deadvey/markdown-webbrowser
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Compare commits

base: deadvey:1b8614b9569d2d72a1781ff122ad27165ac9b801
Branches Tags
deadvey:master deadvey:gemini
...
compare: deadvey:gemini
Branches Tags
deadvey:gemini deadvey:master

10 Commits
1b8614b956 ... gemini

Author SHA1 Message Date
deadvey
ca9d4cfaf9 readme 2025-04-03 23:19:21 +01:00
deadvey
1321e0dcaf works ish 2025-04-03 22:15:38 +01:00
deadvey
a7468d3f40 Gemini version 2025-04-03 21:42:37 +01:00
deadvey
845866ef9d readme 2025-01-26 11:15:34 +00:00
deadvey
5f856f35fe source code viewing 2025-01-26 02:59:06 +00:00
deadvey
7b9d2d6fd3 some stuff 2025-01-26 02:39:43 +00:00
deadvey
f46d3bac42 fixed history functionality, TO DO: arguments, documentation 2025-01-26 02:12:56 +00:00
deadvey
ced5648c01 url parsing works (FINALLY) 2025-01-26 01:44:24 +00:00
deadvey
7b07b6f051 done for the day, got a proper url parser 2025-01-22 01:18:30 +00:00
deadvey
e4b08b45bc ``` 2025-01-21 23:59:18 +00:00
4 changed files with 252 additions and 263 deletions
Expand all files Collapse all files

2
Cargo.toml
View File

@@ -6,6 +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"

29
README.md
View File

@@ -1,11 +1,36 @@
# Markdown web browser
A web browser that let's you browse 'mttp' websites that use markdown as a superior standard to html
Fully static!
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.

393
src/main.rs
View File

@@ -1,192 +1,72 @@
use std::process::{Command};
use std::io::{stdin,stdout,Write};
use colored::Colorize;
use regex::Regex;
use url::{Url, ParseError};
use std::fs;
struct Url {
protocol: String,
hostname: String,
port: u16,
path: String,
}
const DEBUG_MODE: bool = false;
// Import other files
mod parse_gemtext;
fn clear_screen() {
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
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[4].bold().to_string()
}).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
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();
// 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| {
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])
}).to_string();
// Ordered list
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"^([ \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();
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
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");
if DEBUG_MODE == false
{
println!("clearing");
Command::new("clear")
.status()
.expect("Failed to clear screen");
}
// 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: u16, path: &String) -> String {
let full_url_formatted = format!("{}:{}{}", host, port, path);
fn fetch_page(url: &Url) {
let full_url_formatted = format!("{}", url);
// 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: u16, 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 &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());
}
if source == true {
content += &format!("{}", &"Viewing source code".yellow());
}
else {
(content, links) = parse_gemtext::parse_gemtext(content);
}
for _i in 0..screen_width {
print!("—");
}
print!("{}:{}{}\n", host, port, path);
for _i in 0..screen_width {
print!("—");
}
println!("\n\n{}", content);
for _i in 0..screen_width {
print!("—");
}
println!();
for _i in 0..screen_width {
print!("-");
}
print!("{}\n", url);
for _i in 0..screen_width {
print!("-");
}
println!("\n\n{}", content);
for _i in 0..screen_width {
print!("-");
}
println!();
// Return links (you can add link parsing logic)
return links;
// Return links (you can add link parsing logic)
}
return links;
}
fn input() -> String{
@@ -202,59 +82,31 @@ fn input() -> String{
return s;
}
fn parse_url(user_input: String, previous_host: &String) -> Result<Url, u8> {
let mut url = Url {
protocol: "internal".to_string(),
hostname: "home".to_string(),
port: 0,
path: "/".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:\/\/(.*?)\/(.*?)$").unwrap(); // mttp://example.com/index.md
let mttp_regex_no_path = Regex::new(r"^mttp:\/\/(.*?)$").unwrap(); // mttp://example.com
//let accept_this_as_mttp = Regex::new(r"^(.*?)\.(.*?)$").unwrap(); // example.com
//let no_protocol_but_path = Regex::new(r"^(.*?)/(.*?)$").unwrap(); // example.com/index.md
let path_change = Regex::new(r"^/(.*?)$").unwrap(); // /index.md
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(&user_input) {
url.hostname = caps[1].to_string();
url.port = 3477;
url.path = caps[2].to_string();
url.protocol = "mttp".to_string();
Ok(url)
}
else if let Some(caps) = mttp_regex_no_path.captures(&user_input) {
url.hostname = caps[1].to_string();
url.port = 3477;
url.path = "/".to_string();
url.protocol = "mttp".to_string();
Ok(url)
}
else if let Some(caps) = path_change.captures(&user_input) {
url.hostname = previous_host.to_string();
url.port = 3477;
url.path = format!("/{}", caps[1].to_string());
url.protocol = "mttp".to_string();
Ok(url)
}
//else if let Some(caps) = no_protocol_but_path.captures(&user_input) {
// url.hostname = caps[1].to_string();
//url.port = 3477;
//url.path = caps[2].to_string();
//url.protocol = "mttp".to_string();
//Ok(url)
//}
//else if let Some(caps) = accept_this_as_mttp.captures(&user_input) {
// url.hostname = format!("{}{}{}",caps[1].to_string(),".",caps[2].to_string());
//url.port = 3477;
// url.path = "/".to_string();
// url.protocol = "mttp".to_string();
//Ok(url)
//}
else {
open::that(user_input); // Fallback to open it in the web browser
Err(1)
}
}
fn main() {
@@ -262,64 +114,85 @@ fn main() {
println!("Enter a url: ");
let user_input = input();
if user_input == "q" {
if user_input == "q" || user_input == "quit" || user_input == "exit" {
std::process::exit(0);
}
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 url = Url {
protocol: "internal".to_string(),
hostname: "home".to_string(),
port: 0,
path: "/".to_string(),
};
if let Ok(parsed_value) = parse_url(user_input, &"example.com".to_string()) {
url = parsed_value;
history.push(Url {
protocol: url.protocol.clone(),
hostname: url.hostname.clone(),
port: url.port.clone(),
path: url.path.clone(),
});
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 {
if load_page {
links = Vec::new();
links = render_page(history[historical_position].hostname.clone(), history[historical_position].port.clone(), history[historical_position].path.clone());
links = render_page(history[historical_position].clone(), source);
println!("Enter reference number to follow, h for help, or q to quit");
//for i in 0..history.len() {
// println!("{}://{}:{}/{}",history[i].protocol,history[i].hostname, history[i].port, history[i].path);
//}
}
url = history[historical_position].clone();
load_page = false;
let user_input = input();
if user_input == "q" {
break 'mainloop;
}
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 user_input == "s" {
source = ! source; // Flip the boolean to toggle source mode
load_page = true;
}
else if user_input == "i" {
url.path = "/".to_string();
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 >= 1 {
if historical_position > 0 {
historical_position -= 1;
if let Ok(parsed_value) = parse_url(format!("{}://{}/{}",history[historical_position].protocol.clone(), history[historical_position].hostname.clone(),history[historical_position].path.clone()),&url.hostname) {
url = parsed_value;
load_page = true;
}
else {
println!("Invalid url");
}
load_page = true;
}
else {
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\nq: quit\nh: help\nr: reload\ni: visit root index of this host eg: root index of mttp://deadvey.com/blog/4.md is just deadvey.com\nb: go back in history\nox: print the hyprlink of reference x eg: o5 or o24");
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..];
@@ -331,17 +204,12 @@ fn main() {
}
else if let Ok(number) = user_input.parse::<usize>() {
if number < links.len() {
if let Ok(parsed_value) = parse_url(links[number].clone(), &url.hostname) {
if let Ok(parsed_value) = parse_url(links[number].clone(), &url.clone()) {
url = parsed_value;
history.insert(historical_position+1, Url {
protocol: url.protocol.clone(),
hostname: url.hostname.clone(),
port: url.port.clone(),
path: url.path.clone(),
});
for i in historical_position+1..history.len()-1 {
history.remove(i);
for _i in historical_position+1..history.len() {
history.remove(historical_position+1);
}
history.push(url.clone());
historical_position += 1;
load_page = true;
}
@@ -352,8 +220,13 @@ fn main() {
println!("Invalid reference id");
}
}
else if let Ok(parsed_value) = parse_url(user_input, &url.hostname) {
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 {

91
src/parse_gemtext.rs Normal file
View File

@@ -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);
}