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 +1,48 @@
+# Markdown web browser
+A web browser that let's you browse gemini capsules that use gemtext as a superior standard to html.
+
+# Requirements
+- gemget
+
+# Installing
+## Build from source:
+- Clone the repo
+```cargo run -r```
+
+# Help
+Type h in the program to see this text:
+```
+Source code: https://git.javalsai.dynv6.net/deadvey/markdown-webbrowser
+q: quit
+d: debug info
+h: help
+r: reload
+s: view source code of page
+i: visit root index of this host eg: root index of gemini://deadvey.com/blog/4.md is just gemini://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
+- Make pages scrollable
+- Bookmarks
+- 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

@@ -1,150 +1,71 @@
 use std::process::{Command};
 use std::io::{stdin,stdout,Write};
 use colored::Colorize;
-use regex::Regex;
+use url::{Url, ParseError};
+use std::fs;
+
+const DEBUG_MODE: bool = false;
+
+// Import other files
+mod parse_gemtext;
 
 fn clear_screen() {
+    if DEBUG_MODE == false
+    {
+        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
+fn fetch_page(url: &Url) {
+    let full_url_formatted = format!("{}", url);
 
-    for line in page_content.lines() {
-        let mut parsed_line: String = line.to_string();
-        // Bold
-        let bold_regex = Regex::new(r"\*\*(.*?)\*\*").unwrap();
-        parsed_line = bold_regex.replace_all(&parsed_line, |caps: &regex::Captures| {
-            caps[1].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()
-        }).to_string();
-
-        // Horizontal lines
-        let hr_regex = Regex::new(r"^(\*\*\*)|(---)|(___)$").unwrap();
-        parsed_line = hr_regex.replace_all(&parsed_line, |_caps: &regex::Captures| {
-            let mut result: String = "\n".to_string();
-            for _x in 0..screen_width/2 {
-                result += "- ";
-            }
-            result += "\n";
-            result
-        }).to_string();
-
-        // Italics
-        let italic_regex = Regex::new(r"\*(.*?)\*").unwrap();
-        parsed_line = italic_regex.replace_all(&parsed_line, |caps: &regex::Captures| {
-            caps[1].italic().to_string()
-        }).to_string();
-        let italic_regex = Regex::new(r"_(.*?)_").unwrap();
-        parsed_line = italic_regex.replace_all(&parsed_line, |caps: &regex::Captures| {
-            caps[1].italic().to_string()
-        }).to_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())
-        }).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();
-
-        // Unordered list
-        let unordered_list_regex = Regex::new(r"^(-|\+|\*).(.*)").unwrap();
-        parsed_line = unordered_list_regex.replace_all(&parsed_line, "    • $2").to_string();
-
-        // Inline code
-        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();
-
-        // HyperLink
-        let hyperlink_regex = Regex::new(r"(.*?)\[(.*?)\]\((.*?)\)").unwrap();
-        parsed_line = hyperlink_regex.replace_all(&parsed_line, |caps: &regex::Captures| {
-            // Check if the character before the link is not '!'
-            if !caps[1].ends_with('!') {  // caps[1] is everything before the link
-                let result = format!("{}{}[{}]", &caps[1], &caps[2].blue().underline(), hyperlink_number_counter);
-                let url = caps[3].to_string();
-                links.push(url);
-                hyperlink_number_counter += 1;
-                result
-            } else {
-                // If it's an image (starts with !), return the link as is
-                format!("[{}]", &caps[2].green())
-            }
-        }).to_string();
-
-
-        parsed_page_content+=&(parsed_line + "\n");
-    }
-
-    // multiline code
-    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];
-        
-        // Add a tab to each line in the block
-        let indented_code = code_block
-            .lines()
-            .map(|line| format!("\t{}", line)) // Insert tab at the start of each line
-            .collect::<Vec<String>>()
-            .join("\n");
-
-        // Return the formatted block with magenta color
-        format!("{}", indented_code.magenta())
-    }).to_string();
-
-    return (parsed_page_content, links);
-}
-
-fn fetch_page(host: &String, port: &String, path: &String) -> String {
-    let full_url_formatted = format!("{}:{}/{}", host, port, path);
-
-    // Call curl using Com, mand
-    let output = Command::new("curl")
-        .arg(full_url_formatted)
+    let output = Command::new("gemget")
+        .args([full_url_formatted, "-o".to_string(), "/tmp/page".to_string()])
         .output()
-        .expect("Failed to execute curl command");
+        .expect("Failed to execute gemget command");
 
     // Check if the command was successful
-    if output.status.success() {
-        let page: String = String::from_utf8_lossy(&output.stdout).to_string();
-        return page
-    } else {
-        eprintln!("Error:\n{}", String::from_utf8_lossy(&output.stderr));
-        let result: String = "error".to_string();
-        return result
+    if ! output.status.success() {
+        println!("{}\n{:?}\n", "Failed to fetch page:".red(), output);
     } 
 }
 
-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);
+	fetch_page(&url);
 	let mut links = Vec::new();
-    let (screen_width, screen_height) = termion::terminal_size().unwrap();
+        if let Ok(mut content) = fs::read_to_string::<String>("/tmp/page".to_string()) {
+            Command::new("rm")
+                .arg("/tmp/page")
+                .output()
+                .expect("Failed to delete tmp page");
+	    let (screen_width, _screen_height) = termion::terminal_size().unwrap();
+
+            if source == true {
+                    content += &format!("{}", &"Viewing source code".yellow());
+            }
+            else {
+                (content, links) = parse_gemtext::parse_gemtext(content);
+            }
 
-    (content, links) = parse_markdown(content);
-    print!("{}: {}\n", host, path);
             for _i in 0..screen_width {
-       print!("—"); 
+                    print!("-"); 
+            }
+            print!("{}\n", url);
+            for _i in 0..screen_width {
+                    print!("-"); 
             }
             println!("\n\n{}", content);
             for _i in 0..screen_width {
-       print!("—"); 
+                    print!("-"); 
             }
             println!();
+
+            // Return links (you can add link parsing logic)
+        }
         return links;
 }
 
@@ -161,84 +82,160 @@ 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 {
+		format!("gemini://{}/{}",Url::host_str(previous_url).expect("ivalid").to_string(), user_input)
+	};
 
-    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(1965));
+		}
+		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" || user_input == "quit" || user_input == "exit" {
 		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(&"gemini://geminiprotocol.net").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" {
+			}
+                        url = history[historical_position].clone();
+			load_page = false;
+
+			let user_input = input(); 
+			if user_input == "q" {
 				break 'mainloop;
 			}
-        else if link_to_follow == "r" {
+                        else if user_input == "d" {
+                            println!(
+                                "load_page: {}\nhistory: {:?}\nhistorical_postition: {}\nlinks: {:?}\nsource: {}",
+                                load_page,
+                                history,
+                                historical_position,
+                                links,
+                                source
+                            );
+                        }
+			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
+d: debug info
+h: help
+r: reload
+s: view source code of page
+i: visit root index of this host eg: root index of gemini://deadvey.com/blog/4.md is just gemini://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.");
+	}
 }
 

src/parse_gemtext.rs

@@ -0,0 +1,91 @@
+use colored::Colorize;
+use regex::Regex;
+
+pub fn parse_gemtext(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 mut preformatted_code_toggle = false;
+
+        // Regex patterns
+        let preformatted_text_regex = Regex::new(r"^```(.*)").unwrap();
+        let header1_regex = Regex::new(r"^# (.*)").unwrap();
+        let header2_regex = Regex::new(r"^## (.*)").unwrap();
+        let header3_regex = Regex::new(r"^### (.*)").unwrap();
+        let block_quotes_regex = Regex::new(r"^>(.*)").unwrap();
+        let unordered_list_regex = Regex::new(r"^([ \t]+|^)(\*).(.*)").unwrap();
+        let hyperlink_regex = Regex::new(r"=>\s(\S*)\s(.*)").unwrap();
+        let quick_hyperlink_regex = Regex::new(r"=>\s(.*)").unwrap();
+	
+        for line in page_content.lines() {
+            let mut parsed_line: String = line.to_string();
+            let mut remove_line = false;
+
+            // preformatted text
+            parsed_line = preformatted_text_regex.replace_all(&parsed_line, |caps: &regex::Captures| {
+                // Flip the toggle
+                preformatted_code_toggle = ! preformatted_code_toggle; 
+
+                if caps[1] == *""
+                {
+                    remove_line = true;
+                }
+                
+                // Remove the ```
+                format!("{}", &caps[1].magenta())
+            }).to_string();
+
+            if preformatted_code_toggle == false 
+            {
+                // Headers
+                parsed_line = header1_regex.replace_all(&parsed_line, |caps: &regex::Captures| {
+                    format!("{}", &caps[1].blue().bold().underline())
+                }).to_string();
+                parsed_line = header2_regex.replace_all(&parsed_line, |caps: &regex::Captures| {
+                    format!("{}", &caps[1].blue().bold())
+                }).to_string();
+                parsed_line = header3_regex.replace_all(&parsed_line, |caps: &regex::Captures| {
+                    format!("{}", &caps[1].bold())
+                }).to_string();
+
+                // Block quotes
+                parsed_line = block_quotes_regex.replace_all(&parsed_line, |caps: &regex::Captures| {
+                    format!("   | {}", &caps[1].red())
+                }).to_string();
+
+                // Unordered list ([    ]+|^)- (.*)
+                parsed_line = unordered_list_regex.replace_all(&parsed_line, |caps: &regex::Captures| {
+                    format!("{} • {}", &caps[1], &caps[3])
+                }).to_string();
+
+                // HyperLink
+                parsed_line = hyperlink_regex.replace_all(&parsed_line, |caps: &regex::Captures| {
+                    // Check if the character before the link is not '!'
+                    let result = format!("[{}] {}", hyperlink_number_counter, &caps[2].blue().underline());
+                    let url = caps[1].to_string();
+                    links.push(url);
+                    hyperlink_number_counter += 1;
+                    result
+                }).to_string();
+                
+                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!("[{}] {}", hyperlink_number_counter, &caps[1].blue().underline())
+                }).to_string();
+            }
+            else if preformatted_code_toggle == true 
+            {
+                parsed_line = parsed_line.magenta().to_string();
+            }
+
+            if remove_line == false
+            {
+                parsed_page_content+=&(parsed_line + "\n");
+            }
+        }
+
+
+    return (parsed_page_content, links);
+}