feat: add basic webui

This commit is contained in:
2024-12-30 06:39:09 +08:00
parent 608a7fb9c6
commit a4c549e7c3
462 changed files with 35900 additions and 2491 deletions

View File

@@ -0,0 +1,8 @@
FROM mcr.microsoft.com/vscode/devcontainers/rust:0-1
RUN apt-get update && export DEBIAN_FRONTEND=noninteractive \
&& apt-get -y install --no-install-recommends postgresql-client \
&& cargo install sea-orm-cli cargo-insta \
&& chown -R vscode /usr/local/cargo
COPY .env /.env

View File

@@ -0,0 +1,9 @@
{
"name": "Konobangu Recorder",
"dockerComposeFile": "docker-compose.yml",
"service": "app",
"workspaceFolder": "/workspaces/${localWorkspaceFolderBasename}",
"forwardPorts": [
3001
]
}

View File

@@ -0,0 +1,40 @@
version: "3"
services:
app:
build:
context: .
dockerfile: Dockerfile
command: sleep infinity
networks:
- db
- redis
volumes:
- ../..:/workspaces:cached
env_file:
- .env
db:
image: postgres:15.3-alpine
restart: unless-stopped
ports:
- 5432:5432
networks:
- db
volumes:
- postgres-data:/var/lib/postgresql/data
env_file:
- .env
redis:
image: redis:latest
restart: unless-stopped
ports:
- 6379:6379
networks:
- redis
volumes:
postgres-data:
networks:
db:
redis:

107
apps/recorder/.github/workflows/ci.yaml vendored Normal file
View File

@@ -0,0 +1,107 @@
name: CI
on:
push:
branches:
- master
- main
pull_request:
env:
RUST_TOOLCHAIN: stable
TOOLCHAIN_PROFILE: minimal
jobs:
rustfmt:
name: Check Style
runs-on: ubuntu-latest
permissions:
contents: read
steps:
- name: Checkout the code
uses: actions/checkout@v4
- uses: actions-rs/toolchain@v1
with:
profile: ${{ env.TOOLCHAIN_PROFILE }}
toolchain: ${{ env.RUST_TOOLCHAIN }}
override: true
components: rustfmt
- name: Run cargo fmt
uses: actions-rs/cargo@v1
with:
command: fmt
args: --all -- --check
clippy:
name: Run Clippy
runs-on: ubuntu-latest
permissions:
contents: read
steps:
- name: Checkout the code
uses: actions/checkout@v4
- uses: actions-rs/toolchain@v1
with:
profile: ${{ env.TOOLCHAIN_PROFILE }}
toolchain: ${{ env.RUST_TOOLCHAIN }}
override: true
- name: Setup Rust cache
uses: Swatinem/rust-cache@v2
- name: Run cargo clippy
uses: actions-rs/cargo@v1
with:
command: clippy
args: --all-features -- -D warnings -W clippy::pedantic -W clippy::nursery -W rust-2018-idioms
test:
name: Run Tests
runs-on: ubuntu-latest
permissions:
contents: read
services:
redis:
image: redis
options: >-
--health-cmd "redis-cli ping"
--health-interval 10s
--health-timeout 5s
--health-retries 5
ports:
- "6379:6379"
postgres:
image: postgres
env:
POSTGRES_DB: postgress_test
POSTGRES_USER: postgress
POSTGRES_PASSWORD: postgress
ports:
- "5432:5432"
# Set health checks to wait until postgres has started
options: --health-cmd pg_isready
--health-interval 10s
--health-timeout 5s
--health-retries 5
steps:
- name: Checkout the code
uses: actions/checkout@v4
- uses: actions-rs/toolchain@v1
with:
profile: ${{ env.TOOLCHAIN_PROFILE }}
toolchain: ${{ env.RUST_TOOLCHAIN }}
override: true
- name: Setup Rust cache
uses: Swatinem/rust-cache@v2
- name: Run cargo test
uses: actions-rs/cargo@v1
with:
command: test
args: --all-features --all
env:
REDIS_URL: redis://localhost:${{job.services.redis.ports[6379]}}
DATABASE_URL: postgres://postgress:postgress@localhost:5432/postgress_test

17
apps/recorder/.gitignore vendored Normal file
View File

@@ -0,0 +1,17 @@
**/config/local.yaml
**/config/*.local.yaml
# Generated by Cargo
# will have compiled files and executables
debug/
target/
# Remove Cargo.lock from gitignore if creating an executable, leave it for libraries
# More information here https://doc.rust-lang.org/cargo/guide/cargo-toml-vs-cargo-lock.html
Cargo.lock
# These are backup files generated by rustfmt
**/*.rs.bk
# MSVC Windows builds of rustc generate these, which store debugging information
*.pdb

64
apps/recorder/Cargo.toml Normal file
View File

@@ -0,0 +1,64 @@
[package]
name = "recorder"
version = "0.1.0"
edition = "2021"
# See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html
[lib]
name = "recorder"
path = "src/lib.rs"
[[bin]]
name = "recorder_cli"
path = "src/bin/main.rs"
required-features = []
[dependencies]
quirks_path = { path = "../../packages/quirks-path" }
torrent = { path = "../../packages/torrent" }
loco-rs = { version = "0.13" }
serde = { version = "1", features = ["derive"] }
serde_json = "1"
eyre = "0.6"
tokio = { version = "1.42", default-features = false }
async-trait = "0.1.83"
tracing = "0.1"
chrono = "0.4"
sea-orm = { version = "1", features = [
"sqlx-sqlite",
"sqlx-postgres",
"runtime-tokio-rustls",
"macros",
"debug-print"
] }
axum = "0.7.9"
uuid = { version = "1.6.0", features = ["v4"] }
tracing-subscriber = { version = "0.3", features = ["env-filter", "json"] }
sea-orm-migration = { version = "1", features = ["runtime-tokio-rustls"] }
reqwest = "0.12.9"
thiserror = "2"
rss = "2"
bytes = "1.9"
itertools = "0.13.0"
url = "2.5"
fancy-regex = "0.14"
regex = "1.11"
lazy_static = "1.5"
maplit = "1.0.2"
lightningcss = "1.0.0-alpha.61"
html-escape = "0.2.13"
opendal = { version = "0.51.0", features = ["default", "services-fs"] }
zune-image = "0.4.15"
once_cell = "1.20.2"
reqwest-middleware = "0.4.0"
reqwest-retry = "0.7.0"
reqwest-tracing = "0.5.5"
scraper = "0.22.0"
leaky-bucket = "1.1.2"
serde_with = "3"
[dev-dependencies]
serial_test = "3"
loco-rs = { version = "0.13", features = ["testing"] }
insta = { version = "1", features = ["redactions", "yaml", "filters"] }

View File

@@ -0,0 +1,79 @@
#![allow(unused_imports)]
use eyre::Context;
use itertools::Itertools;
use loco_rs::{
app::Hooks,
boot::{BootResult, StartMode},
environment::Environment,
prelude::*,
};
use recorder::{
app::App,
extract::mikan::parse_mikan_rss_items_from_rss_link,
migrations::Migrator,
models::{
subscribers::ROOT_SUBSCRIBER,
subscriptions::{self, SubscriptionCreateFromRssDto},
},
};
use sea_orm_migration::MigratorTrait;
async fn pull_mikan_bangumi_rss(ctx: &AppContext) -> eyre::Result<()> {
let rss_link = "https://mikanani.me/RSS/Bangumi?bangumiId=3416&subgroupid=370";
// let rss_link =
// "https://mikanani.me/RSS/MyBangumi?token=FE9tccsML2nBPUUqpCuJW2uJZydAXCntHJ7RpD9LDP8%3d";
let subscription = if let Some(subscription) = subscriptions::Entity::find()
.filter(subscriptions::Column::SourceUrl.eq(String::from(rss_link)))
.one(&ctx.db)
.await?
{
subscription
} else {
subscriptions::Model::add_subscription(
ctx,
subscriptions::SubscriptionCreateDto::Mikan(SubscriptionCreateFromRssDto {
rss_link: rss_link.to_string(),
display_name: String::from("Mikan Project - 我的番组"),
enabled: Some(true),
}),
1,
)
.await?
};
subscription.pull_subscription(ctx).await?;
Ok(())
}
async fn init() -> eyre::Result<AppContext> {
tracing_subscriber::fmt()
.with_max_level(tracing::Level::INFO)
.with_test_writer()
.init();
let ctx = loco_rs::cli::playground::<App>().await?;
let BootResult {
app_context: ctx, ..
} = loco_rs::boot::run_app::<App>(&StartMode::ServerOnly, ctx).await?;
Migrator::up(&ctx.db, None).await?;
Ok(ctx)
}
#[tokio::main]
async fn main() -> eyre::Result<()> {
let ctx = init().await?;
pull_mikan_bangumi_rss(&ctx).await?;
// let active_model: articles::ActiveModel = ActiveModel {
// title: Set(Some("how to build apps in 3 steps".to_string())),
// content: Set(Some("use Loco: https://loco.rs".to_string())),
// ..Default::default()
// };
// active_model.insert(&ctx.db).await.unwrap();
// let res = articles::Entity::find().all(&ctx.db).await.unwrap();
// println!("{:?}", res);
Ok(())
}

97
apps/recorder/src/app.rs Normal file
View File

@@ -0,0 +1,97 @@
use std::path::Path;
use async_trait::async_trait;
use loco_rs::{
app::{AppContext, Hooks},
boot::{create_app, BootResult, StartMode},
cache,
controller::AppRoutes,
db::truncate_table,
environment::Environment,
prelude::*,
task::Tasks,
Result,
};
use sea_orm::DatabaseConnection;
use crate::{
controllers,
dal::{AppDalClient, AppDalInitalizer},
extract::mikan::{client::AppMikanClientInitializer, AppMikanClient},
migrations::Migrator,
models::entities::subscribers,
workers::subscription_worker::SubscriptionWorker,
};
pub trait AppContextExt {
fn get_dal_client(&self) -> &AppDalClient {
AppDalClient::global()
}
fn get_mikan_client(&self) -> &AppMikanClient {
AppMikanClient::global()
}
}
impl AppContextExt for AppContext {}
pub struct App;
#[async_trait]
impl Hooks for App {
fn app_name() -> &'static str {
env!("CARGO_CRATE_NAME")
}
async fn initializers(_ctx: &AppContext) -> Result<Vec<Box<dyn Initializer>>> {
let initializers: Vec<Box<dyn Initializer>> = vec![
Box::new(AppDalInitalizer),
Box::new(AppMikanClientInitializer),
];
Ok(initializers)
}
fn app_version() -> String {
format!(
"{} ({})",
env!("CARGO_PKG_VERSION"),
option_env!("BUILD_SHA")
.or(option_env!("GITHUB_SHA"))
.unwrap_or("dev")
)
}
async fn boot(mode: StartMode, environment: &Environment) -> Result<BootResult> {
create_app::<Self, Migrator>(mode, environment).await
}
fn routes(_ctx: &AppContext) -> AppRoutes {
AppRoutes::with_default_routes()
.prefix("/api")
.add_route(controllers::subscribers::routes())
}
async fn connect_workers(ctx: &AppContext, queue: &Queue) -> Result<()> {
queue.register(SubscriptionWorker::build(ctx)).await?;
Ok(())
}
async fn after_context(ctx: AppContext) -> Result<AppContext> {
Ok(AppContext {
cache: cache::Cache::new(cache::drivers::inmem::new()).into(),
..ctx
})
}
fn register_tasks(_tasks: &mut Tasks) {}
async fn truncate(db: &DatabaseConnection) -> Result<()> {
truncate_table(db, subscribers::Entity).await?;
Ok(())
}
async fn seed(_db: &DatabaseConnection, _base: &Path) -> Result<()> {
Ok(())
}
}

View File

@@ -0,0 +1,8 @@
use loco_rs::cli;
use recorder::{app::App, migrations::Migrator};
#[tokio::main]
async fn main() -> eyre::Result<()> {
cli::main::<App, Migrator>().await?;
Ok(())
}

View File

@@ -0,0 +1,53 @@
use serde::de::DeserializeOwned;
use crate::{
dal::{config::AppDalConfig, DAL_CONF_KEY},
extract::mikan::{AppMikanConfig, MIKAN_CONF_KEY},
};
pub fn deserialize_key_path_from_json_value<T: DeserializeOwned>(
value: &serde_json::Value,
key_path: &[&str],
) -> Result<Option<T>, loco_rs::Error> {
let mut stack = vec![("", value)];
for key in key_path {
let current = stack.last().unwrap().1;
if let Some(v) = current.get(key) {
stack.push((key, v));
} else {
return Ok(None);
}
}
let result: T = serde_json::from_value(stack.pop().unwrap().1.clone())?;
Ok(Some(result))
}
pub fn deserialize_key_path_from_app_config<T: DeserializeOwned>(
app_config: &loco_rs::config::Config,
key_path: &[&str],
) -> Result<Option<T>, loco_rs::Error> {
let settings = app_config.settings.as_ref();
if let Some(settings) = settings {
deserialize_key_path_from_json_value(settings, key_path)
} else {
Ok(None)
}
}
pub trait AppConfigExt {
fn get_root_conf(&self) -> &loco_rs::config::Config;
fn get_dal_conf(&self) -> loco_rs::Result<Option<AppDalConfig>> {
deserialize_key_path_from_app_config(self.get_root_conf(), &[DAL_CONF_KEY])
}
fn get_mikan_conf(&self) -> loco_rs::Result<Option<AppMikanConfig>> {
deserialize_key_path_from_app_config(self.get_root_conf(), &[MIKAN_CONF_KEY])
}
}
impl AppConfigExt for loco_rs::config::Config {
fn get_root_conf(&self) -> &loco_rs::config::Config {
self
}
}

View File

@@ -0,0 +1 @@
pub mod subscribers;

View File

@@ -0,0 +1,14 @@
use loco_rs::prelude::*;
use crate::{models::entities::subscribers, views::subscribers::CurrentResponse};
async fn current(State(ctx): State<AppContext>) -> Result<impl IntoResponse> {
let subscriber = subscribers::Model::find_root(&ctx).await?;
format::json(CurrentResponse::new(&subscriber))
}
pub fn routes() -> Routes {
Routes::new()
.prefix("subscribers")
.add("/current", get(current))
}

View File

@@ -0,0 +1,201 @@
use std::fmt;
use bytes::Bytes;
use loco_rs::app::{AppContext, Initializer};
use once_cell::sync::OnceCell;
use opendal::{layers::LoggingLayer, services::Fs, Buffer, Operator};
use quirks_path::{Path, PathBuf};
use serde::{Deserialize, Serialize};
use url::Url;
use uuid::Uuid;
use super::AppDalConfig;
use crate::config::AppConfigExt;
// TODO: wait app-context-trait to integrate
#[derive(Clone, Debug, PartialEq, Eq, Serialize, Deserialize)]
#[serde(rename_all = "snake_case")]
pub enum DalContentCategory {
Image,
}
impl AsRef<str> for DalContentCategory {
fn as_ref(&self) -> &str {
match self {
Self::Image => "image",
}
}
}
#[derive(Debug, Clone)]
pub struct AppDalClient {
pub config: AppDalConfig,
}
static APP_DAL_CLIENT: OnceCell<AppDalClient> = OnceCell::new();
pub enum DalStoredUrl {
RelativePath { path: String },
Absolute { url: Url },
}
impl AsRef<str> for DalStoredUrl {
fn as_ref(&self) -> &str {
match &self {
Self::Absolute { url } => url.as_str(),
Self::RelativePath { path } => path,
}
}
}
impl fmt::Display for DalStoredUrl {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
write!(f, "{}", self.as_ref())
}
}
impl AppDalClient {
pub fn new(config: AppDalConfig) -> Self {
Self { config }
}
pub fn global() -> &'static AppDalClient {
APP_DAL_CLIENT
.get()
.expect("Global app dal client is not initialized")
}
pub fn get_fs(&self) -> Fs {
Fs::default().root(
self.config
.data_dir
.as_ref()
.map(|s| s as &str)
.unwrap_or("./data"),
)
}
pub fn create_filename(extname: &str) -> String {
format!("{}{}", Uuid::new_v4(), extname)
}
pub async fn store_object(
&self,
content_category: DalContentCategory,
subscriber_pid: &str,
bucket: Option<&str>,
filename: &str,
data: Bytes,
) -> eyre::Result<DalStoredUrl> {
match content_category {
DalContentCategory::Image => {
let fullname = [
subscriber_pid,
content_category.as_ref(),
bucket.unwrap_or_default(),
filename,
]
.into_iter()
.map(Path::new)
.collect::<PathBuf>();
let fs_op = Operator::new(self.get_fs())?
.layer(LoggingLayer::default())
.finish();
if let Some(dirname) = fullname.parent() {
let dirname = dirname.join("/");
fs_op.create_dir(dirname.as_str()).await?;
}
fs_op.write(fullname.as_str(), data).await?;
Ok(DalStoredUrl::RelativePath {
path: fullname.to_string(),
})
}
}
}
pub async fn exists_object(
&self,
content_category: DalContentCategory,
subscriber_pid: &str,
bucket: Option<&str>,
filename: &str,
) -> eyre::Result<Option<DalStoredUrl>> {
match content_category {
DalContentCategory::Image => {
let fullname = [
subscriber_pid,
content_category.as_ref(),
bucket.unwrap_or_default(),
filename,
]
.into_iter()
.map(Path::new)
.collect::<PathBuf>();
let fs_op = Operator::new(self.get_fs())?
.layer(LoggingLayer::default())
.finish();
if fs_op.exists(fullname.as_str()).await? {
Ok(Some(DalStoredUrl::RelativePath {
path: fullname.to_string(),
}))
} else {
Ok(None)
}
}
}
}
pub async fn load_object(
&self,
content_category: DalContentCategory,
subscriber_pid: &str,
bucket: Option<&str>,
filename: &str,
) -> eyre::Result<Buffer> {
match content_category {
DalContentCategory::Image => {
let fullname = [
subscriber_pid,
content_category.as_ref(),
bucket.unwrap_or_default(),
filename,
]
.into_iter()
.map(Path::new)
.collect::<PathBuf>();
let fs_op = Operator::new(self.get_fs())?
.layer(LoggingLayer::default())
.finish();
let data = fs_op.read(fullname.as_str()).await?;
Ok(data)
}
}
}
}
pub struct AppDalInitalizer;
#[async_trait::async_trait]
impl Initializer for AppDalInitalizer {
fn name(&self) -> String {
String::from("AppDalInitalizer")
}
async fn before_run(&self, app_context: &AppContext) -> loco_rs::Result<()> {
let config = &app_context.config;
let app_dal_conf = config.get_dal_conf()?;
APP_DAL_CLIENT.get_or_init(|| AppDalClient::new(app_dal_conf.unwrap_or_default()));
Ok(())
}
}

View File

@@ -0,0 +1,8 @@
use serde::{Deserialize, Serialize};
pub const DAL_CONF_KEY: &str = "dal";
#[derive(Debug, Clone, PartialEq, Eq, Serialize, Deserialize, Default)]
pub struct AppDalConfig {
pub data_dir: Option<String>,
}

View File

@@ -0,0 +1,4 @@
pub mod client;
pub mod config;
pub use client::{AppDalClient, AppDalInitalizer, DalContentCategory};
pub use config::{AppDalConfig, DAL_CONF_KEY};

View File

@@ -0,0 +1,73 @@
use std::collections::HashMap;
use fancy_regex::Regex as FancyRegex;
use lazy_static::lazy_static;
use maplit::hashmap;
use regex::Regex;
const LANG_ZH_TW: &str = "zh-tw";
const LANG_ZH: &str = "zh";
const LANG_EN: &str = "en";
const LANG_JP: &str = "jp";
lazy_static! {
pub static ref SEASON_REGEX: Regex =
Regex::new(r"(S\|[Ss]eason\s+)(\d+)").expect("Invalid regex");
pub static ref TORRENT_PRASE_RULE_REGS: Vec<FancyRegex> = vec![
FancyRegex::new(
r"(.*) - (\d{1,4}(?!\d|p)|\d{1,4}\.\d{1,2}(?!\d|p))(?:v\d{1,2})?(?: )?(?:END)?(.*)"
)
.unwrap(),
FancyRegex::new(
r"(.*)[\[\ E](\d{1,4}|\d{1,4}\.\d{1,2})(?:v\d{1,2})?(?: )?(?:END)?[\]\ ](.*)"
)
.unwrap(),
FancyRegex::new(r"(.*)\[(?:第)?(\d*\.*\d*)[话集話](?:END)?\](.*)").unwrap(),
FancyRegex::new(r"(.*)第?(\d*\.*\d*)[话話集](?:END)?(.*)").unwrap(),
FancyRegex::new(r"(.*)(?:S\d{2})?EP?(\d+)(.*)").unwrap(),
];
pub static ref SUBTITLE_LANG: Vec<(&'static str, Vec<&'static str>)> = {
vec![
(LANG_ZH_TW, vec!["tc", "cht", "", "zh-tw"]),
(LANG_ZH, vec!["sc", "chs", "", "zh", "zh-cn"]),
(LANG_EN, vec!["en", "eng", ""]),
(LANG_JP, vec!["jp", "jpn", ""]),
]
};
pub static ref BRACKETS_REG: Regex = Regex::new(r"[\[\]()【】()]").unwrap();
pub static ref DIGIT_1PLUS_REG: Regex = Regex::new(r"\d+").unwrap();
pub static ref ZH_NUM_MAP: HashMap<&'static str, i32> = {
hashmap! {
"" => 0,
"" => 1,
"" => 2,
"" => 3,
"" => 4,
"" => 5,
"" => 6,
"" => 7,
"" => 8,
"" => 9,
"" => 10,
"廿" => 20,
"" => 100,
"" => 1000,
"" => 0,
"" => 1,
"" => 2,
"" => 3,
"" => 4,
"" => 5,
"" => 6,
"" => 7,
"" => 8,
"" => 9,
"" => 10,
"" => 20,
"" => 100,
"" => 1000,
}
};
pub static ref ZH_NUM_RE: Regex =
Regex::new(r"[〇一二三四五六七八九十廿百千零壹贰叁肆伍陆柒捌玖拾念佰仟]").unwrap();
}

View File

@@ -0,0 +1,19 @@
use thiserror::Error;
#[derive(Error, Debug)]
pub enum ParseError {
#[error("Parse bangumi season error: {0}")]
BangumiSeasonError(#[from] std::num::ParseIntError),
#[error("Parse file url error: {0}")]
FileUrlError(#[from] url::ParseError),
#[error("Parse {desc} with mime error, expected {expected}, but got {found}")]
MimeError {
desc: String,
expected: String,
found: String,
},
#[error("Parse mikan rss {url} format error")]
MikanRssFormatError { url: String },
#[error("Parse mikan rss item format error, {reason}")]
MikanRssItemFormatError { reason: String },
}

View File

@@ -0,0 +1,3 @@
pub mod styles;
pub use styles::parse_style_attr;

View File

@@ -0,0 +1,6 @@
use lightningcss::declaration::DeclarationBlock;
pub fn parse_style_attr(style_attr: &str) -> Option<DeclarationBlock> {
let result = DeclarationBlock::parse_string(style_attr, Default::default()).ok()?;
Some(result)
}

View File

@@ -0,0 +1,64 @@
use std::ops::Deref;
use loco_rs::app::{AppContext, Initializer};
use once_cell::sync::OnceCell;
use super::{AppMikanConfig, MIKAN_BASE_URL};
use crate::{config::AppConfigExt, fetch::HttpClient};
static APP_MIKAN_CLIENT: OnceCell<AppMikanClient> = OnceCell::new();
pub struct AppMikanClient {
http_client: HttpClient,
base_url: String,
}
impl AppMikanClient {
pub fn new(mut config: AppMikanConfig) -> loco_rs::Result<Self> {
let http_client =
HttpClient::new(config.http_client.take()).map_err(loco_rs::Error::wrap)?;
let base_url = config
.base_url
.unwrap_or_else(|| String::from(MIKAN_BASE_URL));
Ok(Self {
http_client,
base_url,
})
}
pub fn global() -> &'static AppMikanClient {
APP_MIKAN_CLIENT
.get()
.expect("Global mikan http client is not initialized")
}
pub fn base_url(&self) -> &str {
&self.base_url
}
}
impl Deref for AppMikanClient {
type Target = HttpClient;
fn deref(&self) -> &Self::Target {
&self.http_client
}
}
pub struct AppMikanClientInitializer;
#[async_trait::async_trait]
impl Initializer for AppMikanClientInitializer {
fn name(&self) -> String {
"AppMikanClientInitializer".to_string()
}
async fn before_run(&self, app_context: &AppContext) -> loco_rs::Result<()> {
let config = &app_context.config;
let app_mikan_conf = config.get_mikan_conf()?.unwrap_or_default();
APP_MIKAN_CLIENT.get_or_try_init(|| AppMikanClient::new(app_mikan_conf))?;
Ok(())
}
}

View File

@@ -0,0 +1,11 @@
use serde::{Deserialize, Serialize};
use crate::fetch::HttpClientConfig;
pub const MIKAN_CONF_KEY: &str = "mikan";
#[derive(Debug, Clone, PartialEq, Eq, Serialize, Deserialize, Default)]
pub struct AppMikanConfig {
pub http_client: Option<HttpClientConfig>,
pub base_url: Option<String>,
}

View File

@@ -0,0 +1,4 @@
pub const MIKAN_BUCKET_KEY: &str = "mikan";
pub const MIKAN_BASE_URL: &str = "https://mikanani.me/";
pub const MIKAN_UNKNOWN_FANSUB_NAME: &str = "生肉/不明字幕";
pub const MIKAN_UNKNOWN_FANSUB_ID: &str = "202";

View File

@@ -0,0 +1,22 @@
pub mod client;
pub mod config;
pub mod constants;
pub mod rss_parser;
pub mod web_parser;
pub use client::{AppMikanClient, AppMikanClientInitializer};
pub use config::{AppMikanConfig, MIKAN_CONF_KEY};
pub use constants::{MIKAN_BASE_URL, MIKAN_BUCKET_KEY};
pub use rss_parser::{
build_mikan_bangumi_rss_link, build_mikan_subscriber_aggregation_rss_link,
parse_mikan_bangumi_id_from_rss_link, parse_mikan_rss_channel_from_rss_link,
parse_mikan_rss_items_from_rss_link, parse_mikan_subscriber_aggregation_id_from_rss_link,
MikanBangumiAggregationRssChannel, MikanBangumiRssChannel, MikanBangumiRssLink,
MikanRssChannel, MikanRssItem, MikanSubscriberAggregationRssChannel,
MikanSubscriberAggregationRssLink,
};
pub use web_parser::{
build_mikan_bangumi_homepage, build_mikan_episode_homepage,
parse_mikan_bangumi_meta_from_mikan_homepage, parse_mikan_episode_meta_from_mikan_homepage,
MikanBangumiMeta, MikanEpisodeMeta,
};

View File

@@ -0,0 +1,353 @@
use std::ops::Deref;
use chrono::DateTime;
use itertools::Itertools;
use reqwest::IntoUrl;
use serde::{Deserialize, Serialize};
use torrent::core::BITTORRENT_MIME_TYPE;
use url::Url;
use super::{
web_parser::{parse_mikan_episode_id_from_homepage, MikanEpisodeHomepage},
AppMikanClient,
};
use crate::{extract::errors::ParseError, fetch::bytes::download_bytes_with_client};
#[derive(Debug, Clone, Serialize, Deserialize, PartialEq, Eq)]
pub struct MikanRssItem {
pub title: String,
pub homepage: Url,
pub url: Url,
pub content_length: Option<u64>,
pub mime: String,
pub pub_date: Option<i64>,
pub mikan_episode_id: String,
}
#[derive(Debug, Clone, Serialize, Deserialize, PartialEq, Eq)]
pub struct MikanBangumiRssChannel {
pub name: String,
pub url: Url,
pub mikan_bangumi_id: String,
pub mikan_fansub_id: String,
pub items: Vec<MikanRssItem>,
}
#[derive(Debug, Clone, Serialize, Deserialize, PartialEq, Eq)]
pub struct MikanBangumiAggregationRssChannel {
pub name: String,
pub url: Url,
pub mikan_bangumi_id: String,
pub items: Vec<MikanRssItem>,
}
#[derive(Debug, Clone, Serialize, Deserialize, PartialEq, Eq)]
pub struct MikanSubscriberAggregationRssChannel {
pub mikan_aggregation_id: String,
pub url: Url,
pub items: Vec<MikanRssItem>,
}
#[derive(Debug, Clone, Serialize, Deserialize, PartialEq, Eq)]
pub enum MikanRssChannel {
Bangumi(MikanBangumiRssChannel),
BangumiAggregation(MikanBangumiAggregationRssChannel),
SubscriberAggregation(MikanSubscriberAggregationRssChannel),
}
impl MikanRssChannel {
pub fn items(&self) -> &[MikanRssItem] {
match &self {
Self::Bangumi(MikanBangumiRssChannel { items, .. })
| Self::BangumiAggregation(MikanBangumiAggregationRssChannel { items, .. })
| Self::SubscriberAggregation(MikanSubscriberAggregationRssChannel { items, .. }) => {
items
}
}
}
pub fn into_items(self) -> Vec<MikanRssItem> {
match self {
Self::Bangumi(MikanBangumiRssChannel { items, .. })
| Self::BangumiAggregation(MikanBangumiAggregationRssChannel { items, .. })
| Self::SubscriberAggregation(MikanSubscriberAggregationRssChannel { items, .. }) => {
items
}
}
}
pub fn name(&self) -> Option<&str> {
match &self {
Self::Bangumi(MikanBangumiRssChannel { name, .. })
| Self::BangumiAggregation(MikanBangumiAggregationRssChannel { name, .. }) => {
Some(name.as_str())
}
Self::SubscriberAggregation(MikanSubscriberAggregationRssChannel { .. }) => None,
}
}
pub fn url(&self) -> &Url {
match &self {
Self::Bangumi(MikanBangumiRssChannel { url, .. })
| Self::BangumiAggregation(MikanBangumiAggregationRssChannel { url, .. })
| Self::SubscriberAggregation(MikanSubscriberAggregationRssChannel { url, .. }) => url,
}
}
}
impl TryFrom<rss::Item> for MikanRssItem {
type Error = ParseError;
fn try_from(item: rss::Item) -> Result<Self, Self::Error> {
let mime_type = item
.enclosure()
.map(|x| x.mime_type.to_string())
.unwrap_or_default();
if mime_type == BITTORRENT_MIME_TYPE {
let enclosure = item.enclosure.unwrap();
let homepage = item
.link
.ok_or_else(|| ParseError::MikanRssItemFormatError {
reason: String::from("must to have link for homepage"),
})?;
let homepage = Url::parse(&homepage)?;
let enclosure_url = Url::parse(&enclosure.url)?;
let MikanEpisodeHomepage {
mikan_episode_id, ..
} = parse_mikan_episode_id_from_homepage(&homepage).ok_or_else(|| {
ParseError::MikanRssItemFormatError {
reason: String::from("homepage link format invalid"),
}
})?;
Ok(MikanRssItem {
title: item.title.unwrap_or_default(),
homepage,
url: enclosure_url,
content_length: enclosure.length.parse().ok(),
mime: enclosure.mime_type,
pub_date: item
.pub_date
.and_then(|s| DateTime::parse_from_rfc2822(&s).ok())
.map(|s| s.timestamp_millis()),
mikan_episode_id,
})
} else {
Err(ParseError::MimeError {
expected: String::from(BITTORRENT_MIME_TYPE),
found: mime_type,
desc: String::from("MikanRssItem"),
})
}
}
}
#[derive(Debug, Clone)]
pub struct MikanBangumiRssLink {
pub mikan_bangumi_id: String,
pub mikan_fansub_id: Option<String>,
}
#[derive(Debug, Clone)]
pub struct MikanSubscriberAggregationRssLink {
pub mikan_aggregation_id: String,
}
pub fn build_mikan_bangumi_rss_link(
mikan_base_url: &str,
mikan_bangumi_id: &str,
mikan_fansub_id: Option<&str>,
) -> eyre::Result<Url> {
let mut url = Url::parse(mikan_base_url)?;
url.set_path("/RSS/Bangumi");
url.query_pairs_mut()
.append_pair("bangumiId", mikan_bangumi_id);
if let Some(mikan_fansub_id) = mikan_fansub_id {
url.query_pairs_mut()
.append_pair("subgroupid", mikan_fansub_id);
};
Ok(url)
}
pub fn build_mikan_subscriber_aggregation_rss_link(
mikan_base_url: &str,
mikan_aggregation_id: &str,
) -> eyre::Result<Url> {
let mut url = Url::parse(mikan_base_url)?;
url.set_path("/RSS/MyBangumi");
url.query_pairs_mut()
.append_pair("token", mikan_aggregation_id);
Ok(url)
}
pub fn parse_mikan_bangumi_id_from_rss_link(url: &Url) -> Option<MikanBangumiRssLink> {
if url.path() == "/RSS/Bangumi" {
url.query_pairs()
.find(|(k, _)| k == "bangumiId")
.map(|(_, v)| MikanBangumiRssLink {
mikan_bangumi_id: v.to_string(),
mikan_fansub_id: url
.query_pairs()
.find(|(k, _)| k == "subgroupid")
.map(|(_, v)| v.to_string()),
})
} else {
None
}
}
pub fn parse_mikan_subscriber_aggregation_id_from_rss_link(
url: &Url,
) -> Option<MikanSubscriberAggregationRssLink> {
if url.path() == "/RSS/MyBangumi" {
url.query_pairs().find(|(k, _)| k == "token").map(|(_, v)| {
MikanSubscriberAggregationRssLink {
mikan_aggregation_id: v.to_string(),
}
})
} else {
None
}
}
pub async fn parse_mikan_rss_items_from_rss_link(
client: Option<&AppMikanClient>,
url: impl IntoUrl,
) -> eyre::Result<Vec<MikanRssItem>> {
let channel = parse_mikan_rss_channel_from_rss_link(client, url).await?;
Ok(channel.into_items())
}
pub async fn parse_mikan_rss_channel_from_rss_link(
client: Option<&AppMikanClient>,
url: impl IntoUrl,
) -> eyre::Result<MikanRssChannel> {
let http_client = client.map(|s| s.deref());
let bytes = download_bytes_with_client(http_client, url.as_str()).await?;
let channel = rss::Channel::read_from(&bytes[..])?;
let channel_link = Url::parse(channel.link())?;
if let Some(MikanBangumiRssLink {
mikan_bangumi_id,
mikan_fansub_id,
}) = parse_mikan_bangumi_id_from_rss_link(&channel_link)
{
let channel_name = channel.title().replace("Mikan Project - ", "");
let items = channel
.items
.into_iter()
// @TODO log error
.flat_map(MikanRssItem::try_from)
.collect_vec();
if let Some(mikan_fansub_id) = mikan_fansub_id {
Ok(MikanRssChannel::Bangumi(MikanBangumiRssChannel {
name: channel_name,
mikan_bangumi_id,
mikan_fansub_id,
url: channel_link,
items,
}))
} else {
Ok(MikanRssChannel::BangumiAggregation(
MikanBangumiAggregationRssChannel {
name: channel_name,
mikan_bangumi_id,
url: channel_link,
items,
},
))
}
} else if let Some(MikanSubscriberAggregationRssLink {
mikan_aggregation_id,
..
}) = parse_mikan_subscriber_aggregation_id_from_rss_link(&channel_link)
{
let items = channel
.items
.into_iter()
// @TODO log error
.flat_map(MikanRssItem::try_from)
.collect_vec();
return Ok(MikanRssChannel::SubscriberAggregation(
MikanSubscriberAggregationRssChannel {
mikan_aggregation_id,
items,
url: channel_link,
},
));
} else {
return Err(ParseError::MikanRssFormatError {
url: url.as_str().into(),
}
.into());
}
}
#[cfg(test)]
mod tests {
use std::assert_matches::assert_matches;
use torrent::core::BITTORRENT_MIME_TYPE;
use crate::extract::mikan::{
parse_mikan_rss_channel_from_rss_link, MikanBangumiAggregationRssChannel,
MikanBangumiRssChannel, MikanRssChannel,
};
#[tokio::test]
pub async fn test_parse_mikan_rss_channel_from_rss_link() {
{
let bangumi_url = "https://mikanani.me/RSS/Bangumi?bangumiId=3141&subgroupid=370";
let channel = parse_mikan_rss_channel_from_rss_link(None, bangumi_url)
.await
.expect("should get mikan channel from rss url");
assert_matches!(
&channel,
MikanRssChannel::Bangumi(MikanBangumiRssChannel { .. })
);
assert_matches!(&channel.name(), Some("葬送的芙莉莲"));
let items = channel.items();
let first_sub_item = items
.first()
.expect("mikan subscriptions should have at least one subs");
assert_eq!(first_sub_item.mime, BITTORRENT_MIME_TYPE);
assert!(&first_sub_item
.homepage
.as_str()
.starts_with("https://mikanani.me/Home/Episode"));
let name = first_sub_item.title.as_str();
assert!(name.contains("葬送的芙莉莲"));
}
{
let bangumi_url = "https://mikanani.me/RSS/Bangumi?bangumiId=3416";
let channel = parse_mikan_rss_channel_from_rss_link(None, bangumi_url)
.await
.expect("should get mikan channel from rss url");
assert_matches!(
&channel,
MikanRssChannel::BangumiAggregation(MikanBangumiAggregationRssChannel { .. })
);
assert_matches!(&channel.name(), Some("叹气的亡灵想隐退"));
}
}
}

View File

@@ -0,0 +1,493 @@
use std::ops::Deref;
use bytes::Bytes;
use eyre::ContextCompat;
use html_escape::decode_html_entities;
use itertools::Itertools;
use lazy_static::lazy_static;
use lightningcss::{properties::Property, values::image::Image as CSSImage};
use loco_rs::app::AppContext;
use regex::Regex;
use scraper::Html;
use url::Url;
use super::{
parse_mikan_bangumi_id_from_rss_link, AppMikanClient, MikanBangumiRssLink, MIKAN_BUCKET_KEY,
};
use crate::{
app::AppContextExt,
dal::DalContentCategory,
extract::html::parse_style_attr,
fetch::{html::download_html_with_client, image::download_image_with_client},
models::subscribers,
};
#[derive(Clone, Debug, PartialEq)]
pub struct MikanEpisodeMeta {
pub homepage: Url,
pub origin_poster_src: Option<Url>,
pub bangumi_title: String,
pub episode_title: String,
pub fansub: String,
pub mikan_bangumi_id: String,
pub mikan_fansub_id: String,
pub mikan_episode_id: String,
}
#[derive(Clone, Debug, PartialEq)]
pub struct MikanBangumiMeta {
pub homepage: Url,
pub origin_poster_src: Option<Url>,
pub bangumi_title: String,
pub mikan_bangumi_id: String,
pub mikan_fansub_id: Option<String>,
pub fansub: Option<String>,
pub mikan_fansub_candidates: Vec<(String, String)>,
}
#[derive(Clone, Debug, PartialEq)]
pub struct MikanBangumiPosterMeta {
pub origin_poster_src: Url,
pub poster_data: Option<Bytes>,
pub poster_src: Option<String>,
}
#[derive(Clone, Debug, PartialEq)]
pub struct MikanEpisodeHomepage {
pub mikan_episode_id: String,
}
lazy_static! {
static ref MIKAN_TITLE_SEASON: Regex = Regex::new("第.*季").unwrap();
}
pub fn build_mikan_bangumi_homepage(
mikan_base_url: &str,
mikan_bangumi_id: &str,
mikan_fansub_id: Option<&str>,
) -> eyre::Result<Url> {
let mut url = Url::parse(mikan_base_url)?;
url.set_path(&format!("/Home/Bangumi/{mikan_bangumi_id}"));
url.set_fragment(mikan_fansub_id);
Ok(url)
}
pub fn build_mikan_episode_homepage(
mikan_base_url: &str,
mikan_episode_id: &str,
) -> eyre::Result<Url> {
let mut url = Url::parse(mikan_base_url)?;
url.set_path(&format!("/Home/Episode/{mikan_episode_id}"));
Ok(url)
}
pub fn parse_mikan_episode_id_from_homepage(url: &Url) -> Option<MikanEpisodeHomepage> {
if url.path().starts_with("/Home/Episode/") {
let mikan_episode_id = url.path().replace("/Home/Episode/", "");
Some(MikanEpisodeHomepage { mikan_episode_id })
} else {
None
}
}
pub async fn parse_mikan_bangumi_poster_from_origin_poster_src(
client: Option<&AppMikanClient>,
origin_poster_src: Url,
) -> eyre::Result<MikanBangumiPosterMeta> {
let http_client = client.map(|s| s.deref());
let poster_data = download_image_with_client(http_client, origin_poster_src.clone()).await?;
Ok(MikanBangumiPosterMeta {
origin_poster_src,
poster_data: Some(poster_data),
poster_src: None,
})
}
pub async fn parse_mikan_bangumi_poster_from_origin_poster_src_with_cache(
ctx: &AppContext,
origin_poster_src: Url,
subscriber_id: i32,
) -> eyre::Result<MikanBangumiPosterMeta> {
let dal_client = ctx.get_dal_client();
let mikan_client = ctx.get_mikan_client();
let subscriber_pid = &subscribers::Model::find_pid_by_id_with_cache(ctx, subscriber_id).await?;
if let Some(poster_src) = dal_client
.exists_object(
DalContentCategory::Image,
subscriber_pid,
Some(MIKAN_BUCKET_KEY),
&origin_poster_src.path().replace("/images/Bangumi/", ""),
)
.await?
{
return Ok(MikanBangumiPosterMeta {
origin_poster_src,
poster_data: None,
poster_src: Some(poster_src.to_string()),
});
}
let poster_data =
download_image_with_client(Some(mikan_client.deref()), origin_poster_src.clone()).await?;
let poster_str = dal_client
.store_object(
DalContentCategory::Image,
subscriber_pid,
Some(MIKAN_BUCKET_KEY),
&origin_poster_src.path().replace("/images/Bangumi/", ""),
poster_data.clone(),
)
.await?;
Ok(MikanBangumiPosterMeta {
origin_poster_src,
poster_data: Some(poster_data),
poster_src: Some(poster_str.to_string()),
})
}
pub async fn parse_mikan_bangumi_meta_from_mikan_homepage(
client: Option<&AppMikanClient>,
url: Url,
) -> eyre::Result<MikanBangumiMeta> {
let http_client = client.map(|s| s.deref());
let url_host = url.origin().unicode_serialization();
let content = download_html_with_client(http_client, url.as_str()).await?;
let html = Html::parse_document(&content);
let bangumi_fansubs = html
.select(&scraper::Selector::parse(".subgroup-text").unwrap())
.filter_map(|el| {
if let (Some(fansub_id), Some(fansub_name)) = (
el.value()
.attr("id")
.map(|s| decode_html_entities(s).trim().to_string()),
el.select(&scraper::Selector::parse("a:nth-child(1)").unwrap())
.next()
.map(|child| {
let mut s = String::from(
child
.prev_sibling()
.and_then(|t| t.value().as_text())
.map(|s| s.trim())
.unwrap_or_default(),
);
s.extend(child.text());
decode_html_entities(&s).trim().to_string()
}),
) {
Some((fansub_id, fansub_name))
} else {
None
}
})
.collect_vec();
let fansub_info = url.fragment().and_then(|b| {
bangumi_fansubs
.iter()
.find_map(|(id, name)| if id == b { Some((id, name)) } else { None })
});
let bangumi_title = html
.select(&scraper::Selector::parse(".bangumi-title").unwrap())
.next()
.map(|el| {
decode_html_entities(&el.text().collect::<String>())
.trim()
.to_string()
})
.and_then(|title| if title.is_empty() { None } else { Some(title) })
.wrap_err_with(|| {
// todo: error handler
format!("Missing mikan bangumi official title for {}", url)
})?;
let MikanBangumiRssLink {
mikan_bangumi_id, ..
} = html
.select(&scraper::Selector::parse(".bangumi-title > .mikan-rss").unwrap())
.next()
.and_then(|el| el.value().attr("href"))
.as_ref()
.and_then(|s| url.join(s).ok())
.and_then(|rss_link_url| parse_mikan_bangumi_id_from_rss_link(&rss_link_url))
.wrap_err_with(|| {
// todo: error handler
format!("Missing mikan bangumi rss link or error format for {}", url)
})?;
let origin_poster_src = html
.select(&scraper::Selector::parse(".bangumi-poster").unwrap())
.next()
.and_then(|el| el.value().attr("style"))
.as_ref()
.and_then(|s| parse_style_attr(s))
.and_then(|style| {
style.iter().find_map(|(prop, _)| {
match prop {
Property::BackgroundImage(images) => {
for img in images {
if let CSSImage::Url(path) = img {
if let Ok(url) =
Url::parse(&url_host).and_then(|s| s.join(path.url.trim()))
{
return Some(url);
}
}
}
}
Property::Background(backgrounds) => {
for bg in backgrounds {
if let CSSImage::Url(path) = &bg.image {
if let Ok(url) =
Url::parse(&url_host).and_then(|s| s.join(path.url.trim()))
{
return Some(url);
}
}
}
}
_ => {}
}
None
})
})
.map(|mut origin_poster_src| {
origin_poster_src.set_query(None);
origin_poster_src
});
Ok(MikanBangumiMeta {
homepage: url,
bangumi_title,
origin_poster_src,
mikan_bangumi_id,
fansub: fansub_info.map(|s| s.1.to_string()),
mikan_fansub_id: fansub_info.map(|s| s.0.to_string()),
mikan_fansub_candidates: bangumi_fansubs.clone(),
})
}
pub async fn parse_mikan_episode_meta_from_mikan_homepage(
client: Option<&AppMikanClient>,
url: Url,
) -> eyre::Result<MikanEpisodeMeta> {
let http_client = client.map(|s| s.deref());
let url_host = url.origin().unicode_serialization();
let content = download_html_with_client(http_client, url.as_str()).await?;
let html = Html::parse_document(&content);
let bangumi_title = html
.select(&scraper::Selector::parse(".bangumi-title").unwrap())
.next()
.map(|el| {
decode_html_entities(&el.text().collect::<String>())
.trim()
.to_string()
})
.and_then(|title| if title.is_empty() { None } else { Some(title) })
.wrap_err_with(|| {
// todo: error handler
format!("Missing mikan bangumi official title for {}", url)
})?;
let episode_title = html
.select(&scraper::Selector::parse("title").unwrap())
.next()
.map(|el| {
decode_html_entities(&el.text().collect::<String>())
.replace(" - Mikan Project", "")
.trim()
.to_string()
})
.and_then(|title| if title.is_empty() { None } else { Some(title) })
.wrap_err_with(|| {
// todo: error handler
format!("Missing mikan episode official title for {}", url)
})?;
let (mikan_bangumi_id, mikan_fansub_id) = html
.select(&scraper::Selector::parse(".bangumi-title > .mikan-rss").unwrap())
.next()
.and_then(|el| el.value().attr("href"))
.as_ref()
.and_then(|s| url.join(s).ok())
.and_then(|rss_link_url| parse_mikan_bangumi_id_from_rss_link(&rss_link_url))
.and_then(
|MikanBangumiRssLink {
mikan_bangumi_id,
mikan_fansub_id,
..
}| {
mikan_fansub_id.map(|mikan_fansub_id| (mikan_bangumi_id, mikan_fansub_id))
},
)
.wrap_err_with(|| {
// todo: error handler
format!("Missing mikan bangumi rss link or error format for {}", url)
})?;
let fansub = html
.select(&scraper::Selector::parse(".bangumi-info>.magnet-link-wrap").unwrap())
.next()
.map(|el| {
decode_html_entities(&el.text().collect::<String>())
.trim()
.to_string()
})
.wrap_err_with(|| {
// todo: error handler
format!("Missing mikan bangumi fansub name for {}", url)
})?;
let origin_poster_src = html
.select(&scraper::Selector::parse(".bangumi-poster").unwrap())
.next()
.and_then(|el| el.value().attr("style"))
.as_ref()
.and_then(|s| parse_style_attr(s))
.and_then(|style| {
style.iter().find_map(|(prop, _)| {
match prop {
Property::BackgroundImage(images) => {
for img in images {
if let CSSImage::Url(path) = img {
if let Ok(url) =
Url::parse(&url_host).and_then(|s| s.join(path.url.trim()))
{
return Some(url);
}
}
}
}
Property::Background(backgrounds) => {
for bg in backgrounds {
if let CSSImage::Url(path) = &bg.image {
if let Ok(url) =
Url::parse(&url_host).and_then(|s| s.join(path.url.trim()))
{
return Some(url);
}
}
}
}
_ => {}
}
None
})
})
.map(|mut origin_poster_src| {
origin_poster_src.set_query(None);
origin_poster_src
});
let MikanEpisodeHomepage {
mikan_episode_id, ..
} = parse_mikan_episode_id_from_homepage(&url)
.wrap_err_with(|| format!("Failed to extract mikan_episode_id from {}", &url))?;
Ok(MikanEpisodeMeta {
mikan_bangumi_id,
mikan_fansub_id,
bangumi_title,
episode_title,
homepage: url,
origin_poster_src,
fansub,
mikan_episode_id,
})
}
#[cfg(test)]
mod test {
use std::assert_matches::assert_matches;
use url::Url;
use zune_image::{codecs::ImageFormat, image::Image};
use super::{
parse_mikan_bangumi_meta_from_mikan_homepage,
parse_mikan_bangumi_poster_from_origin_poster_src,
parse_mikan_episode_meta_from_mikan_homepage,
};
#[tokio::test]
async fn test_parse_mikan_episode() {
let test_fn = async || -> eyre::Result<()> {
let url_str =
"https://mikanani.me/Home/Episode/475184dce83ea2b82902592a5ac3343f6d54b36a";
let url = Url::parse(url_str)?;
let ep_meta = parse_mikan_episode_meta_from_mikan_homepage(None, url.clone()).await?;
assert_eq!(ep_meta.homepage, url);
assert_eq!(ep_meta.bangumi_title, "葬送的芙莉莲");
assert_eq!(
ep_meta.origin_poster_src,
Some(Url::parse(
"https://mikanani.me/images/Bangumi/202309/5ce9fed1.jpg"
)?)
);
assert_eq!(ep_meta.fansub, "LoliHouse");
assert_eq!(ep_meta.mikan_fansub_id, "370");
assert_eq!(ep_meta.mikan_bangumi_id, "3141");
assert_matches!(ep_meta.origin_poster_src, Some(..));
let bgm_poster = parse_mikan_bangumi_poster_from_origin_poster_src(
None,
ep_meta.origin_poster_src.unwrap(),
)
.await?;
let u8_data = bgm_poster.poster_data.expect("should have poster data");
let image = Image::read(u8_data.to_vec(), Default::default());
assert!(
image.is_ok_and(|img| img
.metadata()
.get_image_format()
.is_some_and(|fmt| matches!(fmt, ImageFormat::JPEG))),
"should start with valid jpeg data magic number"
);
Ok(())
};
test_fn().await.expect("test parse mikan failed");
}
#[tokio::test]
async fn test_parse_mikan_bangumi() {
let test_fn = async || -> eyre::Result<()> {
let url_str = "https://mikanani.me/Home/Bangumi/3416#370";
let url = Url::parse(url_str)?;
let bgm_meta = parse_mikan_bangumi_meta_from_mikan_homepage(None, url.clone()).await?;
assert_eq!(bgm_meta.homepage, url);
assert_eq!(bgm_meta.bangumi_title, "叹气的亡灵想隐退");
assert_eq!(
bgm_meta.origin_poster_src,
Some(Url::parse(
"https://mikanani.me/images/Bangumi/202410/480ef127.jpg"
)?)
);
assert_eq!(bgm_meta.fansub, Some(String::from("LoliHouse")));
assert_eq!(bgm_meta.mikan_fansub_id, Some(String::from("370")));
assert_eq!(bgm_meta.mikan_bangumi_id, "3416");
assert_eq!(
bgm_meta.homepage.as_str(),
"https://mikanani.me/Home/Bangumi/3416#370"
);
assert_eq!(bgm_meta.mikan_fansub_candidates.len(), 6);
Ok(())
};
test_fn().await.expect("test parse mikan failed");
}
}

View File

@@ -0,0 +1,6 @@
pub mod defs;
pub mod errors;
pub mod html;
pub mod mikan;
pub mod rawname;
pub mod torrent;

View File

@@ -0,0 +1,5 @@
pub mod parser;
pub use parser::{
extract_season_from_title_body, parse_episode_meta_from_raw_name, RawEpisodeMeta,
};

View File

@@ -0,0 +1,843 @@
use std::borrow::Cow;
use itertools::Itertools;
use lazy_static::lazy_static;
use regex::Regex;
use serde::{Deserialize, Serialize};
use crate::extract::defs::{DIGIT_1PLUS_REG, ZH_NUM_MAP, ZH_NUM_RE};
const NAME_EXTRACT_REPLACE_ADHOC1_REPLACED: &str = "$1/$2";
lazy_static! {
static ref TITLE_RE: Regex = Regex::new(
r#"(.*|\[.*])( -? \d+|\[\d+]|\[\d+.?[vV]\d]|第\d+[话話集]|\[第?\d+[话話集]]|\[\d+.?END]|[Ee][Pp]?\d+|\[\s*\d+\s*[\-\~]\s*\d+\s*\p{scx=Han}*[话話集]\s*])(.*)"#
).unwrap();
static ref EP_COLLECTION_RE:Regex = Regex::new(r#"\[?\s*\d+\s*[\-\~]\s*\d+\s*\p{scx=Han}*合?[话話集]\s*]?"#).unwrap();
static ref MOVIE_TITLE_RE:Regex = Regex::new(r#"(.*|\[.*])(剧场版|[Mm]ovie|电影)(.*?)$"#).unwrap();
static ref RESOLUTION_RE: Regex = Regex::new(r"1080|720|2160|4K|2K").unwrap();
static ref SOURCE_L1_RE: Regex = Regex::new(r"B-Global|[Bb]aha|[Bb]ilibili|AT-X|W[Ee][Bb][Rr][Ii][Pp]|Sentai|B[Dd][Rr][Ii][Pp]|UHD[Rr][Ii][Pp]|NETFLIX").unwrap();
static ref SOURCE_L2_RE: Regex = Regex::new(r"AMZ|CR|W[Ee][Bb]|B[Dd]").unwrap();
static ref SUB_RE: Regex = Regex::new(r"[简繁日字幕]|CH|BIG5|GB").unwrap();
static ref PREFIX_RE: Regex =
Regex::new(r"[^\w\s\p{Unified_Ideograph}\p{scx=Han}\p{scx=Hira}\p{scx=Kana}-]").unwrap();
static ref EN_BRACKET_SPLIT_RE: Regex = Regex::new(r"[\[\]]").unwrap();
static ref MOVIE_SEASON_EXTRACT_RE: Regex = Regex::new(r"剧场版|Movie|电影").unwrap();
static ref MAIN_TITLE_PREFIX_PROCESS_RE1: Regex = Regex::new(r"新番|月?番").unwrap();
static ref MAIN_TITLE_PREFIX_PROCESS_RE2: Regex = Regex::new(r"[港澳台]{1,3}地区").unwrap();
static ref MAIN_TITLE_PRE_PROCESS_BACKETS_RE: Regex = Regex::new(r"\[.+\]").unwrap();
static ref MAIN_TITLE_PRE_PROCESS_BACKETS_RE_SUB1: Regex = Regex::new(r"^.*?\[").unwrap();
static ref SEASON_EXTRACT_SEASON_ALL_RE: Regex = Regex::new(r"S\d{1,2}|Season \d{1,2}|[第].[季期]|1st|2nd|3rd|\d{1,2}th").unwrap();
static ref SEASON_EXTRACT_SEASON_EN_PREFIX_RE: Regex = Regex::new(r"Season|S").unwrap();
static ref SEASON_EXTRACT_SEASON_EN_NTH_RE: Regex = Regex::new(r"1st|2nd|3rd|\d{1,2}th").unwrap();
static ref SEASON_EXTRACT_SEASON_ZH_PREFIX_RE: Regex = Regex::new(r"[第 ].*[季期(部分)]|部分").unwrap();
static ref SEASON_EXTRACT_SEASON_ZH_PREFIX_SUB_RE: Regex = Regex::new(r"[第季期 ]").unwrap();
static ref NAME_EXTRACT_REMOVE_RE: Regex = Regex::new(r"[(]仅限[港澳台]{1,3}地区[)]").unwrap();
static ref NAME_EXTRACT_SPLIT_RE: Regex = Regex::new(r"/|\s{2}|-\s{2}|\]\[").unwrap();
static ref NAME_EXTRACT_REPLACE_ADHOC1_RE: Regex = Regex::new(r"([\p{scx=Han}\s\(\)]{5,})_([a-zA-Z]{2,})").unwrap();
static ref NAME_JP_TEST: Regex = Regex::new(r"[\p{scx=Hira}\p{scx=Kana}]{2,}").unwrap();
static ref NAME_ZH_TEST: Regex = Regex::new(r"[\p{scx=Han}]{2,}").unwrap();
static ref NAME_EN_TEST: Regex = Regex::new(r"[a-zA-Z]{3,}").unwrap();
static ref TAGS_EXTRACT_SPLIT_RE: Regex = Regex::new(r"[\[\]()_]").unwrap();
static ref CLEAR_SUB_RE: Regex = Regex::new(r"_MP4|_MKV").unwrap();
}
#[derive(Debug, Clone, Serialize, Deserialize, PartialEq, Eq, Default)]
pub struct RawEpisodeMeta {
pub name_en: Option<String>,
pub name_en_no_season: Option<String>,
pub name_jp: Option<String>,
pub name_jp_no_season: Option<String>,
pub name_zh: Option<String>,
pub name_zh_no_season: Option<String>,
pub season: i32,
pub season_raw: Option<String>,
pub episode_index: i32,
pub subtitle: Option<String>,
pub source: Option<String>,
pub fansub: Option<String>,
pub resolution: Option<String>,
}
fn extract_fansub(raw_name: &str) -> Option<&str> {
let mut groups = EN_BRACKET_SPLIT_RE.splitn(raw_name, 3);
groups.nth(1)
}
fn replace_ch_bracket_to_en(raw_name: &str) -> String {
raw_name.replace('【', "[").replace('】', "]")
}
fn title_body_pre_process(title_body: &str, fansub: Option<&str>) -> eyre::Result<String> {
let raw_without_fansub = if let Some(fansub) = fansub {
let fan_sub_re = Regex::new(&format!(".{fansub}."))?;
fan_sub_re.replace_all(title_body, "")
} else {
Cow::Borrowed(title_body)
};
let raw_with_prefix_replaced = PREFIX_RE.replace_all(&raw_without_fansub, "/");
let mut arg_group = raw_with_prefix_replaced
.split('/')
.map(|s| s.trim())
.filter(|s| !s.is_empty())
.collect::<Vec<_>>();
if arg_group.len() == 1 {
arg_group = arg_group.first_mut().unwrap().split(' ').collect();
}
let mut raw = raw_without_fansub.to_string();
for arg in arg_group.iter() {
if (arg_group.len() <= 5 && MAIN_TITLE_PREFIX_PROCESS_RE1.is_match(arg))
|| (MAIN_TITLE_PREFIX_PROCESS_RE2.is_match(arg))
{
let sub = Regex::new(&format!(".{arg}."))?;
raw = sub.replace_all(&raw, "").to_string();
}
}
if let Some(m) = MAIN_TITLE_PRE_PROCESS_BACKETS_RE.find(&raw) {
if m.len() as f32 > (raw.len() as f32) * 0.5 {
let mut raw1 = MAIN_TITLE_PRE_PROCESS_BACKETS_RE_SUB1
.replace(&raw, "")
.chars()
.collect_vec();
while let Some(ch) = raw1.pop() {
if ch == ']' {
break;
}
}
raw = raw1.into_iter().collect();
}
}
Ok(raw.to_string())
}
pub fn extract_season_from_title_body(title_body: &str) -> (String, Option<String>, i32) {
let name_and_season = EN_BRACKET_SPLIT_RE.replace_all(title_body, " ");
let seasons = SEASON_EXTRACT_SEASON_ALL_RE
.find(&name_and_season)
.into_iter()
.map(|s| s.as_str())
.collect_vec();
if seasons.is_empty() {
return (title_body.to_string(), None, 1);
}
let mut season = 1;
let mut season_raw = None;
let name = SEASON_EXTRACT_SEASON_ALL_RE.replace_all(&name_and_season, "");
for s in seasons {
season_raw = Some(s);
if let Some(m) = SEASON_EXTRACT_SEASON_EN_PREFIX_RE.find(s) {
if let Ok(s) = SEASON_EXTRACT_SEASON_ALL_RE
.replace_all(m.as_str(), "")
.parse::<i32>()
{
season = s;
break;
}
}
if let Some(m) = SEASON_EXTRACT_SEASON_EN_NTH_RE.find(s) {
if let Some(s) = DIGIT_1PLUS_REG
.find(m.as_str())
.and_then(|s| s.as_str().parse::<i32>().ok())
{
season = s;
break;
}
}
if let Some(m) = SEASON_EXTRACT_SEASON_ZH_PREFIX_RE.find(s) {
if let Ok(s) = SEASON_EXTRACT_SEASON_ZH_PREFIX_SUB_RE
.replace(m.as_str(), "")
.parse::<i32>()
{
season = s;
break;
}
if let Some(m) = ZH_NUM_RE.find(m.as_str()) {
season = ZH_NUM_MAP[m.as_str()];
break;
}
}
}
(name.to_string(), season_raw.map(|s| s.to_string()), season)
}
fn extract_name_from_title_body_name_section(
title_body_name_section: &str,
) -> (Option<String>, Option<String>, Option<String>) {
let mut name_en = None;
let mut name_zh = None;
let mut name_jp = None;
let replaced1 = NAME_EXTRACT_REMOVE_RE.replace_all(title_body_name_section, "");
let replaced2 = NAME_EXTRACT_REPLACE_ADHOC1_RE
.replace_all(&replaced1, NAME_EXTRACT_REPLACE_ADHOC1_REPLACED);
let trimmed = replaced2.trim();
let mut split = NAME_EXTRACT_SPLIT_RE
.split(trimmed)
.map(|s| s.trim())
.filter(|s| !s.is_empty())
.map(|s| s.to_string())
.collect_vec();
if split.len() == 1 {
let mut split_space = split[0].split(' ').collect_vec();
let mut search_indices = vec![0];
if split_space.len() > 1 {
search_indices.push(split_space.len() - 1);
}
for i in search_indices {
if NAME_ZH_TEST.is_match(split_space[i]) {
let chs = split_space[i];
split_space.remove(i);
split = vec![chs.to_string(), split_space.join(" ")];
break;
}
}
}
for item in split {
if NAME_JP_TEST.is_match(&item) && name_jp.is_none() {
name_jp = Some(item);
} else if NAME_ZH_TEST.is_match(&item) && name_zh.is_none() {
name_zh = Some(item);
} else if NAME_EN_TEST.is_match(&item) && name_en.is_none() {
name_en = Some(item);
}
}
(name_en, name_zh, name_jp)
}
fn extract_episode_index_from_title_episode(title_episode: &str) -> Option<i32> {
DIGIT_1PLUS_REG
.find(title_episode)?
.as_str()
.parse::<i32>()
.ok()
}
fn clear_sub(sub: Option<String>) -> Option<String> {
sub.map(|s| CLEAR_SUB_RE.replace_all(&s, "").to_string())
}
fn extract_tags_from_title_extra(
title_extra: &str,
) -> (Option<String>, Option<String>, Option<String>) {
let replaced = TAGS_EXTRACT_SPLIT_RE.replace_all(title_extra, " ");
let elements = replaced
.split(' ')
.map(|s| s.trim())
.filter(|s| !s.is_empty())
.collect_vec();
let mut sub = None;
let mut resolution = None;
let mut source = None;
for element in elements.iter() {
if SUB_RE.is_match(element) {
sub = Some(element.to_string())
} else if RESOLUTION_RE.is_match(element) {
resolution = Some(element.to_string())
} else if SOURCE_L1_RE.is_match(element) {
source = Some(element.to_string())
}
}
if source.is_none() {
for element in elements {
if SOURCE_L2_RE.is_match(element) {
source = Some(element.to_string())
}
}
}
(clear_sub(sub), resolution, source)
}
pub fn check_is_movie(title: &str) -> bool {
MOVIE_TITLE_RE.is_match(title)
}
pub fn parse_episode_meta_from_raw_name(s: &str) -> eyre::Result<RawEpisodeMeta> {
let raw_title = s.trim();
let raw_title_without_ch_brackets = replace_ch_bracket_to_en(raw_title);
let fansub = extract_fansub(&raw_title_without_ch_brackets);
let movie_capture = check_is_movie(&raw_title_without_ch_brackets);
if let Some(title_re_match_obj) = MOVIE_TITLE_RE
.captures(&raw_title_without_ch_brackets)
.or(TITLE_RE.captures(&raw_title_without_ch_brackets))
{
let mut title_body = title_re_match_obj
.get(1)
.map(|s| s.as_str().trim())
.unwrap_or_else(|| unreachable!("TITLE_RE has at least 3 capture groups"))
.to_string();
let mut title_episode = title_re_match_obj
.get(2)
.map(|s| s.as_str().trim())
.unwrap_or_else(|| unreachable!("TITLE_RE has at least 3 capture groups"));
let title_extra = title_re_match_obj
.get(3)
.map(|s| s.as_str().trim())
.unwrap_or_else(|| unreachable!("TITLE_RE has at least 3 capture groups"));
if movie_capture {
title_body += title_episode;
title_episode = "";
} else if EP_COLLECTION_RE.is_match(title_episode) {
title_episode = "";
}
let title_body = title_body_pre_process(&title_body, fansub)?;
let (name_without_season, season_raw, season) = extract_season_from_title_body(&title_body);
let (name_en, name_zh, name_jp) = extract_name_from_title_body_name_section(&title_body);
let (name_en_no_season, name_zh_no_season, name_jp_no_season) =
extract_name_from_title_body_name_section(&name_without_season);
let episode_index = extract_episode_index_from_title_episode(title_episode).unwrap_or(1);
let (sub, resolution, source) = extract_tags_from_title_extra(title_extra);
Ok(RawEpisodeMeta {
name_en,
name_en_no_season,
name_jp,
name_jp_no_season,
name_zh,
name_zh_no_season,
season,
season_raw,
episode_index,
subtitle: sub,
source,
fansub: fansub.map(|s| s.to_string()),
resolution,
})
} else {
Err(eyre::eyre!(
"Can not parse episode meta from raw filename {}",
raw_title
))
}
}
#[cfg(test)]
mod tests {
use super::{parse_episode_meta_from_raw_name, RawEpisodeMeta};
fn test_raw_ep_parser_case(raw_name: &str, expected: &str) {
let expected: Option<RawEpisodeMeta> = serde_json::from_str(expected).unwrap_or_default();
let found = parse_episode_meta_from_raw_name(raw_name).ok();
if expected != found {
println!(
"expected {} and found {} are not equal",
serde_json::to_string_pretty(&expected).unwrap(),
serde_json::to_string_pretty(&found).unwrap()
)
}
assert_eq!(expected, found);
}
#[test]
fn test_parse_ep_with_all_parts_wrapped() {
test_raw_ep_parser_case(
r#"[新Sub][1月新番][我心里危险的东西 第二季][05][HEVC][10Bit][1080P][简日双语][招募翻译]"#,
r#"{
"name_zh": "我心里危险的东西",
"name_zh_no_season": "我心里危险的东西",
"season": 2,
"season_raw": "第二季",
"episode_index": 5,
"subtitle": "简日双语",
"source": null,
"fansub": "新Sub",
"resolution": "1080P"
}"#,
)
}
#[test]
fn test_parse_ep_with_title_wrapped_by_one_square_bracket_and_season_prefix() {
test_raw_ep_parser_case(
r#"【喵萌奶茶屋】★01月新番★[我内心的糟糕念头 / Boku no Kokoro no Yabai Yatsu][18][1080p][简日双语][招募翻译]"#,
r#"{
"name_en": "Boku no Kokoro no Yabai Yatsu",
"name_en_no_season": "Boku no Kokoro no Yabai Yatsu",
"name_zh": "我内心的糟糕念头",
"name_zh_no_season": "我内心的糟糕念头",
"season": 1,
"season_raw": null,
"episode_index": 18,
"subtitle": "简日双语",
"source": null,
"fansub": "喵萌奶茶屋",
"resolution": "1080p"
}"#,
);
}
#[test]
fn test_parse_ep_with_ep_and_version() {
test_raw_ep_parser_case(
r#"[LoliHouse] 因为不是真正的伙伴而被逐出勇者队伍,流落到边境展开慢活人生 2nd / Shin no Nakama 2nd - 08v2 [WebRip 1080p HEVC-10bit AAC][简繁内封字幕]"#,
r#"{
"name_en": "Shin no Nakama 2nd",
"name_en_no_season": "Shin no Nakama",
"name_zh": "因为不是真正的伙伴而被逐出勇者队伍,流落到边境展开慢活人生 2nd",
"name_zh_no_season": "因为不是真正的伙伴而被逐出勇者队伍,流落到边境展开慢活人生",
"season": 2,
"season_raw": "2nd",
"episode_index": 8,
"subtitle": "简繁内封字幕",
"source": "WebRip",
"fansub": "LoliHouse",
"resolution": "1080p"
}"#,
)
}
#[test]
fn test_parse_ep_with_en_title_only() {
test_raw_ep_parser_case(
r"[动漫国字幕组&LoliHouse] THE MARGINAL SERVICE - 08 [WebRip 1080p HEVC-10bit AAC][简繁内封字幕]",
r#"{
"name_en": "THE MARGINAL SERVICE",
"name_en_no_season": "THE MARGINAL SERVICE",
"season": 1,
"episode_index": 8,
"subtitle": "简繁内封字幕",
"source": "WebRip",
"fansub": "动漫国字幕组&LoliHouse",
"resolution": "1080p"
}"#,
)
}
#[test]
fn test_parse_ep_with_two_zh_title() {
test_raw_ep_parser_case(
r#"[LoliHouse] 事与愿违的不死冒险者 / 非自愿的不死冒险者 / Nozomanu Fushi no Boukensha - 01 [WebRip 1080p HEVC-10bit AAC][简繁内封字幕]"#,
r#"{
"name_en": "Nozomanu Fushi no Boukensha",
"name_en_no_season": "Nozomanu Fushi no Boukensha",
"name_zh": "事与愿违的不死冒险者",
"name_zh_no_season": "事与愿违的不死冒险者",
"season": 1,
"season_raw": null,
"episode_index": 1,
"subtitle": "简繁内封字幕",
"source": "WebRip",
"fansub": "LoliHouse",
"resolution": "1080p"
}"#,
)
}
#[test]
fn test_parse_ep_with_en_zh_jp_titles() {
test_raw_ep_parser_case(
r#"[喵萌奶茶屋&LoliHouse] 碰之道 / ぽんのみち / Pon no Michi - 07 [WebRip 1080p HEVC-10bit AAC][简繁日内封字幕]"#,
r#"{
"name_en": "Pon no Michi",
"name_jp": "ぽんのみち",
"name_zh": "碰之道",
"name_en_no_season": "Pon no Michi",
"name_jp_no_season": "ぽんのみち",
"name_zh_no_season": "碰之道",
"season": 1,
"season_raw": null,
"episode_index": 7,
"subtitle": "简繁日内封字幕",
"source": "WebRip",
"fansub": "喵萌奶茶屋&LoliHouse",
"resolution": "1080p"
}"#,
)
}
#[test]
fn test_parse_ep_with_nth_season() {
test_raw_ep_parser_case(
r#"[ANi] Yowai Character Tomozakikun / 弱角友崎同学 2nd STAGE - 09 [1080P][Baha][WEB-DL][AAC AVC][CHT][MP4]"#,
r#"{
"name_en": "Yowai Character Tomozakikun",
"name_en_no_season": "Yowai Character Tomozakikun",
"name_zh": "弱角友崎同学 2nd STAGE",
"name_zh_no_season": "弱角友崎同学",
"season": 2,
"season_raw": "2nd",
"episode_index": 9,
"subtitle": "CHT",
"source": "Baha",
"fansub": "ANi",
"resolution": "1080P"
}"#,
)
}
#[test]
fn test_parse_ep_with_season_en_and_season_zh() {
test_raw_ep_parser_case(
r#"[豌豆字幕组&LoliHouse] 王者天下 第五季 / Kingdom S5 - 07 [WebRip 1080p HEVC-10bit AAC][简繁外挂字幕]"#,
r#"{
"name_en": "Kingdom S5",
"name_en_no_season": "Kingdom",
"name_zh": "王者天下 第五季",
"name_zh_no_season": "王者天下",
"season": 5,
"season_raw": "第五季",
"episode_index": 7,
"subtitle": "简繁外挂字幕",
"source": "WebRip",
"fansub": "豌豆字幕组&LoliHouse",
"resolution": "1080p"
}"#,
)
}
#[test]
fn test_parse_ep_with_airota_fansub_style_case1() {
test_raw_ep_parser_case(
r#"【千夏字幕组】【爱丽丝与特蕾丝的虚幻工厂_Alice to Therese no Maboroshi Koujou】[剧场版][WebRip_1080p_HEVC][简繁内封][招募新人]"#,
r#"{
"name_en": "Alice to Therese no Maboroshi Koujou",
"name_en_no_season": "Alice to Therese no Maboroshi Koujou",
"name_zh": "爱丽丝与特蕾丝的虚幻工厂",
"name_zh_no_season": "爱丽丝与特蕾丝的虚幻工厂",
"season": 1,
"episode_index": 1,
"subtitle": "简繁内封",
"source": "WebRip",
"fansub": "千夏字幕组",
"resolution": "1080p"
}"#,
)
}
#[test]
fn test_parse_ep_with_airota_fansub_style_case2() {
test_raw_ep_parser_case(
r#"[千夏字幕组&喵萌奶茶屋][电影 轻旅轻营 (摇曳露营) _Yuru Camp Movie][剧场版][UHDRip_2160p_HEVC][繁体][千夏15周年]"#,
r#"{
"name_en": "Yuru Camp Movie",
"name_en_no_season": "Yuru Camp Movie",
"name_zh": "电影 轻旅轻营 (摇曳露营)",
"name_zh_no_season": "电影 轻旅轻营 (摇曳露营)",
"season": 1,
"episode_index": 1,
"subtitle": "繁体",
"source": "UHDRip",
"fansub": "千夏字幕组&喵萌奶茶屋",
"resolution": "2160p"
}"#,
)
}
#[test]
fn test_parse_ep_with_large_episode_style() {
test_raw_ep_parser_case(
r#"[梦蓝字幕组]New Doraemon 哆啦A梦新番[747][2023.02.25][AVC][1080P][GB_JP][MP4]"#,
r#"{
"name_en": "New Doraemon",
"name_en_no_season": "New Doraemon",
"name_zh": "哆啦A梦新番",
"name_zh_no_season": "哆啦A梦新番",
"season": 1,
"episode_index": 747,
"subtitle": "GB",
"fansub": "梦蓝字幕组",
"resolution": "1080P"
}"#,
)
}
#[test]
fn test_parse_ep_with_many_square_brackets_split_title() {
test_raw_ep_parser_case(
r#"【MCE汉化组】[剧场版-摇曳露营][Yuru Camp][Movie][简日双语][1080P][x264 AAC]"#,
r#"{
"name_en": "Yuru Camp",
"name_en_no_season": "Yuru Camp",
"name_zh": "剧场版-摇曳露营",
"name_zh_no_season": "剧场版-摇曳露营",
"season": 1,
"episode_index": 1,
"subtitle": "简日双语",
"fansub": "MCE汉化组",
"resolution": "1080P"
}"#,
)
}
#[test]
fn test_parse_ep_with_implicit_lang_title_sep() {
test_raw_ep_parser_case(
r#"[织梦字幕组][尼尔:机械纪元 NieR Automata Ver1.1a][02集][1080P][AVC][简日双语]"#,
r#"{
"name_en": "NieR Automata Ver1.1a",
"name_en_no_season": "NieR Automata Ver1.1a",
"name_zh": "尼尔:机械纪元",
"name_zh_no_season": "尼尔:机械纪元",
"season": 1,
"episode_index": 2,
"subtitle": "简日双语",
"fansub": "织梦字幕组",
"resolution": "1080P"
}"#,
)
}
#[test]
fn test_parse_ep_with_square_brackets_wrapped_and_space_split() {
test_raw_ep_parser_case(
r#"[天月搬运组][迷宫饭 Delicious in Dungeon][03][日语中字][MKV][1080P][NETFLIX][高画质版]"#,
r#"
{
"name_en": "Delicious in Dungeon",
"name_en_no_season": "Delicious in Dungeon",
"name_zh": "迷宫饭",
"name_zh_no_season": "迷宫饭",
"season": 1,
"episode_index": 3,
"subtitle": "日语中字",
"source": "NETFLIX",
"fansub": "天月搬运组",
"resolution": "1080P"
}
"#,
)
}
#[test]
fn test_parse_ep_with_start_with_brackets_wrapped_season_info_prefix() {
test_raw_ep_parser_case(
r#"[爱恋字幕社][1月新番][迷宫饭][Dungeon Meshi][01][1080P][MP4][简日双语] "#,
r#"{
"name_en": "Dungeon Meshi",
"name_en_no_season": "Dungeon Meshi",
"name_zh": "迷宫饭",
"name_zh_no_season": "迷宫饭",
"season": 1,
"episode_index": 1,
"subtitle": "简日双语",
"fansub": "爱恋字幕社",
"resolution": "1080P"
}"#,
)
}
#[test]
fn test_parse_ep_with_small_no_title_extra_brackets_case() {
test_raw_ep_parser_case(
r#"[ANi] Mahou Shoujo ni Akogarete / 梦想成为魔法少女 [年龄限制版] - 09 [1080P][Baha][WEB-DL][AAC AVC][CHT][MP4]"#,
r#"{
"name_en": "Mahou Shoujo ni Akogarete",
"name_en_no_season": "Mahou Shoujo ni Akogarete",
"name_zh": "梦想成为魔法少女 [年龄限制版]",
"name_zh_no_season": "梦想成为魔法少女 [年龄限制版]",
"season": 1,
"episode_index": 9,
"subtitle": "CHT",
"source": "Baha",
"fansub": "ANi",
"resolution": "1080P"
}"#,
)
}
#[test]
fn test_parse_ep_title_leading_space_style() {
test_raw_ep_parser_case(
r#"[ANi] 16bit 的感动 ANOTHER LAYER - 01 [1080P][Baha][WEB-DL][AAC AVC][CHT][MP4]"#,
r#"{
"name_zh": "16bit 的感动 ANOTHER LAYER",
"name_zh_no_season": "16bit 的感动 ANOTHER LAYER",
"season": 1,
"season_raw": null,
"episode_index": 1,
"subtitle": "CHT",
"source": "Baha",
"fansub": "ANi",
"resolution": "1080P"
}"#,
)
}
#[test]
fn test_parse_ep_title_leading_month_and_wrapped_brackets_style() {
test_raw_ep_parser_case(
r#"【喵萌奶茶屋】★07月新番★[银砂糖师与黑妖精 ~ Sugar Apple Fairy Tale ~][13][1080p][简日双语][招募翻译]"#,
r#"{
"name_en": "~ Sugar Apple Fairy Tale ~",
"name_en_no_season": "~ Sugar Apple Fairy Tale ~",
"name_zh": "银砂糖师与黑妖精",
"name_zh_no_season": "银砂糖师与黑妖精",
"season": 1,
"episode_index": 13,
"subtitle": "简日双语",
"fansub": "喵萌奶茶屋",
"resolution": "1080p"
}"#,
)
}
#[test]
fn test_parse_ep_title_leading_month_style() {
test_raw_ep_parser_case(
r#"【极影字幕社】★4月新番 天国大魔境 Tengoku Daimakyou 第05话 GB 720P MP4字幕社招人内详"#,
r#"{
"name_en": "Tengoku Daimakyou",
"name_en_no_season": "Tengoku Daimakyou",
"name_zh": "天国大魔境",
"name_zh_no_season": "天国大魔境",
"season": 1,
"episode_index": 5,
"subtitle": "字幕社招人内详",
"source": null,
"fansub": "极影字幕社",
"resolution": "720P"
}"#,
)
}
#[test]
fn test_parse_ep_tokusatsu_style() {
test_raw_ep_parser_case(
r#"[MagicStar] 假面骑士Geats / 仮面ライダーギーツ EP33 [WEBDL] [1080p] [TTFC]【生】"#,
r#"{
"name_jp": "仮面ライダーギーツ",
"name_jp_no_season": "仮面ライダーギーツ",
"name_zh": "假面骑士Geats",
"name_zh_no_season": "假面骑士Geats",
"season": 1,
"episode_index": 33,
"source": "WEBDL",
"fansub": "MagicStar",
"resolution": "1080p"
}"#,
)
}
#[test]
fn test_parse_ep_with_multi_lang_zh_title() {
test_raw_ep_parser_case(
r#"[百冬练习组&LoliHouse] BanG Dream! 少女乐团派对☆PICO FEVER / Garupa Pico: Fever! - 26 [WebRip 1080p HEVC-10bit AAC][简繁内封字幕][END] [101.69 MB]"#,
r#"{
"name_en": "Garupa Pico: Fever!",
"name_en_no_season": "Garupa Pico: Fever!",
"name_zh": "BanG Dream! 少女乐团派对☆PICO FEVER",
"name_zh_no_season": "BanG Dream! 少女乐团派对☆PICO FEVER",
"season": 1,
"episode_index": 26,
"subtitle": "简繁内封字幕",
"source": "WebRip",
"fansub": "百冬练习组&LoliHouse",
"resolution": "1080p"
}"#,
)
}
#[test]
fn test_ep_collections() {
test_raw_ep_parser_case(
r#"[奶²&LoliHouse] 蘑菇狗 / Kinokoinu: Mushroom Pup [01-12 精校合集][WebRip 1080p HEVC-10bit AAC][简日内封字幕]"#,
r#"{
"name_en": "Kinokoinu: Mushroom Pup",
"name_en_no_season": "Kinokoinu: Mushroom Pup",
"name_zh": "蘑菇狗",
"name_zh_no_season": "蘑菇狗",
"season": 1,
"episode_index": 1,
"subtitle": "简日内封字幕",
"source": "WebRip",
"fansub": "奶²&LoliHouse",
"resolution": "1080p",
"name": " 蘑菇狗 / Kinokoinu: Mushroom Pup [01-12 精校合集]"
}"#,
);
test_raw_ep_parser_case(
r#"[LoliHouse] 叹气的亡灵想隐退 / Nageki no Bourei wa Intai shitai [01-13 合集][WebRip 1080p HEVC-10bit AAC][简繁内封字幕][Fin]"#,
r#"{
"name_en": "Nageki no Bourei wa Intai shitai",
"name_en_no_season": "Nageki no Bourei wa Intai shitai",
"name_jp": null,
"name_jp_no_season": null,
"name_zh": "叹气的亡灵想隐退",
"name_zh_no_season": "叹气的亡灵想隐退",
"season": 1,
"season_raw": null,
"episode_index": 1,
"subtitle": "简繁内封字幕",
"source": "WebRip",
"fansub": "LoliHouse",
"resolution": "1080p"
}"#,
);
test_raw_ep_parser_case(
r#"[LoliHouse] 精灵幻想记 第二季 / Seirei Gensouki S2 [01-12 合集][WebRip 1080p HEVC-10bit AAC][简繁内封字幕][Fin]"#,
r#"{
"name_en": "Seirei Gensouki S2",
"name_en_no_season": "Seirei Gensouki",
"name_zh": "精灵幻想记 第二季",
"name_zh_no_season": "精灵幻想记",
"season": 2,
"season_raw": "第二季",
"episode_index": 1,
"subtitle": "简繁内封字幕",
"source": "WebRip",
"fansub": "LoliHouse",
"resolution": "1080p"
}"#,
);
test_raw_ep_parser_case(
r#"[喵萌奶茶屋&LoliHouse] 超自然武装当哒当 / 胆大党 / Dandadan [01-12 精校合集][WebRip 1080p HEVC-10bit AAC][简繁日内封字幕][Fin]"#,
r#" {
"name_en": "Dandadan",
"name_en_no_season": "Dandadan",
"name_zh": "超自然武装当哒当",
"name_zh_no_season": "超自然武装当哒当",
"season": 1,
"episode_index": 1,
"subtitle": "简繁日内封字幕",
"source": "WebRip",
"fansub": "喵萌奶茶屋&LoliHouse",
"resolution": "1080p"
}"#,
);
}
// TODO: FIXME
#[test]
fn test_bad_cases() {
test_raw_ep_parser_case(
r#"[7³ACG x 桜都字幕组] 摇曳露营△ 剧场版/映画 ゆるキャン△/Eiga Yuru Camp△ [简繁字幕] BDrip 1080p x265 FLAC 2.0"#,
r#"{
"name_zh": "摇曳露营△剧场版",
"name_zh_no_season": "摇曳露营△剧场版",
"season": 1,
"season_raw": null,
"episode_index": 1,
"subtitle": "简繁字幕",
"source": "BDrip",
"fansub": "7³ACG x 桜都字幕组",
"resolution": "1080p"
}"#,
);
test_raw_ep_parser_case(
r#"【幻樱字幕组】【4月新番】【古见同学有交流障碍症 第二季 Komi-san wa, Komyushou Desu. S02】【22】【GB_MP4】【1920X1080】"#,
r#"{
"name_en": "第二季 Komi-san wa, Komyushou Desu. S02",
"name_en_no_season": "Komi-san wa, Komyushou Desu.",
"name_zh": "古见同学有交流障碍症",
"name_zh_no_season": "古见同学有交流障碍症",
"season": 2,
"season_raw": "第二季",
"episode_index": 22,
"subtitle": "GB",
"fansub": "幻樱字幕组",
"resolution": "1920X1080"
}"#,
);
}
}

View File

@@ -0,0 +1,3 @@
mod parser;
pub use parser::*;

View File

@@ -0,0 +1,316 @@
use eyre::OptionExt;
use fancy_regex::Regex as FancyRegex;
use lazy_static::lazy_static;
use quirks_path::Path;
use regex::Regex;
use serde::{Deserialize, Serialize};
use crate::extract::defs::SUBTITLE_LANG;
lazy_static! {
static ref TORRENT_EP_PARSE_RULES: Vec<FancyRegex> = {
vec![
FancyRegex::new(
r"(.*) - (\d{1,4}(?!\d|p)|\d{1,4}\.\d{1,2}(?!\d|p))(?:v\d{1,2})?(?: )?(?:END)?(.*)",
)
.unwrap(),
FancyRegex::new(
r"(.*)[\[\ E](\d{1,4}|\d{1,4}\.\d{1,2})(?:v\d{1,2})?(?: )?(?:END)?[\]\ ](.*)",
)
.unwrap(),
FancyRegex::new(r"(.*)\[(?:第)?(\d*\.*\d*)[话集話](?:END)?\](.*)").unwrap(),
FancyRegex::new(r"(.*)第?(\d*\.*\d*)[话話集](?:END)?(.*)").unwrap(),
FancyRegex::new(r"(.*)(?:S\d{2})?EP?(\d+)(.*)").unwrap(),
]
};
static ref GET_FANSUB_SPLIT_RE: Regex = Regex::new(r"[\[\]()【】()]").unwrap();
static ref GET_FANSUB_FULL_MATCH_RE: Regex = Regex::new(r"^\d+$").unwrap();
static ref GET_SEASON_AND_TITLE_SUB_RE: Regex = Regex::new(r"([Ss]|Season )\d{1,3}").unwrap();
static ref GET_SEASON_AND_TITLE_FIND_RE: Regex =
Regex::new(r"([Ss]|Season )(\d{1,3})").unwrap();
}
#[derive(Clone, Debug, Serialize, Deserialize, PartialEq, Eq)]
pub struct TorrentEpisodeMediaMeta {
pub fansub: Option<String>,
pub title: String,
pub season: i32,
pub episode_index: i32,
pub extname: String,
}
#[derive(Clone, Debug, Serialize, Deserialize, PartialEq, Eq)]
pub struct TorrentEpisodeSubtitleMeta {
pub media: TorrentEpisodeMediaMeta,
pub lang: Option<String>,
}
fn get_fansub(group_and_title: &str) -> (Option<&str>, &str) {
let n = GET_FANSUB_SPLIT_RE
.split(group_and_title)
.map(|s| s.trim())
.filter(|s| !s.is_empty())
.collect::<Vec<_>>();
match (n.first(), n.get(1)) {
(None, None) => (None, ""),
(Some(n0), None) => (None, *n0),
(Some(n0), Some(n1)) => {
if GET_FANSUB_FULL_MATCH_RE.is_match(n1) {
(None, group_and_title)
} else {
(Some(*n0), *n1)
}
}
_ => unreachable!("vec contains n1 must contains n0"),
}
}
fn get_season_and_title(season_and_title: &str) -> (String, i32) {
let replaced_title = GET_SEASON_AND_TITLE_SUB_RE.replace_all(season_and_title, "");
let title = replaced_title.trim().to_string();
let season = GET_SEASON_AND_TITLE_FIND_RE
.captures(season_and_title)
.map(|m| {
m.get(2)
.unwrap_or_else(|| unreachable!("season regex should have 2 groups"))
.as_str()
.parse::<i32>()
.unwrap_or_else(|_| unreachable!("season should be a number"))
})
.unwrap_or(1);
(title, season)
}
fn get_subtitle_lang(media_name: &str) -> Option<&str> {
let media_name_lower = media_name.to_lowercase();
for (lang, lang_aliases) in SUBTITLE_LANG.iter() {
if lang_aliases
.iter()
.any(|alias| media_name_lower.contains(alias))
{
return Some(lang);
}
}
None
}
pub fn parse_episode_media_meta_from_torrent(
torrent_path: &Path,
torrent_name: Option<&str>,
season: Option<i32>,
) -> eyre::Result<TorrentEpisodeMediaMeta> {
let media_name = torrent_path
.file_name()
.ok_or_else(|| eyre::eyre!("failed to get file name of {}", torrent_path))?;
let mut match_obj = None;
for rule in TORRENT_EP_PARSE_RULES.iter() {
match_obj = if let Some(torrent_name) = torrent_name.as_ref() {
rule.captures(torrent_name)?
} else {
rule.captures(media_name)?
};
if match_obj.is_some() {
break;
}
}
if let Some(match_obj) = match_obj {
let group_season_and_title = match_obj
.get(1)
.ok_or_else(|| eyre::eyre!("should have 1 group"))?
.as_str();
let (fansub, season_and_title) = get_fansub(group_season_and_title);
let (title, season) = if let Some(season) = season {
let (title, _) = get_season_and_title(season_and_title);
(title, season)
} else {
get_season_and_title(season_and_title)
};
let episode_index = match_obj
.get(2)
.ok_or_eyre("should have 2 group")?
.as_str()
.parse::<i32>()
.unwrap_or(1);
let extname = torrent_path
.extension()
.map(|e| format!(".{}", e))
.unwrap_or_default();
Ok(TorrentEpisodeMediaMeta {
fansub: fansub.map(|s| s.to_string()),
title,
season,
episode_index,
extname,
})
} else {
Err(eyre::eyre!(
"failed to parse episode media meta from torrent_path='{}' torrent_name='{:?}'",
torrent_path,
torrent_name
))
}
}
pub fn parse_episode_subtitle_meta_from_torrent(
torrent_path: &Path,
torrent_name: Option<&str>,
season: Option<i32>,
) -> eyre::Result<TorrentEpisodeSubtitleMeta> {
let media_meta = parse_episode_media_meta_from_torrent(torrent_path, torrent_name, season)?;
let media_name = torrent_path
.file_name()
.ok_or_else(|| eyre::eyre!("failed to get file name of {}", torrent_path))?;
let lang = get_subtitle_lang(media_name);
Ok(TorrentEpisodeSubtitleMeta {
media: media_meta,
lang: lang.map(|s| s.to_string()),
})
}
#[cfg(test)]
mod tests {
use quirks_path::Path;
use super::{
parse_episode_media_meta_from_torrent, parse_episode_subtitle_meta_from_torrent,
TorrentEpisodeMediaMeta, TorrentEpisodeSubtitleMeta,
};
#[test]
fn test_lilith_raws_media() {
test_torrent_ep_parser(
r#"[Lilith-Raws] Boku no Kokoro no Yabai Yatsu - 01 [Baha][WEB-DL][1080p][AVC AAC][CHT][MP4].mp4"#,
r#"{"fansub": "Lilith-Raws", "title": "Boku no Kokoro no Yabai Yatsu", "season": 1, "episode_index": 1, "extname": ".mp4"}"#,
);
}
#[test]
fn test_sakurato_media() {
test_torrent_ep_parser(
r#"[Sakurato] Tonikaku Kawaii S2 [03][AVC-8bit 1080p AAC][CHS].mp4"#,
r#"{"fansub": "Sakurato", "title": "Tonikaku Kawaii", "season": 2, "episode_index": 3, "extname": ".mp4"}"#,
)
}
#[test]
fn test_lolihouse_media() {
test_torrent_ep_parser(
r#"[SweetSub&LoliHouse] Heavenly Delusion - 08 [WebRip 1080p HEVC-10bit AAC ASSx2].mkv"#,
r#"{"fansub": "SweetSub&LoliHouse", "title": "Heavenly Delusion", "season": 1, "episode_index": 8, "extname": ".mkv"}"#,
)
}
#[test]
fn test_sbsub_media() {
test_torrent_ep_parser(
r#"[SBSUB][CONAN][1082][V2][1080P][AVC_AAC][CHS_JP](C1E4E331).mp4"#,
r#"{"fansub": "SBSUB", "title": "CONAN", "season": 1, "episode_index": 1082, "extname": ".mp4"}"#,
)
}
#[test]
fn test_non_fansub_media() {
test_torrent_ep_parser(
r#"海盗战记 (2019) S04E11.mp4"#,
r#"{"title": "海盗战记 (2019)", "season": 4, "episode_index": 11, "extname": ".mp4"}"#,
)
}
#[test]
fn test_non_fansub_media_with_dirname() {
test_torrent_ep_parser(
r#"海盗战记/海盗战记 S01E01.mp4"#,
r#"{"title": "海盗战记", "season": 1, "episode_index": 1, "extname": ".mp4"}"#,
);
}
#[test]
fn test_non_fansub_tc_subtitle() {
test_torrent_ep_parser(
r#"海盗战记 S01E08.zh-tw.ass"#,
r#"{"media": { "title": "海盗战记", "season": 1, "episode_index": 8, "extname": ".ass" }, "lang": "zh-tw"}"#,
);
}
#[test]
fn test_non_fansub_sc_subtitle() {
test_torrent_ep_parser(
r#"海盗战记 S01E01.SC.srt"#,
r#"{ "media": { "title": "海盗战记", "season": 1, "episode_index": 1, "extname": ".srt" }, "lang": "zh" }"#,
)
}
#[test]
fn test_non_fansub_media_with_season_zero() {
test_torrent_ep_parser(
r#"水星的魔女(2022) S00E19.mp4"#,
r#"{"fansub": null,"title": "水星的魔女(2022)","season": 0,"episode_index": 19,"extname": ".mp4"}"#,
)
}
#[test]
fn test_shimian_fansub_media() {
test_torrent_ep_parser(
r#"【失眠搬运组】放学后失眠的你-Kimi wa Houkago Insomnia - 06 [bilibili - 1080p AVC1 CHS-JP].mp4"#,
r#"{"fansub": "失眠搬运组","title": "放学后失眠的你-Kimi wa Houkago Insomnia","season": 1,"episode_index": 6,"extname": ".mp4"}"#,
)
}
pub fn test_torrent_ep_parser(raw_name: &str, expected: &str) {
let extname = Path::new(raw_name)
.extension()
.map(|e| format!(".{}", e))
.unwrap_or_default()
.to_lowercase();
if extname == ".srt" || extname == ".ass" {
let expected: Option<TorrentEpisodeSubtitleMeta> = serde_json::from_str(expected).ok();
let found_raw =
parse_episode_subtitle_meta_from_torrent(Path::new(raw_name), None, None);
let found = found_raw.as_ref().ok().cloned();
if expected != found {
if found_raw.is_ok() {
println!(
"expected {} and found {} are not equal",
serde_json::to_string_pretty(&expected).unwrap(),
serde_json::to_string_pretty(&found).unwrap()
)
} else {
println!(
"expected {} and found {:#?} are not equal",
serde_json::to_string_pretty(&expected).unwrap(),
found_raw
)
}
}
assert_eq!(expected, found);
} else {
let expected: Option<TorrentEpisodeMediaMeta> = serde_json::from_str(expected).ok();
let found_raw = parse_episode_media_meta_from_torrent(Path::new(raw_name), None, None);
let found = found_raw.as_ref().ok().cloned();
if expected != found {
if found_raw.is_ok() {
println!(
"expected {} and found {} are not equal",
serde_json::to_string_pretty(&expected).unwrap(),
serde_json::to_string_pretty(&found).unwrap()
)
} else {
println!(
"expected {} and found {:#?} are not equal",
serde_json::to_string_pretty(&expected).unwrap(),
found_raw
)
}
}
assert_eq!(expected, found);
}
}
}

View File

@@ -0,0 +1,24 @@
use bytes::Bytes;
use reqwest::IntoUrl;
use super::{core::DEFAULT_HTTP_CLIENT_USER_AGENT, HttpClient};
pub async fn download_bytes<T: IntoUrl>(url: T) -> eyre::Result<Bytes> {
let request_client = reqwest::Client::builder()
.user_agent(DEFAULT_HTTP_CLIENT_USER_AGENT)
.build()?;
let bytes = request_client.get(url).send().await?.bytes().await?;
Ok(bytes)
}
pub async fn download_bytes_with_client<T: IntoUrl>(
client: Option<&HttpClient>,
url: T,
) -> eyre::Result<Bytes> {
if let Some(client) = client {
let bytes = client.get(url).send().await?.bytes().await?;
Ok(bytes)
} else {
download_bytes(url).await
}
}

View File

@@ -0,0 +1,96 @@
use std::{ops::Deref, time::Duration};
use axum::http::Extensions;
use leaky_bucket::RateLimiter;
use reqwest::{ClientBuilder, Request, Response};
use reqwest_middleware::{
ClientBuilder as ClientWithMiddlewareBuilder, ClientWithMiddleware, Next,
};
use reqwest_retry::{policies::ExponentialBackoff, RetryTransientMiddleware};
use reqwest_tracing::TracingMiddleware;
use serde::{Deserialize, Serialize};
use serde_with::serde_as;
use super::DEFAULT_HTTP_CLIENT_USER_AGENT;
#[serde_as]
#[derive(Debug, Clone, PartialEq, Eq, Serialize, Deserialize, Default)]
pub struct HttpClientConfig {
pub exponential_backoff_max_retries: Option<u32>,
pub leaky_bucket_max_tokens: Option<u32>,
pub leaky_bucket_initial_tokens: Option<u32>,
pub leaky_bucket_refill_tokens: Option<u32>,
#[serde_as(as = "Option<serde_with::DurationMilliSeconds>")]
pub leaky_bucket_refill_interval: Option<Duration>,
pub user_agent: Option<String>,
}
pub struct HttpClient {
client: ClientWithMiddleware,
}
impl Deref for HttpClient {
type Target = ClientWithMiddleware;
fn deref(&self) -> &Self::Target {
&self.client
}
}
pub struct RateLimiterMiddleware {
rate_limiter: RateLimiter,
}
#[async_trait::async_trait]
impl reqwest_middleware::Middleware for RateLimiterMiddleware {
async fn handle(
&self,
req: Request,
extensions: &'_ mut Extensions,
next: Next<'_>,
) -> reqwest_middleware::Result<Response> {
self.rate_limiter.acquire_one().await;
next.run(req, extensions).await
}
}
impl HttpClient {
pub fn new(config: Option<HttpClientConfig>) -> reqwest::Result<Self> {
let mut config = config.unwrap_or_default();
let retry_policy = ExponentialBackoff::builder()
.build_with_max_retries(config.exponential_backoff_max_retries.take().unwrap_or(3));
let rate_limiter = RateLimiter::builder()
.max(config.leaky_bucket_max_tokens.take().unwrap_or(3) as usize)
.initial(
config
.leaky_bucket_initial_tokens
.take()
.unwrap_or_default() as usize,
)
.refill(config.leaky_bucket_refill_tokens.take().unwrap_or(1) as usize)
.interval(
config
.leaky_bucket_refill_interval
.take()
.unwrap_or_else(|| Duration::from_millis(500)),
)
.build();
let client = ClientBuilder::new()
.user_agent(
config
.user_agent
.take()
.unwrap_or_else(|| DEFAULT_HTTP_CLIENT_USER_AGENT.to_owned()),
)
.build()?;
Ok(Self {
client: ClientWithMiddlewareBuilder::new(client)
.with(TracingMiddleware::default())
.with(RateLimiterMiddleware { rate_limiter })
.with(RetryTransientMiddleware::new_with_policy(retry_policy))
.build(),
})
}
}

View File

@@ -0,0 +1 @@
pub const DEFAULT_HTTP_CLIENT_USER_AGENT: &str = "Wget/1.13.4 (linux-gnu)";

View File

@@ -0,0 +1,23 @@
use reqwest::IntoUrl;
use super::{core::DEFAULT_HTTP_CLIENT_USER_AGENT, HttpClient};
pub async fn download_html<U: IntoUrl>(url: U) -> eyre::Result<String> {
let request_client = reqwest::Client::builder()
.user_agent(DEFAULT_HTTP_CLIENT_USER_AGENT)
.build()?;
let content = request_client.get(url).send().await?.text().await?;
Ok(content)
}
pub async fn download_html_with_client<T: IntoUrl>(
client: Option<&HttpClient>,
url: T,
) -> eyre::Result<String> {
if let Some(client) = client {
let content = client.get(url).send().await?.text().await?;
Ok(content)
} else {
download_html(url).await
}
}

View File

@@ -0,0 +1,18 @@
use bytes::Bytes;
use reqwest::IntoUrl;
use super::{
bytes::{download_bytes, download_bytes_with_client},
HttpClient,
};
pub async fn download_image<U: IntoUrl>(url: U) -> eyre::Result<Bytes> {
download_bytes(url).await
}
pub async fn download_image_with_client<T: IntoUrl>(
client: Option<&HttpClient>,
url: T,
) -> eyre::Result<Bytes> {
download_bytes_with_client(client, url).await
}

View File

@@ -0,0 +1,11 @@
pub mod bytes;
pub mod client;
pub mod core;
pub mod html;
pub mod image;
pub use core::DEFAULT_HTTP_CLIENT_USER_AGENT;
pub use bytes::download_bytes;
pub use client::{HttpClient, HttpClientConfig};
pub use image::download_image;

13
apps/recorder/src/lib.rs Normal file
View File

@@ -0,0 +1,13 @@
#![feature(duration_constructors, assert_matches, unboxed_closures)]
pub mod app;
pub mod config;
pub mod controllers;
pub mod dal;
pub mod extract;
pub mod fetch;
pub mod migrations;
pub mod models;
pub mod tasks;
pub mod views;
pub mod workers;

View File

@@ -0,0 +1,373 @@
use std::collections::HashSet;
use sea_orm::{DeriveIden, Statement};
use sea_orm_migration::prelude::{extension::postgres::IntoTypeRef, *};
use crate::migrations::extension::postgres::Type;
#[derive(DeriveIden)]
pub enum GeneralIds {
CreatedAt,
UpdatedAt,
}
#[derive(DeriveIden)]
pub enum Subscribers {
Table,
Id,
Pid,
DisplayName,
DownloaderId,
BangumiConf,
}
#[derive(DeriveIden)]
pub enum Subscriptions {
Table,
Id,
DisplayName,
SubscriberId,
Category,
SourceUrl,
Enabled,
}
#[derive(DeriveIden)]
pub enum Bangumi {
Table,
Id,
MikanBangumiId,
DisplayName,
SubscriptionId,
SubscriberId,
RawName,
Season,
SeasonRaw,
Fansub,
MikanFansubId,
Filter,
RssLink,
PosterLink,
SavePath,
Deleted,
Homepage,
Extra,
}
#[derive(DeriveIden)]
pub enum Episodes {
Table,
Id,
MikanEpisodeId,
RawName,
DisplayName,
BangumiId,
SubscriptionId,
SubscriberId,
DownloadId,
SavePath,
Resolution,
Season,
SeasonRaw,
Fansub,
PosterLink,
EpisodeIndex,
Homepage,
Subtitle,
Deleted,
Source,
Extra,
}
#[derive(DeriveIden)]
pub enum Downloads {
Table,
Id,
OriginalName,
DisplayName,
SubscriptionId,
Status,
CurrSize,
AllSize,
Mime,
Url,
Homepage,
SavePath,
}
#[derive(DeriveIden)]
pub enum Downloaders {
Table,
Id,
Category,
Endpoint,
Password,
Username,
SubscriberId,
SavePath,
}
macro_rules! create_postgres_enum_for_active_enum {
($manager: expr, $active_enum: expr, $($enum_value:expr),+) => {
{
use sea_orm::ActiveEnum;
let values = [$($enum_value,)+].map(|v| ActiveEnum::to_value(&v));
($manager).create_postgres_enum_for_active_enum($active_enum, values)
}
};
}
#[async_trait::async_trait]
pub trait CustomSchemaManagerExt {
async fn create_postgres_auto_update_ts_fn(&self, col_name: &str) -> Result<(), DbErr>;
async fn create_postgres_auto_update_ts_fn_for_col<C: IntoIden + 'static + Send>(
&self,
col: C,
) -> Result<(), DbErr> {
let column_ident = col.into_iden();
self.create_postgres_auto_update_ts_fn(&column_ident.to_string())
.await?;
Ok(())
}
async fn create_postgres_auto_update_ts_trigger(
&self,
tab_name: &str,
col_name: &str,
) -> Result<(), DbErr>;
async fn create_postgres_auto_update_ts_trigger_for_col<
T: IntoIden + 'static + Send,
C: IntoIden + 'static + Send,
>(
&self,
tab: T,
col: C,
) -> Result<(), DbErr> {
let column_ident = col.into_iden();
let table_ident = tab.into_iden();
self.create_postgres_auto_update_ts_trigger(
&table_ident.to_string(),
&column_ident.to_string(),
)
.await?;
Ok(())
}
async fn drop_postgres_auto_update_ts_fn(&self, col_name: &str) -> Result<(), DbErr>;
async fn drop_postgres_auto_update_ts_fn_for_col<C: IntoIden + Send>(
&self,
col: C,
) -> Result<(), DbErr> {
let column_ident = col.into_iden();
self.drop_postgres_auto_update_ts_fn(&column_ident.to_string())
.await?;
Ok(())
}
async fn drop_postgres_auto_update_ts_trigger(
&self,
tab_name: &str,
col_name: &str,
) -> Result<(), DbErr>;
async fn drop_postgres_auto_update_ts_trigger_for_col<
T: IntoIden + 'static + Send,
C: IntoIden + 'static + Send,
>(
&self,
tab: T,
col: C,
) -> Result<(), DbErr> {
let column_ident = col.into_iden();
let table_ident = tab.into_iden();
self.drop_postgres_auto_update_ts_trigger(
&table_ident.to_string(),
&column_ident.to_string(),
)
.await?;
Ok(())
}
async fn create_postgres_enum_for_active_enum<
E: IntoTypeRef + IntoIden + Send + Clone,
I: IntoIterator<Item = String> + Send,
>(
&self,
enum_name: E,
values: I,
) -> Result<(), DbErr>;
async fn add_postgres_enum_values_for_active_enum<
E: IntoTypeRef + IntoIden + Send + Clone,
I: IntoIterator<Item = String> + Send,
>(
&self,
enum_name: E,
values: I,
) -> Result<(), DbErr>;
async fn drop_postgres_enum_for_active_enum<E: IntoTypeRef + IntoIden + Send + Clone>(
&self,
enum_name: E,
) -> Result<(), DbErr>;
async fn if_postgres_enum_exists<E: IntoTypeRef + IntoIden + Send + Clone>(
&self,
enum_name: E,
) -> Result<bool, DbErr>;
async fn get_postgres_enum_values<E: IntoTypeRef + IntoIden + Send + Clone>(
&self,
enum_name: E,
) -> Result<HashSet<String>, DbErr>;
}
#[async_trait::async_trait]
impl CustomSchemaManagerExt for SchemaManager<'_> {
async fn create_postgres_auto_update_ts_fn(&self, col_name: &str) -> Result<(), DbErr> {
let sql = format!(
"CREATE OR REPLACE FUNCTION update_{col_name}_column() RETURNS TRIGGER AS $$ BEGIN \
NEW.{col_name} = current_timestamp; RETURN NEW; END; $$ language 'plpgsql';"
);
self.get_connection()
.execute(Statement::from_string(self.get_database_backend(), sql))
.await?;
Ok(())
}
async fn create_postgres_auto_update_ts_trigger(
&self,
tab_name: &str,
col_name: &str,
) -> Result<(), DbErr> {
let sql = format!(
"CREATE OR REPLACE TRIGGER update_{tab_name}_{col_name}_column_trigger BEFORE UPDATE \
ON {tab_name} FOR EACH ROW EXECUTE PROCEDURE update_{col_name}_column();"
);
self.get_connection()
.execute(Statement::from_string(self.get_database_backend(), sql))
.await?;
Ok(())
}
async fn drop_postgres_auto_update_ts_fn(&self, col_name: &str) -> Result<(), DbErr> {
let sql = format!("DROP FUNCTION IF EXISTS update_{col_name}_column();");
self.get_connection()
.execute(Statement::from_string(self.get_database_backend(), sql))
.await?;
Ok(())
}
async fn drop_postgres_auto_update_ts_trigger(
&self,
tab_name: &str,
col_name: &str,
) -> Result<(), DbErr> {
let sql = format!(
"DROP TRIGGER IF EXISTS update_{tab_name}_{col_name}_column_trigger ON {tab_name};"
);
self.get_connection()
.execute(Statement::from_string(self.get_database_backend(), sql))
.await?;
Ok(())
}
async fn create_postgres_enum_for_active_enum<
E: IntoTypeRef + IntoIden + Send + Clone,
I: IntoIterator<Item = String> + Send,
>(
&self,
enum_name: E,
values: I,
) -> Result<(), DbErr> {
let existed = self.if_postgres_enum_exists(enum_name.clone()).await?;
if !existed {
let idents = values.into_iter().map(Alias::new).collect::<Vec<_>>();
self.create_type(Type::create().as_enum(enum_name).values(idents).to_owned())
.await?;
} else {
self.add_postgres_enum_values_for_active_enum(enum_name, values)
.await?;
}
Ok(())
}
async fn add_postgres_enum_values_for_active_enum<
E: IntoTypeRef + IntoIden + Send + Clone,
I: IntoIterator<Item = String> + Send,
>(
&self,
enum_name: E,
values: I,
) -> Result<(), DbErr> {
let exists_values = self.get_postgres_enum_values(enum_name.clone()).await?;
let to_add_values = values
.into_iter()
.filter(|v| !exists_values.contains(v as &str))
.collect::<Vec<_>>();
if to_add_values.is_empty() {
return Ok(());
}
let mut type_alter = Type::alter().name(enum_name);
for v in to_add_values {
type_alter = type_alter.add_value(Alias::new(v));
}
self.alter_type(type_alter.to_owned()).await?;
Ok(())
}
async fn drop_postgres_enum_for_active_enum<E: IntoTypeRef + IntoIden + Send + Clone>(
&self,
enum_name: E,
) -> Result<(), DbErr> {
if self.if_postgres_enum_exists(enum_name.clone()).await? {
self.drop_type(Type::drop().name(enum_name).to_owned())
.await?;
}
Ok(())
}
async fn if_postgres_enum_exists<E: IntoTypeRef + IntoIden + Send + Clone>(
&self,
enum_name: E,
) -> Result<bool, DbErr> {
let enum_name: String = enum_name.into_iden().to_string();
let sql = format!("SELECT 1 FROM pg_type WHERE typname = '{enum_name}'");
let result = self
.get_connection()
.query_one(Statement::from_string(self.get_database_backend(), sql))
.await?;
Ok(result.is_some())
}
async fn get_postgres_enum_values<E: IntoTypeRef + IntoIden + Send + Clone>(
&self,
enum_name: E,
) -> Result<HashSet<String>, DbErr> {
let enum_name: String = enum_name.into_iden().to_string();
let sql = format!(
"SELECT pg_enum.enumlabel AS enumlabel FROM pg_type JOIN pg_enum ON pg_enum.enumtypid \
= pg_type.oid WHERE pg_type.typname = '{enum_name}';"
);
let results = self
.get_connection()
.query_all(Statement::from_string(self.get_database_backend(), sql))
.await?;
let mut items = HashSet::new();
for r in results {
items.insert(r.try_get::<String>("", "enumlabel")?);
}
Ok(items)
}
}

View File

@@ -0,0 +1,284 @@
use loco_rs::schema::jsonb_null;
use sea_orm_migration::{prelude::*, schema::*};
use super::defs::{
Bangumi, CustomSchemaManagerExt, Episodes, GeneralIds, Subscribers, Subscriptions,
};
use crate::models::{
subscribers::ROOT_SUBSCRIBER,
subscriptions::{self, SubscriptionCategoryEnum},
};
#[derive(DeriveMigrationName)]
pub struct Migration;
#[async_trait::async_trait]
impl MigrationTrait for Migration {
async fn up(&self, manager: &SchemaManager) -> Result<(), DbErr> {
manager
.create_postgres_auto_update_ts_fn_for_col(GeneralIds::UpdatedAt)
.await?;
manager
.create_table(
table_auto(Subscribers::Table)
.col(pk_auto(Subscribers::Id))
.col(string_len_uniq(Subscribers::Pid, 64))
.col(string(Subscribers::DisplayName))
.col(jsonb_null(Subscribers::BangumiConf))
.to_owned(),
)
.await?;
manager
.create_postgres_auto_update_ts_trigger_for_col(
Subscribers::Table,
GeneralIds::UpdatedAt,
)
.await?;
let insert = Query::insert()
.into_table(Subscribers::Table)
.columns([Subscribers::Pid, Subscribers::DisplayName])
.values_panic([ROOT_SUBSCRIBER.into(), ROOT_SUBSCRIBER.into()])
.to_owned();
manager.exec_stmt(insert).await?;
create_postgres_enum_for_active_enum!(
manager,
subscriptions::SubscriptionCategoryEnum,
subscriptions::SubscriptionCategory::Mikan,
subscriptions::SubscriptionCategory::Manual
)
.await?;
manager
.create_table(
table_auto(Subscriptions::Table)
.col(pk_auto(Subscriptions::Id))
.col(string(Subscriptions::DisplayName))
.col(integer(Subscriptions::SubscriberId))
.col(text(Subscriptions::SourceUrl))
.col(boolean(Subscriptions::Enabled))
.col(enumeration(
Subscriptions::Category,
subscriptions::SubscriptionCategoryEnum,
subscriptions::SubscriptionCategory::iden_values(),
))
.foreign_key(
ForeignKey::create()
.name("fk_subscriptions_subscriber_id")
.from(Subscriptions::Table, Subscriptions::SubscriberId)
.to(Subscribers::Table, Subscribers::Id)
.on_update(ForeignKeyAction::Restrict)
.on_delete(ForeignKeyAction::Cascade),
)
.to_owned(),
)
.await?;
manager
.create_postgres_auto_update_ts_trigger_for_col(
Subscriptions::Table,
GeneralIds::UpdatedAt,
)
.await?;
manager
.create_table(
table_auto(Bangumi::Table)
.col(pk_auto(Bangumi::Id))
.col(text_null(Bangumi::MikanBangumiId))
.col(integer(Bangumi::SubscriptionId))
.col(integer(Bangumi::SubscriberId))
.col(text(Bangumi::DisplayName))
.col(text(Bangumi::RawName))
.col(integer(Bangumi::Season))
.col(text_null(Bangumi::SeasonRaw))
.col(text_null(Bangumi::Fansub))
.col(text_null(Bangumi::MikanFansubId))
.col(jsonb_null(Bangumi::Filter))
.col(text_null(Bangumi::RssLink))
.col(text_null(Bangumi::PosterLink))
.col(text_null(Bangumi::SavePath))
.col(boolean(Bangumi::Deleted).default(false))
.col(text_null(Bangumi::Homepage))
.col(jsonb_null(Bangumi::Extra))
.foreign_key(
ForeignKey::create()
.name("fk_bangumi_subscription_id")
.from(Bangumi::Table, Bangumi::SubscriptionId)
.to(Subscriptions::Table, Subscriptions::Id)
.on_update(ForeignKeyAction::Restrict)
.on_delete(ForeignKeyAction::Cascade),
)
.foreign_key(
ForeignKey::create()
.name("fk_bangumi_subscriber_id")
.from(Bangumi::Table, Bangumi::SubscriberId)
.to(Subscribers::Table, Subscribers::Id)
.on_update(ForeignKeyAction::Restrict)
.on_delete(ForeignKeyAction::Cascade),
)
.to_owned(),
)
.await?;
manager
.create_index(
Index::create()
.if_not_exists()
.name("idx_bangumi_mikan_bangumi_id")
.table(Bangumi::Table)
.col(Bangumi::MikanBangumiId)
.to_owned(),
)
.await?;
manager
.create_index(
Index::create()
.if_not_exists()
.name("idx_bangumi_mikan_fansub_id")
.table(Bangumi::Table)
.col(Bangumi::MikanFansubId)
.to_owned(),
)
.await?;
manager
.create_postgres_auto_update_ts_trigger_for_col(Bangumi::Table, GeneralIds::UpdatedAt)
.await?;
manager
.create_table(
table_auto(Episodes::Table)
.col(pk_auto(Episodes::Id))
.col(text_null(Episodes::MikanEpisodeId))
.col(text(Episodes::RawName))
.col(text(Episodes::DisplayName))
.col(integer(Episodes::BangumiId))
.col(integer(Episodes::SubscriptionId))
.col(integer(Episodes::SubscriberId))
.col(text_null(Episodes::SavePath))
.col(text_null(Episodes::Resolution))
.col(integer(Episodes::Season))
.col(text_null(Episodes::SeasonRaw))
.col(text_null(Episodes::Fansub))
.col(text_null(Episodes::PosterLink))
.col(integer(Episodes::EpisodeIndex))
.col(text_null(Episodes::Homepage))
.col(text_null(Episodes::Subtitle))
.col(boolean(Episodes::Deleted).default(false))
.col(text_null(Episodes::Source))
.col(jsonb_null(Episodes::Extra))
.foreign_key(
ForeignKey::create()
.name("fk_episodes_subscription_id")
.from(Episodes::Table, Episodes::SubscriptionId)
.to(Subscriptions::Table, Subscriptions::Id)
.on_update(ForeignKeyAction::Restrict)
.on_delete(ForeignKeyAction::Cascade),
)
.foreign_key(
ForeignKey::create()
.name("fk_episodes_bangumi_id")
.from(Episodes::Table, Episodes::BangumiId)
.to(Bangumi::Table, Bangumi::Id)
.on_update(ForeignKeyAction::Restrict)
.on_delete(ForeignKeyAction::Cascade),
)
.foreign_key(
ForeignKey::create()
.name("fk_episodes_subscriber_id")
.from(Episodes::Table, Episodes::SubscriberId)
.to(Subscribers::Table, Subscribers::Id)
.on_update(ForeignKeyAction::Restrict)
.on_delete(ForeignKeyAction::Cascade),
)
.to_owned(),
)
.await?;
manager
.create_index(
Index::create()
.if_not_exists()
.name("idx_episodes_mikan_episode_id")
.table(Episodes::Table)
.col(Episodes::MikanEpisodeId)
.to_owned(),
)
.await?;
manager
.create_index(
Index::create()
.if_not_exists()
.name("idx_episodes_bangumi_id_mikan_episode_id")
.table(Episodes::Table)
.col(Episodes::BangumiId)
.col(Episodes::MikanEpisodeId)
.unique()
.to_owned(),
)
.await?;
manager
.create_postgres_auto_update_ts_trigger_for_col(Episodes::Table, GeneralIds::UpdatedAt)
.await?;
Ok(())
}
async fn down(&self, manager: &SchemaManager) -> Result<(), DbErr> {
manager
.drop_table(Table::drop().table(Episodes::Table).to_owned())
.await?;
manager
.drop_postgres_auto_update_ts_trigger_for_col(Episodes::Table, GeneralIds::UpdatedAt)
.await?;
manager
.drop_table(Table::drop().table(Bangumi::Table).to_owned())
.await?;
manager
.drop_postgres_auto_update_ts_trigger_for_col(Bangumi::Table, GeneralIds::UpdatedAt)
.await?;
manager
.drop_table(Table::drop().table(Subscriptions::Table).to_owned())
.await?;
manager
.drop_postgres_auto_update_ts_trigger_for_col(
Subscriptions::Table,
GeneralIds::UpdatedAt,
)
.await?;
manager
.drop_table(Table::drop().table(Subscribers::Table).to_owned())
.await?;
manager
.drop_postgres_auto_update_ts_trigger_for_col(Subscribers::Table, GeneralIds::UpdatedAt)
.await?;
manager
.drop_postgres_enum_for_active_enum(subscriptions::SubscriptionCategoryEnum)
.await?;
manager
.drop_postgres_auto_update_ts_fn_for_col(GeneralIds::UpdatedAt)
.await?;
manager
.drop_postgres_enum_for_active_enum(SubscriptionCategoryEnum)
.await?;
Ok(())
}
}

View File

@@ -0,0 +1,128 @@
use loco_rs::schema::table_auto;
use sea_orm_migration::{prelude::*, schema::*};
use super::defs::*;
use crate::models::prelude::{
downloads::{DownloadMimeEnum, DownloadStatusEnum},
DownloadMime, DownloadStatus,
};
#[derive(DeriveMigrationName)]
pub struct Migration;
#[async_trait::async_trait]
impl MigrationTrait for Migration {
async fn up(&self, manager: &SchemaManager) -> Result<(), DbErr> {
create_postgres_enum_for_active_enum!(
manager,
DownloadMimeEnum,
DownloadMime::BitTorrent,
DownloadMime::OctetStream
)
.await?;
create_postgres_enum_for_active_enum!(
manager,
DownloadStatusEnum,
DownloadStatus::Downloading,
DownloadStatus::Paused,
DownloadStatus::Pending,
DownloadStatus::Completed,
DownloadStatus::Failed,
DownloadStatus::Deleted
)
.await?;
manager
.create_table(
table_auto(Downloads::Table)
.col(pk_auto(Downloads::Id))
.col(string(Downloads::OriginalName))
.col(string(Downloads::DisplayName))
.col(integer(Downloads::SubscriptionId))
.col(enumeration(
Downloads::Status,
DownloadStatusEnum,
DownloadMime::iden_values(),
))
.col(enumeration(
Downloads::Mime,
DownloadMimeEnum,
DownloadMime::iden_values(),
))
.col(big_unsigned(Downloads::AllSize))
.col(big_unsigned(Downloads::CurrSize))
.col(text(Downloads::Url))
.col(text_null(Downloads::Homepage))
.col(text_null(Downloads::SavePath))
.foreign_key(
ForeignKey::create()
.name("fk_downloads_subscription_id")
.from(Downloads::Table, Downloads::SubscriptionId)
.to(Subscriptions::Table, Subscriptions::Id)
.on_update(ForeignKeyAction::Restrict)
.on_delete(ForeignKeyAction::Cascade),
)
.to_owned(),
)
.await?;
manager
.create_index(
Index::create()
.if_not_exists()
.name("idx_downloads_url")
.table(Downloads::Table)
.col(Downloads::Url)
.to_owned(),
)
.await?;
manager
.alter_table(
Table::alter()
.table(Episodes::Table)
.add_column_if_not_exists(integer_null(Episodes::DownloadId))
.add_foreign_key(
TableForeignKey::new()
.name("fk_episodes_download_id")
.from_tbl(Episodes::Table)
.from_col(Episodes::DownloadId)
.to_tbl(Downloads::Table)
.to_col(Downloads::Id)
.on_update(ForeignKeyAction::Restrict)
.on_delete(ForeignKeyAction::SetNull),
)
.to_owned(),
)
.await?;
Ok(())
}
async fn down(&self, manager: &SchemaManager) -> Result<(), DbErr> {
manager
.alter_table(
Table::alter()
.table(Episodes::Table)
.drop_foreign_key(Alias::new("fk_episodes_download_id"))
.drop_column(Episodes::DownloadId)
.to_owned(),
)
.await?;
manager
.drop_table(Table::drop().table(Downloads::Table).to_owned())
.await?;
manager
.drop_postgres_enum_for_active_enum(DownloadMimeEnum)
.await?;
manager
.drop_postgres_enum_for_active_enum(DownloadStatusEnum)
.await?;
Ok(())
}
}

View File

@@ -0,0 +1,102 @@
use sea_orm_migration::{prelude::*, schema::*};
use crate::{
migrations::defs::{CustomSchemaManagerExt, Downloaders, GeneralIds, Subscribers},
models::{downloaders::DownloaderCategoryEnum, prelude::DownloaderCategory},
};
#[derive(DeriveMigrationName)]
pub struct Migration;
#[async_trait::async_trait]
impl MigrationTrait for Migration {
async fn up(&self, manager: &SchemaManager) -> Result<(), DbErr> {
create_postgres_enum_for_active_enum!(
manager,
DownloaderCategoryEnum,
DownloaderCategory::QBittorrent
)
.await?;
manager
.create_table(
table_auto(Downloaders::Table)
.col(pk_auto(Downloaders::Id))
.col(text(Downloaders::Endpoint))
.col(string_null(Downloaders::Username))
.col(string_null(Downloaders::Password))
.col(enumeration(
Downloaders::Category,
DownloaderCategoryEnum,
DownloaderCategory::iden_values(),
))
.col(text(Downloaders::SavePath))
.col(integer(Downloaders::SubscriberId))
.foreign_key(
ForeignKey::create()
.name("fk_downloader_subscriber_id")
.from_tbl(Downloaders::Table)
.from_col(Downloaders::SubscriberId)
.to_tbl(Subscribers::Table)
.to_col(Subscribers::Id)
.on_delete(ForeignKeyAction::Cascade)
.on_update(ForeignKeyAction::Restrict),
)
.to_owned(),
)
.await?;
manager
.create_postgres_auto_update_ts_trigger_for_col(
Downloaders::Table,
GeneralIds::UpdatedAt,
)
.await?;
manager
.alter_table(
Table::alter()
.table(Subscribers::Table)
.add_column_if_not_exists(integer_null(Subscribers::DownloaderId))
.add_foreign_key(
TableForeignKey::new()
.name("fk_subscribers_downloader_id")
.from_tbl(Subscribers::Table)
.from_col(Subscribers::DownloaderId)
.to_tbl(Downloaders::Table)
.to_col(Downloaders::Id)
.on_delete(ForeignKeyAction::SetNull)
.on_update(ForeignKeyAction::Restrict),
)
.to_owned(),
)
.await?;
Ok(())
}
async fn down(&self, manager: &SchemaManager) -> Result<(), DbErr> {
manager
.alter_table(
Table::alter()
.table(Subscribers::Table)
.drop_foreign_key(Alias::new("fk_subscribers_downloader_id"))
.drop_column(Subscribers::DownloaderId)
.to_owned(),
)
.await?;
manager
.drop_postgres_auto_update_ts_trigger_for_col(Downloaders::Table, GeneralIds::UpdatedAt)
.await?;
manager
.drop_table(Table::drop().table(Downloaders::Table).to_owned())
.await?;
manager
.drop_postgres_enum_for_active_enum(DownloaderCategoryEnum)
.await?;
Ok(())
}
}

View File

@@ -0,0 +1,20 @@
pub use sea_orm_migration::prelude::*;
#[macro_use]
pub mod defs;
pub mod m20220101_000001_init;
pub mod m20240224_082543_add_downloads;
pub mod m20240225_060853_subscriber_add_downloader;
pub struct Migrator;
#[async_trait::async_trait]
impl MigratorTrait for Migrator {
fn migrations() -> Vec<Box<dyn MigrationTrait>> {
vec![
Box::new(m20220101_000001_init::Migration),
Box::new(m20240224_082543_add_downloads::Migration),
Box::new(m20240225_060853_subscriber_add_downloader::Migration),
]
}
}

View File

@@ -0,0 +1,45 @@
use loco_rs::app::AppContext;
use sea_orm::{entity::prelude::*, ActiveValue, TryIntoModel};
pub use super::entities::bangumi::*;
impl Model {
pub async fn get_or_insert_from_mikan<F>(
ctx: &AppContext,
subscriber_id: i32,
subscription_id: i32,
mikan_bangumi_id: String,
mikan_fansub_id: String,
f: F,
) -> eyre::Result<Model>
where
F: AsyncFnOnce(&mut ActiveModel) -> eyre::Result<()>,
{
let db = &ctx.db;
if let Some(existed) = Entity::find()
.filter(
Column::MikanBangumiId
.eq(Some(mikan_bangumi_id.clone()))
.and(Column::MikanFansubId.eq(Some(mikan_fansub_id.clone()))),
)
.one(db)
.await?
{
Ok(existed)
} else {
let mut bgm = ActiveModel {
mikan_bangumi_id: ActiveValue::Set(Some(mikan_bangumi_id)),
mikan_fansub_id: ActiveValue::Set(Some(mikan_fansub_id)),
subscription_id: ActiveValue::Set(subscription_id),
subscriber_id: ActiveValue::Set(subscriber_id),
..Default::default()
};
f(&mut bgm).await?;
let bgm: Model = bgm.save(db).await?.try_into_model()?;
Ok(bgm)
}
}
}
#[async_trait::async_trait]
impl ActiveModelBehavior for ActiveModel {}

View File

@@ -0,0 +1,17 @@
use sea_orm::prelude::*;
use url::Url;
pub use crate::models::entities::downloaders::*;
#[async_trait::async_trait]
impl ActiveModelBehavior for ActiveModel {}
impl Model {
pub fn get_endpoint(&self) -> String {
self.endpoint.clone()
}
pub fn endpoint_url(&self) -> Result<Url, url::ParseError> {
let url = Url::parse(&self.endpoint)?;
Ok(url)
}
}

View File

@@ -0,0 +1,27 @@
use sea_orm::{prelude::*, ActiveValue};
use crate::extract::mikan::MikanRssItem;
pub use crate::models::entities::downloads::*;
#[async_trait::async_trait]
impl ActiveModelBehavior for ActiveModel {}
impl ActiveModel {
pub fn from_mikan_rss_item(m: MikanRssItem, subscription_id: i32) -> Self {
let _ = Self {
origin_name: ActiveValue::Set(m.title.clone()),
display_name: ActiveValue::Set(m.title),
subscription_id: ActiveValue::Set(subscription_id),
status: ActiveValue::Set(DownloadStatus::Pending),
mime: ActiveValue::Set(DownloadMime::BitTorrent),
url: ActiveValue::Set(m.url.to_string()),
curr_size: ActiveValue::Set(m.content_length.as_ref().map(|_| 0)),
all_size: ActiveValue::Set(m.content_length),
homepage: ActiveValue::Set(Some(m.homepage.to_string())),
..Default::default()
};
todo!()
}
}
impl Model {}

View File

@@ -0,0 +1,80 @@
use sea_orm::{entity::prelude::*, FromJsonQueryResult};
use serde::{Deserialize, Serialize};
#[derive(Clone, Debug, PartialEq, Eq, Serialize, Deserialize, FromJsonQueryResult)]
pub struct BangumiFilter {
pub name: Option<Vec<String>>,
pub group: Option<Vec<String>>,
}
#[derive(Clone, Debug, PartialEq, Eq, Serialize, Deserialize, FromJsonQueryResult)]
pub struct BangumiExtra {
pub name_zh: Option<String>,
pub s_name_zh: Option<String>,
pub name_en: Option<String>,
pub s_name_en: Option<String>,
pub name_jp: Option<String>,
pub s_name_jp: Option<String>,
}
#[derive(Clone, Debug, PartialEq, DeriveEntityModel, Eq, Serialize, Deserialize)]
#[sea_orm(table_name = "bangumi")]
pub struct Model {
pub created_at: DateTime,
pub updated_at: DateTime,
#[sea_orm(primary_key)]
pub id: i32,
pub mikan_bangumi_id: Option<String>,
pub subscription_id: i32,
pub subscriber_id: i32,
pub display_name: String,
pub raw_name: String,
pub season: i32,
pub season_raw: Option<String>,
pub fansub: Option<String>,
pub mikan_fansub_id: Option<String>,
pub filter: Option<BangumiFilter>,
pub rss_link: Option<String>,
pub poster_link: Option<String>,
pub save_path: Option<String>,
#[sea_orm(default = "false")]
pub deleted: bool,
pub homepage: Option<String>,
pub extra: Option<BangumiExtra>,
}
#[derive(Copy, Clone, Debug, EnumIter, DeriveRelation)]
pub enum Relation {
#[sea_orm(
belongs_to = "super::subscriptions::Entity",
from = "Column::SubscriptionId",
to = "super::subscriptions::Column::Id"
)]
Subscription,
#[sea_orm(
belongs_to = "super::subscribers::Entity",
from = "Column::SubscriberId",
to = "super::subscribers::Column::Id"
)]
Subscriber,
#[sea_orm(has_many = "super::episodes::Entity")]
Episode,
}
impl Related<super::episodes::Entity> for Entity {
fn to() -> RelationDef {
Relation::Episode.def()
}
}
impl Related<super::subscriptions::Entity> for Entity {
fn to() -> RelationDef {
Relation::Subscription.def()
}
}
impl Related<super::subscribers::Entity> for Entity {
fn to() -> RelationDef {
Relation::Subscriber.def()
}
}

View File

@@ -0,0 +1,45 @@
use sea_orm::entity::prelude::*;
use serde::{Deserialize, Serialize};
#[derive(
Clone, Debug, PartialEq, Eq, EnumIter, DeriveActiveEnum, DeriveDisplay, Serialize, Deserialize,
)]
#[sea_orm(rs_type = "String", db_type = "Enum", enum_name = "downloader_type")]
#[serde(rename_all = "snake_case")]
pub enum DownloaderCategory {
#[sea_orm(string_value = "qbittorrent")]
QBittorrent,
}
#[derive(Clone, Debug, PartialEq, DeriveEntityModel, Eq, Serialize, Deserialize)]
#[sea_orm(table_name = "downloaders")]
pub struct Model {
#[sea_orm(column_type = "Timestamp")]
pub created_at: DateTime,
#[sea_orm(column_type = "Timestamp")]
pub updated_at: DateTime,
#[sea_orm(primary_key)]
pub id: i32,
pub category: DownloaderCategory,
pub endpoint: String,
pub password: String,
pub username: String,
pub subscriber_id: i32,
pub save_path: String,
}
#[derive(Copy, Clone, Debug, EnumIter, DeriveRelation)]
pub enum Relation {
#[sea_orm(
belongs_to = "super::subscribers::Entity",
from = "Column::SubscriberId",
to = "super::subscribers::Column::Id"
)]
Subscriber,
}
impl Related<super::subscribers::Entity> for Entity {
fn to() -> RelationDef {
Relation::Subscriber.def()
}
}

View File

@@ -0,0 +1,78 @@
use sea_orm::entity::prelude::*;
use serde::{Deserialize, Serialize};
#[derive(
Clone, Debug, PartialEq, Eq, EnumIter, DeriveActiveEnum, DeriveDisplay, Serialize, Deserialize,
)]
#[sea_orm(rs_type = "String", db_type = "Enum", enum_name = "download_status")]
#[serde(rename_all = "snake_case")]
pub enum DownloadStatus {
#[sea_orm(string_value = "pending")]
Pending,
#[sea_orm(string_value = "downloading")]
Downloading,
#[sea_orm(string_value = "paused")]
Paused,
#[sea_orm(string_value = "completed")]
Completed,
#[sea_orm(string_value = "failed")]
Failed,
#[sea_orm(string_value = "deleted")]
Deleted,
}
#[derive(
Clone, Debug, PartialEq, Eq, EnumIter, DeriveActiveEnum, DeriveDisplay, Serialize, Deserialize,
)]
#[sea_orm(rs_type = "String", db_type = "Enum", enum_name = "download_mime")]
pub enum DownloadMime {
#[sea_orm(string_value = "application/octet-stream")]
#[serde(rename = "application/octet-stream")]
OctetStream,
#[sea_orm(string_value = "application/x-bittorrent")]
#[serde(rename = "application/x-bittorrent")]
BitTorrent,
}
#[derive(Clone, Debug, PartialEq, DeriveEntityModel, Eq, Serialize, Deserialize)]
#[sea_orm(table_name = "downloads")]
pub struct Model {
pub created_at: DateTime,
pub updated_at: DateTime,
#[sea_orm(primary_key)]
pub id: i32,
pub origin_name: String,
pub display_name: String,
pub subscription_id: i32,
pub status: DownloadStatus,
pub mime: DownloadMime,
pub url: String,
pub all_size: Option<u64>,
pub curr_size: Option<u64>,
pub homepage: Option<String>,
pub save_path: Option<String>,
}
#[derive(Copy, Clone, Debug, EnumIter, DeriveRelation)]
pub enum Relation {
#[sea_orm(
belongs_to = "super::subscriptions::Entity",
from = "Column::SubscriptionId",
to = "super::subscriptions::Column::Id"
)]
Subscription,
#[sea_orm(has_many = "super::episodes::Entity")]
Episode,
}
impl Related<super::subscriptions::Entity> for Entity {
fn to() -> RelationDef {
Relation::Subscription.def()
}
}
impl Related<super::episodes::Entity> for Entity {
fn to() -> RelationDef {
Relation::Episode.def()
}
}

View File

@@ -0,0 +1,95 @@
//! `SeaORM` Entity. Generated by sea-orm-codegen 0.12.2
use sea_orm::{entity::prelude::*, FromJsonQueryResult};
use serde::{Deserialize, Serialize};
#[derive(Clone, Debug, PartialEq, Eq, Serialize, Deserialize, FromJsonQueryResult, Default)]
pub struct EpisodeExtra {
pub name_zh: Option<String>,
pub s_name_zh: Option<String>,
pub name_en: Option<String>,
pub s_name_en: Option<String>,
pub name_jp: Option<String>,
pub s_name_jp: Option<String>,
}
#[derive(Clone, Debug, PartialEq, DeriveEntityModel, Eq, Serialize, Deserialize)]
#[sea_orm(table_name = "episodes")]
pub struct Model {
pub created_at: DateTime,
pub updated_at: DateTime,
#[sea_orm(primary_key)]
pub id: i32,
#[sea_orm(indexed)]
pub mikan_episode_id: Option<String>,
pub raw_name: String,
pub display_name: String,
pub bangumi_id: i32,
pub subscription_id: i32,
pub subscriber_id: i32,
pub download_id: Option<i32>,
pub save_path: Option<String>,
pub resolution: Option<String>,
pub season: i32,
pub season_raw: Option<String>,
pub fansub: Option<String>,
pub poster_link: Option<String>,
pub episode_index: i32,
pub homepage: Option<String>,
pub subtitle: Option<Vec<String>>,
#[sea_orm(default = "false")]
pub deleted: bool,
pub source: Option<String>,
pub extra: EpisodeExtra,
}
#[derive(Copy, Clone, Debug, EnumIter, DeriveRelation)]
pub enum Relation {
#[sea_orm(
belongs_to = "super::bangumi::Entity",
from = "Column::BangumiId",
to = "super::bangumi::Column::Id"
)]
Bangumi,
#[sea_orm(
belongs_to = "super::downloads::Entity",
from = "Column::DownloadId",
to = "super::downloads::Column::Id"
)]
Downloads,
#[sea_orm(
belongs_to = "super::subscriptions::Entity",
from = "Column::SubscriptionId",
to = "super::subscriptions::Column::Id"
)]
Subscriptions,
#[sea_orm(
belongs_to = "super::subscribers::Entity",
from = "Column::SubscriberId",
to = "super::subscribers::Column::Id"
)]
Subscriber,
}
impl Related<super::bangumi::Entity> for Entity {
fn to() -> RelationDef {
Relation::Bangumi.def()
}
}
impl Related<super::downloads::Entity> for Entity {
fn to() -> RelationDef {
Relation::Downloads.def()
}
}
impl Related<super::subscriptions::Entity> for Entity {
fn to() -> RelationDef {
Relation::Subscriptions.def()
}
}
impl Related<super::subscribers::Entity> for Entity {
fn to() -> RelationDef {
Relation::Subscriber.def()
}
}

View File

@@ -0,0 +1,7 @@
//! `SeaORM` Entity. Generated by sea-orm-codegen 0.12.4
pub mod bangumi;
pub mod downloads;
pub mod episodes;
pub mod subscribers;
pub mod subscriptions;
pub mod downloaders;

View File

@@ -0,0 +1,63 @@
//! `SeaORM` Entity. Generated by sea-orm-codegen 0.12.2
use sea_orm::{entity::prelude::*, FromJsonQueryResult};
use serde::{Deserialize, Serialize};
#[derive(Clone, Debug, PartialEq, Eq, Serialize, Deserialize, FromJsonQueryResult)]
pub struct SubscriberBangumiConfig {
pub leading_group_tag: Option<bool>,
}
#[derive(Clone, Debug, PartialEq, DeriveEntityModel, Eq, Serialize, Deserialize)]
#[sea_orm(table_name = "subscribers")]
pub struct Model {
pub created_at: DateTime,
pub updated_at: DateTime,
#[sea_orm(primary_key)]
pub id: i32,
#[sea_orm(unique)]
pub pid: String,
pub display_name: String,
pub downloader_id: Option<i32>,
pub bangumi_conf: Option<SubscriberBangumiConfig>,
}
#[derive(Copy, Clone, Debug, EnumIter, DeriveRelation)]
pub enum Relation {
#[sea_orm(has_many = "super::subscriptions::Entity")]
Subscription,
#[sea_orm(
belongs_to = "super::downloaders::Entity",
from = "Column::DownloaderId",
to = "super::downloaders::Column::Id"
)]
Downloader,
#[sea_orm(has_many = "super::bangumi::Entity")]
Bangumi,
#[sea_orm(has_many = "super::episodes::Entity")]
Episode,
}
impl Related<super::subscriptions::Entity> for Entity {
fn to() -> RelationDef {
Relation::Subscription.def()
}
}
impl Related<super::downloaders::Entity> for Entity {
fn to() -> RelationDef {
Relation::Downloader.def()
}
}
impl Related<super::bangumi::Entity> for Entity {
fn to() -> RelationDef {
Relation::Bangumi.def()
}
}
impl Related<super::episodes::Entity> for Entity {
fn to() -> RelationDef {
Relation::Episode.def()
}
}

View File

@@ -0,0 +1,66 @@
use sea_orm::entity::prelude::*;
use serde::{Deserialize, Serialize};
#[derive(
Clone, Debug, PartialEq, Eq, EnumIter, DeriveActiveEnum, Serialize, Deserialize, DeriveDisplay,
)]
#[sea_orm(
rs_type = "String",
db_type = "Enum",
enum_name = "subscription_category"
)]
#[serde(rename_all = "snake_case")]
pub enum SubscriptionCategory {
#[sea_orm(string_value = "mikan")]
Mikan,
#[sea_orm(string_value = "manual")]
Manual,
}
#[derive(Clone, Debug, PartialEq, DeriveEntityModel, Eq, Serialize, Deserialize)]
#[sea_orm(table_name = "subscriptions")]
pub struct Model {
#[sea_orm(column_type = "Timestamp")]
pub created_at: DateTime,
#[sea_orm(column_type = "Timestamp")]
pub updated_at: DateTime,
#[sea_orm(primary_key)]
pub id: i32,
pub display_name: String,
pub subscriber_id: i32,
pub category: SubscriptionCategory,
pub source_url: String,
pub enabled: bool,
}
#[derive(Copy, Clone, Debug, EnumIter, DeriveRelation)]
pub enum Relation {
#[sea_orm(
belongs_to = "super::subscribers::Entity",
from = "Column::SubscriberId",
to = "super::subscribers::Column::Id"
)]
Subscriber,
#[sea_orm(has_many = "super::bangumi::Entity")]
Bangumi,
#[sea_orm(has_many = "super::episodes::Entity")]
Episodes,
}
impl Related<super::subscribers::Entity> for Entity {
fn to() -> RelationDef {
Relation::Subscriber.def()
}
}
impl Related<super::bangumi::Entity> for Entity {
fn to() -> RelationDef {
Relation::Bangumi.def()
}
}
impl Related<super::episodes::Entity> for Entity {
fn to() -> RelationDef {
Relation::Episodes.def()
}
}

View File

@@ -0,0 +1,104 @@
use std::sync::Arc;
use loco_rs::app::AppContext;
use sea_orm::{entity::prelude::*, sea_query::OnConflict, ActiveValue};
use super::bangumi;
pub use super::entities::episodes::*;
use crate::{
app::AppContextExt,
extract::{
mikan::{build_mikan_episode_homepage, MikanEpisodeMeta},
rawname::parse_episode_meta_from_raw_name,
},
};
#[derive(Clone, Debug, PartialEq)]
pub struct MikanEpsiodeCreation {
pub episode: MikanEpisodeMeta,
pub bangumi: Arc<bangumi::Model>,
}
impl Model {
pub async fn add_episodes(
ctx: &AppContext,
creations: impl IntoIterator<Item = MikanEpsiodeCreation>,
) -> eyre::Result<()> {
let db = &ctx.db;
let new_episode_active_modes = creations
.into_iter()
.map(|cr| ActiveModel::from_mikan_episode_meta(ctx, cr))
.inspect(|result| {
if let Err(e) = result {
tracing::warn!("Failed to create episode: {:?}", e);
}
})
.flatten();
Entity::insert_many(new_episode_active_modes)
.on_conflict(
OnConflict::columns([Column::BangumiId, Column::MikanEpisodeId])
.do_nothing()
.to_owned(),
)
.on_empty_do_nothing()
.exec(db)
.await?;
Ok(())
}
}
impl ActiveModel {
pub fn from_mikan_episode_meta(
ctx: &AppContext,
creation: MikanEpsiodeCreation,
) -> eyre::Result<Self> {
let item = creation.episode;
let bgm = creation.bangumi;
let raw_meta = parse_episode_meta_from_raw_name(&item.episode_title)
.inspect_err(|e| {
tracing::warn!("Failed to parse episode meta: {:?}", e);
})
.ok()
.unwrap_or_default();
let homepage = build_mikan_episode_homepage(
ctx.get_mikan_client().base_url(),
&item.mikan_episode_id,
)?;
Ok(Self {
mikan_episode_id: ActiveValue::Set(Some(item.mikan_episode_id)),
raw_name: ActiveValue::Set(item.episode_title.clone()),
display_name: ActiveValue::Set(item.episode_title.clone()),
bangumi_id: ActiveValue::Set(bgm.id),
subscription_id: ActiveValue::Set(bgm.subscription_id),
subscriber_id: ActiveValue::Set(bgm.subscriber_id),
resolution: ActiveValue::Set(raw_meta.resolution),
season: ActiveValue::Set(if raw_meta.season > 0 {
raw_meta.season
} else {
bgm.season
}),
season_raw: ActiveValue::Set(raw_meta.season_raw.or_else(|| bgm.season_raw.clone())),
fansub: ActiveValue::Set(raw_meta.fansub.or_else(|| bgm.fansub.clone())),
poster_link: ActiveValue::Set(bgm.poster_link.clone()),
episode_index: ActiveValue::Set(raw_meta.episode_index),
homepage: ActiveValue::Set(Some(homepage.to_string())),
subtitle: ActiveValue::Set(raw_meta.subtitle.map(|s| vec![s])),
source: ActiveValue::Set(raw_meta.source),
extra: ActiveValue::Set(EpisodeExtra {
name_zh: raw_meta.name_zh,
name_en: raw_meta.name_en,
name_jp: raw_meta.name_jp,
s_name_en: raw_meta.name_en_no_season,
s_name_jp: raw_meta.name_jp_no_season,
s_name_zh: raw_meta.name_zh_no_season,
}),
..Default::default()
})
}
}
#[async_trait::async_trait]
impl ActiveModelBehavior for ActiveModel {}

View File

@@ -0,0 +1,10 @@
pub mod bangumi;
pub mod downloaders;
pub mod downloads;
pub mod entities;
pub mod episodes;
pub mod notifications;
pub mod prelude;
pub mod query;
pub mod subscribers;
pub mod subscriptions;

View File

@@ -0,0 +1,8 @@
use serde::{Deserialize, Serialize};
#[derive(Debug, Clone, PartialEq, Eq, Serialize, Deserialize)]
pub struct Notification {
season: i32,
episode_size: u32,
poster_url: Option<String>,
}

View File

@@ -0,0 +1,8 @@
pub use super::{
bangumi::{self, Entity as Bangumi},
downloaders::{self, DownloaderCategory, Entity as Downloader},
downloads::{self, DownloadMime, DownloadStatus, Entity as Download},
episodes::{self, Entity as Episode},
subscribers::{self, Entity as Subscriber},
subscriptions::{self, Entity as Subscription, SubscriptionCategory},
};

View File

@@ -0,0 +1,26 @@
use sea_orm::{
prelude::Expr,
sea_query::{Alias, IntoColumnRef, IntoTableRef, Query, SelectStatement},
Value,
};
pub fn filter_values_in<
I: IntoIterator<Item = T>,
T: Into<Value>,
R: IntoTableRef,
C: IntoColumnRef + Copy,
>(
tbl_ref: R,
col_ref: C,
values: I,
) -> SelectStatement {
Query::select()
.expr(Expr::col((Alias::new("t"), Alias::new("column1"))))
.from_values(values, Alias::new("t"))
.left_join(
tbl_ref,
Expr::col((Alias::new("t"), Alias::new("column1"))).equals(col_ref),
)
.and_where(Expr::col(col_ref).is_not_null())
.to_owned()
}

View File

@@ -0,0 +1,97 @@
use loco_rs::{
app::AppContext,
model::{ModelError, ModelResult},
};
use sea_orm::{entity::prelude::*, ActiveValue, TransactionTrait};
use serde::{Deserialize, Serialize};
pub use super::entities::subscribers::*;
pub const ROOT_SUBSCRIBER: &str = "konobangu";
#[derive(Debug, Deserialize, Serialize)]
pub struct SubscriberIdParams {
pub id: String,
}
#[async_trait::async_trait]
impl ActiveModelBehavior for ActiveModel {
async fn before_save<C>(self, _db: &C, insert: bool) -> Result<Self, DbErr>
where
C: ConnectionTrait,
{
if insert {
let mut this = self;
this.pid = ActiveValue::Set(Uuid::new_v4().to_string());
Ok(this)
} else {
Ok(self)
}
}
}
impl Model {
/// finds a user by the provided pid
///
/// # Errors
///
/// When could not find user or DB query error
pub async fn find_by_pid(ctx: &AppContext, pid: &str) -> ModelResult<Self> {
let db = &ctx.db;
let parse_uuid = Uuid::parse_str(pid).map_err(|e| ModelError::Any(e.into()))?;
let subscriber = Entity::find()
.filter(Column::Pid.eq(parse_uuid))
.one(db)
.await?;
subscriber.ok_or_else(|| ModelError::EntityNotFound)
}
pub async fn find_by_id(ctx: &AppContext, id: i32) -> ModelResult<Self> {
let db = &ctx.db;
let subscriber = Entity::find_by_id(id).one(db).await?;
subscriber.ok_or_else(|| ModelError::EntityNotFound)
}
pub async fn find_pid_by_id_with_cache(ctx: &AppContext, id: i32) -> eyre::Result<String> {
let db = &ctx.db;
let cache = &ctx.cache;
let pid = cache
.get_or_insert(&format!("subscriber-id2pid::{}", id), async {
let subscriber = Entity::find_by_id(id)
.one(db)
.await?
.ok_or_else(|| loco_rs::Error::string(&format!("No such pid for id {}", id)))?;
Ok(subscriber.pid)
})
.await?;
Ok(pid)
}
pub async fn find_root(ctx: &AppContext) -> ModelResult<Self> {
Self::find_by_pid(ctx, ROOT_SUBSCRIBER).await
}
/// Asynchronously creates a user with a password and saves it to the
/// database.
///
/// # Errors
///
/// When could not save the user into the DB
pub async fn create_root(ctx: &AppContext) -> ModelResult<Self> {
let db = &ctx.db;
let txn = db.begin().await?;
let user = ActiveModel {
display_name: ActiveValue::set(ROOT_SUBSCRIBER.to_string()),
pid: ActiveValue::set(ROOT_SUBSCRIBER.to_string()),
..Default::default()
}
.insert(&txn)
.await?;
txn.commit().await?;
Ok(user)
}
}

View File

@@ -0,0 +1,228 @@
use std::{collections::HashSet, sync::Arc};
use itertools::Itertools;
use loco_rs::app::AppContext;
use sea_orm::{entity::prelude::*, ActiveValue};
use serde::{Deserialize, Serialize};
pub use super::entities::subscriptions::{self, *};
use super::{bangumi, episodes, query::filter_values_in};
use crate::{
app::AppContextExt,
extract::{
mikan::{
build_mikan_bangumi_homepage, build_mikan_bangumi_rss_link,
parse_mikan_bangumi_meta_from_mikan_homepage,
parse_mikan_episode_meta_from_mikan_homepage, parse_mikan_rss_channel_from_rss_link,
web_parser::{
parse_mikan_bangumi_poster_from_origin_poster_src_with_cache,
MikanBangumiPosterMeta,
},
},
rawname::extract_season_from_title_body,
},
models::episodes::MikanEpsiodeCreation,
};
#[derive(Debug, Clone, Serialize, Deserialize, PartialEq, Eq)]
pub struct SubscriptionCreateFromRssDto {
pub rss_link: String,
pub display_name: String,
pub enabled: Option<bool>,
}
#[derive(Debug, Clone, Serialize, Deserialize, PartialEq, Eq)]
#[serde(tag = "category")]
pub enum SubscriptionCreateDto {
Mikan(SubscriptionCreateFromRssDto),
}
#[async_trait::async_trait]
impl ActiveModelBehavior for ActiveModel {}
impl ActiveModel {
pub fn from_create_dto(create_dto: SubscriptionCreateDto, subscriber_id: i32) -> Self {
match create_dto {
SubscriptionCreateDto::Mikan(create_dto) => {
Self::from_rss_create_dto(SubscriptionCategory::Mikan, create_dto, subscriber_id)
}
}
}
fn from_rss_create_dto(
category: SubscriptionCategory,
create_dto: SubscriptionCreateFromRssDto,
subscriber_id: i32,
) -> Self {
Self {
display_name: ActiveValue::Set(create_dto.display_name),
enabled: ActiveValue::Set(create_dto.enabled.unwrap_or(false)),
subscriber_id: ActiveValue::Set(subscriber_id),
category: ActiveValue::Set(category),
source_url: ActiveValue::Set(create_dto.rss_link),
..Default::default()
}
}
}
impl Model {
pub async fn add_subscription(
ctx: &AppContext,
create_dto: SubscriptionCreateDto,
subscriber_id: i32,
) -> eyre::Result<Self> {
let db = &ctx.db;
let subscription = ActiveModel::from_create_dto(create_dto, subscriber_id);
Ok(subscription.insert(db).await?)
}
pub async fn toggle_iters(
ctx: &AppContext,
ids: impl Iterator<Item = i32>,
enabled: bool,
) -> eyre::Result<()> {
let db = &ctx.db;
Entity::update_many()
.col_expr(Column::Enabled, Expr::value(enabled))
.filter(Column::Id.is_in(ids))
.exec(db)
.await?;
Ok(())
}
pub async fn delete_iters(
ctx: &AppContext,
ids: impl Iterator<Item = i32>,
) -> eyre::Result<()> {
let db = &ctx.db;
Entity::delete_many()
.filter(Column::Id.is_in(ids))
.exec(db)
.await?;
Ok(())
}
pub async fn pull_subscription(&self, ctx: &AppContext) -> eyre::Result<()> {
match &self.category {
SubscriptionCategory::Mikan => {
let mikan_client = ctx.get_mikan_client();
let channel =
parse_mikan_rss_channel_from_rss_link(Some(mikan_client), &self.source_url)
.await?;
let items = channel.into_items();
let db = &ctx.db;
let items = items.into_iter().collect_vec();
let mut stmt = filter_values_in(
episodes::Entity,
episodes::Column::MikanEpisodeId,
items
.iter()
.map(|s| Value::from(s.mikan_episode_id.clone())),
);
stmt.and_where(Expr::col(episodes::Column::SubscriberId).eq(self.subscriber_id));
let builder = &db.get_database_backend();
let old_rss_item_mikan_episode_ids_set = db
.query_all(builder.build(&stmt))
.await?
.into_iter()
.flat_map(|qs| qs.try_get_by_index(0))
.collect::<HashSet<String>>();
let new_rss_items = items
.into_iter()
.filter(|item| {
!old_rss_item_mikan_episode_ids_set.contains(&item.mikan_episode_id)
})
.collect_vec();
let mut new_metas = vec![];
for new_rss_item in new_rss_items.iter() {
new_metas.push(
parse_mikan_episode_meta_from_mikan_homepage(
Some(mikan_client),
new_rss_item.homepage.clone(),
)
.await?,
);
}
let new_mikan_bangumi_groups = new_metas
.into_iter()
.into_group_map_by(|s| (s.mikan_bangumi_id.clone(), s.mikan_fansub_id.clone()));
for ((mikan_bangumi_id, mikan_fansub_id), new_ep_metas) in new_mikan_bangumi_groups
{
let mikan_base_url = ctx.get_mikan_client().base_url();
let bgm_homepage = build_mikan_bangumi_homepage(
mikan_base_url,
&mikan_bangumi_id,
Some(&mikan_fansub_id),
)?;
let bgm_rss_link = build_mikan_bangumi_rss_link(
mikan_base_url,
&mikan_bangumi_id,
Some(&mikan_fansub_id),
)?;
let bgm = Arc::new(
bangumi::Model::get_or_insert_from_mikan(
ctx,
self.subscriber_id,
self.id,
mikan_bangumi_id.to_string(),
mikan_fansub_id.to_string(),
async |am| -> eyre::Result<()> {
let bgm_meta = parse_mikan_bangumi_meta_from_mikan_homepage(
Some(mikan_client),
bgm_homepage.clone(),
)
.await?;
let bgm_name = bgm_meta.bangumi_title;
let (_, bgm_season_raw, bgm_season) =
extract_season_from_title_body(&bgm_name);
am.raw_name = ActiveValue::Set(bgm_name.clone());
am.display_name = ActiveValue::Set(bgm_name);
am.season = ActiveValue::Set(bgm_season);
am.season_raw = ActiveValue::Set(bgm_season_raw);
am.rss_link = ActiveValue::Set(Some(bgm_rss_link.to_string()));
am.homepage = ActiveValue::Set(Some(bgm_homepage.to_string()));
am.fansub = ActiveValue::Set(bgm_meta.fansub);
if let Some(origin_poster_src) = bgm_meta.origin_poster_src {
if let MikanBangumiPosterMeta {
poster_src: Some(poster_src),
..
} = parse_mikan_bangumi_poster_from_origin_poster_src_with_cache(
ctx,
origin_poster_src,
self.subscriber_id,
)
.await?
{
am.poster_link = ActiveValue::Set(Some(poster_src))
}
}
Ok(())
},
)
.await?,
);
episodes::Model::add_episodes(
ctx,
new_ep_metas.into_iter().map(|item| MikanEpsiodeCreation {
episode: item,
bangumi: bgm.clone(),
}),
)
.await?;
}
Ok(())
}
_ => todo!(),
}
}
}

View File

@@ -0,0 +1 @@

View File

@@ -0,0 +1 @@
pub mod subscribers;

View File

@@ -0,0 +1,19 @@
use serde::{Deserialize, Serialize};
use crate::models::entities::subscribers;
#[derive(Debug, Deserialize, Serialize)]
pub struct CurrentResponse {
pub pid: String,
pub display_name: String,
}
impl CurrentResponse {
#[must_use]
pub fn new(user: &subscribers::Model) -> Self {
Self {
pid: user.pid.to_string(),
display_name: user.display_name.to_string(),
}
}
}

View File

@@ -0,0 +1 @@
pub mod subscription_worker;

View File

@@ -0,0 +1,31 @@
use loco_rs::prelude::*;
use serde::{Deserialize, Serialize};
use crate::models::subscriptions;
pub struct SubscriptionWorker {
pub ctx: AppContext,
}
#[derive(Debug, Serialize, Deserialize, Clone)]
pub struct SubscriptionWorkerArgs {
pub subscription: subscriptions::Model,
}
#[async_trait]
impl BackgroundWorker<SubscriptionWorkerArgs> for SubscriptionWorker {
fn build(ctx: &AppContext) -> Self {
Self { ctx: ctx.clone() }
}
async fn perform(&self, _args: SubscriptionWorkerArgs) -> Result<()> {
println!("================================================");
let _db = &self.ctx.db;
let _storage = &self.ctx.storage;
println!("================================================");
Ok(())
}
}

View File

@@ -0,0 +1,3 @@
mod models;
mod requests;
mod tasks;

View File

@@ -0,0 +1 @@
mod subscribers;

View File

@@ -0,0 +1,7 @@
---
source: tests/models/subscribers.rs
expression: non_existing_subscriber_results
---
Err(
EntityNotFound,
)

View File

@@ -0,0 +1,13 @@
---
source: tests/models/subscribers.rs
expression: existing_subscriber
---
Ok(
Model {
created_at: 2023-11-12T12:34:56.789,
updated_at: 2023-11-12T12:34:56.789,
id: 1,
pid: "11111111-1111-1111-1111-111111111111",
display_name: "user1"
},
)

View File

@@ -0,0 +1,27 @@
use insta::assert_debug_snapshot;
use loco_rs::testing;
use recorder::{app::App, models::subscribers::Model};
use serial_test::serial;
macro_rules! configure_insta {
($($expr:expr),*) => {
let mut settings = insta::Settings::clone_current();
settings.set_prepend_module_to_snapshot(false);
settings.set_snapshot_suffix("users");
let _guard = settings.bind_to_scope();
};
}
#[tokio::test]
#[serial]
async fn can_find_by_pid() {
// configure_insta!();
//
// let boot = testing::boot_test::<App>().await.unwrap();
// testing::seed::<App>(&boot.app_context.db).await.unwrap();
//
// let existing_subscriber =
// Model::find_by_pid(&boot.app_context, "11111111-1111-1111-1111-111111111111").await;
//
// assert_debug_snapshot!(existing_subscriber);
}

View File

@@ -0,0 +1 @@
mod subscribers;

View File

@@ -0,0 +1,33 @@
#![allow(unused_imports)]
use insta::{assert_debug_snapshot, with_settings};
use loco_rs::testing;
use recorder::app::App;
use serial_test::serial;
// TODO: see how to dedup / extract this to app-local test utils
// not to framework, because that would require a runtime dep on insta
// macro_rules! configure_insta {
// ($($expr:expr),*) => {
// let mut settings = insta::Settings::clone_current();
// settings.set_prepend_module_to_snapshot(false);
// settings.set_snapshot_suffix("user_request");
// let _guard = settings.bind_to_scope();
// };
// }
#[tokio::test]
#[serial]
async fn can_get_current_user() {
// configure_insta!();
//
// testing::request::<App, _, _>(|request, _ctx| async move {
// let response = request.get("/api/user/current").await;
//
// with_settings!({
// filters => testing::cleanup_user_model()
// }, {
// assert_debug_snapshot!((response.status_code(),
// response.text())); });
// })
// .await;
}

View File

@@ -0,0 +1 @@
pub mod seed;

View File

@@ -0,0 +1,42 @@
//! This task implements data seeding functionality for initializing new
//! development/demo environments.
//!
//! # Example
//!
//! Run the task with the following command:
//! ```sh
//! cargo run task
//! ```
//!
//! To override existing data and reset the data structure, use the following
//! command with the `refresh:true` argument:
//! ```sh
//! cargo run task seed_data refresh:true
//! ```
#![allow(unused_imports)]
use loco_rs::{db, prelude::*};
use recorder::{app::App, migrations::Migrator};
#[allow(clippy::module_name_repetitions)]
pub struct SeedData;
#[async_trait]
impl Task for SeedData {
fn task(&self) -> TaskInfo {
TaskInfo {
name: "seed_data".to_string(),
detail: "Task for seeding data".to_string(),
}
}
async fn run(&self, _app_context: &AppContext, _vars: &task::Vars) -> Result<()> {
// let refresh = vars.cli.get("refresh").is_some_and(|refresh| refresh ==
// "true");
//
// if refresh {
// db::reset::<Migrator>(&app_context.db).await?;
// }
// let path = std::path::Path::new("src/fixtures");
// db::run_app_seed::<App>(&app_context.db, path).await?;
Ok(())
}
}