Merge pull request #18 from robatipoor/master

lazy init regex refcator some code cargo clippy and cargo fmt
pull/22/head
Vincent Flyson 5 years ago committed by GitHub
commit 1e0e6347a8
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23

@ -13,3 +13,5 @@ mime-sniffer = "0.1.2"
regex = "1.2.1"
reqwest = "0.9.20"
url = "2.1.0"
lazy_static = "1.3.0"

@ -13,7 +13,7 @@ If compared to saving websites with `wget -mpk`, `monolith` embeds all assets as
### Installation
$ git clone https://github.com/Y2Z/monolith.git
$ cd monolith
$ cargo install
$ cargo install --path .
### Usage
$ monolith https://lyrics.github.io/db/p/portishead/dummy/roads/ > portishead-roads-lyrics.html

@ -1,13 +1,11 @@
extern crate html5ever;
use http::{is_valid_url, resolve_url, retrieve_asset};
use std::default::Default;
use std::io;
use http::{is_url, retrieve_asset, resolve_url};
use self::html5ever::parse_document;
use self::html5ever::rcdom::{Handle, NodeData, RcDom};
use self::html5ever::tendril::TendrilSink;
use self::html5ever::serialize::{SerializeOpts, serialize};
use html5ever::parse_document;
use html5ever::rcdom::{Handle, NodeData, RcDom};
use html5ever::serialize::{serialize, SerializeOpts};
use html5ever::tendril::TendrilSink;
enum NodeMatch {
Icon,
@ -19,19 +17,37 @@ enum NodeMatch {
Other,
}
static PNG_PIXEL: &str = "data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAQAAAC1HAwCAAAAC0lEQVR42mNkYAAAAAYAAjCB0C8AAAAASUVORK5CYII=";
const PNG_PIXEL: &str = "data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAQAAAC1HAwCAAAAC0lEQVR42mNkYAAAAAYAAjCB0C8AAAAASUVORK5CYII=";
static JS_DOM_EVENT_ATTRS: [&str; 21] = [
const JS_DOM_EVENT_ATTRS: [&str; 21] = [
// Input
"onfocus", "onblur", "onselect", "onchange", "onsubmit", "onreset", "onkeydown", "onkeypress", "onkeyup",
"onfocus",
"onblur",
"onselect",
"onchange",
"onsubmit",
"onreset",
"onkeydown",
"onkeypress",
"onkeyup",
// Mouse
"onmouseover", "onmouseout", "onmousedown", "onmouseup", "onmousemove",
"onmouseover",
"onmouseout",
"onmousedown",
"onmouseup",
"onmousemove",
// Click
"onclick", "ondblclick",
"onclick",
"ondblclick",
// Load
"onload", "onunload", "onabort", "onerror", "onresize",
"onload",
"onunload",
"onabort",
"onerror",
"onresize",
];
#[allow(clippy::cognitive_complexity)]
pub fn walk_and_embed_assets(url: &str, node: &Handle, opt_no_js: bool, opt_no_images: bool) {
match node.data {
NodeData::Document => {
@ -39,28 +55,24 @@ pub fn walk_and_embed_assets(url: &str, node: &Handle, opt_no_js: bool, opt_no_i
for child in node.children.borrow().iter() {
walk_and_embed_assets(&url, child, opt_no_js, opt_no_images);
}
},
}
NodeData::Doctype {
name: _,
public_id: _,
system_id: _,
} => {},
NodeData::Doctype { .. } => {}
NodeData::Text { contents: _, } => {},
NodeData::Text { .. } => {}
NodeData::Comment { contents: _, } => {
NodeData::Comment { .. } => {
// Note: in case of opt_no_js being set to true, there's no need to worry about
// getting rid of comments that may contain scripts, e.g. <!--[if IE]><script>...
// since that's not part of W3C standard and gets ignored by browsers other than IE [5, 9]
},
}
NodeData::Element {
ref name,
ref attrs,
..
} => {
let ref mut attrs_mut = attrs.borrow_mut();
let attrs_mut = &mut attrs.borrow_mut();
let mut found = NodeMatch::Other;
if &name.local == "link" {
@ -95,7 +107,7 @@ pub fn walk_and_embed_assets(url: &str, node: &Handle, opt_no_js: bool, opt_no_i
attr.value.push_slice(favicon_datauri.unwrap().as_str());
}
}
},
}
NodeMatch::Image => {
for attr in attrs_mut.iter_mut() {
if &attr.name.local == "src" {
@ -110,7 +122,7 @@ pub fn walk_and_embed_assets(url: &str, node: &Handle, opt_no_js: bool, opt_no_i
}
}
}
},
}
NodeMatch::Anchor => {
for attr in attrs_mut.iter_mut() {
if &attr.name.local == "href" {
@ -124,17 +136,18 @@ pub fn walk_and_embed_assets(url: &str, node: &Handle, opt_no_js: bool, opt_no_i
attr.value.push_slice(href_full_url.unwrap().as_str());
}
}
},
}
NodeMatch::StyleSheet => {
for attr in attrs_mut.iter_mut() {
if &attr.name.local == "href" {
let href_full_url = resolve_url(&url, &attr.value.to_string());
let css_datauri = retrieve_asset(&href_full_url.unwrap(), true, "text/css");
let css_datauri =
retrieve_asset(&href_full_url.unwrap(), true, "text/css");
attr.value.clear();
attr.value.push_slice(css_datauri.unwrap().as_str());
}
}
},
}
NodeMatch::Script => {
if opt_no_js {
// Get rid of src and inner content of SCRIPT tags
@ -148,18 +161,22 @@ pub fn walk_and_embed_assets(url: &str, node: &Handle, opt_no_js: bool, opt_no_i
for attr in attrs_mut.iter_mut() {
if &attr.name.local == "src" {
let src_full_url = resolve_url(&url, &attr.value.to_string());
let js_datauri = retrieve_asset(&src_full_url.unwrap(), true, "application/javascript");
let js_datauri = retrieve_asset(
&src_full_url.unwrap(),
true,
"application/javascript",
);
attr.value.clear();
attr.value.push_slice(js_datauri.unwrap().as_str());
}
}
}
},
}
NodeMatch::Form => {
for attr in attrs_mut.iter_mut() {
if &attr.name.local == "action" {
// Do not touch action props which are set to a URL
if is_url(&attr.value) {
if is_valid_url(&attr.value) {
continue;
}
@ -168,8 +185,8 @@ pub fn walk_and_embed_assets(url: &str, node: &Handle, opt_no_js: bool, opt_no_i
attr.value.push_slice(href_full_url.unwrap().as_str());
}
}
},
NodeMatch::Other => {},
}
NodeMatch::Other => {}
}
if opt_no_js {
@ -179,13 +196,13 @@ pub fn walk_and_embed_assets(url: &str, node: &Handle, opt_no_js: bool, opt_no_i
attr.value.clear();
}
}
}
}
// Dig deeper
for child in node.children.borrow().iter() {
walk_and_embed_assets(&url, child, opt_no_js, opt_no_images);
}
},
}
NodeData::ProcessingInstruction { .. } => unreachable!(),
}

@ -1,53 +1,59 @@
extern crate regex;
extern crate reqwest;
extern crate url;
use self::regex::Regex;
use self::reqwest::Client;
use self::reqwest::header::CONTENT_TYPE;
use regex::Regex;
use reqwest::header::CONTENT_TYPE;
use reqwest::Client;
use std::time::Duration;
use self::url::{Url, ParseError};
use url::{ParseError, Url};
use utils::data_to_dataurl;
pub fn is_url(path: &str) -> bool {
let re = Regex::new(r"^https?://").unwrap();
re.is_match(path)
lazy_static! {
static ref REGEX_URL: Regex = Regex::new(r"^https?://").unwrap();
}
pub fn resolve_url(from: &str, to: &str) -> Result<String, ParseError> {
let mut result = String::new();
pub fn is_valid_url(path: &str) -> bool {
REGEX_URL.is_match(path)
}
if is_url(to) { // (anything, http://site.com/css/main.css)
result = to.to_string();
pub fn resolve_url(from: &str, to: &str) -> Result<String, ParseError> {
let result = if is_valid_url(to) {
// (anything, http://site.com/css/main.css)
to.to_string()
} else {
if is_url(from) { // It's a remote resource (HTTP)
if to.chars().nth(0) == Some('/') { // (http://site.com/article/1, /...?)
let mut re = String::new();
if is_valid_url(from) {
// It's a remote resource (HTTP)
if to.chars().nth(0) == Some('/') {
// (http://site.com/article/1, /...?)
let from_url = Url::parse(from)?;
if to.chars().nth(1) == Some('/') { // (http://site.com/article/1, //images/1.png)
result.push_str(from_url.scheme());
result.push_str(":");
result.push_str(to);
} else { // (http://site.com/article/1, /css/main.css)
result.push_str(from_url.scheme());
result.push_str("://");
result.push_str(from_url.host_str().unwrap());
result.push_str(to);
if to.chars().nth(1) == Some('/') {
// (http://site.com/article/1, //images/1.png)
re.push_str(from_url.scheme());
re.push_str(":");
re.push_str(to);
} else {
// (http://site.com/article/1, /css/main.css)
re.push_str(from_url.scheme());
re.push_str("://");
re.push_str(from_url.host_str().unwrap());
re.push_str(to);
}
} else { // (http://site.com, css/main.css)
} else {
// (http://site.com, css/main.css)
// TODO improve to ensure no // or /// ever happen
result.push_str(from);
result.push_str("/");
result.push_str(to);
re.push_str(from);
re.push_str("/");
re.push_str(to);
}
} else { // It's a local resource (fs)
} else {
// It's a local resource (fs)
// TODO improve to ensure no // or /// ever happen
// TODO for fs use basepath instead of $from
result.push_str(from);
result.push_str("/");
result.push_str(to);
re.push_str(from);
re.push_str("/");
re.push_str(to);
}
}
re
};
Ok(result)
}
@ -55,11 +61,15 @@ pub fn resolve_url(from: &str, to: &str) -> Result<String, ParseError> {
pub fn url_is_data(url: &str) -> Result<bool, String> {
match Url::parse(url) {
Ok(parsed_url) => Ok(parsed_url.scheme() == "data"),
Err(err) => return Err(format!("{}", err)),
Err(err) => Err(format!("{}", err)),
}
}
pub fn retrieve_asset(url: &str, as_dataurl: bool, as_mime: &str) -> Result<String, reqwest::Error> {
pub fn retrieve_asset(
url: &str,
as_dataurl: bool,
as_mime: &str,
) -> Result<String, reqwest::Error> {
if url_is_data(&url).unwrap() {
Ok(url.to_string())
} else {
@ -67,10 +77,7 @@ pub fn retrieve_asset(url: &str, as_dataurl: bool, as_mime: &str) -> Result<Stri
.timeout(Duration::from_secs(10))
.build()
.unwrap();
let mut response = client
.get(url)
.send()
.unwrap();
let mut response = client.get(url).send().unwrap();
if as_dataurl {
// Convert response into a byte array
@ -78,13 +85,15 @@ pub fn retrieve_asset(url: &str, as_dataurl: bool, as_mime: &str) -> Result<Stri
response.copy_to(&mut data)?;
// Attempt to obtain MIME type by reading the Content-Type header
let mut mimetype = as_mime;
if as_mime == "" {
mimetype = response.headers()
let mimetype = if as_mime == "" {
response
.headers()
.get(CONTENT_TYPE)
.and_then(|header| header.to_str().ok())
.unwrap_or(&as_mime);
}
.unwrap_or(&as_mime)
} else {
as_mime
};
Ok(data_to_dataurl(&mimetype, &data))
} else {
@ -99,37 +108,66 @@ mod tests {
#[test]
fn test_is_url() {
assert!(is_url("https://www.rust-lang.org/"));
assert!(is_url("http://kernel.org"));
assert!(!is_url("./index.html"));
assert!(!is_url("some-local-page.htm"));
assert!(!is_url("ftp://1.2.3.4/www/index.html"));
assert!(!is_url("data:text/html;base64,V2VsY29tZSBUbyBUaGUgUGFydHksIDxiPlBhbDwvYj4h"));
assert!(is_valid_url("https://www.rust-lang.org/"));
assert!(is_valid_url("http://kernel.org"));
assert!(!is_valid_url("./index.html"));
assert!(!is_valid_url("some-local-page.htm"));
assert!(!is_valid_url("ftp://1.2.3.4/www/index.html"));
assert!(!is_valid_url(
"data:text/html;base64,V2VsY29tZSBUbyBUaGUgUGFydHksIDxiPlBhbDwvYj4h"
));
}
#[test]
fn test_resolve_url() -> Result<(), ParseError> {
let resolved_url = resolve_url("https://www.kernel.org", "../category/signatures.html")?;
assert_eq!(resolved_url.as_str(), "https://www.kernel.org/../category/signatures.html");
assert_eq!(
resolved_url.as_str(),
"https://www.kernel.org/../category/signatures.html"
);
let resolved_url = resolve_url("https://www.kernel.org", "category/signatures.html")?;
assert_eq!(resolved_url.as_str(), "https://www.kernel.org/category/signatures.html");
let resolved_url = resolve_url("saved_page.htm", "https://www.kernel.org/category/signatures.html")?;
assert_eq!(resolved_url.as_str(), "https://www.kernel.org/category/signatures.html");
let resolved_url = resolve_url("https://www.kernel.org", "//www.kernel.org/theme/images/logos/tux.png")?;
assert_eq!(resolved_url.as_str(), "https://www.kernel.org/theme/images/logos/tux.png");
let resolved_url = resolve_url("https://www.kernel.org/category/signatures.html", "/theme/images/logos/tux.png")?;
assert_eq!(resolved_url.as_str(), "https://www.kernel.org/theme/images/logos/tux.png");
assert_eq!(
resolved_url.as_str(),
"https://www.kernel.org/category/signatures.html"
);
let resolved_url = resolve_url(
"saved_page.htm",
"https://www.kernel.org/category/signatures.html",
)?;
assert_eq!(
resolved_url.as_str(),
"https://www.kernel.org/category/signatures.html"
);
let resolved_url = resolve_url(
"https://www.kernel.org",
"//www.kernel.org/theme/images/logos/tux.png",
)?;
assert_eq!(
resolved_url.as_str(),
"https://www.kernel.org/theme/images/logos/tux.png"
);
let resolved_url = resolve_url(
"https://www.kernel.org/category/signatures.html",
"/theme/images/logos/tux.png",
)?;
assert_eq!(
resolved_url.as_str(),
"https://www.kernel.org/theme/images/logos/tux.png"
);
Ok(())
}
#[test]
fn test_url_is_data() {
assert!(url_is_data("data:text/html;base64,V2VsY29tZSBUbyBUaGUgUGFydHksIDxiPlBhbDwvYj4h").unwrap_or(false));
assert!(
url_is_data("data:text/html;base64,V2VsY29tZSBUbyBUaGUgUGFydHksIDxiPlBhbDwvYj4h")
.unwrap_or(false)
);
assert!(!url_is_data("https://kernel.org").unwrap_or(false));
assert!(!url_is_data("//kernel.org").unwrap_or(false));
}

@ -1,3 +1,10 @@
#[macro_use]
extern crate lazy_static;
extern crate html5ever;
extern crate regex;
extern crate reqwest;
extern crate url;
pub mod html;
pub mod http;
pub mod utils;

@ -2,20 +2,22 @@
extern crate clap;
extern crate monolith;
use clap::{Arg, App};
use monolith::http::{is_url, retrieve_asset};
use monolith::html::{walk_and_embed_assets, html_to_dom, print_dom};
use clap::{App, Arg};
use monolith::html::{html_to_dom, print_dom, walk_and_embed_assets};
use monolith::http::{is_valid_url, retrieve_asset};
fn main() {
let command = App::new("monolith")
.version(crate_version!())
.author(crate_authors!())
.about(crate_description!())
.arg(Arg::with_name("url")
.required(true)
.takes_value(true)
.index(1)
.help("URL to download"))
.arg(
Arg::with_name("url")
.required(true)
.takes_value(true)
.index(1)
.help("URL to download"),
)
.args_from_usage("-j, --no-js 'Excludes JavaScript'")
.args_from_usage("-i, --no-images 'Removes images'")
.get_matches();
@ -25,7 +27,7 @@ fn main() {
let opt_no_js = command.is_present("no-js");
let opt_no_img = command.is_present("no-images");
if is_url(arg_target) {
if is_valid_url(arg_target) {
let data = retrieve_asset(&arg_target, false, "");
let dom = html_to_dom(&data.unwrap());

@ -5,14 +5,11 @@ use self::base64::encode;
use self::mime_sniffer::MimeTypeSniffer;
pub fn data_to_dataurl(mime: &str, data: &[u8]) -> String {
let mimetype: String;
if mime == "" {
mimetype = detect_mimetype(data);
let mimetype = if mime == "" {
detect_mimetype(data)
} else {
mimetype = mime.to_string();
}
mime.to_string()
};
format!("data:{};base64,{}", mimetype, encode(data))
}

Loading…
Cancel
Save