewayIntents::from_bits_truncate(intents_u64);
    if intents_truncated.bits() != intents_u64 {
-        println!("WARN: Intents integer got truncated from {} to {}!", intents_u64, intents_truncated.bits())
+        warn!("Intents integer got truncated from {} to {}!", intents_u64, intents_truncated.bits())
    };
    BotEnv {
        intents: intents_truncated,
Author	SHA1	Message	Date
Oliver Atkinson	4d3a41db43	fix copied code to be applicable to this codebase	2024-07-25 13:17:03 -06:00
Oliver Atkinson	54c533abff	re-write scrape code so it actually works	2024-07-25 13:16:46 -06:00
Oliver Atkinson	b7c9a9b0e3	Fix message counting logic	2024-07-25 13:15:31 -06:00
Oliver Atkinson	79638ed324	add tracing	2024-07-25 10:56:45 -06:00