diff --git a/Cargo.lock b/Cargo.lock
index b845814..59c9470 100644
--- a/Cargo.lock
+++ b/Cargo.lock
@@ -114,16 +114,26 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "5b63caa9aa9397e2d9480a9b13673856c78d8ac123288526c37d7839f2a86990"
[[package]]
-name = "env_logger"
-version = "0.10.2"
+name = "env_filter"
+version = "0.1.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "4cd405aab171cb85d6735e5c8d9db038c17d3ca007a4d2c25f337935c3d90580"
+checksum = "186e05a59d4c50738528153b83b0b0194d3a29507dfec16eccd4b342903397d0"
dependencies = [
- "humantime",
- "is-terminal",
"log",
"regex",
- "termcolor",
+]
+
+[[package]]
+name = "env_logger"
+version = "0.11.7"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "c3716d7a920fb4fac5d84e9d4bce8ceb321e9414b4409da61b07b75c1e3d0697"
+dependencies = [
+ "anstream",
+ "anstyle",
+ "env_filter",
+ "jiff",
+ "log",
]
[[package]]
@@ -168,29 +178,6 @@ version = "0.5.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "2304e00983f87ffb38b55b444b5e3b60a884b5d30c0fca7d82fe33449bbe55ea"
-[[package]]
-name = "hermit-abi"
-version = "0.5.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "fbd780fe5cc30f81464441920d82ac8740e2e46b29a6fad543ddd075229ce37e"
-
-[[package]]
-name = "humantime"
-version = "2.2.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "9b112acc8b3adf4b107a8ec20977da0273a8c386765a3ec0229bd500a1443f9f"
-
-[[package]]
-name = "is-terminal"
-version = "0.4.16"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "e04d7f318608d35d4b61ddd75cbdaee86b023ebe2bd5a66ee0915f0bf93095a9"
-dependencies = [
- "hermit-abi",
- "libc",
- "windows-sys",
-]
-
[[package]]
name = "is_terminal_polyfill"
version = "1.70.1"
@@ -204,10 +191,28 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "4a5f13b858c8d314ee3e8f639011f7ccefe71f97f96e50151fb991f267928e2c"
[[package]]
-name = "libc"
-version = "0.2.171"
+name = "jiff"
+version = "0.2.5"
source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "c19937216e9d3aa9956d9bb8dfc0b0c8beb6058fc4f7a4dc4d850edf86a237d6"
+checksum = "c102670231191d07d37a35af3eb77f1f0dbf7a71be51a962dcd57ea607be7260"
+dependencies = [
+ "jiff-static",
+ "log",
+ "portable-atomic",
+ "portable-atomic-util",
+ "serde",
+]
+
+[[package]]
+name = "jiff-static"
+version = "0.2.5"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "4cdde31a9d349f1b1f51a0b3714a5940ac022976f4b49485fc04be052b183b4c"
+dependencies = [
+ "proc-macro2",
+ "quote",
+ "syn",
+]
[[package]]
name = "libsqlite3-sys"
@@ -244,13 +249,18 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "7edddbd0b52d732b21ad9a5fab5c704c14cd949e5e9a1ec5929a24fded1b904c"
[[package]]
-name = "pretty_env_logger"
-version = "0.5.0"
+name = "portable-atomic"
+version = "1.11.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "865724d4dbe39d9f3dd3b52b88d859d66bcb2d6a0acfd5ea68a65fb66d4bdc1c"
+checksum = "350e9b48cbc6b0e028b0473b114454c6316e57336ee184ceab6e53f72c178b3e"
+
+[[package]]
+name = "portable-atomic-util"
+version = "0.2.4"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "d8a2f0d8d040d7848a709caf78912debcc3f33ee4b3cac47d73d1e1069e83507"
dependencies = [
- "env_logger",
- "log",
+ "portable-atomic",
]
[[package]]
@@ -363,11 +373,12 @@ name = "smogon-stats"
version = "0.1.0"
dependencies = [
"clap",
+ "env_logger",
"log",
- "pretty_env_logger",
"rusqlite",
"serde",
"serde_json",
+ "thiserror",
]
[[package]]
@@ -388,12 +399,23 @@ dependencies = [
]
[[package]]
-name = "termcolor"
-version = "1.4.1"
+name = "thiserror"
+version = "2.0.12"
source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "06794f8f6c5c898b3275aebefa6b8a1cb24cd2c6c79397ab15774837a0bc5755"
+checksum = "567b8a2dae586314f7be2a752ec7474332959c6460e02bde30d702a66d488708"
dependencies = [
- "winapi-util",
+ "thiserror-impl",
+]
+
+[[package]]
+name = "thiserror-impl"
+version = "2.0.12"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "7f7cf42b4507d8ea322120659672cf1b9dbb93f8f2d4ecfd6e51350ff5b17a1d"
+dependencies = [
+ "proc-macro2",
+ "quote",
+ "syn",
]
[[package]]
@@ -414,15 +436,6 @@ version = "0.2.15"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "accd4ea62f7bb7a82fe23066fb0957d48ef677f6eeb8215f372f52e48bb32426"
-[[package]]
-name = "winapi-util"
-version = "0.1.9"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "cf221c93e13a30d793f7645a0e7762c55d169dbb0a49671918a2319d289b10bb"
-dependencies = [
- "windows-sys",
-]
-
[[package]]
name = "windows-sys"
version = "0.59.0"
diff --git a/Cargo.toml b/Cargo.toml
index f6b3d22..151630b 100644
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -2,11 +2,14 @@
name = "smogon-stats"
version = "0.1.0"
edition = "2024"
+authors = ["mehbark <terezi@pyrope.net>"]
+description = "turn https://smogon.com/stats/ chaos json files into https://sqlite.org databases"
[dependencies]
clap = { version = "4.5.32", features = ["derive"] }
+env_logger = "0.11.7"
log = "0.4.27"
-pretty_env_logger = "0.5.0"
rusqlite = "0.34.0"
serde = { version = "1.0.219", features = ["derive"] }
serde_json = "1.0.140"
+thiserror = "2.0.12"
diff --git a/src/main.rs b/src/main.rs
index 64f7a62..8031bae 100644
--- a/src/main.rs
+++ b/src/main.rs
@@ -1,22 +1,82 @@
// https://www.smogon.com/stats/2025-02/chaos/
-use std::{collections::HashMap, error::Error, fs::File, path::PathBuf};
+use std::{
+ collections::HashMap,
+ fs::{self, File},
+ io,
+ path::PathBuf,
+ process,
+};
use clap::Parser;
use rusqlite::Connection;
use serde::Deserialize;
+use thiserror::Error;
-fn main() -> Result<(), Box<dyn Error>> {
- let Args {
+fn main() {
+ env_logger::Builder::from_env(env_logger::Env::default().default_filter_or("warn")).init();
+ let config = Config::parse();
+ if let Err(e) = run(config) {
+ log::error!("{e}");
+ process::exit(1);
+ };
+}
+
+fn run(
+ Config {
input_file,
output_file,
- } = Args::parse();
- pretty_env_logger::init_timed();
+ }: Config,
+) -> Result<(), AppError> {
+ if fs::exists(&output_file)? {
+ log::warn!("{output_file:?} already exists: creating the db will probably fail");
+ }
+
+ // parsing the stats is slow but doing it first avoids creating empty tables
+ log::info!("parsing stats");
+ let input_file = File::open(input_file)?;
+ let stats: Stats = serde_json::from_reader(input_file)?;
+
+ log::info!(
+ "meta: {}: {} battles, cutoff: {}",
+ stats.info.metagame,
+ stats.info.battle_count,
+ stats.info.cutoff
+ );
log::info!("opening connection");
let mut conn = Connection::open(output_file)?;
+
+ create_tables(&mut conn)?;
+ insert_stats(&mut conn, &stats)?;
+
+ Ok(())
+}
+
+#[derive(clap::Parser)]
+#[command(about)]
+struct Config {
+ #[arg()]
+ input_file: PathBuf,
+ #[arg(short, long = "output")]
+ output_file: PathBuf,
+}
+
+#[derive(Error, Debug)]
+enum AppError {
+ #[error("Error creating sqlite db: {0}")]
+ Sql(#[from] rusqlite::Error),
+ #[error("IO error: {0}")]
+ Io(#[from] io::Error),
+ #[error("Error reading JSON: {0}")]
+ Deserialize(#[from] serde_json::Error),
+}
+
+fn create_tables(conn: &mut Connection) -> rusqlite::Result<()> {
log::info!("creating tables");
- // TODO: checks and counters, specifically KO or switch
+ // we could do IF NOT EXISTS, but these are meant to be ephemeral+read-only;
+ // i'd like to be able to evolve and improve the schema, so y'all'll have to
+ // take the ten seconds to delete and remake the db SORRY! things'll change!
conn.execute_batch(
"
BEGIN;
@@ -58,32 +118,24 @@ fn main() -> Result<(), Box<dyn Error>> {
);
COMMIT;
",
- )?;
-
- log::info!("parsing stats");
- let input_file = File::open(input_file)?;
- let stats: Stats = serde_json::from_reader(input_file)?;
- log::info!(
- "meta: {}: {} battles, cutoff: {}",
- stats.info.metagame,
- stats.info.battle_count,
- stats.info.cutoff
- );
-
- insert_stats(&mut conn, &stats)?;
-
- Ok(())
+ )
}
fn insert_stats(conn: &mut Connection, stats: &Stats) -> rusqlite::Result<()> {
+ let mon_count = stats.data.len();
+ let mon_count_digits = mon_count.to_string().len();
+
for (i, (mon, data)) in stats.data.iter().enumerate() {
- log::debug!("Processing mon #{i} ({mon})…");
+ log::debug!(
+ "Processing mon {:mon_count_digits$}/{mon_count} ({mon})",
+ i + 1
+ );
let tx = conn.transaction()?;
// normalizing with mon_count gives us data that is much easier to work
- // with. for example, if pikachu has 10k raw count and thunderbolt is used
- // 9k times (weighted), we'd like 0.9 so that we can say pikachu runs
- // thunderbolt 90% of the time.
+ // with. for example, if pikachu has 10k count (weighted) and thunderbolt
+ // is used 9k times (weighted), we'd like 0.9 so that we can say pikachu
+ // runs thunderbolt 90% of the time.
// HACK: we get the weighted mon count by summing the ability usage
// there's is 1000% a better way to do this
@@ -144,14 +196,6 @@ fn insert_stats(conn: &mut Connection, stats: &Stats) -> rusqlite::Result<()> {
Ok(())
}
-#[derive(clap::Parser)]
-struct Args {
- #[arg()]
- input_file: PathBuf,
- #[arg(short, long = "output")]
- output_file: PathBuf,
-}
-
#[derive(Debug, Deserialize, PartialEq, Eq, Hash, Clone, Copy)]
#[serde(rename_all = "lowercase")]
enum Type {