readme

.gitignore

@@ -0,0 +1,3 @@
+Cargo.lock
+target
+*.swp

Cargo.toml

@@ -6,5 +6,6 @@ edition = "2021"
 [dependencies]
 colored = "2.2.0"
 regex = "1.11.1"
-url = "2.5.4"
 termion = "4.0.3"
+open = "5.3.2"
+url = "2.5.4"

README.md

@@ -0,0 +1,42 @@
+# Markdown web browser
+A web browser that let's you browse 'mttp' websites that use markdown as a superior standard to html
+Fully static!
+
+# Getting a website on this
+The default port is 3477, though you can use any port as long as you specify it in the url.
+You need a <!DOCTYPE md> tag at the start of any markdown files so the browser know's which files are markdown and which are other generic text files.
+
+# Help
+Type h in the program to see this text:
+```
+Source code: https://git.javalsai.dynv6.net/deadvey/markdown-webbrowser
+q: quit
+h: help
+r: reload
+s: view source code of page
+i: visit root index of this host eg: root index of mttp://deadvey.com/blog/4.md is just deadvey.com
+b: go back in history
+f: go forward in history
+ox: print the hyprlink of reference x eg: o5 or o24
+[url]: follow the inputed url
+```
+
+# Example:
+![screenshot](/images/screenshot.png)
+
+# TO DO
+- Properly comment it because I can't lie I can't even follow it.
+- Make it memory safe, it crashes if the input is unexpected.
+- Use treesitter instead of Regex, because, reasons.
+    - "wtf deadvey" - [error](https://git.javalsai.dynv6.net/ErrorNoInternet) 
+    - "Don't use Regex to parse **ANYTHING**" - [error](https://git.javalsai.dynv6.net/ErrorNoInternet)
+    - "use treesitter" - [error](https://git.javalsai.dynv6.net/ErrorNoInternet)
+    - "yeah, definitley use treesitter" - [error](https://git.javalsai.dynv6.net/ErrorNoInternet)
+    - "use treesitter" - [error](https://git.javalsai.dynv6.net/ErrorNoInternet)
+    - "or glow" - [error](https://git.javalsai.dynv6.net/ErrorNoInternet)
+    - "I found another markdown to terminal converter" - [error](https://git.javalsai.dynv6.net/ErrorNoInternet)
+    - "ban [for using regex]"  - [error](https://git.javalsai.dynv6.net/ErrorNoInternet)
+    - "use treesitter" - [error](https://git.javalsai.dynv6.net/ErrorNoInternet)
+- Get a catchier name, 'markdown web browser' sounds kind of lame.
+
+![error yapping](/images/error_yapping.gif)

src/main.rs

@@ -2,30 +2,33 @@ use std::process::{Command};
 use std::io::{stdin,stdout,Write};
 use colored::Colorize;
 use regex::Regex;
+use url::{Url, ParseError};
 
 fn clear_screen() {
+    println!("clearing");
     Command::new("clear")
         .status()
         .expect("Failed to clear screen");
-    //println!("clearing");
 }
 
 fn parse_markdown(page_content: String) -> (String, Vec<String>) {
 	let mut parsed_page_content: String = "".to_string();
 	let mut hyperlink_number_counter: u64 = 0;
 	let mut links: Vec<String> = Vec::new();
-    let (screen_width, screen_height) = termion::terminal_size().unwrap(); // So the horizontal line (<hr/>) spans the whole console
+	let (screen_width, _screen_height) = termion::terminal_size().unwrap(); // So the horizontal line (<hr/>) spans the whole console
 	
     for line in page_content.lines() {
         let mut parsed_line: String = line.to_string();
         // Bold
-        let bold_regex = Regex::new(r"\*\*(.*?)\*\*").unwrap();
+        let bold_regex = Regex::new(r"((\*\*)|(__))(.*?)((\*\*)|(__))").unwrap();
         parsed_line = bold_regex.replace_all(&parsed_line, |caps: &regex::Captures| {
-            caps[1].bold().to_string()
+            caps[4].bold().to_string()
         }).to_string();
-        let bold_regex = Regex::new(r"__(.*?)__").unwrap();
-        parsed_line = bold_regex.replace_all(&parsed_line, |caps: &regex::Captures| {
-            caps[1].bold().to_string()
+
+        // Strikethrough
+        let strikethrough_regex = Regex::new(r"~~(.*?)~~").unwrap();
+        parsed_line = strikethrough_regex.replace_all(&parsed_line, |caps: &regex::Captures| {
+            caps[1].strikethrough().to_string()
         }).to_string();
 
         // Horizontal lines
@@ -39,6 +42,12 @@ fn parse_markdown(page_content: String) -> (String, Vec<String>) {
             result
         }).to_string();
 
+        // html br tag support
+        let br_regex = Regex::new(r"(.*?)<br/>(.*?)").unwrap();
+        parsed_line = br_regex.replace_all(&parsed_line, |caps: &regex::Captures| {
+            format!("{}{}{}", &caps[1], "\n", &caps[2])
+        }).to_string();
+
         // Italics
         let italic_regex = Regex::new(r"\*(.*?)\*").unwrap();
         parsed_line = italic_regex.replace_all(&parsed_line, |caps: &regex::Captures| {
@@ -52,19 +61,23 @@ fn parse_markdown(page_content: String) -> (String, Vec<String>) {
         // Block quotes
         let block_quotes_regex = Regex::new(r"^>(.*)").unwrap();
         parsed_line = block_quotes_regex.replace_all(&parsed_line, |caps: &regex::Captures| {
-            format!("   | {}", &caps[1].on_black())
+            format!("   | {}", &caps[1])
         }).to_string();
 
         // Ordered list
-        let ordered_list_regex = Regex::new(r"^([0-9]+)\.(.*)").unwrap();
-        parsed_line = ordered_list_regex.replace_all(&parsed_line, "  $1. $2").to_string();
+        let ordered_list_regex = Regex::new(r"^([ \t]+|^)([0-9]+)\. (.*)").unwrap();
+        parsed_line = ordered_list_regex.replace_all(&parsed_line, |caps: &regex::Captures| {
+            format!("{} {}. {}", &caps[1], &caps[2], &caps[3])
+        }).to_string();
 
-        // Unordered list
-        let unordered_list_regex = Regex::new(r"^(-|\+|\*).(.*)").unwrap();
-        parsed_line = unordered_list_regex.replace_all(&parsed_line, "    • $2").to_string();
+        // Unordered list ([    ]+|^)- (.*)
+        let unordered_list_regex = Regex::new(r"^([ \t]+|^)(-|\+|\*).(.*)").unwrap();
+        parsed_line = unordered_list_regex.replace_all(&parsed_line, |caps: &regex::Captures| {
+            format!("{} • {}", &caps[1], &caps[3])
+        }).to_string();
 
         // Inline code
-        let inline_code_regex = Regex::new(r"`(.*)`").unwrap();
+        let inline_code_regex = Regex::new(r"`([^`]+?)`").unwrap();
         parsed_line = inline_code_regex.replace_all(&parsed_line, |caps: &regex::Captures| {
             format!("{}", &caps[1].magenta())
         }).to_string();
@@ -81,16 +94,27 @@ fn parse_markdown(page_content: String) -> (String, Vec<String>) {
                 result
             } else {
                 // If it's an image (starts with !), return the link as is
-                format!("[{}]", &caps[2].green())
+                let url = caps[3].to_string();
+                links.push(url);
+                hyperlink_number_counter += 1;
+                format!("({})[{}]", &caps[2].green(), hyperlink_number_counter)
             }
         }).to_string();
         
+        let quick_hyperlink_regex = Regex::new(r"<(.*:\/\/.*)>").unwrap();
+        parsed_line = quick_hyperlink_regex.replace_all(&parsed_line, |caps: &regex::Captures| {
+            hyperlink_number_counter += 1;
+            let url = caps[1].to_string();
+            links.push(url);
+            format!("{}[{}]", &caps[1].blue().underline(), hyperlink_number_counter)
+        }).to_string();
+
 
         parsed_page_content+=&(parsed_line + "\n");
     }
 
     // multiline code
-    let multiline_code_regex = Regex::new(r"(?ms)%%%((.*?\n)+?)%%%").unwrap();
+    let multiline_code_regex = Regex::new(r"(?ms)```((.*?\n)+?)```").unwrap();
     parsed_page_content = multiline_code_regex.replace_all(&parsed_page_content, |caps: &regex::Captures| {
         // Capture the code inside the %% blocks
         let code_block = &caps[1];
@@ -109,8 +133,8 @@ fn parse_markdown(page_content: String) -> (String, Vec<String>) {
     return (parsed_page_content, links);
 }
 
-fn fetch_page(host: &String, port: &String, path: &String) -> String {
-    let full_url_formatted = format!("{}:{}/{}", host, port, path);
+fn fetch_page(url: &Url) -> String {
+    let full_url_formatted = format!("{}", url);
 
     // Call curl using Com, mand
     let output = Command::new("curl")
@@ -129,14 +153,26 @@ fn fetch_page(host: &String, port: &String, path: &String) -> String {
     }
 }
 
-fn render_page(host: String, port: String, path: String) -> Vec<String> {
+fn render_page(url: Url, source: bool) -> Vec<String> {
 	clear_screen();
-    let mut content = fetch_page(&host, &port, &path);
+	let mut content = fetch_page(&url);
 	let mut links = Vec::new();
-    let (screen_width, screen_height) = termion::terminal_size().unwrap();
+	let (screen_width, _screen_height) = termion::terminal_size().unwrap();
 
-    (content, links) = parse_markdown(content);
-    print!("{}: {}\n", host, path);
+	if source == true {
+		content += &format!("{}", &"Viewing source code".yellow());
+	}
+	else if &content[..13] == "<!DOCTYPE md>" {
+		(content, links) = parse_markdown((&content[13..]).to_string());
+	}
+	else {
+		content += &format!("{}", &"Warning: This page is invalid markdown, it should contain <!DOCTYPE md> at the very start of the file, showing raw text".yellow());
+	}
+
+	for _i in 0..screen_width {
+		print!("—"); 
+	}
+	print!("{}\n", url);
 	for _i in 0..screen_width {
 		print!("—"); 
 	}
@@ -145,6 +181,8 @@ fn render_page(host: String, port: String, path: String) -> Vec<String> {
 		print!("—"); 
 	}
 	println!();
+
+	// Return links (you can add link parsing logic)
 	return links;   
 }
 
@@ -161,84 +199,152 @@ fn input() -> String{
     return s;
 }
 
-fn parse_url(url: String, previous_host: &String) -> (String, String, String) {
-    let mut host: String = previous_host.to_string();
-    let mut port: String = "3477".to_string();
-    let mut path: String = "/".to_string();
+fn parse_url(user_input: String, previous_url: &Url) -> Result<Url, ParseError> {
+	println!("user input: {}",user_input);
+	println!("previous url: {:?}",previous_url);
+	let to_parse =  if user_input.contains("://")  {
+		println!("Contains different scheme or is a path");
+		user_input
+	}
+	else if user_input[..1] == *"/" {
+		format!("http://{}/{}",Url::host_str(previous_url).expect("ivalid").to_string(), user_input)
+	}
+	else {
+		println!("prepending scheme to user input");
+		format!("http://{}", user_input)  // Prepend 'mttp://' if no scheme is found
+	};
 
-    let mttp_regex = Regex::new(r"^mttp:\/\/(.*?)\/(.*?)$").unwrap(); 
-    let mttp_regex_no_path = Regex::new(r"^mttp:\/\/(.*?)$").unwrap(); 
-    let accept_this_as_mttp = Regex::new(r"^(.*?)$").unwrap();
-    let path_change = Regex::new(r"^/(.*?)$").unwrap();
+	println!("Parsing: {}", to_parse);
+	if let Ok(mut url) = Url::parse(&to_parse) {
+		if url.port() == None {
+			let _ = url.set_port(Some(3477));
+		}
+		println!("{:?}",url);
+		println!("{}",url.as_str());
+		println!("parsed successfully");
+		return Ok(url)
+	}
+	else {
+		return Err(ParseError::InvalidDomainCharacter)
+	}
 
-    if let Some(caps) = mttp_regex.captures(&url) {
-        host = caps[1].to_string(); 
-        port = "3477".to_string();   
-        path = caps[2].to_string();
-    }
-    else if let Some(caps) = mttp_regex_no_path.captures(&url) {
-        host = caps[1].to_string(); 
-        port = "3477".to_string();   
-        path = "/".to_string();
-    }
-    else if let Some(caps) = path_change.captures(&url) {
-        println!("path change");
-        host = previous_host.to_string(); 
-        port = "3477".to_string();   
-        path = caps[1].to_string();
-    }
-    else if let Some(caps) = accept_this_as_mttp.captures(&url) {
-        host = caps[1].to_string(); 
-        port = "3477".to_string();   
-        path = "/".to_string();
-    }
-    println!("{}:{}/{}",host,port,path);
-
-    return (host, port, path);
 }
 
 fn main() {
 	clear_screen();
 	println!("Enter a url: ");
-    let url = input();
+	let user_input = input();
 
-    if url == "q" {
+	if user_input == "q" {
 		std::process::exit(0);
 	}
-
-    let (mut host, mut port, mut path) = parse_url(url, &"example.com".to_string()); // Must pass
-                                                                                    // 'previous
-                                                                                    // host'
-
+	let mut load_page: bool = true;
+	let mut history: Vec<Url> = Vec::new();
+	let mut historical_position: usize = 0;
+	let mut links: Vec<String> = Vec::new();
+	let mut source: bool = false; // Wether to view source of markdown page or rendered version
+	if let Ok(mut url) = parse_url(user_input, &Url::parse(&"http://deadvey.com").unwrap()) { // Change this and make internal pages ;)
+		history.push(url.clone());
 		'mainloop: loop {
-        let links = render_page(host.clone(), port.clone(), path.clone());   
-
-        println!("{}:{}/{}", host, port, path);
-
+			if load_page {
+				links = render_page(history[historical_position].clone(), source);
 				println!("Enter reference number to follow, h for help, or q to quit");
-        let link_to_follow = input(); 
-        if link_to_follow == "q" {
+			}
+			load_page = false;
+
+			let user_input = input(); 
+			if user_input == "q" {
 				break 'mainloop;
 			}
-        else if link_to_follow == "r" {
+			else if user_input == "r" {
+				load_page = true;
 				continue;
 			}
-        else if link_to_follow == "h" {
-            println!("
-                Source code: https://git.javalsai.dynv6.net/deadvey/markdown-webbrowser\n
-                q: quit
-                h: help
-                r: reload
-                i: visit root index of this host eg: root index of mttp://deadvey.com/blog/4.md is just deadvey.com
-                b: go back in history
-            ");
+			else if user_input == "s" {
+				source = ! source; // Flip the boolean to toggle source mode
+				load_page = true;
+			}
+			else if user_input == "i" {
+				let _ = url.set_path("/"); 				
+				for _i in historical_position+1..history.len() {
+					history.remove(historical_position+1);
+				}
+				history.push(url.clone());
+				historical_position += 1;
+				load_page = true;
+			}
+			else if user_input == "b" {
+				if historical_position > 0 {
+					historical_position -= 1;
+					load_page = true;
 				}
 				else {
-            let number: usize = link_to_follow.parse::<usize>().unwrap();
-
-            (host, port, path) = parse_url(links[number].clone(), &host);
-            println!("{}:{}/{}", host, port, path);
+					println!("At start of history");
 				}
 			}
+			else if user_input == "f" {
+				if historical_position < history.len()-1 {
+					historical_position += 1;
+					load_page = true;
+				}
+				else {
+					println!("At end of history");
+				}
+			}
+			else if user_input == "h" {
+				println!("Source code: https://git.javalsai.dynv6.net/deadvey/markdown-webbrowser
+q: quit
+h: help
+r: reload
+s: view source code of page
+i: visit root index of this host eg: root index of mttp://deadvey.com/blog/4.md is just deadvey.com
+b: go back in history
+f: go forward in history
+ox: print the hyprlink of reference x eg: o5 or o24
+[url]: follow the inputed url");
+			}
+			else if user_input.chars().nth(0).unwrap() == 'o' {
+			    let number_str = &user_input[1..];
+			    if let Ok(number) =  number_str.parse::<usize>() {
+				println!("{}", links[number]);
+			    } else {
+				println!("error");
+			    }
+			}
+			else if let Ok(number) = user_input.parse::<usize>() {
+				if number < links.len() {
+					if let Ok(parsed_value) = parse_url(links[number].clone(), &url.clone()) {
+						url = parsed_value;
+						for _i in historical_position+1..history.len() {
+							history.remove(historical_position+1);
+						}
+						history.push(url.clone());
+						historical_position += 1;
+						load_page = true;
+					}
+					else {
+						println!("Invalid url\nAttempting to open url in web browser");
+					}
+				} else {
+					println!("Invalid reference id");
+				}
+			}
+			else if let Ok(parsed_value) = parse_url(user_input, &url.clone()) {
+				url = parsed_value;
+				for _i in historical_position+1..history.len() {
+					history.remove(historical_position+1);
+				}
+				history.push(url.clone());
+				historical_position += 1;
+				load_page = true;
+			}
+			else {
+				println!("Invalid input");
+			}
+		}
+	}
+	else {
+		println!("Invalid mttp url, try mttp:// at the start of your input.");
+	}
 }