Files
markdownblog/markdown_backend/src/markdown.rs
rattatwinko 21f13ef8ae Enhance blog features and improve backend functionality
- Added a VS Code-style editor with YAML frontmatter support and live preview.
- Implemented force reparse functionality for immediate updates of posts.
- Improved directory scanning with error handling and automatic directory creation.
- Introduced new CLI commands for cache management: `reinterpret-all` and `reparse-post`.
- Enhanced logging for better debugging and monitoring of the Rust backend.
- Updated README to reflect new features and improvements.
2025-07-05 22:23:58 +02:00

911 lines
35 KiB
Rust
Raw Blame History

This file contains invisible Unicode characters

This file contains invisible Unicode characters that are indistinguishable to humans but may be processed differently by a computer. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

//
// src/markdown.rs
// Written by: @rattatwinko
//
use std::fs;
use std::path::{Path, PathBuf};
use std::collections::HashMap;
use std::sync::RwLock;
use std::time::Instant;
use std::sync::mpsc::channel;
use std::collections::VecDeque;
use chrono::{DateTime, Utc};
use serde::{Deserialize, Serialize};
use pulldown_cmark::{Parser, Options, html, Event, Tag, CowStr};
use gray_matter::engine::YAML;
use gray_matter::Matter;
use slug::slugify;
use notify::{RecursiveMode, RecommendedWatcher, Watcher, Config};
use syntect::highlighting::ThemeSet;
use syntect::parsing::SyntaxSet;
use syntect::html::highlighted_html_for_string;
use once_cell::sync::Lazy;
use serde_json;
use sysinfo::{System, RefreshKind, CpuRefreshKind, ProcessRefreshKind};
use regex::Regex;
// Constants
const POSTS_CACHE_PATH: &str = "./cache/posts_cache.json";
const POST_STATS_PATH: &str = "./cache/post_stats.json";
const MAX_FILE_SIZE: usize = 2 * 1024 * 1024; // 10MB
const PARSING_TIMEOUT_SECS: u64 = 6000;
const MAX_LOG_ENTRIES: usize = 1000;
const PARSER_LOGS_PATH: &str = "./cache/parser_logs.json";
// Data structures
#[derive(Debug, Deserialize, Clone, Serialize)]
pub struct PostFrontmatter {
pub title: String,
pub date: String,
pub tags: Option<Vec<String>>,
pub summary: Option<String>,
}
// Post Data Structures
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct Post {
pub slug: String,
pub title: String,
pub date: String,
pub tags: Vec<String>,
pub summary: Option<String>,
pub content: String,
pub created_at: String,
pub author: String,
}
// Data Structure for Posts Statistics
#[derive(Debug, Clone, Serialize, Deserialize, Default)]
pub struct PostStats {
pub slug: String,
pub cache_hits: u64,
pub cache_misses: u64,
pub last_interpret_time_ms: u128,
pub last_compile_time_ms: u128,
pub last_cpu_usage_percent: f32,
pub last_cache_status: String, // "hit" or "miss"
}
// Data Structures for Health Reporting
#[derive(Debug, Serialize)]
pub struct HealthReport {
pub posts_dir_exists: bool,
pub posts_count: usize,
pub cache_file_exists: bool,
pub cache_stats_file_exists: bool,
pub cache_readable: bool,
pub cache_stats_readable: bool,
pub cache_post_count: Option<usize>,
pub cache_stats_count: Option<usize>,
pub errors: Vec<String>,
}
// Log Data Structure (frontend related)
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct LogEntry {
pub timestamp: String,
pub level: String, // "info", "warning", "error"
pub message: String,
pub slug: Option<String>,
pub details: Option<String>,
}
// Static caches
static POST_CACHE: Lazy<RwLock<HashMap<String, Post>>> = Lazy::new(|| RwLock::new(HashMap::new()));
static ALL_POSTS_CACHE: Lazy<RwLock<Option<Vec<Post>>>> = Lazy::new(|| RwLock::new(None));
static POST_STATS: Lazy<RwLock<HashMap<String, PostStats>>> = Lazy::new(|| RwLock::new(HashMap::new()));
static PARSER_LOGS: Lazy<RwLock<VecDeque<LogEntry>>> = Lazy::new(|| RwLock::new(VecDeque::new()));
// Ammonia HTML sanitizer configuration
static AMMONIA: Lazy<ammonia::Builder<'static>> = Lazy::new(|| {
let mut builder = ammonia::Builder::default();
// Add allowed attributes for various HTML tags
builder.add_tag_attributes("h1", &["style", "id"]);
builder.add_tag_attributes("h2", &["style", "id"]);
builder.add_tag_attributes("h3", &["style", "id"]);
builder.add_tag_attributes("h4", &["style", "id"]);
builder.add_tag_attributes("h5", &["style", "id"]);
builder.add_tag_attributes("h6", &["style", "id"]);
builder.add_tag_attributes("p", &["style"]);
builder.add_tag_attributes("span", &["style"]);
builder.add_tag_attributes("strong", &["style"]);
builder.add_tag_attributes("em", &["style"]);
builder.add_tag_attributes("b", &["style"]);
builder.add_tag_attributes("i", &["style"]);
builder.add_tag_attributes("u", &["style"]);
builder.add_tag_attributes("mark", &["style"]);
builder.add_tag_attributes("small", &["style"]);
builder.add_tag_attributes("abbr", &["style"]);
builder.add_tag_attributes("cite", &["style"]);
builder.add_tag_attributes("q", &["style"]);
builder.add_tag_attributes("code", &["style"]);
builder.add_tag_attributes("pre", &["style"]);
builder.add_tag_attributes("kbd", &["style"]);
builder.add_tag_attributes("samp", &["style"]);
builder.add_tag_attributes("section", &["style"]);
builder.add_tag_attributes("article", &["style"]);
builder.add_tag_attributes("header", &["style"]);
builder.add_tag_attributes("footer", &["style"]);
builder.add_tag_attributes("main", &["style"]);
builder.add_tag_attributes("aside", &["style"]);
builder.add_tag_attributes("nav", &["style"]);
builder.add_tag_attributes("ul", &["style"]);
builder.add_tag_attributes("ol", &["style"]);
builder.add_tag_attributes("li", &["style"]);
builder.add_tag_attributes("dl", &["style"]);
builder.add_tag_attributes("dt", &["style"]);
builder.add_tag_attributes("dd", &["style"]);
builder.add_tag_attributes("table", &["style"]);
builder.add_tag_attributes("thead", &["style"]);
builder.add_tag_attributes("tbody", &["style"]);
builder.add_tag_attributes("tfoot", &["style"]);
builder.add_tag_attributes("tr", &["style"]);
builder.add_tag_attributes("td", &["style"]);
builder.add_tag_attributes("th", &["style"]);
builder.add_tag_attributes("a", &["style"]);
builder.add_tag_attributes("img", &["style"]);
builder.add_tag_attributes("video", &["style"]);
builder.add_tag_attributes("audio", &["style"]);
builder.add_tag_attributes("source", &["style"]);
builder.add_tag_attributes("iframe", &["style"]);
builder.add_tag_attributes("sup", &["style"]);
builder.add_tag_attributes("sub", &["style"]);
builder.add_tag_attributes("time", &["style"]);
builder.add_tag_attributes("var", &["style"]);
builder.add_tag_attributes("del", &["style"]);
builder.add_tag_attributes("ins", &["style"]);
builder.add_tag_attributes("br", &["style"]);
builder.add_tag_attributes("wbr", &["style"]);
builder.add_tag_attributes("form", &["style"]);
builder.add_tag_attributes("input", &["style"]);
builder.add_tag_attributes("textarea", &["style"]);
builder.add_tag_attributes("select", &["style"]);
builder.add_tag_attributes("option", &["style"]);
builder.add_tag_attributes("button", &["style"]);
builder.add_tag_attributes("label", &["style"]);
builder.add_tag_attributes("fieldset", &["style"]);
builder.add_tag_attributes("legend", &["style"]);
builder.add_tag_attributes("blockquote", &["style"]);
builder.add_tag_attributes("font", &["style"]);
builder.add_tag_attributes("center", &["style"]);
builder.add_tag_attributes("big", &["style"]);
builder.add_tag_attributes("tt", &["style"]);
// Add class attribute for div
builder.add_tag_attributes("div", &["style", "class"]);
builder
});
// Helper functions
fn ensure_cache_directory() {
let cache_dir = PathBuf::from("./cache");
if !cache_dir.exists() {
if let Err(e) = fs::create_dir_all(&cache_dir) {
eprintln!("Failed to create cache directory: {}", e);
add_log("error", &format!("Failed to create cache directory: {}", e), None, None);
} else {
add_log("info", "Created cache directory: ./cache", None, None);
}
}
}
fn get_posts_directory() -> PathBuf {
let is_docker = std::env::var("DOCKER_CONTAINER").is_ok()
|| std::env::var("KUBERNETES_SERVICE_HOST").is_ok()
|| std::path::Path::new("/.dockerenv").exists();
let candidates = if is_docker {
vec![
"/app/docker", // Docker volume mount point (highest priority in Docker)
"/app/posts", // Fallback in Docker
"./posts",
"../posts",
"/posts",
"/docker"
]
} else {
vec![
"./posts",
"../posts",
"/posts",
"/docker",
"/app/docker" // Lower priority for non-Docker environments
]
};
for candidate in candidates.iter() {
let path = PathBuf::from(candidate);
if path.exists() && path.is_dir() {
add_log("info", &format!("Using posts directory: {:?}", path), None, None);
return path;
}
}
// Fallback: create ./posts if it doesn't exist
let fallback_path = PathBuf::from("./posts");
if !fallback_path.exists() {
if let Err(e) = fs::create_dir_all(&fallback_path) {
add_log("error", &format!("Failed to create posts directory: {}", e), None, None);
} else {
add_log("info", "Created posts directory: ./posts", None, None);
}
}
fallback_path
}
// Function to find Markdown files with improved reliability
fn find_markdown_files(dir: &Path) -> std::io::Result<Vec<PathBuf>> {
let mut files = Vec::new();
let mut errors = Vec::new();
if !dir.exists() {
let error_msg = format!("Directory does not exist: {:?}", dir);
add_log("error", &error_msg, None, None);
return Err(std::io::Error::new(std::io::ErrorKind::NotFound, error_msg));
}
if !dir.is_dir() {
let error_msg = format!("Path is not a directory: {:?}", dir);
add_log("error", &error_msg, None, None);
return Err(std::io::Error::new(std::io::ErrorKind::InvalidInput, error_msg));
}
// Try to read directory with retry logic
let entries = match fs::read_dir(dir) {
Ok(entries) => entries,
Err(e) => {
add_log("error", &format!("Failed to read directory {:?}: {}", dir, e), None, None);
return Err(e);
}
};
for entry_result in entries {
match entry_result {
Ok(entry) => {
let path = entry.path();
// Skip hidden files and directories
if let Some(name) = path.file_name() {
if name.to_string_lossy().starts_with('.') {
continue;
}
}
if path.is_dir() {
// Recursively scan subdirectories
match find_markdown_files(&path) {
Ok(subfiles) => files.extend(subfiles),
Err(e) => {
let error_msg = format!("Error scanning subdirectory {:?}: {}", path, e);
add_log("warning", &error_msg, None, None);
errors.push(error_msg);
}
}
} else if path.extension().map(|e| e == "md").unwrap_or(false) {
// Verify the file is readable
match fs::metadata(&path) {
Ok(metadata) => {
if metadata.is_file() {
files.push(path);
}
}
Err(e) => {
let error_msg = format!("Cannot access file {:?}: {}", path, e);
add_log("warning", &error_msg, None, None);
errors.push(error_msg);
}
}
}
}
Err(e) => {
let error_msg = format!("Error reading directory entry: {}", e);
add_log("warning", &error_msg, None, None);
errors.push(error_msg);
}
}
}
// Log summary
add_log("info", &format!("Found {} markdown files in {:?}", files.len(), dir), None, None);
if !errors.is_empty() {
add_log("warning", &format!("Encountered {} errors during directory scan", errors.len()), None, None);
}
Ok(files)
}
// Generate a SlugPath.
fn path_to_slug(file_path: &Path, posts_dir: &Path) -> String {
let relative_path = file_path.strip_prefix(posts_dir).unwrap_or(file_path);
let without_ext = relative_path.with_extension("");
without_ext.to_string_lossy()
.replace(std::path::MAIN_SEPARATOR, "::")
.replace("/", "::")
.replace("\\", "::")
}
// Slugify the Path
fn slug_to_path(slug: &str, posts_dir: &Path) -> PathBuf {
let parts: Vec<&str> = slug.split("::").collect();
if parts.len() == 1 {
posts_dir.join(format!("{}.md", parts[0]))
} else {
let mut path = posts_dir.to_path_buf();
for (i, part) in parts.iter().enumerate() {
if i == parts.len() - 1 {
path = path.join(format!("{}.md", part));
} else {
path = path.join(part);
}
}
path
}
}
// Look at the Markdown File and generate a Creation Date based upon gathered things.
fn get_file_creation_date(path: &Path) -> std::io::Result<DateTime<Utc>> {
let metadata = fs::metadata(path)?;
match metadata.created() {
Ok(created) => Ok(DateTime::<Utc>::from(created)),
Err(_) => {
let modified = metadata.modified()?;
Ok(DateTime::<Utc>::from(modified))
}
}
}
// The Frontend expects a plain old string that will be used for the anchor
// something like this -> #i-am-a-heading
// This creates a crossreference for Links that scroll to said heading
fn process_anchor_links(content: &str) -> String {
let re = regex::Regex::new(r"\[([^\]]+)\]\(#([^)]+)\)").unwrap();
re.replace_all(content, |caps: &regex::Captures| {
let link_text = &caps[1];
let anchor = &caps[2];
let slugified = slugify(anchor);
format!("[{}](#{})", link_text, slugified)
}).to_string()
}
// Here we just remove the Emoji if it is in the heading.
// Example "🏳️‍🌈 Hi!" will turn into "#hi"
fn strip_emojis(s: &str) -> String {
s.chars()
.filter(|c| {
let c = *c as u32;
!( (c >= 0x1F600 && c <= 0x1F64F) // Emoticons
|| (c >= 0x1F300 && c <= 0x1F5FF) // Misc Symbols and Pictographs
|| (c >= 0x1F680 && c <= 0x1F6FF) // Transport and Map
|| (c >= 0x2600 && c <= 0x26FF) // Misc symbols
|| (c >= 0x2700 && c <= 0x27BF) // Dingbats
|| (c >= 0x1F900 && c <= 0x1F9FF) // Supplemental Symbols and Pictographs
|| (c >= 0x1FA70 && c <= 0x1FAFF) // Symbols and Pictographs Extended-A
|| (c >= 0x1F1E6 && c <= 0x1F1FF) // Regional Indicator Symbols
)
})
.collect()
}
// This is a obsolete Function for Custom Tags for HTML
// Example usage in Text: <warning />
fn process_custom_tags(content: &str) -> String {
let mut processed = content.to_string();
// Handle simple tags without parameters
let simple_tags = [
("<mytag />", "<div class=\"custom-tag mytag\">This is my custom tag content!</div>"),
("<warning />", "<div class=\"custom-tag warning\" style=\"background: #fff3cd; border: 1px solid #ffeaa7; padding: 1rem; border-radius: 4px; margin: 1rem 0;\">⚠️ Warning: This is a custom warning tag!</div>"),
("<info />", "<div class=\"custom-tag info\" style=\"background: #d1ecf1; border: 1px solid #bee5eb; padding: 1rem; border-radius: 4px; margin: 1rem 0;\"> Info: This is a custom info tag!</div>"),
("<success />", "<div class=\"custom-tag success\" style=\"background: #d4edda; border: 1px solid #c3e6cb; padding: 1rem; border-radius: 4px; margin: 1rem 0;\">✅ Success: This is a custom success tag!</div>"),
("<error />", "<div class=\"custom-tag error\" style=\"background: #f8d7da; border: 1px solid #f5c6cb; padding: 1rem; border-radius: 4px; margin: 1rem 0;\">❌ Error: This is a custom error tag!</div>"),
];
for (tag, replacement) in simple_tags.iter() {
processed = processed.replace(tag, replacement);
}
// Handle tags with parameters
let tag_with_params = Regex::new(r"<(\w+)\s+([^>]*?[a-zA-Z0-9=])[^>]*/>").unwrap();
processed = tag_with_params.replace_all(&processed, |caps: &regex::Captures| {
let tag_name = &caps[1];
let params = &caps[2];
match tag_name {
"mytag" => {
format!("<div class=\"custom-tag mytag\" data-params=\"{}\">Custom content with params: {}</div>", params, params)
},
"alert" => {
if params.contains("type=\"warning\"") {
"<div class=\"custom-tag alert warning\" style=\"background: #fff3cd; border: 1px solid #ffeaa7; padding: 1rem; border-radius: 4px; margin: 1rem 0;\">⚠️ Warning Alert!</div>".to_string()
} else if params.contains("type=\"error\"") {
"<div class=\"custom-tag alert error\" style=\"background: #f8d7da; border: 1px solid #f5c6cb; padding: 1rem; border-radius: 4px; margin: 1rem 0;\">❌ Error Alert!</div>".to_string()
} else {
"<div class=\"custom-tag alert info\" style=\"background: #d1ecf1; border: 1px solid #bee5eb; padding: 1rem; border-radius: 4px; margin: 1rem 0;\"> Info Alert!</div>".to_string()
}
},
_ => format!("<div class=\"custom-tag {}\">Unknown custom tag: {}</div>", tag_name, tag_name)
}
}).to_string();
processed
}
// Logging functions
fn add_log(level: &str, message: &str, slug: Option<&str>, details: Option<&str>) {
let timestamp = chrono::Utc::now().to_rfc3339();
let log_entry = LogEntry {
timestamp,
level: level.to_string(),
message: message.to_string(),
slug: slug.map(|s| s.to_string()),
details: details.map(|s| s.to_string()),
};
{
let mut logs = PARSER_LOGS.write().unwrap();
logs.push_back(log_entry.clone());
// Keep only the last MAX_LOG_ENTRIES
while logs.len() > MAX_LOG_ENTRIES {
logs.pop_front();
}
// Write logs to disk
let _ = save_parser_logs_to_disk_inner(&logs);
}
}
fn save_parser_logs_to_disk_inner(logs: &VecDeque<LogEntry>) -> std::io::Result<()> {
ensure_cache_directory();
let logs_vec: Vec<_> = logs.iter().cloned().collect();
let json = serde_json::to_string(&logs_vec)?;
std::fs::write(PARSER_LOGS_PATH, json)?;
Ok(())
}
pub fn load_parser_logs_from_disk() {
if let Ok(data) = std::fs::read_to_string(PARSER_LOGS_PATH) {
if let Ok(logs_vec) = serde_json::from_str::<Vec<LogEntry>>(&data) {
let mut logs = PARSER_LOGS.write().unwrap();
logs.clear();
for entry in logs_vec {
logs.push_back(entry);
}
}
}
}
// Main public functions
pub fn rsparseinfo() -> String {
let _ = get_all_posts();
let stats = POST_STATS.read().unwrap();
let values: Vec<&PostStats> = stats.values().collect();
if values.is_empty() {
"[]".to_string()
} else {
serde_json::to_string(&values).unwrap_or_else(|_| "[]".to_string())
}
}
// This Function gets the Post by its Slugified Version.
// This is basically only used for Caching (loading from it).
pub fn get_post_by_slug(slug: &str) -> Result<Post, Box<dyn std::error::Error>> {
add_log("info", "Starting post parsing", Some(slug), None);
let mut sys = System::new_with_specifics(RefreshKind::new().with_processes(ProcessRefreshKind::everything()).with_cpu(CpuRefreshKind::everything()));
sys.refresh_processes();
let pid = sysinfo::get_current_pid()?;
let before_cpu = sys.process(pid).map(|p| p.cpu_usage()).unwrap_or(0.0);
let start = Instant::now();
let mut stats = POST_STATS.write().unwrap();
let entry = stats.entry(slug.to_string()).or_insert_with(|| PostStats {
slug: slug.to_string(),
..Default::default()
});
// Try cache first
if let Some(post) = POST_CACHE.read().unwrap().get(slug).cloned() {
entry.cache_hits += 1;
entry.last_interpret_time_ms = 0;
entry.last_compile_time_ms = 0;
entry.last_cache_status = "hit".to_string();
sys.refresh_process(pid);
entry.last_cpu_usage_percent = sys.process(pid).map(|p| p.cpu_usage()).unwrap_or(0.0) - before_cpu;
add_log("info", "Cache hit", Some(slug), None);
return Ok(post);
}
entry.cache_misses += 1;
entry.last_cache_status = "miss".to_string();
drop(stats);
let posts_dir = get_posts_directory();
let file_path = slug_to_path(slug, &posts_dir);
if !file_path.exists() {
let error_msg = format!("File not found: {:?}", file_path);
add_log("error", &error_msg, Some(slug), None);
return Err(error_msg.into());
}
let file_content = fs::read_to_string(&file_path)?;
add_log("info", &format!("File loaded: {} bytes", file_content.len()), Some(slug), None);
if file_content.len() > MAX_FILE_SIZE {
let error_msg = format!("File too large: {} bytes (max: {} bytes)", file_content.len(), MAX_FILE_SIZE);
add_log("error", &error_msg, Some(slug), None);
return Err(error_msg.into());
}
let matter = Matter::<YAML>::new();
let result = matter.parse(&file_content);
let front: PostFrontmatter = if let Some(data) = result.data {
match data.deserialize() {
Ok(front) => front,
Err(e) => {
let error_msg = format!("Failed to deserialize frontmatter: {}", e);
add_log("error", &error_msg, Some(slug), None);
return Err(error_msg.into());
}
}
} else {
add_log("error", "No frontmatter found", Some(slug), None);
return Err("No frontmatter found".into());
};
let created_at = get_file_creation_date(&file_path)?;
let processed_markdown = process_anchor_links(&result.content);
let processed_markdown = process_custom_tags(&processed_markdown);
add_log("info", "Starting markdown parsing", Some(slug), Some(&format!("Content length: {} chars", processed_markdown.len())));
let parser = Parser::new_ext(&processed_markdown, Options::all());
let mut html_output = String::new();
let mut heading_text = String::new();
let mut in_heading = false;
let mut heading_level = 0;
let mut in_code_block = false;
let mut code_block_lang = String::new();
let mut code_block_content = String::new();
let mut events = Vec::new();
let ss = SyntaxSet::load_defaults_newlines();
let ts = ThemeSet::load_defaults();
let theme = &ts.themes["base16-ocean.dark"];
let start_parsing = Instant::now();
let mut event_count = 0;
for event in parser {
event_count += 1;
if start_parsing.elapsed().as_secs() > PARSING_TIMEOUT_SECS {
let error_msg = "Parsing timeout - file too large";
add_log("error", error_msg, Some(slug), Some(&format!("Processed {} events", event_count)));
return Err(error_msg.into());
}
match &event {
Event::Start(Tag::Heading(level, _, _)) => {
in_heading = true;
heading_level = *level as usize;
heading_text.clear();
},
Event::End(Tag::Heading(_, _, _)) => {
in_heading = false;
let heading_no_emoji = strip_emojis(&heading_text);
let id = slugify(&heading_no_emoji);
let style = "color: #2d3748; margin-top: 1.5em; margin-bottom: 0.5em;";
events.push(Event::Html(CowStr::Boxed(format!("<h{lvl} id=\"{id}\" style=\"{style}\">", lvl=heading_level, id=id, style=style).into_boxed_str())));
events.push(Event::Text(CowStr::Boxed(heading_text.clone().into_boxed_str())));
events.push(Event::Html(CowStr::Boxed(format!("</h{lvl}>", lvl=heading_level).into_boxed_str())));
},
Event::Text(text) if in_heading => {
heading_text.push_str(text);
},
Event::Start(Tag::CodeBlock(kind)) => {
in_code_block = true;
code_block_content.clear();
code_block_lang = match kind {
pulldown_cmark::CodeBlockKind::Fenced(lang) => lang.to_string(),
pulldown_cmark::CodeBlockKind::Indented => String::new(),
};
},
Event::End(Tag::CodeBlock(_)) => {
in_code_block = false;
let highlighted = if !code_block_lang.is_empty() {
if let Some(syntax) = ss.find_syntax_by_token(&code_block_lang) {
highlighted_html_for_string(&code_block_content, &ss, syntax, theme).unwrap_or_else(|_| format!("<pre style=\"background: #2d2d2d; color: #f8f8f2; padding: 1em; border-radius: 6px; overflow-x: auto;\"><code style=\"background: none;\">{}</code></pre>", html_escape::encode_text(&code_block_content)))
} else {
format!("<pre style=\"background: #2d2d2d; color: #f8f8f2; padding: 1em; border-radius: 6px; overflow-x: auto;\"><code style=\"background: none;\">{}</code></pre>", html_escape::encode_text(&code_block_content))
}
} else {
format!("<pre style=\"background: #2d2d2d; color: #f8f8f2; padding: 1em; border-radius: 6px; overflow-x: auto;\"><code style=\"background: none;\">{}</code></pre>", html_escape::encode_text(&code_block_content))
};
events.push(Event::Html(CowStr::Boxed(highlighted.into_boxed_str())));
},
Event::Text(text) if in_code_block => {
code_block_content.push_str(text);
},
_ if !in_heading && !in_code_block => {
events.push(event);
},
_ => {},
}
}
add_log("info", "Markdown parsing completed", Some(slug), Some(&format!("Processed {} events", event_count)));
html::push_html(&mut html_output, events.into_iter());
let sanitized_html = AMMONIA.clean(&html_output).to_string();
let interpret_time = start.elapsed();
let compile_start = Instant::now();
let post = Post {
slug: slug.to_string(),
title: front.title,
date: front.date,
tags: front.tags.unwrap_or_default(),
summary: front.summary,
content: sanitized_html,
created_at: created_at.to_rfc3339(),
author: std::env::var("BLOG_OWNER").unwrap_or_else(|_| "Anonymous".to_string()),
};
let compile_time = compile_start.elapsed();
// Insert into cache
// If this no worky , programm fucky wucky? - Check Logs
POST_CACHE.write().unwrap().insert(slug.to_string(), post.clone());
// Update stats
let mut stats = POST_STATS.write().unwrap();
let entry = stats.entry(slug.to_string()).or_insert_with(|| PostStats {
slug: slug.to_string(),
..Default::default()
});
entry.last_interpret_time_ms = interpret_time.as_millis();
entry.last_compile_time_ms = compile_time.as_millis();
sys.refresh_process(pid);
entry.last_cpu_usage_percent = sys.process(pid).map(|p| p.cpu_usage()).unwrap_or(0.0) - before_cpu;
add_log("info", "Post parsing completed successfully", Some(slug), Some(&format!("Interpret: {}ms, Compile: {}ms", interpret_time.as_millis(), compile_time.as_millis())));
Ok(post)
}
pub fn get_all_posts() -> Result<Vec<Post>, Box<dyn std::error::Error>> {
// Try cache first
if let Some(posts) = ALL_POSTS_CACHE.read().unwrap().clone() {
return Ok(posts);
}
let posts_dir = get_posts_directory();
let markdown_files = find_markdown_files(&posts_dir)?;
let mut posts = Vec::new();
for file_path in markdown_files {
let slug = path_to_slug(&file_path, &posts_dir);
if let Ok(post) = get_post_by_slug(&slug) {
POST_CACHE.write().unwrap().insert(slug.clone(), post.clone());
posts.push(post);
}
}
posts.sort_by(|a, b| b.created_at.cmp(&a.created_at));
*ALL_POSTS_CACHE.write().unwrap() = Some(posts.clone());
Ok(posts)
}
pub fn get_posts_by_tag(tag: &str) -> Result<Vec<Post>, Box<dyn std::error::Error>> {
let all_posts = get_all_posts()?;
Ok(all_posts.into_iter().filter(|p| p.tags.contains(&tag.to_string())).collect())
}
pub fn watch_posts<F: Fn() + Send + 'static>(on_change: F) -> notify::Result<RecommendedWatcher> {
let (tx, rx) = channel();
let mut watcher = RecommendedWatcher::new(tx, Config::default())?;
watcher.watch(get_posts_directory().as_path(), RecursiveMode::Recursive)?;
std::thread::spawn(move || {
loop {
match rx.recv() {
Ok(_event) => {
POST_CACHE.write().unwrap().clear();
*ALL_POSTS_CACHE.write().unwrap() = None;
on_change();
},
Err(e) => {
eprintln!("watch error: {:?}", e);
break;
}
}
}
});
Ok(watcher)
}
pub fn load_post_cache_from_disk() {
if let Ok(data) = fs::read_to_string(POSTS_CACHE_PATH) {
if let Ok(map) = serde_json::from_str::<HashMap<String, Post>>(&data) {
*POST_CACHE.write().unwrap() = map;
}
}
if let Ok(data) = fs::read_to_string(POST_STATS_PATH) {
if let Ok(map) = serde_json::from_str::<HashMap<String, PostStats>>(&data) {
*POST_STATS.write().unwrap() = map;
}
}
}
pub fn save_post_cache_to_disk() {
ensure_cache_directory();
if let Ok(map) = serde_json::to_string(&*POST_CACHE.read().unwrap()) {
let _ = fs::write(POSTS_CACHE_PATH, map);
}
if let Ok(map) = serde_json::to_string(&*POST_STATS.read().unwrap()) {
let _ = fs::write(POST_STATS_PATH, map);
}
}
pub fn checkhealth() -> HealthReport {
let mut errors = Vec::new();
let posts_dir = get_posts_directory();
let posts_dir_exists = posts_dir.exists() && posts_dir.is_dir();
let mut posts_count = 0;
if posts_dir_exists {
match std::fs::read_dir(&posts_dir) {
Ok(entries) => {
posts_count = entries.filter_map(|e| e.ok())
.filter(|e| e.path().extension().map(|ext| ext == "md").unwrap_or(false))
.count();
},
Err(e) => errors.push(format!("Failed to read posts dir: {}", e)),
}
} else {
errors.push("Posts directory does not exist".to_string());
}
let cache_file_exists = Path::new(POSTS_CACHE_PATH).exists();
let cache_stats_file_exists = Path::new(POST_STATS_PATH).exists();
let (mut cache_readable, mut cache_post_count) = (false, None);
if cache_file_exists {
match std::fs::read_to_string(POSTS_CACHE_PATH) {
Ok(data) => {
match serde_json::from_str::<HashMap<String, Post>>(&data) {
Ok(map) => {
cache_readable = true;
cache_post_count = Some(map.len());
},
Err(e) => errors.push(format!("Cache file not valid JSON: {}", e)),
}
},
Err(e) => errors.push(format!("Failed to read cache file: {}", e)),
}
}
let (mut cache_stats_readable, mut cache_stats_count) = (false, None);
if cache_stats_file_exists {
match std::fs::read_to_string(POST_STATS_PATH) {
Ok(data) => {
match serde_json::from_str::<HashMap<String, PostStats>>(&data) {
Ok(map) => {
cache_stats_readable = true;
cache_stats_count = Some(map.len());
},
Err(e) => errors.push(format!("Cache stats file not valid JSON: {}", e)),
}
},
Err(e) => errors.push(format!("Failed to read cache stats file: {}", e)),
}
}
HealthReport {
posts_dir_exists,
posts_count,
cache_file_exists,
cache_stats_file_exists,
cache_readable,
cache_stats_readable,
cache_post_count,
cache_stats_count,
errors,
}
}
pub fn get_parser_logs() -> Vec<LogEntry> {
// Always reload from disk to ensure up-to-date logs
load_parser_logs_from_disk();
let logs = PARSER_LOGS.read().unwrap();
logs.iter().cloned().collect()
}
pub fn clear_parser_logs() {
PARSER_LOGS.write().unwrap().clear();
if let Err(e) = save_parser_logs_to_disk_inner(&VecDeque::new()) {
eprintln!("Failed to save empty logs to disk: {}", e);
}
}
// Force reinterpret all posts by clearing cache and re-parsing
pub fn force_reinterpret_all_posts() -> Result<Vec<Post>, Box<dyn std::error::Error>> {
add_log("info", "Starting force reinterpret of all posts", None, None);
// Clear all caches
POST_CACHE.write().unwrap().clear();
ALL_POSTS_CACHE.write().unwrap().take();
POST_STATS.write().unwrap().clear();
add_log("info", "Cleared all caches", None, None);
// Get posts directory and find all markdown files
let posts_dir = get_posts_directory();
let markdown_files = find_markdown_files(&posts_dir)?;
add_log("info", &format!("Found {} markdown files to reinterpret", markdown_files.len()), None, None);
let mut posts = Vec::new();
let mut success_count = 0;
let mut error_count = 0;
for file_path in markdown_files {
let slug = path_to_slug(&file_path, &posts_dir);
match get_post_by_slug(&slug) {
Ok(post) => {
posts.push(post);
success_count += 1;
add_log("info", &format!("Successfully reinterpreted: {}", slug), Some(&slug), None);
}
Err(e) => {
error_count += 1;
add_log("error", &format!("Failed to reinterpret {}: {}", slug, e), Some(&slug), None);
}
}
}
// Update the all posts cache
ALL_POSTS_CACHE.write().unwrap().replace(posts.clone());
// Save cache to disk
save_post_cache_to_disk();
add_log("info", &format!("Force reinterpret completed. Success: {}, Errors: {}", success_count, error_count), None, None);
Ok(posts)
}
// Force reparse a single post by clearing its cache and re-parsing
pub fn force_reparse_single_post(slug: &str) -> Result<Post, Box<dyn std::error::Error>> {
add_log("info", &format!("Starting force reparse of post: {}", slug), Some(slug), None);
// Clear this specific post from all caches
POST_CACHE.write().unwrap().remove(slug);
POST_STATS.write().unwrap().remove(slug);
// Clear the all posts cache since it might contain this post
ALL_POSTS_CACHE.write().unwrap().take();
add_log("info", &format!("Cleared cache for post: {}", slug), Some(slug), None);
// Re-parse the post
let post = get_post_by_slug(slug)?;
// Update the all posts cache with the new post
let mut all_posts_cache = ALL_POSTS_CACHE.write().unwrap();
if let Some(ref mut posts) = *all_posts_cache {
// Remove old version if it exists
posts.retain(|p| p.slug != slug);
// Add new version
posts.push(post.clone());
// Sort by creation date
posts.sort_by(|a, b| b.created_at.cmp(&a.created_at));
}
// Save cache to disk
save_post_cache_to_disk();
add_log("info", &format!("Successfully reparsed post: {}", slug), Some(slug), None);
Ok(post)
}