summaryrefslogtreecommitdiff
path: root/_prototype/converter
diff options
context:
space:
mode:
Diffstat (limited to '_prototype/converter')
-rw-r--r--_prototype/converter/Cargo.toml11
-rw-r--r--_prototype/converter/src/bin/mdialogc.rs95
-rw-r--r--_prototype/converter/src/error.rs121
-rw-r--r--_prototype/converter/src/lib.rs5
-rw-r--r--_prototype/converter/src/macros.rs33
-rw-r--r--_prototype/converter/src/parse.rs1303
-rw-r--r--_prototype/converter/src/syntax_checker.rs221
-rw-r--r--_prototype/converter/src/utils.rs1
-rw-r--r--_prototype/converter/src/utils/path_fmt.rs123
-rw-r--r--_prototype/converter/usage.txt6
-rw-r--r--_prototype/converter/version.txt2
11 files changed, 1921 insertions, 0 deletions
diff --git a/_prototype/converter/Cargo.toml b/_prototype/converter/Cargo.toml
new file mode 100644
index 0000000..ce44af9
--- /dev/null
+++ b/_prototype/converter/Cargo.toml
@@ -0,0 +1,11 @@
+[package]
+name = "markdialog_converter"
+edition = "2024"
+version.workspace = true
+
+[dependencies]
+colored = "3.0"
+strip-ansi-escapes = "0.2.1"
+unicode-width = "0.2"
+regex = "1.12"
+sha2 = "0.10"
diff --git a/_prototype/converter/src/bin/mdialogc.rs b/_prototype/converter/src/bin/mdialogc.rs
new file mode 100644
index 0000000..ebe7804
--- /dev/null
+++ b/_prototype/converter/src/bin/mdialogc.rs
@@ -0,0 +1,95 @@
+use markdialog_converter::{
+ error::{Exit, handle_exit},
+ parse::parse,
+ special_argument, special_flag,
+};
+use std::{path::PathBuf, str::FromStr};
+
+fn process() -> Result<(), Exit> {
+ let mut args: Vec<String> = std::env::args().skip(1).collect();
+
+ let help = special_flag!(args, "--help") || special_flag!(args, "-h");
+ let version = special_flag!(args, "--version") || special_flag!(args, "-v");
+
+ if version {
+ let version = include_str!("../../version.txt");
+ println!("{}", version.trim());
+ return Err(Exit::Code(0));
+ }
+
+ if help || args.len() < 1 {
+ let usage = include_str!("../../usage.txt");
+ println!("{}", usage.trim());
+ return Err(Exit::Code(0));
+ }
+
+ let input_file = get_input_file(&mut args)?;
+ let output_ir_file = get_output_ir_file(&mut args)?.unwrap_or_else(|| {
+ let mut path = input_file.clone();
+ if let Some(file_name) = path.file_name() {
+ let mut new_name = std::ffi::OsString::new();
+ new_name.push(file_name);
+ // Change extension to .dialog
+ path.set_extension("dialog");
+ } else {
+ path.set_file_name("ir.dialog");
+ }
+ path
+ });
+
+ parse(input_file, output_ir_file)?;
+
+ Ok(())
+}
+
+fn get_input_file(args: &mut Vec<String>) -> Result<PathBuf, Exit> {
+ let input = match special_argument!(args, "--input") {
+ Some(i) => i,
+ None => match special_argument!(args, "-i") {
+ Some(i) => i,
+ None => {
+ eprintln!("Missing required input argument. Use --input or -i.");
+ std::process::exit(2);
+ }
+ },
+ };
+
+ let input_file = PathBuf::from_str(&input).map_err(|_| {
+ eprintln!("Invalid file path `{}`!", input);
+ Exit::Code(2)
+ })?;
+
+ Ok(input_file)
+}
+
+fn get_output_ir_file(args: &mut Vec<String>) -> Result<Option<PathBuf>, Exit> {
+ let input = match special_argument!(args, "--output") {
+ Some(i) => Some(i),
+ None => match special_argument!(args, "-o") {
+ Some(i) => Some(i),
+ None => None,
+ },
+ };
+
+ match input {
+ Some(i) => {
+ let input_file = PathBuf::from_str(&i).map_err(|_| {
+ eprintln!("Invalid file path `{}`!", i);
+ return Exit::Code(2);
+ })?;
+ Ok(Some(input_file))
+ }
+ None => Ok(None),
+ }
+}
+
+fn main() {
+ // Init colored
+ #[cfg(windows)]
+ colored::control::set_virtual_terminal(true).unwrap();
+
+ match process() {
+ Ok(_) => {}
+ Err(e) => handle_exit(e),
+ }
+}
diff --git a/_prototype/converter/src/error.rs b/_prototype/converter/src/error.rs
new file mode 100644
index 0000000..b594165
--- /dev/null
+++ b/_prototype/converter/src/error.rs
@@ -0,0 +1,121 @@
+use std::{i64, path::PathBuf, process::exit};
+
+use colored::Colorize;
+use unicode_width::UnicodeWidthStr;
+
+#[derive(Debug)]
+pub enum Exit {
+ Code(i32),
+ IoError(std::io::Error),
+ FileNotFound(PathBuf),
+ SyntaxError {
+ content: String,
+ reason: String,
+ line: i64,
+ begin: i64,
+ end: i64,
+ },
+ DuplicateMarker(String),
+ CycleDependency(PathBuf),
+}
+
+impl From<std::io::Error> for Exit {
+ fn from(error: std::io::Error) -> Self {
+ Exit::IoError(error)
+ }
+}
+
+pub fn handle_exit(e: Exit) {
+ match e {
+ Exit::Code(code) => exit(code),
+ Exit::IoError(error) => print_parse_error(error.to_string()),
+ Exit::FileNotFound(path_buf) => {
+ eprintln!("File `{}` not found!", path_buf.display());
+ exit(1)
+ }
+ Exit::SyntaxError {
+ content,
+ reason,
+ line,
+ begin,
+ end,
+ } => {
+ print_syntax_error(content, reason, line, begin, end);
+ }
+ Exit::DuplicateMarker(marker) => {
+ eprintln!("Duplicate marker `{}` found!", marker);
+ exit(1)
+ }
+ Exit::CycleDependency(dialog) => {
+ eprintln!("Dialog `{}` depends on itself!", dialog.display());
+ exit(1)
+ }
+ }
+}
+
+fn print_parse_error(content: impl AsRef<str>) {
+ eprintln!("Parse Error !");
+ eprintln!("{}", content.as_ref().trim());
+ exit(1);
+}
+
+macro_rules! line {
+ ($line:expr, $N:expr) => {
+ if $line + $N <= 0 {
+ " ".to_string()
+ } else {
+ ($line + $N).to_string()
+ }
+ };
+}
+
+pub fn print_syntax_error(content: String, reason: String, line: i64, begin: i64, end: i64) {
+ let content_len = content.width() as i64;
+ let end = end.clamp(begin, content_len);
+
+ eprintln!("{}", "Parse Failed: Syntax Error".bright_yellow());
+ eprintln!("{}{}", line!(line, -1), "|");
+
+ let before: String = content.chars().take(begin.max(0) as usize).collect();
+ let highlight_len = (end - begin).max(1) as usize;
+ let highlight: String = content
+ .chars()
+ .skip(begin.max(0) as usize)
+ .take(highlight_len)
+ .collect();
+ let after: String = content
+ .chars()
+ .skip((begin.max(0) + highlight_len as i64) as usize)
+ .collect();
+
+ eprintln!(
+ "{}{} {}{}{}",
+ line.to_string().cyan(),
+ "|".cyan(),
+ before.cyan(),
+ highlight.bright_cyan(),
+ after.cyan()
+ );
+
+ let prefix_chars: String = content.chars().take(begin.max(0) as usize).collect();
+ let prefix_width = prefix_chars.width() as usize;
+
+ eprintln!(
+ "{}{} {}",
+ line!(line, 1),
+ "|",
+ format!(
+ "{}{}____ {}",
+ " ".repeat(prefix_width),
+ "^".repeat(((end - begin).max(1)) as usize),
+ reason
+ )
+ .bright_cyan()
+ );
+ eprintln!("{}{}", line!(line, 2), "|");
+ eprintln!(
+ "{}",
+ "Please fix the issue and run the program again".bright_yellow()
+ );
+ exit(1);
+}
diff --git a/_prototype/converter/src/lib.rs b/_prototype/converter/src/lib.rs
new file mode 100644
index 0000000..d7caac3
--- /dev/null
+++ b/_prototype/converter/src/lib.rs
@@ -0,0 +1,5 @@
+pub mod error;
+pub mod macros;
+pub mod parse;
+pub mod syntax_checker;
+pub mod utils;
diff --git a/_prototype/converter/src/macros.rs b/_prototype/converter/src/macros.rs
new file mode 100644
index 0000000..894b3f4
--- /dev/null
+++ b/_prototype/converter/src/macros.rs
@@ -0,0 +1,33 @@
+#[macro_export]
+macro_rules! special_flag {
+ ($args:expr, $flag:expr) => {{
+ let flag = $flag;
+ let found = $args.iter().any(|arg| arg == flag);
+ $args.retain(|arg| arg != flag);
+ found
+ }};
+}
+
+#[macro_export]
+macro_rules! special_argument {
+ ($args:expr, $flag:expr) => {{
+ let flag = $flag;
+ let mut value: Option<String> = None;
+ let mut i = 0;
+ while i < $args.len() {
+ if $args[i] == flag {
+ if i + 1 < $args.len() {
+ value = Some($args[i + 1].clone());
+ $args.remove(i + 1);
+ $args.remove(i);
+ } else {
+ value = None;
+ $args.remove(i);
+ }
+ break;
+ }
+ i += 1;
+ }
+ value
+ }};
+}
diff --git a/_prototype/converter/src/parse.rs b/_prototype/converter/src/parse.rs
new file mode 100644
index 0000000..292a761
--- /dev/null
+++ b/_prototype/converter/src/parse.rs
@@ -0,0 +1,1303 @@
+use std::path::PathBuf;
+
+use crate::{
+ error::Exit,
+ syntax_checker::{check_duplicate_marker, check_markdown_syntax},
+};
+
+pub fn parse(input: PathBuf, ir_output: PathBuf) -> Result<(), Exit> {
+ let result = std::fs::read_to_string(&input)?;
+
+ check_markdown_syntax(&result)?;
+
+ let result = unwrap_includes::proc(result, input)?;
+
+ check_duplicate_marker(&result)?;
+
+ let result = markdown_cleanup::proc(result)?;
+ let result = markdown_jump_fix::proc(result)?;
+ let result = markdown_marker_rename::proc(result)?;
+ let result = markdown_struct_build::proc(result)?;
+ let result = markdown_strip_invalid_jump::proc(result)?;
+ let result = markdown_convert_image::proc(result)?;
+ let result = markdown_apply_codes::proc(result)?;
+ let result = markdown_split_and_encode::proc(result)?;
+
+ std::fs::write(&ir_output, result)?;
+ Ok(())
+}
+
+pub mod unwrap_includes {
+ use crate::{error::Exit, utils::path_fmt::format_path};
+ use regex::Regex;
+ use std::collections::HashMap;
+ use std::path::{Path, PathBuf};
+
+ /// Expand text includes of [[markdown]] (searches for markdown.md) and image paths
+ pub fn proc(input: String, self_path: PathBuf) -> Result<String, Exit> {
+ let mut stack = Vec::<PathBuf>::new();
+ let mut cache = HashMap::<String, PathBuf>::new();
+ let mut img_cache = HashMap::<String, PathBuf>::new();
+ let root_path = self_path.clone();
+ expand_recursive(
+ input,
+ &self_path,
+ &root_path,
+ &mut stack,
+ &mut cache,
+ &mut img_cache,
+ )
+ }
+
+ fn expand_recursive(
+ content: String,
+ current_path: &Path,
+ root_path: &Path,
+ stack: &mut Vec<PathBuf>,
+ cache: &mut HashMap<String, PathBuf>,
+ img_cache: &mut HashMap<String, PathBuf>,
+ ) -> Result<String, Exit> {
+ let mut output = String::new();
+ let mut in_code_block = false;
+ let obsidian_image_re = Regex::new(r"!\[\[([^\]]+)\]\]").unwrap();
+
+ let current_norm = format_path(current_path)?;
+
+ if stack.contains(&current_norm) {
+ return Err(Exit::CycleDependency(current_norm));
+ }
+
+ stack.push(current_norm.clone());
+
+ for line in content.lines() {
+ if line.trim().starts_with("```") {
+ in_code_block = !in_code_block;
+ output.push_str(line);
+ output.push('\n');
+ continue;
+ }
+
+ if in_code_block {
+ output.push_str(line);
+ output.push('\n');
+ continue;
+ }
+
+ if let Some(include_name) = extract_include(line) {
+ let include_abs = if let Some(cached_path) = cache.get(include_name) {
+ cached_path.clone()
+ } else {
+ let base_dir = current_path.parent().unwrap();
+ let mut queue = vec![base_dir.to_path_buf()];
+ let mut visited = std::collections::HashSet::new();
+ let mut found_path = None;
+
+ while let Some(dir) = queue.pop() {
+ if visited.contains(&dir) {
+ continue;
+ }
+ visited.insert(dir.clone());
+
+ let candidate1 = dir.join(include_name);
+ if candidate1.exists()
+ && candidate1.extension().map_or(false, |ext| ext == "md")
+ {
+ found_path = Some(format_path(&candidate1)?);
+ break;
+ }
+
+ // Try add .md extension
+ let candidate2 = dir.join(format!("{}.md", include_name));
+ if candidate2.exists() {
+ found_path = Some(format_path(&candidate2)?);
+ break;
+ }
+
+ if let Some(parent) = dir.parent() {
+ queue.push(parent.to_path_buf());
+ }
+
+ if let Ok(entries) = std::fs::read_dir(&dir) {
+ for entry in entries.flatten() {
+ if let Ok(file_type) = entry.file_type() {
+ if file_type.is_dir() {
+ queue.push(entry.path());
+ }
+ }
+ }
+ }
+ }
+
+ match found_path {
+ Some(path) => {
+ cache.insert(include_name.to_string(), path.clone());
+ path
+ }
+ None => {
+ return Err(Exit::FileNotFound(
+ format!(
+ "{} or {}.md (searched from {:?})",
+ include_name, include_name, base_dir
+ )
+ .into(),
+ ));
+ }
+ }
+ };
+
+ let include_content = std::fs::read_to_string(&include_abs).map_err(|e| {
+ if e.kind() == std::io::ErrorKind::NotFound {
+ Exit::FileNotFound(include_abs.clone())
+ } else {
+ Exit::IoError(e)
+ }
+ })?;
+
+ let expanded = expand_recursive(
+ include_content,
+ &include_abs,
+ root_path,
+ stack,
+ cache,
+ img_cache,
+ )?;
+ output.push_str(&expanded);
+ } else {
+ // Process image links
+ let mut processed_line = line.to_string();
+
+ // First process Obsidian image syntax ![[...]]
+ let mut obsidian_matches: Vec<(usize, usize, String)> = Vec::new();
+
+ for caps in obsidian_image_re.captures_iter(line) {
+ let full_match = caps.get(0).unwrap();
+ let image_ref = caps.get(1).unwrap().as_str();
+
+ // Try to get from img_cache first
+ let image_abs = if let Some(cached_path) = img_cache.get(image_ref) {
+ cached_path.clone()
+ } else {
+ // Start search from root directory for Obsidian syntax
+ let root_dir = root_path.parent().unwrap();
+ let mut queue = vec![root_dir.to_path_buf()];
+ let mut visited = std::collections::HashSet::new();
+ let mut found_image_path = None;
+
+ while let Some(dir) = queue.pop() {
+ if visited.contains(&dir) {
+ continue;
+ }
+ visited.insert(dir.clone());
+
+ // Check if file exists in this directory
+ let candidate = dir.join(image_ref);
+ if candidate.exists() {
+ found_image_path = Some(format_path(&candidate)?);
+ break;
+ }
+
+ // Add parent directory to queue (breadth-first upward)
+ if let Some(parent) = dir.parent() {
+ queue.push(parent.to_path_buf());
+ }
+
+ // Add subdirectories to queue (breadth-first downward)
+ if let Ok(entries) = std::fs::read_dir(&dir) {
+ for entry in entries.flatten() {
+ if let Ok(file_type) = entry.file_type() {
+ if file_type.is_dir() {
+ queue.push(entry.path());
+ }
+ }
+ }
+ }
+ }
+
+ match found_image_path {
+ Some(path) => {
+ img_cache.insert(image_ref.to_string(), path.clone());
+ path
+ }
+ None => {
+ // If image not found, keep the original syntax
+ continue;
+ }
+ }
+ };
+
+ // Store the replacement information
+ let start = full_match.start();
+ let end = full_match.end();
+ let new_image_markdown = format!("![]({})", image_abs.display());
+ obsidian_matches.push((start, end, new_image_markdown));
+ }
+
+ // Apply Obsidian replacements in reverse order to maintain correct indices
+ for (start, end, replacement) in obsidian_matches.into_iter().rev() {
+ processed_line.replace_range(start..end, &replacement);
+ }
+
+ output.push_str(&processed_line);
+ output.push('\n');
+ }
+ }
+
+ stack.pop();
+
+ Ok(output)
+ }
+
+ fn extract_include(line: &str) -> Option<&str> {
+ line.trim()
+ .strip_prefix("[[")
+ .and_then(|s| s.strip_suffix("]]"))
+ }
+}
+
+pub mod markdown_cleanup {
+ use crate::error::Exit;
+
+ /// Clean Markdown
+ /// 1. Remove blockquotes
+ /// 2. Remove empty lines
+ /// 3. Trim each line
+ pub fn proc(i: String) -> Result<String, Exit> {
+ let lines = i.lines();
+ let mut cleaned = Vec::new();
+
+ for line in lines {
+ if line.starts_with('>') {
+ continue;
+ }
+ let trimmed = line.trim();
+ if trimmed.is_empty() {
+ continue;
+ }
+ cleaned.push(trimmed.to_string());
+ }
+
+ Ok(cleaned.join("\n"))
+ }
+
+ #[cfg(test)]
+ mod test_clean_markdown {
+ use super::*;
+
+ #[test]
+ fn test_clean_markdown_removes_blockquotes() {
+ let input = "> This is a blockquote\nNormal text\n> Another blockquote".to_string();
+ let expected = "Normal text".to_string();
+ let Ok(result) = proc(input) else {
+ panic!("Parse error!");
+ };
+ assert_eq!(result, expected);
+ }
+
+ #[test]
+ fn test_clean_markdown_removes_empty_lines() {
+ let input = "Line 1\n\n\nLine 2\n\n".to_string();
+ let expected = "Line 1\nLine 2".to_string();
+ let Ok(result) = proc(input) else {
+ panic!("Parse error!");
+ };
+ assert_eq!(result, expected);
+ }
+
+ #[test]
+ fn test_clean_markdown_trims_lines() {
+ let input = " Line 1 \n\tLine 2\t\n".to_string();
+ let expected = "Line 1\nLine 2".to_string();
+ let Ok(result) = proc(input) else {
+ panic!("Parse error!");
+ };
+ assert_eq!(result, expected);
+ }
+
+ #[test]
+ fn test_clean_markdown_combined() {
+ let input = "> Blockquote\n\n Line 1 \n> Another\n\nLine 2\n\n".to_string();
+ let expected = "Line 1\nLine 2".to_string();
+ let Ok(result) = proc(input) else {
+ panic!("Parse error!");
+ };
+ assert_eq!(result, expected);
+ }
+
+ #[test]
+ fn test_clean_markdown_empty_input() {
+ let input = "".to_string();
+ let expected = "".to_string();
+ let Ok(result) = proc(input) else {
+ panic!("Parse error!");
+ };
+ assert_eq!(result, expected);
+ }
+
+ #[test]
+ fn test_clean_markdown_only_blockquotes() {
+ let input = "> Quote 1\n> Quote 2".to_string();
+ let expected = "".to_string();
+ let Ok(result) = proc(input) else {
+ panic!("Parse error!");
+ };
+ assert_eq!(result, expected);
+ }
+
+ #[test]
+ fn test_clean_markdown_only_whitespace() {
+ let input = " \n\t\n ".to_string();
+ let expected = "".to_string();
+ let Ok(result) = proc(input) else {
+ panic!("Parse error!");
+ };
+ assert_eq!(result, expected);
+ }
+ }
+}
+
+pub mod markdown_jump_fix {
+ use crate::error::Exit;
+
+ /// Fix jump syntax in each line
+ /// 1. Correct the following syntax
+ /// ```ignore
+ /// - It's [Item](#Mark)
+ /// > corrected to
+ /// - It's Item [](#Mark)
+ /// ```
+ ///
+ /// 2. If there are multiple options, take the first one
+ /// ```ignore
+ /// - There might be two options: [A](#A) and [B](#B)!
+ /// > corrected to
+ /// - There might be two options: A and B! [](#A)
+ /// ```
+ pub fn proc(i: String) -> Result<String, Exit> {
+ let mut result = String::new();
+
+ for line in i.lines() {
+ let (processed_content, first_link_dest) = process_line_content(line);
+ let processed_line = format_line_with_link(processed_content, first_link_dest);
+ let final_line = convert_ordered_list_marker(processed_line);
+
+ result.push_str(&final_line);
+ result.push('\n');
+ }
+
+ if result.ends_with('\n') {
+ result.pop();
+ }
+
+ Ok(result)
+ }
+
+ pub fn process_line_content(line: &str) -> (String, Option<String>) {
+ // Check if line is an image line (starts with "![")
+ if line.starts_with("![") {
+ // Return the original line unchanged with no link destination
+ return (line.to_string(), None);
+ }
+
+ let mut processed = String::new();
+ let mut chars = line.chars().peekable();
+ let mut first_link_dest = None;
+ let mut has_link = false;
+
+ while let Some(ch) = chars.next() {
+ if ch == '[' {
+ if let Some((link_text, link_dest, remaining_chars)) = helper_parse_link(&mut chars)
+ {
+ processed.push_str(&link_text);
+ if !has_link {
+ first_link_dest = Some(link_dest);
+ has_link = true;
+ }
+ chars = remaining_chars;
+ continue;
+ } else {
+ // Invalid
+ processed.push(ch);
+ }
+ } else {
+ processed.push(ch);
+ }
+ }
+
+ (processed, first_link_dest)
+ }
+
+ pub fn helper_parse_link<'a>(
+ chars: &mut std::iter::Peekable<std::str::Chars<'a>>,
+ ) -> Option<(String, String, std::iter::Peekable<std::str::Chars<'a>>)> {
+ let mut link_text = String::new();
+
+ while let Some(&ch) = chars.peek() {
+ chars.next();
+ if ch == ']' {
+ break;
+ }
+ link_text.push(ch);
+ }
+
+ if chars.next() != Some('(') || chars.next() != Some('#') {
+ return None;
+ }
+
+ let mut link_dest = String::new();
+ while let Some(ch) = chars.next() {
+ if ch == ')' {
+ break;
+ }
+ link_dest.push(ch);
+ }
+
+ let cleaned_dest = link_dest.trim().replace(' ', "").replace('#', "");
+
+ Some((link_text, cleaned_dest, chars.clone()))
+ }
+
+ pub fn format_line_with_link(content: String, link_dest: Option<String>) -> String {
+ match link_dest {
+ Some(dest) if !dest.trim().is_empty() => {
+ format!("{} [](#{})", content.trim_end(), dest.trim())
+ .trim()
+ .to_string()
+ }
+ _ => content,
+ }
+ }
+
+ pub fn convert_ordered_list_marker(line: String) -> String {
+ let trimmed = line.trim_start();
+
+ if let Some(_rest) = trimmed.strip_prefix(|c: char| c.is_ascii_digit()) {
+ let mut chars = trimmed.chars();
+ let mut digit_count = 0;
+
+ while let Some(c) = chars.next() {
+ if c.is_ascii_digit() {
+ digit_count += 1;
+ } else {
+ break;
+ }
+ }
+
+ if digit_count > 0 {
+ let rest_after_digits = &trimmed[digit_count..];
+ if let Some(content) = rest_after_digits.strip_prefix(". ") {
+ return format!("- {}", content);
+ }
+ }
+ }
+
+ line
+ }
+
+ #[cfg(test)]
+ mod test_fix_mark_jump {
+ use super::*;
+
+ #[test]
+ fn test_fix_mark_jump_single_link() {
+ let input = "- It's [Item](#Mark)".to_string();
+ let expected = "- It's Item [](#Mark)".to_string();
+ let Ok(result) = proc(input) else {
+ panic!("Parse error!");
+ };
+ assert_eq!(result, expected);
+ }
+
+ #[test]
+ fn test_fix_mark_jump_multiple_links_takes_first() {
+ let input = "- There might be two options: [A](#A) and [B](#B)!".to_string();
+ let expected = "- There might be two options: A and B! [](#A)".to_string();
+ let Ok(result) = proc(input) else {
+ panic!("Parse error!");
+ };
+ assert_eq!(result, expected);
+ }
+
+ #[test]
+ fn test_fix_mark_jump_no_link() {
+ let input = "- Just a normal line".to_string();
+ let expected = "- Just a normal line".to_string();
+ let Ok(result) = proc(input) else {
+ panic!("Parse error!");
+ };
+ assert_eq!(result, expected);
+ }
+
+ #[test]
+ fn test_fix_mark_jump_empty_line() {
+ let input = "".to_string();
+ let expected = "".to_string();
+ let Ok(result) = proc(input) else {
+ panic!("Parse error!");
+ };
+ assert_eq!(result, expected);
+ }
+
+ #[test]
+ fn test_fix_mark_jump_multiple_lines() {
+ let input = "- First [Item](#First)\n- Second [Item](#Second)".to_string();
+ let expected = "- First Item [](#First)\n- Second Item [](#Second)".to_string();
+ let Ok(result) = proc(input) else {
+ panic!("Parse error!");
+ };
+ assert_eq!(result, expected);
+ }
+
+ #[test]
+ fn test_fix_mark_jump_link_at_end() {
+ let input = "- End with [link](#target)".to_string();
+ let expected = "- End with link [](#target)".to_string();
+ let Ok(result) = proc(input) else {
+ panic!("Parse error!");
+ };
+ assert_eq!(result, expected);
+ }
+
+ #[test]
+ fn test_fix_mark_jump_link_at_beginning() {
+ let input = "- [Start](#target) with link".to_string();
+ let expected = "- Start with link [](#target)".to_string();
+ let Ok(result) = proc(input) else {
+ panic!("Parse error!");
+ };
+ assert_eq!(result, expected);
+ }
+
+ #[test]
+ fn test_fix_mark_jump_link_in_middle() {
+ let input = "- Text [middle](#target) text".to_string();
+ let expected = "- Text middle text [](#target)".to_string();
+ let Ok(result) = proc(input) else {
+ panic!("Parse error!");
+ };
+ assert_eq!(result, expected);
+ }
+
+ #[test]
+ fn test_fix_mark_jump_ordered_list_conversion() {
+ let input = "1. [Item](#target)".to_string();
+ let expected = "- Item [](#target)".to_string();
+ let Ok(result) = proc(input) else {
+ panic!("Parse error!");
+ };
+ assert_eq!(result, expected);
+ }
+
+ #[test]
+ fn test_fix_mark_jump_ordered_list_multiple_digits() {
+ let input = "10. [Tenth](#target) item".to_string();
+ let expected = "- Tenth item [](#target)".to_string();
+ let Ok(result) = proc(input) else {
+ panic!("Parse error!");
+ };
+ assert_eq!(result, expected);
+ }
+
+ #[test]
+ fn test_fix_mark_jump_mixed_ordered_and_unordered() {
+ let input = "1. [First](#first)\n- [Second](#second)\n2. [Third](#third)".to_string();
+ let expected =
+ "- First [](#first)\n- Second [](#second)\n- Third [](#third)".to_string();
+ let Ok(result) = proc(input) else {
+ panic!("Parse error!");
+ };
+ assert_eq!(result, expected);
+ }
+
+ #[test]
+ fn test_fix_mark_jump_invalid_link_format() {
+ let input = "- Invalid [link format".to_string();
+ let expected = "- Invalid [".to_string();
+ let Ok(result) = proc(input) else {
+ panic!("Parse error!");
+ };
+ assert_eq!(result, expected);
+ }
+
+ #[test]
+ fn test_fix_mark_jump_link_with_spaces_in_target() {
+ let input = "- Link [text](# target#)".to_string();
+ let expected = "- Link text [](#target)".to_string();
+ let Ok(result) = proc(input) else {
+ panic!("Parse error!");
+ };
+ assert_eq!(result, expected);
+ }
+
+ #[test]
+ fn test_fix_mark_jump_empty_link_text() {
+ let input = "- [](#target)".to_string();
+ let expected = "- [](#target)".to_string();
+ let Ok(result) = proc(input) else {
+ panic!("Parse error!");
+ };
+ assert_eq!(result, expected);
+ }
+
+ #[test]
+ fn test_fix_mark_jump_only_whitespace() {
+ let input = " ".to_string();
+ let expected = " ".to_string();
+ let Ok(result) = proc(input) else {
+ panic!("Parse error!");
+ };
+ assert_eq!(result, expected);
+ }
+
+ #[test]
+ fn test_fix_mark_jump_complex_multiple_links() {
+ let input = "- Choose [A](#A), [B](#B), or [C](#C)!".to_string();
+ let expected = "- Choose A, B, or C! [](#A)".to_string();
+ let Ok(result) = proc(input) else {
+ panic!("Parse error!");
+ };
+ assert_eq!(result, expected);
+ }
+ }
+}
+
+pub mod markdown_marker_rename {
+ use regex::Regex;
+ use sha2::{Digest, Sha256};
+
+ use crate::error::Exit;
+
+ /// Replace marker names: replace heading text and link anchors with corresponding SHA256
+ ///
+ /// Example:
+ /// ```ignore
+ /// # Original text
+ /// # Chapter Title
+ /// - Jump to [Chapter Title](#Chapter Title)
+ ///
+ /// # After processing
+ /// # a1b2c3d4
+ /// - Jump to [](#a1b2c3d4)
+ /// ```
+ pub fn proc(i: String) -> Result<String, Exit> {
+ let mut result = i;
+
+ let heading_re = Regex::new(r"^(#{1,5})\s+(.+)$").unwrap();
+ let mut heading_map = std::collections::HashMap::new();
+
+ for line in result.lines() {
+ if let Some(caps) = heading_re.captures(line) {
+ let heading_text = caps[2].trim().to_string();
+ let hash = format!("{:x}", Sha256::digest(heading_text.as_bytes()));
+ let short_hash = &hash[..8];
+ heading_map.insert(heading_text, short_hash.to_string());
+ }
+ }
+
+ let mut lines: Vec<String> = Vec::new();
+ for line in result.lines() {
+ if let Some(caps) = heading_re.captures(line) {
+ let level = &caps[1];
+ let heading_text = caps[2].trim();
+
+ if let Some(hash) = heading_map.get(heading_text) {
+ lines.push(format!("{} {}", level, hash));
+ } else {
+ lines.push(line.to_string());
+ }
+ } else {
+ lines.push(line.to_string());
+ }
+ }
+ result = lines.join("\n");
+
+ let link_re = Regex::new(r"\[\]\(#([^)]+)\)").unwrap();
+ result = link_re
+ .replace_all(&result, |caps: &regex::Captures| {
+ let anchor_name = &caps[1];
+ if let Some(hash) = heading_map.get(anchor_name) {
+ format!("[](#{})", hash)
+ } else {
+ let hash = format!("{:x}", Sha256::digest(anchor_name.as_bytes()));
+ let short_hash = &hash[..8];
+ format!("[](#{})", short_hash)
+ }
+ })
+ .to_string();
+
+ Ok(result)
+ }
+}
+
+pub mod markdown_struct_build {
+ use regex::Regex;
+
+ use crate::error::Exit;
+
+ /// Split content into Step + Sentence structure
+ pub fn proc(input: String) -> Result<String, Exit> {
+ let mut result = String::new();
+ let mut current_marker = String::new();
+ let mut current_step_id = 0;
+ let mut current_character = String::new();
+ let mut has_no_switch_flag = false;
+
+ let mut code_record_mode = false;
+ let mut option_record_mode = false;
+
+ let mut sentences_buffer = String::new();
+ for line in input.split("\n") {
+ // Record code
+ if code_record_mode {
+ // If code block marker is found again, end code recording
+ if line.starts_with("```") && code_record_mode {
+ sentences_buffer.push_str("\n");
+ code_record_mode = false;
+ continue;
+ }
+ sentences_buffer.push_str(format!("`{}`", line).as_str());
+ continue;
+ }
+
+ // Record options
+ if option_record_mode {
+ // Still an option, continue appending
+ if line.starts_with("- ") {
+ let (sentence, next) = get_jump_from_line(line);
+ let next = if let Some(next) = next {
+ format!("->[#{}_0]", next)
+ } else {
+ next_flag(current_marker.as_str(), current_step_id)
+ };
+ let option_line = format!(
+ "{}[{}]{}",
+ character(&current_character, has_no_switch_flag),
+ sentence,
+ next
+ );
+ sentences_buffer.push_str(option_line.as_str());
+ sentences_buffer.push('\n');
+ continue;
+ } else {
+ // When ending option recording, create and advance one Step
+ result.push_str(step_line(current_marker.as_str(), current_step_id).as_str());
+ result.push('\n');
+ result.push_str(sentences_buffer.as_str());
+ sentences_buffer.clear();
+ current_step_id += 1;
+ // Clean "Has no switch flag"
+ has_no_switch_flag = false;
+ // Close option mode
+ option_record_mode = false;
+ // Do not continue here, proceed to process subsequent content
+ }
+ }
+
+ // Refresh heading
+ if is_marker(line) {
+ current_marker = read_maker(line).to_string();
+ current_step_id = 0;
+ continue;
+ }
+
+ // Refresh character
+ if is_character(line) {
+ let (character, no_switch_flag) = read_character(line);
+ current_character = character.to_string();
+ has_no_switch_flag = no_switch_flag;
+ continue;
+ }
+
+ // Image recording
+ if line.starts_with('!') {
+ sentences_buffer.push_str(line);
+ sentences_buffer.push('\n');
+ continue;
+ }
+
+ // Start code recording
+ if line.starts_with("```") && !code_record_mode {
+ code_record_mode = true;
+ continue;
+ }
+
+ // Option recording
+ if line.starts_with("- ") {
+ let (sentence, next) = get_jump_from_line(line);
+ let next = if let Some(next) = next {
+ format!("->[#{}_0]", next)
+ } else {
+ next_flag(current_marker.as_str(), current_step_id)
+ };
+ let option_line = format!(
+ "{}[{}]{}",
+ character(&current_character, has_no_switch_flag),
+ sentence,
+ next
+ );
+ sentences_buffer.push_str(option_line.as_str());
+ sentences_buffer.push('\n');
+
+ // Start option recording mode
+ if !option_record_mode {
+ option_record_mode = true;
+ }
+ continue;
+ }
+
+ // Normal sentence
+ let (sentence, next) = get_jump_from_line(line);
+ let next = if let Some(next) = next {
+ format!("->[#{}_0]", next)
+ } else {
+ next_flag(current_marker.as_str(), current_step_id)
+ };
+ let sentence_line = format!(
+ "{}[{}]{}",
+ character(&current_character, has_no_switch_flag),
+ sentence,
+ next
+ );
+ has_no_switch_flag = false;
+
+ // Create and advance one Step
+ result.push_str(step_line(current_marker.as_str(), current_step_id).as_str());
+ result.push('\n');
+ result.push_str(sentences_buffer.as_str());
+ sentences_buffer.clear();
+ result.push_str(sentence_line.as_str());
+ result.push('\n');
+ current_step_id += 1;
+ }
+
+ Ok(result)
+ }
+
+ pub fn character(character: &str, has_no_switch_flag: bool) -> String {
+ let flag = if has_no_switch_flag { "*" } else { "" };
+ format!("[{}{}{}]:", &flag, character, &flag)
+ }
+
+ pub fn step_name(marker: &str, current_id: i64) -> String {
+ format!("{}_{}", marker, current_id)
+ }
+
+ pub fn step_line(marker: &str, current_id: i64) -> String {
+ format!("@@@@@@@@@@ {}_{}", marker, current_id)
+ }
+
+ pub fn next_flag(marker: &str, current_id: i64) -> String {
+ format!("->[#{}_{}]", marker, current_id + 1)
+ }
+
+ pub fn is_marker(line: &str) -> bool {
+ line.starts_with("# ")
+ || line.starts_with("## ")
+ || line.starts_with("### ")
+ || line.starts_with("#### ")
+ || line.starts_with("##### ")
+ }
+
+ pub fn read_maker(line: &str) -> &str {
+ let trimmed = line.trim_start();
+ if trimmed.starts_with('#') {
+ if trimmed.starts_with("# ")
+ || trimmed.starts_with("## ")
+ || trimmed.starts_with("### ")
+ || trimmed.starts_with("#### ")
+ || trimmed.starts_with("##### ")
+ {
+ let parts: Vec<&str> = trimmed.splitn(2, ' ').collect();
+ if parts.len() == 2 {
+ return parts[1].trim();
+ }
+ }
+ }
+ ""
+ }
+
+ pub fn is_character(line: &str) -> bool {
+ line.starts_with("######")
+ }
+
+ pub fn read_character(line: &str) -> (&str, bool) {
+ let trimmed = line.trim_start();
+ if trimmed.starts_with("######") {
+ let parts: Vec<&str> = trimmed.splitn(2, ' ').collect();
+ if parts.len() == 2 {
+ let character = parts[1].trim();
+ if character.starts_with('*') && character.ends_with('*') {
+ let trimmed = character.trim_matches('*');
+ return (trimmed.trim(), true);
+ } else {
+ return (character.trim(), false);
+ }
+ }
+ }
+ ("", false)
+ }
+
+ pub fn get_jump_from_line(line: &str) -> (String, Option<String>) {
+ let pattern = r"\[\]\(#([^)]+)\)$";
+ let re = Regex::new(pattern).unwrap();
+
+ if let Some(caps) = re.captures(line.trim_end()) {
+ let target = caps.get(1).unwrap().as_str();
+ let line_without_jump = line
+ .trim_end()
+ .replace(&format!(" [](#{})", target), "")
+ .to_string();
+ return (
+ line_without_jump.trim_start_matches("- ").to_string(),
+ Some(format!("{}", target)),
+ );
+ }
+
+ (line.trim_start_matches("- ").to_string(), None)
+ }
+}
+
+pub mod markdown_strip_invalid_jump {
+ use crate::error::Exit;
+ use regex::Regex;
+
+ /// Strip all jumps that have not appeared
+ pub fn proc(input: String) -> Result<String, Exit> {
+ let lines: Vec<&str> = input.lines().collect();
+ let mut valid_ids = std::collections::HashSet::new();
+
+ for line in &lines {
+ if line.starts_with("@@@@@@@@@@ ") {
+ let id = line.trim_start_matches("@@@@@@@@@@ ").trim();
+ valid_ids.insert(id.to_string());
+ }
+ }
+
+ let mut result_lines = Vec::new();
+ let link_re = Regex::new(r"\[#([^)]+)\]").unwrap();
+
+ for line in lines {
+ let processed_line = link_re.replace_all(line, |caps: &regex::Captures| {
+ let id = &caps[1];
+ if valid_ids.contains(id) {
+ format!("[#{}]", id)
+ } else {
+ "[]".to_string()
+ }
+ });
+ result_lines.push(processed_line.to_string());
+ }
+
+ Ok(result_lines.join("\n"))
+ }
+}
+
+pub mod markdown_convert_image {
+ use regex::Regex;
+
+ use crate::error::Exit;
+
+ /// Convert image lines to code lines
+ pub fn proc(input: String) -> Result<String, Exit> {
+ let mut result = String::new();
+ let lines: Vec<&str> = input.lines().collect();
+ let image_re = Regex::new(r"^!\[[^\]]*\]\(([^)]+)\)$").unwrap();
+
+ for line in lines {
+ if let Some(caps) = image_re.captures(line) {
+ let image_path = caps.get(1).unwrap().as_str();
+ result.push_str(&format!("`image \"{}\"`\n", image_path));
+ } else {
+ result.push_str(line);
+ result.push('\n');
+ }
+ }
+
+ // Remove trailing newline if present
+ if result.ends_with('\n') {
+ result.pop();
+ }
+
+ Ok(result)
+ }
+}
+
+pub mod markdown_apply_codes {
+ use crate::error::Exit;
+
+ /// Apply code lines to sentences
+ pub fn proc(input: String) -> Result<String, Exit> {
+ let mut out = String::new();
+ let lines: Vec<&str> = input.lines().collect();
+
+ let mut i = 0;
+ while i < lines.len() {
+ let line = lines[i];
+
+ if !line.trim_start().starts_with('`') {
+ out.push_str(line);
+ out.push('\n');
+ i += 1;
+ continue;
+ }
+
+ let mut code_buf = String::new();
+ while i < lines.len() && {
+ let line: &str = lines[i];
+ line.trim_start().starts_with('`')
+ } {
+ code_buf.push_str(lines[i].trim());
+ i += 1;
+ }
+
+ if i >= lines.len()
+ || !{
+ let line: &str = lines[i];
+ line.trim_start().starts_with('[')
+ }
+ {
+ continue;
+ }
+
+ if i + 1 < lines.len() && {
+ let line: &str = lines[i + 1];
+ line.trim_start().starts_with('[')
+ } {
+ continue;
+ }
+
+ let merged = merge_code_into_sentence(&code_buf, lines[i]);
+ out.push_str(&merged);
+ out.push('\n');
+ i += 1;
+ }
+
+ Ok(out)
+ }
+
+ fn merge_code_into_sentence(code: &str, sentence: &str) -> String {
+ if let Some(start) = sentence.find(":[") {
+ if let Some(_) = sentence[start + 2..].find(']') {
+ let content_start = start + 2;
+
+ let mut result = String::new();
+ result.push_str(&sentence[..content_start]);
+ result.push_str(code);
+ result.push_str(&sentence[content_start..]);
+ return result;
+ }
+ }
+
+ sentence.to_string()
+ }
+}
+
+pub mod markdown_split_and_encode {
+ use crate::error::Exit;
+
+ /// Split sentences into embeddable tokens and perform Unicode encoding
+ pub fn proc(input: String) -> Result<String, Exit> {
+ let mut result = String::new();
+ let lines: Vec<&str> = input.lines().collect();
+
+ for line in lines {
+ if line.starts_with('[') && line.contains("]:[") && line.contains("]->[") {
+ if let Some(start) = line.find("]:[") {
+ if let Some(end) = line.find("]->[") {
+ let content = &line[start + 3..end];
+ let processed_content = process_sentence_content(content);
+
+ let suffix = &line[end + 1..];
+
+ let char_end = start;
+ let char_start = 1;
+ let character = &line[char_start..char_end];
+ let encoded_character = encode_unicode(character);
+
+ // Build the new line with encoded character and processed content
+ let new_line =
+ format!("[{}]:{}{}", encoded_character, processed_content, suffix);
+ result.push_str(&format!("{}\n", new_line));
+ continue;
+ }
+ }
+ }
+ result.push_str(&format!("{}\n", line));
+ }
+
+ if result.ends_with('\n') {
+ result.pop();
+ }
+
+ Ok(result)
+ }
+
+ fn process_sentence_content(content: &str) -> String {
+ let mut result = String::new();
+ let mut chars = content.chars().peekable();
+ let mut current_text = String::new();
+ let mut in_code = false;
+ let mut in_bold = false;
+ let mut in_italic = false;
+ let mut code_buffer = String::new();
+ let mut backticks_count = 0;
+
+ while let Some(ch) = chars.next() {
+ match ch {
+ '`' => {
+ backticks_count += 1;
+ if backticks_count == 1 {
+ // Start of code block
+ if !current_text.is_empty() {
+ let encoded_text = encode_unicode(&current_text);
+ result.push_str(&format!("[text:[{}]]", encoded_text));
+ current_text.clear();
+ }
+ code_buffer.push(ch);
+ in_code = true;
+ } else if backticks_count == 2 && in_code {
+ // End of code block
+ code_buffer.push(ch);
+ let encoded_code = encode_unicode(&code_buffer);
+ result.push_str(&format!("[code:[{}]]", encoded_code));
+ code_buffer.clear();
+ backticks_count = 0;
+ in_code = false;
+ } else if backticks_count == 1 && !in_code {
+ // Single backtick in text
+ current_text.push(ch);
+ }
+ }
+ '*' => {
+ if in_code {
+ code_buffer.push(ch);
+ continue;
+ }
+
+ // Check for bold
+ if chars.peek() == Some(&'*') {
+ chars.next(); // Consume the second '*'
+
+ // Check for bold_italic (***)
+ if chars.peek() == Some(&'*') {
+ chars.next(); // Consume the third '*'
+
+ if in_bold && in_italic {
+ // End bold_italic
+ if !current_text.is_empty() {
+ let encoded_text = encode_unicode(&current_text);
+ result.push_str(&format!("[bold_italic:[{}]]", encoded_text));
+ current_text.clear();
+ }
+ in_bold = false;
+ in_italic = false;
+ } else {
+ // Start bold_italic
+ if !current_text.is_empty() {
+ let encoded_text = encode_unicode(&current_text);
+ result.push_str(&format!("[text:[{}]]", encoded_text));
+ current_text.clear();
+ }
+ in_bold = true;
+ in_italic = true;
+ }
+ } else {
+ // Handle ** (bold)
+ if in_bold && !in_italic {
+ // End bold
+ if !current_text.is_empty() {
+ let encoded_text = encode_unicode(&current_text);
+ result.push_str(&format!("[bold:[{}]]", encoded_text));
+ current_text.clear();
+ }
+ in_bold = false;
+ } else if in_italic {
+ // Currently in italic, encountering ** means we need to end italic and start bold_italic
+ if !current_text.is_empty() {
+ let encoded_text = encode_unicode(&current_text);
+ result.push_str(&format!("[italic:[{}]]", encoded_text));
+ current_text.clear();
+ }
+ in_italic = false;
+ in_bold = true;
+ } else {
+ // Start bold
+ if !current_text.is_empty() {
+ let encoded_text = encode_unicode(&current_text);
+ result.push_str(&format!("[text:[{}]]", encoded_text));
+ current_text.clear();
+ }
+ in_bold = true;
+ }
+ }
+ } else {
+ // Single * (italic)
+ if in_italic && !in_bold {
+ // End italic
+ if !current_text.is_empty() {
+ let encoded_text = encode_unicode(&current_text);
+ result.push_str(&format!("[italic:[{}]]", encoded_text));
+ current_text.clear();
+ }
+ in_italic = false;
+ } else if in_bold {
+ // Currently in bold, encountering * means we need to start bold_italic
+ if !current_text.is_empty() {
+ let encoded_text = encode_unicode(&current_text);
+ result.push_str(&format!("[bold:[{}]]", encoded_text));
+ current_text.clear();
+ }
+ in_italic = true;
+ } else {
+ // Start italic
+ if !current_text.is_empty() {
+ let encoded_text = encode_unicode(&current_text);
+ result.push_str(&format!("[text:[{}]]", encoded_text));
+ current_text.clear();
+ }
+ in_italic = true;
+ }
+ }
+ }
+ _ => {
+ if in_code {
+ code_buffer.push(ch);
+ } else {
+ current_text.push(ch);
+ }
+ }
+ }
+ }
+
+ // Handle any remaining text
+ if !code_buffer.is_empty() {
+ let encoded_code = encode_unicode(&code_buffer);
+ result.push_str(&format!("[code:[{}]]", encoded_code));
+ }
+
+ if !current_text.is_empty() {
+ let style = match (in_bold, in_italic) {
+ (true, true) => "bold_italic",
+ (true, false) => "bold",
+ (false, true) => "italic",
+ (false, false) => "text",
+ };
+ let encoded_text = encode_unicode(&current_text);
+ result.push_str(&format!("[{}:[{}]]", style, encoded_text));
+ }
+
+ result
+ }
+
+ fn encode_unicode(s: &str) -> String {
+ let mut result = String::new();
+ for ch in s.chars() {
+ let code = ch as u32;
+ if code <= 0x7F {
+ result.push(ch);
+ } else {
+ result.push_str(&format!("\\u{:X}", code));
+ }
+ }
+ result
+ }
+}
diff --git a/_prototype/converter/src/syntax_checker.rs b/_prototype/converter/src/syntax_checker.rs
new file mode 100644
index 0000000..52021c7
--- /dev/null
+++ b/_prototype/converter/src/syntax_checker.rs
@@ -0,0 +1,221 @@
+use regex::Regex;
+
+use crate::error::Exit;
+
+pub fn check_markdown_syntax(i: &String) -> Result<(), Exit> {
+ let mut stack = Vec::new();
+ let lines: Vec<&str> = i.lines().collect();
+ let mut anchors = Vec::new();
+ let mut heading_ids = Vec::new();
+
+ for (line_num, line) in lines.iter().enumerate() {
+ let line_num = line_num as i64 + 1;
+
+ // Check for headings to collect anchor IDs
+ if line.starts_with('#') {
+ let heading_text = line.trim_start_matches('#').trim();
+ let id = heading_text
+ .to_lowercase()
+ .chars()
+ .filter(|c| c.is_alphanumeric() || *c == '-' || *c == '_')
+ .collect::<String>();
+ if !id.is_empty() {
+ heading_ids.push(id);
+ }
+ }
+
+ let mut chars = line.chars().enumerate().peekable();
+ while let Some((pos, ch)) = chars.next() {
+ let pos = pos as i64 + 1;
+
+ match ch {
+ '[' => {
+ // Check if it's a link or image
+ let is_image = chars.peek().map(|&(_, c)| c) == Some('!');
+ if is_image {
+ chars.next(); // Skip '!'
+ }
+ stack.push(('['.to_string(), line_num, pos, is_image));
+ }
+ ']' => {
+ if let Some((last, _l, b, is_image)) = stack.pop() {
+ if last != "[" {
+ return Err(Exit::SyntaxError {
+ content: line.to_string(),
+ reason: format!(
+ "Mismatched bracket: expected '[' but found '{}'",
+ last
+ ),
+ line: line_num,
+ begin: b,
+ end: pos,
+ });
+ }
+ // Check if it's followed by '(' for a link
+ if chars.peek().map(|&(_, c)| c) == Some('(') {
+ chars.next(); // Skip '('
+ // Look for closing ')'
+ let mut found = false;
+ let mut anchor_started = false;
+ let mut anchor = String::new();
+ while let Some((_, c)) = chars.next() {
+ if c == ')' {
+ found = true;
+ break;
+ }
+ if c == '#' && !anchor_started {
+ anchor_started = true;
+ continue;
+ }
+ if anchor_started {
+ anchor.push(c);
+ }
+ }
+ if !found {
+ return Err(Exit::SyntaxError {
+ content: line.to_string(),
+ reason: "Link parentheses not closed".to_string(),
+ line: line_num,
+ begin: pos,
+ end: pos,
+ });
+ }
+ if !anchor.is_empty() {
+ // Remove whitespace from anchor
+ let anchor = anchor.replace(|c: char| c.is_whitespace(), "");
+ anchors.push((anchor, line_num, pos));
+ }
+ } else if !is_image {
+ // It's a reference link, collect the anchor
+ // Check for anchor like [](#anchor)
+ if chars.peek().map(|&(_, c)| c) == Some('(') {
+ chars.next(); // Skip '('
+ if chars.peek().map(|&(_, c)| c) == Some('#') {
+ chars.next(); // Skip '#'
+ let mut anchor = String::new();
+ while let Some(&(_, c)) = chars.peek() {
+ if c == ')' {
+ break;
+ }
+ anchor.push(c);
+ chars.next();
+ }
+ if !anchor.is_empty() {
+ // Remove whitespace from anchor
+ let anchor =
+ anchor.replace(|c: char| c.is_whitespace(), "");
+ anchors.push((anchor, line_num, pos));
+ }
+ }
+ }
+ }
+ } else {
+ return Err(Exit::SyntaxError {
+ content: line.to_string(),
+ reason: "Unmatched ']'".to_string(),
+ line: line_num,
+ begin: pos,
+ end: pos,
+ });
+ }
+ }
+ '(' => {
+ // Check for standalone anchor like (#anchor)
+ if chars.peek().map(|&(_, c)| c) == Some('#') {
+ chars.next(); // Skip '#'
+ let mut anchor = String::new();
+ while let Some(&(_, c)) = chars.peek() {
+ if c == ')' {
+ break;
+ }
+ anchor.push(c);
+ chars.next();
+ }
+ if !anchor.is_empty() {
+ // Remove whitespace from anchor
+ let anchor = anchor.replace(|c: char| c.is_whitespace(), "");
+ anchors.push((anchor, line_num, pos));
+ }
+ } else {
+ stack.push(('('.to_string(), line_num, pos, false));
+ }
+ }
+ ')' => {
+ if let Some((last, _l, b, _)) = stack.pop() {
+ if last != "(" {
+ return Err(Exit::SyntaxError {
+ content: line.to_string(),
+ reason: format!(
+ "Mismatched parenthesis: expected '(' but found '{}'",
+ last
+ ),
+ line: line_num,
+ begin: b,
+ end: pos,
+ });
+ }
+ } else {
+ return Err(Exit::SyntaxError {
+ content: line.to_string(),
+ reason: "Unmatched ')'".to_string(),
+ line: line_num,
+ begin: pos,
+ end: pos,
+ });
+ }
+ }
+ '`' => {
+ // Check for backticks
+ let mut count = 1;
+ while chars.peek().map(|&(_, c)| c) == Some('`') {
+ count += 1;
+ chars.next();
+ }
+ let marker = "`".repeat(count);
+
+ if let Some((last, _, _, _)) = stack.last() {
+ if last == &marker {
+ stack.pop();
+ } else {
+ stack.push((marker.clone(), line_num, pos, false));
+ }
+ } else {
+ stack.push((marker, line_num, pos, false));
+ }
+ }
+ _ => {}
+ }
+ }
+ }
+
+ // Check for unclosed brackets/parentheses
+ if let Some((last, line, begin, _)) = stack.pop() {
+ return Err(Exit::SyntaxError {
+ content: lines[(line - 1) as usize].to_string(),
+ reason: format!("Unclosed '{}'", last),
+ line,
+ begin,
+ end: begin,
+ });
+ }
+
+ Ok(())
+}
+
+/// Check for duplicate markers
+pub fn check_duplicate_marker(input: &String) -> Result<(), Exit> {
+ let mut seen = std::collections::HashSet::new();
+ let heading_re = Regex::new(r"^(#{1,5})\s+(.+)$").unwrap();
+
+ for line in input.lines() {
+ if let Some(caps) = heading_re.captures(line) {
+ let heading_text = caps[2].trim().to_string();
+ if seen.contains(&heading_text) {
+ return Err(Exit::DuplicateMarker(heading_text));
+ }
+ seen.insert(heading_text);
+ }
+ }
+
+ Ok(())
+}
diff --git a/_prototype/converter/src/utils.rs b/_prototype/converter/src/utils.rs
new file mode 100644
index 0000000..0fbb516
--- /dev/null
+++ b/_prototype/converter/src/utils.rs
@@ -0,0 +1 @@
+pub mod path_fmt;
diff --git a/_prototype/converter/src/utils/path_fmt.rs b/_prototype/converter/src/utils/path_fmt.rs
new file mode 100644
index 0000000..8750db6
--- /dev/null
+++ b/_prototype/converter/src/utils/path_fmt.rs
@@ -0,0 +1,123 @@
+use std::path::{Path, PathBuf};
+
+/// Normalize an input path string into a canonical, platform‑agnostic form.
+///
+/// This function removes ANSI escape sequences, unifies separators to `/`,
+/// collapses duplicate slashes, strips unfriendly characters (`*`, `?`, `"`, `<`, `>`, `|`),
+/// resolves simple `..` components, and preserves a trailing slash when present.
+///
+/// See examples below for the exact normalization behavior.
+///
+/// # Examples
+///
+/// ```
+/// # use string_proc::format_path::format_path_str;
+/// use std::io::Error;
+///
+/// # fn main() -> Result<(), Error> {
+/// assert_eq!(format_path_str("C:\\Users\\\\test")?, "C:/Users/test");
+/// assert_eq!(
+/// format_path_str("/path/with/*unfriendly?chars")?,
+/// "/path/with/unfriendlychars"
+/// );
+/// assert_eq!(format_path_str("\x1b[31m/path\x1b[0m")?, "/path");
+/// assert_eq!(format_path_str("/home/user/dir/")?, "/home/user/dir/");
+/// assert_eq!(
+/// format_path_str("/home/user/file.txt")?,
+/// "/home/user/file.txt"
+/// );
+/// assert_eq!(
+/// format_path_str("/home/my_user/DOCS/JVCS_TEST/Workspace/../Vault/")?,
+/// "/home/my_user/DOCS/JVCS_TEST/Vault/"
+/// );
+/// assert_eq!(format_path_str("./home/file.txt")?, "home/file.txt");
+/// assert_eq!(format_path_str("./home/path/")?, "home/path/");
+/// assert_eq!(format_path_str("./")?, "");
+/// # Ok(())
+/// # }
+/// ```
+pub fn format_path_str(path: impl Into<String>) -> Result<String, std::io::Error> {
+ let path_str = path.into();
+ let ends_with_slash = path_str.ends_with('/');
+
+ // ANSI Strip
+ let cleaned = strip_ansi_escapes::strip(&path_str);
+ let path_without_ansi = String::from_utf8(cleaned)
+ .map_err(|e| std::io::Error::new(std::io::ErrorKind::InvalidData, e))?;
+
+ let path_with_forward_slash = path_without_ansi.replace('\\', "/");
+ let mut result = String::new();
+ let mut prev_char = '\0';
+
+ for c in path_with_forward_slash.chars() {
+ if c == '/' && prev_char == '/' {
+ continue;
+ }
+ result.push(c);
+ prev_char = c;
+ }
+
+ let unfriendly_chars = ['*', '?', '"', '<', '>', '|'];
+ result = result
+ .chars()
+ .filter(|c| !unfriendly_chars.contains(c))
+ .collect();
+
+ // Handle ".." path components
+ let path_buf = PathBuf::from(&result);
+ let normalized_path = normalize_path(&path_buf);
+ result = normalized_path.to_string_lossy().replace('\\', "/");
+
+ // Restore trailing slash if original path had one
+ if ends_with_slash && !result.ends_with('/') {
+ result.push('/');
+ }
+
+ // Special case: when result is only "./", return ""
+ if result == "./" {
+ return Ok(String::new());
+ }
+
+ Ok(result)
+}
+
+/// Normalize path by resolving ".." components without requiring file system access
+fn normalize_path(path: &Path) -> PathBuf {
+ let mut components = Vec::new();
+
+ for component in path.components() {
+ match component {
+ std::path::Component::ParentDir => {
+ if !components.is_empty() {
+ components.pop();
+ }
+ }
+ std::path::Component::CurDir => {
+ // Skip current directory components
+ }
+ _ => {
+ components.push(component);
+ }
+ }
+ }
+
+ if components.is_empty() {
+ PathBuf::from(".")
+ } else {
+ components.iter().collect()
+ }
+}
+
+/// Format a [`PathBuf`] into its canonical string form and convert it back.
+///
+/// This is a convenience wrapper around [`format_path_str`], preserving
+/// the semantics of [`PathBuf`] while applying the same normalization rules:
+/// - normalize separators to `/`
+/// - remove duplicated separators
+/// - strip ANSI escape sequences
+/// - remove unfriendly characters (`*`, `?`, etc.)
+/// - resolve simple `..` segments
+pub fn format_path(path: impl Into<PathBuf>) -> Result<PathBuf, std::io::Error> {
+ let path_str = format_path_str(path.into().display().to_string())?;
+ Ok(PathBuf::from(path_str))
+}
diff --git a/_prototype/converter/usage.txt b/_prototype/converter/usage.txt
new file mode 100644
index 0000000..e03897e
--- /dev/null
+++ b/_prototype/converter/usage.txt
@@ -0,0 +1,6 @@
+mdialogc -i <FILE> -o <FILE>
+ -i --input <FILE> Input file (Markdown / MarkDialog source)
+ -o, --output <FILE> Output file (Generated IR / result)
+
+ -h, --help Show this content
+ -v, --version Show version
diff --git a/_prototype/converter/version.txt b/_prototype/converter/version.txt
new file mode 100644
index 0000000..032f755
--- /dev/null
+++ b/_prototype/converter/version.txt
@@ -0,0 +1,2 @@
+mdialogc (0.1.0)
+Mark Dialog Converter - Copyright © 2026 Weicao-CatilGrass
generated by cgit (git 2.34.1) at 2026-05-09 10:35:02 +0000