Fix spelling :3

with this commit mlem now requires nightly to build

Cargo.lock

@@ -350,6 +350,15 @@ version = "0.14.3"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "290f1a1d9242c78d09ce40a5e87e7554ee637af1351968159f4952f028f75604"
 
+[[package]]
+name = "html-escape"
+version = "0.2.13"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "6d1ad449764d627e22bfd7cd5e8868264fc9236e07c752972b4080cd351cb476"
+dependencies = [
+ "utf8-width",
+]
+
 [[package]]
 name = "html5ever"
 version = "0.26.0"
@@ -512,6 +521,7 @@ version = "0.2.0"
 dependencies = [
  "chrono",
  "derive_more",
+ "html-escape",
  "markdown",
  "regex",
  "scraper",
@@ -519,6 +529,7 @@ dependencies = [
  "serde_json",
  "tera",
  "toml",
+ "xml-builder",
 ]
 
 [[package]]
@@ -1193,6 +1204,12 @@ version = "0.7.6"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "09cc8ee72d2a9becf2f2febe0205bbed8fc6615b7cb429ad062dc7b7ddd036a9"
 
+[[package]]
+name = "utf8-width"
+version = "0.1.7"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "86bd8d4e895da8537e5315b8254664e6b769c4ff3db18321b297a1e7004392e3"
+
 [[package]]
 name = "version_check"
 version = "0.9.4"
@@ -1432,6 +1449,12 @@ dependencies = [
  "memchr",
 ]
 
+[[package]]
+name = "xml-builder"
+version = "0.5.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "efc4f1a86af7800dfc4056c7833648ea4515ae21502060b5c98114d828f5333b"
+
 [[package]]
 name = "zerocopy"
 version = "0.7.32"

Cargo.toml

@@ -15,3 +15,5 @@ chrono = "0.4"
 derive_more = "0.99"
 scraper = "0.19"
 serde_json = "1.0"
+xml-builder = "0.5"
+html-escape = "0.2"

src/config.rs

@@ -1,15 +1,30 @@
-use serde::{Deserialize,Serialize};
+use serde::{Deserialize, Serialize};
 use std::fs::read_to_string;
 
-#[derive(Deserialize,Serialize,Debug)]
+#[derive(Deserialize, Serialize, Debug)]
 pub struct Config {
     pub output_dir: String,
     pub src_dir: String,
     pub templates_dir: String,
     pub emoji_config: Option<EmojiConfig>,
+    pub syndication: Option<Syndication>,
+}
+#[derive(Deserialize, Serialize, Debug)]
+pub struct Syndication {
+    pub title: String,
+    pub link: String,
+    pub blog_root: String,
+    pub icon: Option<String>,
+    pub subtitle: Option<String>,
+    pub author: String,
+    pub atom: Option<AtomConfig>,
 }
 
-#[derive(Deserialize,Serialize,Debug)]
+#[derive(Deserialize, Serialize, Debug)]
+pub struct AtomConfig {
+    pub enabled: bool,
+}
+#[derive(Deserialize, Serialize, Debug)]
 pub struct EmojiConfig {
     pub emoji_web_directory: String,
     pub emoji_local_directory: String,
@@ -17,7 +32,13 @@ pub struct EmojiConfig {
 
 impl Default for Config {
     fn default() -> Self {
-        Config { output_dir: "output".to_string() , src_dir: "md_src".to_string(), templates_dir: "templates".to_string(), emoji_config: None }
+        Config {
+            output_dir: "output".to_string(),
+            src_dir: "md_src".to_string(),
+            templates_dir: "templates".to_string(),
+            emoji_config: None,
+            syndication: None,
+        }
     }
 }
 pub fn read_config() -> Config {

src/emoji.rs

@@ -2,10 +2,7 @@ use regex::Regex;
 use std::{ffi::OsString, fs::read_dir};
 
 use crate::config;
-pub fn emoji_pass(
-    markdown: &str,
-    emoji_config: &Option<config::EmojiConfig>,
-) -> String {
+pub fn emoji_pass(markdown: &str, emoji_config: &Option<config::EmojiConfig>) -> String {
     if emoji_config.is_none() {
         return markdown.to_string();
     }
@@ -14,13 +11,15 @@ pub fn emoji_pass(
     let re_emojis = Regex::new(r":\w+:").unwrap();
 
     for emoji in re_emojis.find_iter(&markdown.clone()) {
-        let emoji_file_name =
-            get_emoji_file_name(&get_emoji_name(emoji.as_str()), &emoji_config.as_ref().unwrap().emoji_local_directory);
+        let emoji_file_name = get_emoji_file_name(
+            &get_emoji_name(emoji.as_str()),
+            &emoji_config.as_ref().unwrap().emoji_local_directory,
+        );
         if emoji_file_name.is_none() {
             continue;
         }
         let html_string = format!(
-            "<img class=\"emoji\" src=\"{}/{}\"><img>",
+            "<img class=\"emoji\" src=\"{}/{}\"></img>",
             emoji_config.as_ref().unwrap().emoji_web_directory,
             emoji_file_name.unwrap().to_str().unwrap()
         );

src/index.rs

@@ -1,51 +1,65 @@
 use derive_more::Constructor;
 use serde::Serialize;
 use serde_json::value::Value;
-use std::collections::HashMap;
+use std::{collections::HashMap, fmt::Debug};
 use tera::{Context, Tera};
 
 #[derive(Constructor, Debug, Serialize)]
 pub struct BlogPost {
     pub title: String,
     pub human_date: String,
-    pub sort_date: i64,  // Sort date = unix timestamp
-    pub content: String, // Unformatted Content of blog post can be used to display a preview or reading time estimate. Will be html when assigned but later turned into raw text
+    pub sort_date: i64, // Sort date = unix timestamp
+    pub last_updated: i64,
+    pub content: String,
     pub output_file_name: String,
 }
 
-pub fn generate(mut blog_posts: Vec<BlogPost>, template_dir: &String) -> String {
-    for post in &mut blog_posts {
-        post.content = get_unformatted_text(post.content.clone());
-    }
+pub fn generate(blog_posts: Vec<BlogPost>, template_dir: &String) -> String {
 
     let mut tera = Tera::new(&format!("{}/*", template_dir)).unwrap();
     tera.autoescape_on(vec![]);
     tera.register_filter("truncate", truncate);
+    tera.register_filter("get_unformatted_content", get_unformatted_text);
+    tera.register_filter("remove_headers", remove_headers);
     let mut context = Context::new();
     context.insert("blog_posts", &blog_posts);
     tera.render("index.html", &context).unwrap()
 }
 
-fn get_unformatted_text(html: String) -> String {
-    let frag = scraper::Html::parse_fragment(&html);
+fn get_unformatted_text(html: &Value, _: &HashMap<String, Value>) -> Result<Value, tera::Error> {
+    let frag = scraper::Html::parse_fragment(&html.as_str().unwrap());
     let mut unformatted_text = String::new();
     for node in frag.tree {
         if let scraper::node::Node::Text(text) = node {
             unformatted_text.push_str(&text);
         }
     }
-    unformatted_text
+    Ok(Value::String(unformatted_text))
+
+}
+
+fn remove_headers(html: &Value, _: &HashMap<String, Value> ) -> Result<Value, tera::Error> {
+    let frag = scraper::Html::parse_fragment(html.as_str().unwrap());
+    let mut mut_frag = frag.clone();
+    let headers = vec!["h1","h2","h3","h4","h5","h6"];
+    for node in frag.tree.nodes() {
+        if let scraper::Node::Element(element) = node.value() {
+            if headers.contains(&element.name.local.to_string().as_str()) {
+                mut_frag.tree.get_mut(node.id()).unwrap().detach();
+            }
+        }
+    }
+
+    Ok(Value::String(mut_frag.html()))
 }
 
 fn truncate(value: &Value, args: &HashMap<String, Value>) -> Result<Value, tera::Error> {
     let mut value = value.as_str().unwrap().to_string();
-    let new_len:usize = args.get("len").unwrap().as_str().unwrap().parse().unwrap();
-    value
-        .truncate(new_len);
+    let new_len: usize = args.get("len").unwrap().as_str().unwrap().parse().unwrap();
+    value.truncate(new_len);
     Ok(Value::String(value.to_string()))
 }
 
-
 #[cfg(test)]
 mod tests {
     use crate::index::*;
@@ -54,13 +68,6 @@ mod tests {
         let mut args: HashMap<String, Value> = HashMap::new();
         args.insert("len".to_string(), Value::String("4".to_string()));
         let truncated_string = truncate(&Value::String("Meow Nya".to_string()), &args).unwrap();
-        assert_eq!(truncated_string.as_str().unwrap(),"Meow");
-    }
-
-    #[test]
-    fn unformat_html() {
-        let html_src = "<p>Meow nyaaa<em> UwU</em></p>".to_string();
-        let unformatted_text = get_unformatted_text(html_src);
-        assert_eq!(unformatted_text,"Meow nyaaa UwU");
+        assert_eq!(truncated_string.as_str().unwrap(), "Meow");
     }
 }

src/lib.rs

@@ -1,3 +1,4 @@
+#![feature(str_split_remainder)]
 use chrono::prelude::*;
 use derive_more::Constructor;
 use markdown::{to_html_with_options, CompileOptions, Options};
@@ -12,14 +13,17 @@ pub fn read_src_files(src_dir: &str) -> Vec<SrcMD> {
     let mut files: Vec<SrcMD> = Vec::new();
     for file in read_dir(src_dir).expect("Cant read src dir") {
         let file = file.unwrap();
-        let kv = get_kv(read_to_string(file.path()).unwrap()).0;
-        let title = kv.get("title").unwrap();
-        let date = kv.get("date").unwrap();
+        let frontmatter = get_frontmatter(read_to_string(file.path()).unwrap()).0;
+        let title = frontmatter
+            .get("title")
+            .expect("Error getting field: title from frontmatter");
+        let date = frontmatter
+            .get("date")
+            .expect("Error getting field: date from frontmatter");
         files.push(SrcMD::new(
             file.path(),
             title.to_string(),
-            DateTime::parse_from_str(&format!("{date} 00:00:00 +00:00"), "%Y-%m-%d %H:%M:%S %z")
-                .unwrap(),
+            DateTime::parse_from_rfc3339(date).unwrap(),
             file.path().file_stem().unwrap().to_os_string(),
         ))
     }
@@ -32,7 +36,7 @@ pub fn write_to_fs(html: String, output_dir: &String, file_name: &String) {
         .unwrap_or_else(|_| panic!("Error writing {file_name}"));
 }
 
-#[derive(Constructor)]
+#[derive(Constructor, Debug)]
 pub struct SrcMD {
     pub path: PathBuf,
     pub title: String,
@@ -40,14 +44,15 @@ pub struct SrcMD {
     pub file_name: OsString,
 }
 
-pub fn generate_blog_entry(markdown: String, template_dir: &String) -> (String, String) {
+pub fn generate_blog_entry(
+    markdown: String,
+    template_dir: &String,
+    frontmatter: &mut HashMap<String, String>,
+) -> Option<(String, String)> {
     let markdown = markdown.clone();
-
     let mut tera = Tera::new(&format!("{template_dir}/*")).unwrap();
     tera.autoescape_on(vec![]);
 
-    let (mut key_value, markdown) = get_kv(markdown);
-
     let html_markdown = to_html_with_options(
         &markdown,
         &Options {
@@ -61,22 +66,20 @@ pub fn generate_blog_entry(markdown: String, template_dir: &String) -> (String,
     )
     .unwrap();
 
-    key_value.insert("blog_content".to_string(), html_markdown.clone());
+    frontmatter.insert("blog_content".to_string(), html_markdown.clone());
 
-    let context = Context::from_serialize(&key_value).unwrap();
+    let context = Context::from_serialize(&frontmatter).unwrap();
 
-    let templated_html = tera
-        .render(key_value.get("template").unwrap(), &context)
-        .unwrap();
-    (templated_html, html_markdown)
+    let templated_html = tera.render(frontmatter.get("template")?, &context).unwrap();
+    Some((templated_html, html_markdown))
 }
 
-pub fn get_kv(markdown: String) -> (HashMap<String, String>, String) {
-    let re_key_value = Regex::new(r"(?ms)---(.*)---(?:\n)").unwrap();
+pub fn get_frontmatter(markdown: String) -> (HashMap<String, String>, String) {
+    let re_key_value = Regex::new(r"(?msU)---(.*)---(?:\n)").unwrap();
 
     let key_value_string = re_key_value
         .find(markdown.as_str())
-        .expect("Can't find key value map in markdown");
+        .expect("Can't find frontmatter in markdown");
 
     let content_markdown = re_key_value
         .replace(markdown.clone().as_str(), "")
@@ -88,6 +91,12 @@ pub fn get_kv(markdown: String) -> (HashMap<String, String>, String) {
         if line == "---" {
             continue;
         }
+        let second_value = {
+            let mut line_iter = line.split(':');
+            line_iter.next();
+
+            line_iter.remainder().unwrap().trim().to_string()
+        };
         key_value.insert(
             line.split(':')
                 .collect::<Vec<&str>>()
@@ -95,12 +104,7 @@ pub fn get_kv(markdown: String) -> (HashMap<String, String>, String) {
                 .unwrap()
                 .trim()
                 .to_string(),
-            line.split(':')
-                .collect::<Vec<&str>>()
-                .get(1)
-                .unwrap()
-                .trim()
-                .to_string(),
+            second_value,
         );
     }
 

src/main.rs

@@ -2,10 +2,11 @@ use std::fs::read_to_string;
 pub mod config;
 pub mod emoji;
 pub mod index;
+pub mod syndication;
 use mlem::*;
 use std::env;
 
-fn main(){
+fn main() {
     let args: Vec<String> = env::args().collect();
 
     if args.len() == 1 {
@@ -24,27 +25,46 @@ fn generate() {
     let mut post_index: Vec<index::BlogPost> = Vec::new();
 
     for file in raw_files {
-        let mut markdown = read_to_string(file.path).expect("File does not exist");
-        markdown = emoji::emoji_pass(
-            &markdown,
-            &config.emoji_config,
-        );
+        let markdown = read_to_string(file.path).expect("File does not exist");
+        let (mut frontmatter, mut markdown) = get_frontmatter(markdown);
+        markdown = emoji::emoji_pass(&markdown, &config.emoji_config);
 
-        let (html, index_content) = generate_blog_entry(markdown, &config.templates_dir);
+        let (html, index_content) =
+            generate_blog_entry(markdown, &config.templates_dir, &mut frontmatter)
+                .unwrap_or_else(|| panic!("Error generating entry {}", &file.title));
         write_to_fs(
             html,
             &config.output_dir,
             &file.file_name.clone().into_string().unwrap(),
         );
 
+        let last_updated: Option<&String> = frontmatter.get("updated_at");
+        let updated = {
+            if last_updated.is_some() {
+                chrono::DateTime::parse_from_rfc3339(last_updated.unwrap())
+                    .unwrap()
+                    .timestamp()
+            } else {
+                file.date.clone().timestamp()
+            }
+        };
+
         post_index.push(index::BlogPost::new(
             file.title,
             file.date.format("%Y-%m-%d").to_string(),
             file.date.timestamp(),
+            updated,
             index_content,
             format!("{}.html", file.file_name.to_str().unwrap()),
         ));
     }
+    if config.syndication.is_some() {
+        if config.syndication.as_ref().unwrap().atom.is_some() {
+            let atom = syndication::atom::generate(config.syndication.unwrap(), &post_index);
+            std::fs::write(format!("{}/atom.xml", config.output_dir), atom)
+                .unwrap_or_else(|_| panic!("Error writing atom feed"));
+        }
+    }
     let index = index::generate(post_index, &config.templates_dir);
 
     std::fs::write(format!("{}/index.html", config.output_dir), index)
@@ -56,7 +76,7 @@ fn create_default_env() {
     let config = config::Config::default();
     let toml_config = toml::to_string(&config).unwrap();
     std::fs::write("mlem.toml", toml_config).unwrap();
-    println!("Creating default directorys");
+    println!("Creating default directories");
     std::fs::create_dir(config.output_dir).unwrap();
     std::fs::create_dir(config.src_dir).unwrap();
     std::fs::create_dir(config.templates_dir).unwrap();

src/syndication/atom.rs

@@ -0,0 +1,99 @@
+use xml_builder::{XMLBuilder, XMLElement, XMLVersion};
+
+use crate::{xml_tag_attribute, xml_tag_text};
+
+pub fn generate(
+    config: crate::config::Syndication,
+    post_index: &Vec<crate::index::BlogPost>,
+) -> String {
+    let mut xml = XMLBuilder::new()
+        .version(XMLVersion::XML1_0)
+        .encoding("UTF-8".into())
+        .build();
+
+    let mut feed = XMLElement::new("feed");
+    feed.add_attribute("xmlns", "http://www.w3.org/2005/Atom");
+
+    xml_tag_text!("id", config.link.clone(), feed);
+
+    xml_tag_text!("title", config.title, feed);
+
+    xml_tag_attribute!("link", feed, ("href", &config.link));
+
+    let mut author = XMLElement::new("author");
+    let mut name = XMLElement::new("name");
+    name.add_text(config.author).unwrap();
+    author.add_child(name).unwrap();
+    feed.add_child(author).unwrap();
+
+    let last_update = chrono::DateTime::from_timestamp(last_update(post_index), 0).unwrap();
+    xml_tag_text!("updated", last_update.to_rfc3339(), feed);
+
+    if config.subtitle.is_some() {
+        xml_tag_text!("subtitle", config.subtitle.unwrap(), feed);
+    }
+
+    if config.icon.is_some() {
+        xml_tag_text!("icon", config.icon.unwrap(), feed);
+    }
+
+    for post in post_index {
+        let mut entry = XMLElement::new("entry");
+
+        xml_tag_text!("title", post.title.clone(), entry);
+
+        xml_tag_attribute!(
+            "link",
+            entry,
+            (
+                "href",
+                &format!(
+                    "{}{}/{}",
+                    &config.link, &config.blog_root, &post.output_file_name
+                )
+            )
+        );
+
+        xml_tag_text!(
+            "id",
+            format!(
+                "{}{}/{}",
+                &config.link, &config.blog_root, &post.output_file_name
+            ),
+            entry
+        );
+        let last_update = chrono::DateTime::from_timestamp(post.last_updated, 0)
+            .unwrap()
+            .to_rfc3339();
+        xml_tag_text!("updated", last_update, entry);
+
+        let mut content = XMLElement::new("content");
+        content.add_attribute("type", "html");
+        content
+            .add_text(crate::syndication::escape_html(post.content.clone()))
+            .unwrap();
+        entry.add_child(content).unwrap();
+
+        feed.add_child(entry).unwrap();
+    }
+
+    xml.set_root_element(feed);
+    let mut writer: Vec<u8> = Vec::new();
+    xml.generate(&mut writer).unwrap();
+
+
+
+
+    String::from_utf8(writer).unwrap()
+}
+
+fn last_update(post_index: &Vec<crate::index::BlogPost>) -> i64 {
+    let mut last_timestamp: i64 = i64::MIN;
+
+    for post in post_index {
+        if post.sort_date > last_timestamp {
+            last_timestamp = post.sort_date;
+        }
+    }
+    last_timestamp
+}

src/syndication/mod.rs

@@ -0,0 +1,26 @@
+pub mod atom;
+
+#[macro_export]
+macro_rules! xml_tag_text {
+    ($a: expr, $b: expr, $c: expr) => {
+        let mut tag = XMLElement::new($a);
+        tag.add_text($b).unwrap();
+        $c.add_child(tag).unwrap();
+    };
+}
+
+#[macro_export]
+macro_rules! xml_tag_attribute {
+    ($a: expr, $b: expr, $( $c: expr),+ ) => {
+        let mut tag = XMLElement::new($a);
+        $(
+            tag.add_attribute($c.0,$c.1);
+        )+
+        $b.add_child(tag).unwrap();
+
+    };
+}
+
+fn escape_html(unescaped_html: String) -> String {
+    html_escape::encode_text(&unescaped_html).to_string()
+}