add rss

2024-05-02 19:23:20 +03:00 · 2024-05-02 19:23:20 +03:00 · 457692f766
commit 457692f766
parent 086ddb7665
7 changed files with 263 additions and 10 deletions
--- a/Cargo.lock
+++ b/Cargo.lock
@ -128,6 +128,19 @@ dependencies = [
 "syn 2.0.60",
 ]

+[[package]]
+name = "atom_syndication"
+version = "0.12.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "571832dcff775e26562e8e6930cd483de5587301d40d3a3b85d532b6383e15a7"
+dependencies = [
+ "chrono",
+ "derive_builder",
+ "diligent-date-parser",
+ "never",
+ "quick-xml 0.30.0",
+]
+
 [[package]]
 name = "autocfg"
 version = "1.2.0"
@ -290,6 +303,7 @@ dependencies = [
 "comrak",
 "console-subscriber",
 "fronma",
+ "rss",
 "scc",
 "serde",
 "syntect",
@ -300,6 +314,7 @@ dependencies = [
 "tower-http",
 "tracing",
 "tracing-subscriber",
+ "url",
 "zstd",
 ]

@ -576,12 +591,30 @@ version = "1.4.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "322ef0094744e63628e6f0eb2295517f79276a5b342a4c2ff3042566ca181d4e"

+[[package]]
+name = "diligent-date-parser"
+version = "0.1.4"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "f6cf7fe294274a222363f84bcb63cdea762979a0443b4cf1f4f8fd17c86b1182"
+dependencies = [
+ "chrono",
+]
+
 [[package]]
 name = "either"
 version = "1.11.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "a47c1c47d2f5964e29c61246e81db715514cd532db6b5116a25ea3c03d6780a2"

+[[package]]
+name = "encoding_rs"
+version = "0.8.34"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "b45de904aa0b010bce2ab45264d0631681847fa7b6f2eaa7dab7619943bc4f59"
+dependencies = [
+ "cfg-if",
+]
+
 [[package]]
 name = "entities"
 version = "1.0.1"
@ -918,6 +951,16 @@ version = "1.0.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "b9e0384b61958566e926dc50660321d12159025e767c18e043daf26b70104c39"

+[[package]]
+name = "idna"
+version = "0.5.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "634d9b1461af396cad843f47fdba5597a4f9e6ddd4bfb6ff5d85028c25cb12f6"
+dependencies = [
+ "unicode-bidi",
+ "unicode-normalization",
+]
+
 [[package]]
 name = "indenter"
 version = "0.3.3"
@ -1061,6 +1104,12 @@ dependencies = [
 "windows-sys 0.48.0",
 ]

+[[package]]
+name = "never"
+version = "0.1.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "c96aba5aa877601bb3f6dd6a63a969e1f82e60646e81e71b14496995e9853c91"
+
 [[package]]
 name = "nom"
 version = "7.1.3"
@ -1208,7 +1257,7 @@ dependencies = [
 "base64",
 "indexmap 2.2.6",
 "line-wrap",
- "quick-xml",
+ "quick-xml 0.31.0",
 "serde",
 "time",
 ]
@ -1266,6 +1315,16 @@ dependencies = [
 "prost",
 ]

+[[package]]
+name = "quick-xml"
+version = "0.30.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "eff6510e86862b57b210fd8cbe8ed3f0d7d600b9c2863cd4549a2e033c66e956"
+dependencies = [
+ "encoding_rs",
+ "memchr",
+]
+
 [[package]]
 name = "quick-xml"
 version = "0.31.0"
@ -1358,6 +1417,18 @@ version = "0.8.3"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "adad44e29e4c806119491a7f06f03de4d1af22c3a680dd47f1e6e179439d1f56"

+[[package]]
+name = "rss"
+version = "2.0.7"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "f7b2c77eb4450d7d5f98df52c381cd6c4e19b75dad9209a9530b85a44510219a"
+dependencies = [
+ "atom_syndication",
+ "derive_builder",
+ "never",
+ "quick-xml 0.30.0",
+]
+
 [[package]]
 name = "rustc-demangle"
 version = "0.1.23"
@ -1652,6 +1723,21 @@ dependencies = [
 "time-core",
 ]

+[[package]]
+name = "tinyvec"
+version = "1.6.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "87cc5ceb3875bb20c2890005a4e226a4651264a5c75edb2421b52861a0a0cb50"
+dependencies = [
+ "tinyvec_macros",
+]
+
+[[package]]
+name = "tinyvec_macros"
+version = "0.1.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "1f3ccbac311fea05f86f61904b462b55fb3df8837a366dfc601a0161d0532f20"
+
 [[package]]
 name = "tokio"
 version = "1.37.0"
@ -1929,18 +2015,45 @@ dependencies = [
 "version_check",
 ]

+[[package]]
+name = "unicode-bidi"
+version = "0.3.15"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "08f95100a766bf4f8f28f90d77e0a5461bbdb219042e7679bebe79004fed8d75"
+
 [[package]]
 name = "unicode-ident"
 version = "1.0.12"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "3354b9ac3fae1ff6755cb6db53683adb661634f67557942dea4facebec0fee4b"

+[[package]]
+name = "unicode-normalization"
+version = "0.1.23"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "a56d1686db2308d901306f92a263857ef59ea39678a5458e7cb17f01415101f5"
+dependencies = [
+ "tinyvec",
+]
+
 [[package]]
 name = "unicode_categories"
 version = "0.1.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "39ec24b3121d976906ece63c9daad25b85969647682eee313cb5779fdd69e14e"

+[[package]]
+name = "url"
+version = "2.5.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "31e6302e3bb753d46e83516cae55ae196fc0c309407cf11ab35cc51a4c2a4633"
+dependencies = [
+ "form_urlencoded",
+ "idna",
+ "percent-encoding",
+ "serde",
+]
+
 [[package]]
 name = "valuable"
 version = "0.1.0"
--- a/Cargo.toml
+++ b/Cargo.toml
@ -37,6 +37,7 @@ comrak = { version = "0.22.0", features = [
 ], default-features = false }
 console-subscriber = { version = "0.2.0", optional = true }
 fronma = "0.2.0"
+rss = "2.0.7"
 scc = { version = "2.1.0", features = ["serde"] }
 serde = { version = "1.0.197", features = ["derive"] }
 syntect = "5.2.0"
@ -56,4 +57,5 @@ tower-http = { version = "0.5.2", features = [
 ], default-features = false }
 tracing = "0.1.40"
 tracing-subscriber = { version = "0.3.18", features = ["env-filter"] }
+url = { version = "2.5.0", features = ["serde"] }
 zstd = { version = "0.13.1", default-features = false }
--- a/README.md
+++ b/README.md
@ -11,7 +11,7 @@ blazingly fast markdown blog software written in rust memory safe

 ## TODO

- [ ] RSS
+- [x] RSS
 - [x] finish writing this document
 - [x] document config
 - [ ] extend syntect options
@ -27,6 +27,7 @@ blazingly fast markdown blog software written in rust memory safe
 - [x] clean up imports and require less features
 - [ ] improve home page
 - [x] tags
+- [ ] multi-language support
 - [x] be blazingly fast
 - [x] 100+ MiB binary size

@ -39,6 +40,11 @@ title = "bingus-blog"  # title of the website
 description = "blazingly fast markdown blog software written in rust memory safe" # description of the website
 raw_access = true      # allow users to see the raw markdown of a post

+[rss]
+enable = false         # serve an rss field under /feed.xml
+                       # this may be a bit resource intensive
+link = "https://..."   # public url of the blog, required if rss is enabled
+
 [dirs]
 posts = "posts"        # where posts are stored
 media = "media"        # directory served under /media/
--- a/src/config.rs
+++ b/src/config.rs
@ -6,6 +6,7 @@ use color_eyre::eyre::{bail, Context, Result};
 use serde::{Deserialize, Serialize};
 use tokio::io::{AsyncReadExt, AsyncWriteExt};
 use tracing::{error, info};
+use url::Url;

 use crate::ranged_i128_visitor::RangedI128Visitor;

@ -50,6 +51,12 @@ pub struct DirsConfig {
    pub media: PathBuf,
 }

+#[derive(Serialize, Deserialize, Debug, Clone)]
+pub struct RssConfig {
+    pub enable: bool,
+    pub link: Url,
+}
+
 #[derive(Serialize, Deserialize, Debug, Clone)]
 #[serde(default)]
 pub struct Config {
@ -57,6 +64,7 @@ pub struct Config {
    pub description: String,
    pub raw_access: bool,
    pub num_posts: usize,
+    pub rss: RssConfig,
    pub dirs: DirsConfig,
    pub http: HttpConfig,
    pub render: RenderConfig,
@ -70,6 +78,16 @@ impl Default for Config {
            description: "blazingly fast markdown blog software written in rust memory safe".into(),
            raw_access: true,
            num_posts: 5,
+            // i have a love-hate relationship with serde
+            // it was engimatic at first, but then i started actually using it
+            // writing my own serialize and deserialize implementations.. spending
+            // a lot of time in the docs trying to understand each and every option..
+            // now with this knowledge i can do stuff like this! (see rss field)
+            // and i'm proud to say that it still makes 0 sense.
+            rss: RssConfig {
+                enable: false,
+                link: Url::parse("http://example.com").unwrap(),
+            },
            dirs: Default::default(),
            http: Default::default(),
            render: Default::default(),
--- a/src/error.rs
+++ b/src/error.rs
@ -53,6 +53,10 @@ pub type AppResult<T> = Result<T, AppError>;
 pub enum AppError {
    #[error("failed to fetch post: {0}")]
    PostError(#[from] PostError),
+    #[error("rss is disabled")]
+    RssDisabled,
+    #[error(transparent)]
+    UrlError(#[from] url::ParseError),
 }

 impl From<std::io::Error> for AppError {
@ -75,7 +79,8 @@ impl IntoResponse for AppError {
                PostError::NotFound(_) => StatusCode::NOT_FOUND,
                _ => StatusCode::INTERNAL_SERVER_ERROR,
            },
-            //_ => StatusCode::INTERNAL_SERVER_ERROR,
+            AppError::RssDisabled => StatusCode::FORBIDDEN,
+            AppError::UrlError(_) => StatusCode::INTERNAL_SERVER_ERROR,
        };
        (
            status_code,
--- a/src/main.rs
+++ b/src/main.rs
@ -18,11 +18,13 @@ use std::time::Duration;

 use askama_axum::Template;
 use axum::extract::{Path, Query, State};
-use axum::http::Request;
+use axum::http::{header, Request};
 use axum::response::{IntoResponse, Redirect, Response};
 use axum::routing::{get, Router};
 use axum::Json;
 use color_eyre::eyre::{self, Context};
+use error::AppError;
+use rss::{Category, ChannelBuilder, ItemBuilder};
 use serde::Deserialize;
 use tokio::io::{AsyncReadExt, AsyncWriteExt};
 use tokio::net::TcpListener;
@ -78,7 +80,7 @@ async fn index(
 ) -> AppResult<IndexTemplate> {
    let posts = state
        .posts
-        .get_max_n_posts_with_optional_tag_sorted(query.num_posts, query.tag.as_ref())
+        .get_max_n_post_metadata_with_optional_tag_sorted(query.num_posts, query.tag.as_ref())
        .await?;

    Ok(IndexTemplate {
@ -94,12 +96,68 @@ async fn all_posts(
 ) -> AppResult<Json<Vec<PostMetadata>>> {
    let posts = state
        .posts
-        .get_max_n_posts_with_optional_tag_sorted(query.num_posts, query.tag.as_ref())
+        .get_max_n_post_metadata_with_optional_tag_sorted(query.num_posts, query.tag.as_ref())
        .await?;

    Ok(Json(posts))
 }

+async fn rss(
+    State(state): State<ArcState>,
+    Query(query): Query<QueryParams>,
+) -> AppResult<Response> {
+    if !state.config.rss.enable {
+        return Err(AppError::RssDisabled);
+    }
+
+    let posts = state
+        .posts
+        .get_max_n_posts_with_optional_tag_sorted(query.num_posts, query.tag.as_ref())
+        .await?;
+
+    let mut channel = ChannelBuilder::default();
+    channel
+        .title(&state.config.title)
+        .link(state.config.rss.link.to_string())
+        .description(&state.config.description);
+    //TODO: .language()
+
+    for (metadata, content, _) in posts {
+        channel.item(
+            ItemBuilder::default()
+                .title(metadata.title)
+                .description(metadata.description)
+                .author(metadata.author)
+                .categories(
+                    metadata
+                        .tags
+                        .into_iter()
+                        .map(|tag| Category {
+                            name: tag,
+                            domain: None,
+                        })
+                        .collect::<Vec<Category>>(),
+                )
+                .pub_date(metadata.created_at.map(|date| date.to_rfc2822()))
+                .content(content)
+                .link(
+                    state
+                        .config
+                        .rss
+                        .link
+                        .join(&format!("/posts/{}", metadata.name))?
+                        .to_string(),
+                )
+                .build(),
+        );
+    }
+
+    let body = channel.build().to_string();
+    drop(channel);
+
+    Ok(([(header::CONTENT_TYPE, "text/xml")], body).into_response())
+}
+
 async fn post(State(state): State<ArcState>, Path(name): Path<String>) -> AppResult<Response> {
    if name.ends_with(".md") && state.config.raw_access {
        let mut file = tokio::fs::OpenOptions::new()
@ -250,6 +308,7 @@ async fn main() -> eyre::Result<()> {
        )
        .route("/posts/:name", get(post))
        .route("/posts", get(all_posts))
+        .route("/feed.xml", get(rss))
        .nest_service("/static", ServeDir::new("static").precompressed_gzip())
        .nest_service("/media", ServeDir::new("media"))
        .layer(
--- a/src/post/mod.rs
+++ b/src/post/mod.rs
@ -200,7 +200,36 @@ impl PostManager {
        Ok(posts)
    }

-    pub async fn get_max_n_posts_with_optional_tag_sorted(
+    pub async fn get_all_posts_filtered(
+        &self,
+        filter: impl Fn(&PostMetadata, &str) -> bool,
+    ) -> Result<Vec<(PostMetadata, String, RenderStats)>, PostError> {
+        let mut posts = Vec::new();
+
+        let mut read_dir = fs::read_dir(&self.dir).await?;
+        while let Some(entry) = read_dir.next_entry().await? {
+            let path = entry.path();
+            let stat = fs::metadata(&path).await?;
+
+            if stat.is_file() && path.extension().is_some_and(|ext| ext == "md") {
+                let name = path
+                    .clone()
+                    .file_stem()
+                    .unwrap()
+                    .to_string_lossy()
+                    .to_string();
+
+                let post = self.get_post(&name).await?;
+                if filter(&post.0, &post.1) {
+                    posts.push(post);
+                }
+            }
+        }
+
+        Ok(posts)
+    }
+
+    pub async fn get_max_n_post_metadata_with_optional_tag_sorted(
        &self,
        n: Option<usize>,
        tag: Option<&String>,
@ -210,13 +239,34 @@ impl PostManager {
                !tag.is_some_and(|tag| !metadata.tags.contains(tag))
            })
            .await?;
-        posts.sort_unstable_by_key(|metadata| metadata.created_at.unwrap_or_default());
-
+        // we still want some semblance of order if created_at is None so sort by mtime as well
+        posts.sort_unstable_by_key(|metadata| metadata.modified_at.unwrap_or_default());
+        posts.sort_by_key(|metadata| metadata.created_at.unwrap_or_default());
+        posts.reverse();
        if let Some(n) = n {
-            posts = Vec::from(&posts[posts.len().saturating_sub(n)..]);
+            posts.truncate(n);
        }

+        Ok(posts)
+    }
+
+    pub async fn get_max_n_posts_with_optional_tag_sorted(
+        &self,
+        n: Option<usize>,
+        tag: Option<&String>,
+    ) -> Result<Vec<(PostMetadata, String, RenderStats)>, PostError> {
+        let mut posts = self
+            .get_all_posts_filtered(|metadata, _| {
+                !tag.is_some_and(|tag| !metadata.tags.contains(tag))
+            })
+            .await?;
+        posts.sort_unstable_by_key(|(metadata, ..)| metadata.modified_at.unwrap_or_default());
+        posts.sort_by_key(|(metadata, ..)| metadata.created_at.unwrap_or_default());
        posts.reverse();
+        if let Some(n) = n {
+            posts.truncate(n);
+        }
+
        Ok(posts)
    }