steam data to the disk
This commit is contained in:
		
							
								
								
									
										1
									
								
								Cargo.lock
									
									
									
										generated
									
									
									
								
							
							
						
						
									
										1
									
								
								Cargo.lock
									
									
									
										generated
									
									
									
								
							@@ -1966,6 +1966,7 @@ name = "internet_mapper"
 | 
				
			|||||||
version = "0.1.0"
 | 
					version = "0.1.0"
 | 
				
			||||||
dependencies = [
 | 
					dependencies = [
 | 
				
			||||||
 "base64 0.22.1",
 | 
					 "base64 0.22.1",
 | 
				
			||||||
 | 
					 "futures-util",
 | 
				
			||||||
 "html5ever 0.29.1",
 | 
					 "html5ever 0.29.1",
 | 
				
			||||||
 "metrics",
 | 
					 "metrics",
 | 
				
			||||||
 "metrics-exporter-prometheus",
 | 
					 "metrics-exporter-prometheus",
 | 
				
			||||||
 
 | 
				
			|||||||
@@ -5,12 +5,13 @@ edition = "2021"
 | 
				
			|||||||
 | 
					
 | 
				
			||||||
[dependencies]
 | 
					[dependencies]
 | 
				
			||||||
base64 = "0.22.1"
 | 
					base64 = "0.22.1"
 | 
				
			||||||
 | 
					futures-util = "0.3.31"
 | 
				
			||||||
html5ever = "0.29"
 | 
					html5ever = "0.29"
 | 
				
			||||||
metrics = "0.24.1"
 | 
					metrics = "0.24.1"
 | 
				
			||||||
metrics-exporter-prometheus = { version = "0.16.2", features=["http-listener"]}
 | 
					metrics-exporter-prometheus = { version = "0.16.2", features=["http-listener"]}
 | 
				
			||||||
# minio = "0.1.0"
 | 
					# minio = "0.1.0"
 | 
				
			||||||
minio = {git="https://github.com/minio/minio-rs.git", rev = "c28f576"}
 | 
					minio = {git="https://github.com/minio/minio-rs.git", rev = "c28f576"}
 | 
				
			||||||
reqwest = { version = "0.12", features = ["gzip", "default", "rustls-tls"] }
 | 
					reqwest = { version = "0.12", features = ["gzip", "default", "rustls-tls", "stream"] }
 | 
				
			||||||
serde = { version = "1.0", features = ["derive"] }
 | 
					serde = { version = "1.0", features = ["derive"] }
 | 
				
			||||||
surrealdb = "2.2"
 | 
					surrealdb = "2.2"
 | 
				
			||||||
tokio = { version="1.41.0", features = ["full"] }
 | 
					tokio = { version="1.41.0", features = ["full"] }
 | 
				
			||||||
 
 | 
				
			|||||||
@@ -1,14 +1,10 @@
 | 
				
			|||||||
use std::{ffi::OsStr, path::PathBuf};
 | 
					use std::{ffi::OsStr, io::ErrorKind, path::PathBuf};
 | 
				
			||||||
 | 
					
 | 
				
			||||||
use tokio::fs;
 | 
					use tokio::fs;
 | 
				
			||||||
use tracing::{debug, error, instrument, trace, warn};
 | 
					use tracing::{error, trace};
 | 
				
			||||||
use url::Url;
 | 
					use url::Url;
 | 
				
			||||||
 | 
					
 | 
				
			||||||
#[instrument(skip(data))]
 | 
					pub fn as_path(url: &Url) -> PathBuf {
 | 
				
			||||||
/// Returns whether or not the saved file should be parsed.
 | 
					 | 
				
			||||||
/// If the file is just data, like an image, it doesn't need to be parsed.
 | 
					 | 
				
			||||||
/// If it's html, then it does need to be parsed.
 | 
					 | 
				
			||||||
pub async fn store(data: &str, url: &Url) -> bool {
 | 
					 | 
				
			||||||
    // extract data from url to save it accurately
 | 
					    // extract data from url to save it accurately
 | 
				
			||||||
    let url_path = PathBuf::from("./downloaded/".to_string() + url.domain().unwrap_or("UnknownDomain") + url.path());
 | 
					    let url_path = PathBuf::from("./downloaded/".to_string() + url.domain().unwrap_or("UnknownDomain") + url.path());
 | 
				
			||||||
 | 
					
 | 
				
			||||||
@@ -24,20 +20,42 @@ pub async fn store(data: &str, url: &Url) -> bool {
 | 
				
			|||||||
        (url_path.clone(), "index.html".into())
 | 
					        (url_path.clone(), "index.html".into())
 | 
				
			||||||
    };
 | 
					    };
 | 
				
			||||||
 | 
					
 | 
				
			||||||
    let should_parse = filename.ends_with(".html");
 | 
					    let mut path = PathBuf::new();
 | 
				
			||||||
 | 
					    path = path.join(basepath);
 | 
				
			||||||
 | 
					    path = path.join(filename);
 | 
				
			||||||
 | 
					
 | 
				
			||||||
    debug!("Writing at: {:?} {:?}", basepath, filename);
 | 
					    path
 | 
				
			||||||
 | 
					}
 | 
				
			||||||
 | 
					
 | 
				
			||||||
 | 
					pub async fn init(filename: &PathBuf) -> Option<fs::File> {
 | 
				
			||||||
 | 
					    let file = async || tokio::fs::OpenOptions::new()
 | 
				
			||||||
 | 
					        .append(true)
 | 
				
			||||||
 | 
					        .create(true)
 | 
				
			||||||
 | 
					        .open(&filename).await;
 | 
				
			||||||
 | 
					
 | 
				
			||||||
 | 
					    match file().await {
 | 
				
			||||||
 | 
					        Ok(ok) => Some(ok),
 | 
				
			||||||
 | 
					        Err(err) => {
 | 
				
			||||||
 | 
					            // the file/folder isn't found
 | 
				
			||||||
 | 
					            if err.kind() == ErrorKind::NotFound {
 | 
				
			||||||
 | 
					                if let Some(parent ) = &filename.parent() {
 | 
				
			||||||
                    // create the folders
 | 
					                    // create the folders
 | 
				
			||||||
    if let Err(err) = fs::create_dir_all(&basepath).await {
 | 
					                    if let Err(err) = fs::create_dir_all(&parent).await {
 | 
				
			||||||
        error!("Dir creation: {err} {:?}", basepath);
 | 
					                        error!("Dir creation: {err} {:?}", filename);
 | 
				
			||||||
 | 
					                        eprintln!("{}", err)
 | 
				
			||||||
 | 
					                    } else if let Ok(ok) = file().await {
 | 
				
			||||||
 | 
					                        return Some(ok);
 | 
				
			||||||
 | 
					                    }
 | 
				
			||||||
                } else {
 | 
					                } else {
 | 
				
			||||||
        if let Err(err) = fs::write(&basepath.join(filename), data).await {
 | 
					                    error!("Couldn't get file's parents: {:?}", &filename);
 | 
				
			||||||
            error!("File creation: {err} {:?}", url_path);
 | 
					                }
 | 
				
			||||||
 | 
					            } else {
 | 
				
			||||||
 | 
					                error!("File creation: {err} {:?}", filename);
 | 
				
			||||||
 | 
					            }
 | 
				
			||||||
 | 
					            // we don't care about other errors, we can't/shouldn't fix them
 | 
				
			||||||
 | 
					            None
 | 
				
			||||||
        }
 | 
					        }
 | 
				
			||||||
    }
 | 
					    }
 | 
				
			||||||
 | 
					 | 
				
			||||||
    should_parse
 | 
					 | 
				
			||||||
}
 | 
					}
 | 
				
			||||||
 | 
					
 | 
				
			||||||
fn valid_file_extension(take: &&OsStr) -> bool {
 | 
					fn valid_file_extension(take: &&OsStr) -> bool {
 | 
				
			||||||
 
 | 
				
			|||||||
							
								
								
									
										90
									
								
								src/main.rs
									
									
									
									
									
								
							
							
						
						
									
										90
									
								
								src/main.rs
									
									
									
									
									
								
							@@ -1,9 +1,17 @@
 | 
				
			|||||||
#![feature(ip_from)]
 | 
					#![feature(ip_from)]
 | 
				
			||||||
 | 
					#![feature(async_closure)]
 | 
				
			||||||
 | 
					#![warn(clippy::expect_used)]
 | 
				
			||||||
 | 
					#![deny(clippy::unwrap_used)]
 | 
				
			||||||
 | 
					
 | 
				
			||||||
extern crate html5ever;
 | 
					extern crate html5ever;
 | 
				
			||||||
 | 
					
 | 
				
			||||||
 | 
					use futures_util::StreamExt;
 | 
				
			||||||
 | 
					
 | 
				
			||||||
use std::{
 | 
					use std::{
 | 
				
			||||||
    collections::HashSet, fs::File, io::Read, net::{IpAddr, Ipv4Addr}
 | 
					    collections::HashSet,
 | 
				
			||||||
 | 
					    fs::File,
 | 
				
			||||||
 | 
					    io::Read,
 | 
				
			||||||
 | 
					    net::{IpAddr, Ipv4Addr},
 | 
				
			||||||
};
 | 
					};
 | 
				
			||||||
 | 
					
 | 
				
			||||||
use db::{connect, Website};
 | 
					use db::{connect, Website};
 | 
				
			||||||
@@ -11,21 +19,19 @@ use metrics::{counter, gauge};
 | 
				
			|||||||
use metrics_exporter_prometheus::PrometheusBuilder;
 | 
					use metrics_exporter_prometheus::PrometheusBuilder;
 | 
				
			||||||
use serde::Deserialize;
 | 
					use serde::Deserialize;
 | 
				
			||||||
use surrealdb::{engine::remote::ws::Client, Surreal};
 | 
					use surrealdb::{engine::remote::ws::Client, Surreal};
 | 
				
			||||||
use tokio::task::JoinSet;
 | 
					use tokio::{io::AsyncWriteExt, task::JoinSet};
 | 
				
			||||||
use tracing::{debug, error, info, instrument, level_filters::LevelFilter, trace, trace_span};
 | 
					use tracing::{debug, error, info, instrument, level_filters::LevelFilter, trace, trace_span};
 | 
				
			||||||
use tracing_subscriber::{fmt, layer::SubscriberExt, EnvFilter, Layer, Registry};
 | 
					use tracing_subscriber::{fmt, layer::SubscriberExt, EnvFilter, Layer, Registry};
 | 
				
			||||||
 | 
					
 | 
				
			||||||
mod db;
 | 
					mod db;
 | 
				
			||||||
mod parser;
 | 
					 | 
				
			||||||
mod filesystem;
 | 
					mod filesystem;
 | 
				
			||||||
 | 
					mod parser;
 | 
				
			||||||
 | 
					
 | 
				
			||||||
const GET_METRIC: &str = "total_gets";
 | 
					const GET_METRIC: &str = "total_gets";
 | 
				
			||||||
const GET_IN_FLIGHT: &str = "gets_in_flight";
 | 
					const GET_IN_FLIGHT: &str = "gets_in_flight";
 | 
				
			||||||
const SITES_CRAWLED: &str = "pages_crawled";
 | 
					const SITES_CRAWLED: &str = "pages_crawled";
 | 
				
			||||||
const BEING_PROCESSED: &str = "pages_being_processed";
 | 
					const BEING_PROCESSED: &str = "pages_being_processed";
 | 
				
			||||||
 | 
					
 | 
				
			||||||
const BATCH_SIZE: usize = 2;
 | 
					 | 
				
			||||||
 | 
					 | 
				
			||||||
#[derive(Deserialize)]
 | 
					#[derive(Deserialize)]
 | 
				
			||||||
struct Config {
 | 
					struct Config {
 | 
				
			||||||
    surreal_ns: String,
 | 
					    surreal_ns: String,
 | 
				
			||||||
@@ -37,6 +43,7 @@ struct Config {
 | 
				
			|||||||
    crawl_filter: String,
 | 
					    crawl_filter: String,
 | 
				
			||||||
    start_url: String,
 | 
					    start_url: String,
 | 
				
			||||||
    budget: usize,
 | 
					    budget: usize,
 | 
				
			||||||
 | 
					    batch_size: usize,
 | 
				
			||||||
}
 | 
					}
 | 
				
			||||||
 | 
					
 | 
				
			||||||
#[tokio::main]
 | 
					#[tokio::main]
 | 
				
			||||||
@@ -111,7 +118,8 @@ async fn main() {
 | 
				
			|||||||
    let span = trace_span!("Loop");
 | 
					    let span = trace_span!("Loop");
 | 
				
			||||||
    let span = span.enter();
 | 
					    let span = span.enter();
 | 
				
			||||||
    while crawled < config.budget {
 | 
					    while crawled < config.budget {
 | 
				
			||||||
        let uncrawled = get_uncrawled_links(&db, config.budget - crawled, config.crawl_filter.clone()).await;
 | 
					        let uncrawled =
 | 
				
			||||||
 | 
					            get_uncrawled_links(&db, config.budget - crawled, config.crawl_filter.clone(), &config).await;
 | 
				
			||||||
        if uncrawled.is_empty() {
 | 
					        if uncrawled.is_empty() {
 | 
				
			||||||
            info!("Had more budget but finished crawling everything.");
 | 
					            info!("Had more budget but finished crawling everything.");
 | 
				
			||||||
            return;
 | 
					            return;
 | 
				
			||||||
@@ -137,14 +145,15 @@ async fn main() {
 | 
				
			|||||||
    }
 | 
					    }
 | 
				
			||||||
    drop(span);
 | 
					    drop(span);
 | 
				
			||||||
 | 
					
 | 
				
			||||||
    if let Ok(mut ok) = db.query("count(select id from website where crawled = true)").await {
 | 
					    if let Ok(mut ok) = db
 | 
				
			||||||
 | 
					        .query("count(select id from website where crawled = true)")
 | 
				
			||||||
 | 
					        .await
 | 
				
			||||||
 | 
					    {
 | 
				
			||||||
        let res = ok.take::<Option<usize>>(0);
 | 
					        let res = ok.take::<Option<usize>>(0);
 | 
				
			||||||
        if let Ok(i) = res {
 | 
					        if let Ok(Some(n)) = res {
 | 
				
			||||||
            if let Some(n) = i {
 | 
					 | 
				
			||||||
            info!("Total crawled pages now equals {n}");
 | 
					            info!("Total crawled pages now equals {n}");
 | 
				
			||||||
        }
 | 
					        }
 | 
				
			||||||
    }
 | 
					    }
 | 
				
			||||||
    }
 | 
					 | 
				
			||||||
 | 
					
 | 
				
			||||||
    info!("Done");
 | 
					    info!("Done");
 | 
				
			||||||
}
 | 
					}
 | 
				
			||||||
@@ -153,7 +162,6 @@ async fn main() {
 | 
				
			|||||||
/// Downloads and crawls and stores a webpage.
 | 
					/// Downloads and crawls and stores a webpage.
 | 
				
			||||||
/// It is acceptable to clone `db`, `reqwest`, and `s3` because they all use `Arc`s internally. - Noted by Oliver
 | 
					/// It is acceptable to clone `db`, `reqwest`, and `s3` because they all use `Arc`s internally. - Noted by Oliver
 | 
				
			||||||
async fn process(mut site: Website, db: Surreal<Client>, reqwest: reqwest::Client) {
 | 
					async fn process(mut site: Website, db: Surreal<Client>, reqwest: reqwest::Client) {
 | 
				
			||||||
    
 | 
					 | 
				
			||||||
    // METRICS
 | 
					    // METRICS
 | 
				
			||||||
    trace!("Process: {}", &site.site);
 | 
					    trace!("Process: {}", &site.site);
 | 
				
			||||||
    // Build the request
 | 
					    // Build the request
 | 
				
			||||||
@@ -165,60 +173,75 @@ async fn process(mut site: Website, db: Surreal<Client>, reqwest: reqwest::Clien
 | 
				
			|||||||
 | 
					
 | 
				
			||||||
    // Send the http request (get)
 | 
					    // Send the http request (get)
 | 
				
			||||||
    if let Ok(response) = request_builder.send().await {
 | 
					    if let Ok(response) = request_builder.send().await {
 | 
				
			||||||
 | 
					 | 
				
			||||||
        // TODO if this will fail if the object we are downloading is
 | 
					 | 
				
			||||||
        // larger than the memory of the device it's running on.
 | 
					 | 
				
			||||||
        // We should store it *as* we download it then parse it in-place.
 | 
					 | 
				
			||||||
        // Get body from response
 | 
					        // Get body from response
 | 
				
			||||||
        let data = response
 | 
					 | 
				
			||||||
            .text()
 | 
					 | 
				
			||||||
            .await
 | 
					 | 
				
			||||||
            .expect("Failed to read http response's body!");
 | 
					 | 
				
			||||||
 | 
					
 | 
				
			||||||
        // METRICS
 | 
					        let path = filesystem::as_path(&site.site);
 | 
				
			||||||
        g.decrement(1);
 | 
					 | 
				
			||||||
        counter!(GET_METRIC).increment(1);
 | 
					 | 
				
			||||||
 | 
					
 | 
				
			||||||
        // Store document
 | 
					        // make sure that the file is good to go
 | 
				
			||||||
        let should_parse = filesystem::store(&data, &site.site).await;
 | 
					        if let Some(mut file) = filesystem::init(&path).await {
 | 
				
			||||||
 | 
					            let should_parse = path.to_string_lossy().ends_with(".html");
 | 
				
			||||||
 | 
					            let mut buf: Vec<u8> = Vec::new();
 | 
				
			||||||
 | 
					
 | 
				
			||||||
 | 
					            // stream the response onto the disk
 | 
				
			||||||
 | 
					            let mut stream = response.bytes_stream();
 | 
				
			||||||
 | 
					            while let Some(data) = stream.next().await {
 | 
				
			||||||
 | 
					                match data {
 | 
				
			||||||
 | 
					                    Ok(data) => {
 | 
				
			||||||
 | 
					                        debug!("Writing at: {:?}", path);
 | 
				
			||||||
 | 
					                        let _ = file.write_all(&data).await;
 | 
				
			||||||
 | 
					                        // If we are going to parse this file later, we will save it
 | 
				
			||||||
 | 
					                        // into memory as well as the disk.
 | 
				
			||||||
 | 
					                        if should_parse {
 | 
				
			||||||
 | 
					                            data.iter().for_each(|f| buf.push(*f));
 | 
				
			||||||
 | 
					                        }
 | 
				
			||||||
 | 
					                    },
 | 
				
			||||||
 | 
					                    Err(err) => {
 | 
				
			||||||
 | 
					                        eprintln!("{}", err)
 | 
				
			||||||
 | 
					                    },
 | 
				
			||||||
 | 
					                }
 | 
				
			||||||
 | 
					            }
 | 
				
			||||||
 | 
					
 | 
				
			||||||
            if should_parse {
 | 
					            if should_parse {
 | 
				
			||||||
                // Parse document and get relationships
 | 
					                // Parse document and get relationships
 | 
				
			||||||
            let sites = parser::parse(&site, &data).await;
 | 
					                let sites = parser::parse(&site, &buf).await;
 | 
				
			||||||
        
 | 
					 | 
				
			||||||
                // De-duplicate this list
 | 
					                // De-duplicate this list
 | 
				
			||||||
                let prev_len = sites.len();
 | 
					                let prev_len = sites.len();
 | 
				
			||||||
            let set = sites.into_iter().fold(HashSet::new(), |mut set,item| {
 | 
					                let set = sites.into_iter().fold(HashSet::new(), |mut set, item| {
 | 
				
			||||||
                    set.insert(item);
 | 
					                    set.insert(item);
 | 
				
			||||||
                    set
 | 
					                    set
 | 
				
			||||||
                });
 | 
					                });
 | 
				
			||||||
                let de_dupe_sites: Vec<Website> = set.into_iter().collect();
 | 
					                let de_dupe_sites: Vec<Website> = set.into_iter().collect();
 | 
				
			||||||
                let diff = prev_len - de_dupe_sites.len();
 | 
					                let diff = prev_len - de_dupe_sites.len();
 | 
				
			||||||
                trace!("Saved {diff} from being entered into the db by de-duping");
 | 
					                trace!("Saved {diff} from being entered into the db by de-duping");
 | 
				
			||||||
 | 
					 | 
				
			||||||
                // Store all the other sites so that we can link to them.
 | 
					                // Store all the other sites so that we can link to them.
 | 
				
			||||||
                let _ = Website::store_all(de_dupe_sites, &db).await;
 | 
					                let _ = Website::store_all(de_dupe_sites, &db).await;
 | 
				
			||||||
            }
 | 
					            }
 | 
				
			||||||
 | 
					
 | 
				
			||||||
 | 
					            // METRICS
 | 
				
			||||||
 | 
					            g.decrement(1);
 | 
				
			||||||
 | 
					            counter!(GET_METRIC).increment(1);
 | 
				
			||||||
 | 
					
 | 
				
			||||||
            // update self in db
 | 
					            // update self in db
 | 
				
			||||||
            site.set_crawled();
 | 
					            site.set_crawled();
 | 
				
			||||||
            Website::store_all(vec![site], &db).await;
 | 
					            Website::store_all(vec![site], &db).await;
 | 
				
			||||||
 | 
					        } else {
 | 
				
			||||||
 | 
					            error!("File failed to cooperate: {:?}", path);
 | 
				
			||||||
 | 
					        }
 | 
				
			||||||
    } else {
 | 
					    } else {
 | 
				
			||||||
        error!("Failed to get: {}", &site.site);
 | 
					        error!("Failed to get: {}", &site.site);
 | 
				
			||||||
    }
 | 
					    }
 | 
				
			||||||
}
 | 
					}
 | 
				
			||||||
 | 
					
 | 
				
			||||||
/// Returns uncrawled links
 | 
					/// Returns uncrawled links
 | 
				
			||||||
#[instrument(skip(db))]
 | 
					#[instrument(skip(db, config))]
 | 
				
			||||||
async fn get_uncrawled_links(
 | 
					async fn get_uncrawled_links(
 | 
				
			||||||
    db: &Surreal<Client>,
 | 
					    db: &Surreal<Client>,
 | 
				
			||||||
    mut count: usize,
 | 
					    mut count: usize,
 | 
				
			||||||
    filter: String,
 | 
					    filter: String,
 | 
				
			||||||
 | 
					    config: &Config,
 | 
				
			||||||
) -> Vec<Website> {
 | 
					) -> Vec<Website> {
 | 
				
			||||||
 | 
					    if count > config.batch_size {
 | 
				
			||||||
    if count > BATCH_SIZE {
 | 
					        count = config.batch_size;
 | 
				
			||||||
        count = BATCH_SIZE;
 | 
					 | 
				
			||||||
    }
 | 
					    }
 | 
				
			||||||
 | 
					
 | 
				
			||||||
    debug!("Getting uncrawled links");
 | 
					    debug!("Getting uncrawled links");
 | 
				
			||||||
@@ -233,4 +256,3 @@ async fn get_uncrawled_links(
 | 
				
			|||||||
        .take(0)
 | 
					        .take(0)
 | 
				
			||||||
        .expect("Returned websites couldn't be parsed")
 | 
					        .expect("Returned websites couldn't be parsed")
 | 
				
			||||||
}
 | 
					}
 | 
				
			||||||
 | 
					 | 
				
			||||||
 
 | 
				
			|||||||
@@ -1,5 +1,4 @@
 | 
				
			|||||||
use std::default::Default;
 | 
					use std::default::Default;
 | 
				
			||||||
use std::str::FromStr;
 | 
					 | 
				
			||||||
 | 
					
 | 
				
			||||||
use html5ever::tokenizer::{BufferQueue, TokenizerResult};
 | 
					use html5ever::tokenizer::{BufferQueue, TokenizerResult};
 | 
				
			||||||
use html5ever::tokenizer::{StartTag, TagToken};
 | 
					use html5ever::tokenizer::{StartTag, TagToken};
 | 
				
			||||||
@@ -63,12 +62,14 @@ impl TokenSink for Website {
 | 
				
			|||||||
 | 
					
 | 
				
			||||||
#[instrument(skip_all)]
 | 
					#[instrument(skip_all)]
 | 
				
			||||||
/// Parses the passed site and returns all the sites it links to.
 | 
					/// Parses the passed site and returns all the sites it links to.
 | 
				
			||||||
pub async fn parse(site: &Website, data: &str) -> Vec<Website> {
 | 
					pub async fn parse(site: &Website, data: &[u8]) -> Vec<Website> {
 | 
				
			||||||
 | 
					    debug!("Parsing {}", site.site.to_string());
 | 
				
			||||||
    // prep work
 | 
					    // prep work
 | 
				
			||||||
    let mut other_sites: Vec<Website> = Vec::new();
 | 
					    let mut other_sites: Vec<Website> = Vec::new();
 | 
				
			||||||
 | 
					
 | 
				
			||||||
    // change data into something that can be tokenized
 | 
					    // change data into something that can be tokenized
 | 
				
			||||||
    let chunk = Tendril::from_str(data).expect("Failed to parse string into Tendril!");
 | 
					    let s: Result<Tendril<fmt::UTF8>, ()> = Tendril::try_from_byte_slice(data);
 | 
				
			||||||
 | 
					    if let Ok(chunk) = s {
 | 
				
			||||||
        // create buffer of tokens and push our input into it
 | 
					        // create buffer of tokens and push our input into it
 | 
				
			||||||
        let token_buffer = BufferQueue::default();
 | 
					        let token_buffer = BufferQueue::default();
 | 
				
			||||||
        token_buffer.push_back(
 | 
					        token_buffer.push_back(
 | 
				
			||||||
@@ -86,6 +87,9 @@ pub async fn parse(site: &Website, data: &str) -> Vec<Website> {
 | 
				
			|||||||
        }
 | 
					        }
 | 
				
			||||||
        assert!(token_buffer.is_empty());
 | 
					        assert!(token_buffer.is_empty());
 | 
				
			||||||
        tokenizer.end();
 | 
					        tokenizer.end();
 | 
				
			||||||
 | 
					    } else {
 | 
				
			||||||
 | 
					        warn!("Tendril failed to parse on: {}", site.site.to_string());
 | 
				
			||||||
 | 
					    }
 | 
				
			||||||
 | 
					
 | 
				
			||||||
    other_sites
 | 
					    other_sites
 | 
				
			||||||
}
 | 
					}
 | 
				
			||||||
@@ -107,7 +111,7 @@ fn try_get_url(parent: &Url, link: &str) -> Option<Url> {
 | 
				
			|||||||
                match Url::parse(&format!("{scheme}://{}", link)) {
 | 
					                match Url::parse(&format!("{scheme}://{}", link)) {
 | 
				
			||||||
                    Ok(url) => Some(url),
 | 
					                    Ok(url) => Some(url),
 | 
				
			||||||
                    Err(err) => {
 | 
					                    Err(err) => {
 | 
				
			||||||
                        error!("Failed parsing realative scheme url: {}", err);
 | 
					                        error!("Failed parsing relative scheme url: {}", err);
 | 
				
			||||||
                        None
 | 
					                        None
 | 
				
			||||||
                    }
 | 
					                    }
 | 
				
			||||||
                }
 | 
					                }
 | 
				
			||||||
@@ -117,10 +121,13 @@ fn try_get_url(parent: &Url, link: &str) -> Option<Url> {
 | 
				
			|||||||
                match e {
 | 
					                match e {
 | 
				
			||||||
                    url::ParseError::RelativeUrlWithoutBase => {
 | 
					                    url::ParseError::RelativeUrlWithoutBase => {
 | 
				
			||||||
                        // Is: scheme://host:port
 | 
					                        // Is: scheme://host:port
 | 
				
			||||||
                        let origin = parent.origin().ascii_serialization();
 | 
					                        let mut origin = parent.origin().ascii_serialization();
 | 
				
			||||||
 | 
					                        if !origin.ends_with('/') && !link.starts_with('/') {
 | 
				
			||||||
 | 
					                            origin += "/";
 | 
				
			||||||
 | 
					                        }
 | 
				
			||||||
                        let url = origin.clone() + link;
 | 
					                        let url = origin.clone() + link;
 | 
				
			||||||
 | 
					
 | 
				
			||||||
                        trace!("Built `{url}` from `{origin} + {}`", link.to_string());
 | 
					                        trace!("Built `{url}` from `{origin} + `{}`", link.to_string());
 | 
				
			||||||
 | 
					
 | 
				
			||||||
                        if let Ok(url) = Url::parse(&url) {
 | 
					                        if let Ok(url) = Url::parse(&url) {
 | 
				
			||||||
                            trace!("Saved relative url `{}` AS: `{}`", link, url);
 | 
					                            trace!("Saved relative url `{}` AS: `{}`", link, url);
 | 
				
			||||||
 
 | 
				
			|||||||
		Reference in New Issue
	
	Block a user