Add command line param to exclude instances from crawl

This commit is contained in:
Felix Ableitner 2021-06-17 16:48:45 +02:00
parent fefd708777
commit f110554993
3 changed files with 17 additions and 14 deletions

View file

@ -1,6 +1,6 @@
use crate::federated_instances::GetSiteResponse;
use crate::node_info::NodeInfo;
use crate::{EXCLUDE_INSTANCES, REQUEST_TIMEOUT};
use crate::REQUEST_TIMEOUT;
use anyhow::anyhow;
use anyhow::Error;
use futures::try_join;
@ -10,6 +10,7 @@ use std::collections::VecDeque;
pub async fn crawl(
start_instances: Vec<String>,
exclude: Vec<String>,
max_depth: i32,
) -> Result<(Vec<InstanceDetails>, i32), Error> {
let mut pending_instances: VecDeque<CrawlInstance> = start_instances
@ -21,9 +22,7 @@ pub async fn crawl(
let mut failed_instances = 0;
while let Some(current_instance) = pending_instances.pop_back() {
crawled_instances.push(current_instance.domain.clone());
if current_instance.depth > max_depth
|| EXCLUDE_INSTANCES.contains(&&**&current_instance.domain)
{
if current_instance.depth > max_depth || exclude.contains(&current_instance.domain) {
continue;
}
match fetch_instance_details(&current_instance.domain).await {
@ -48,7 +47,7 @@ pub async fn crawl(
// Sort by active monthly users descending
instance_details.sort_by_key(|i| i.users_active_month);
instance_details.reverse();
Ok((instance_details, failed_instances))
}

View file

@ -7,9 +7,5 @@ pub mod node_info;
pub const REQUEST_TIMEOUT: Duration = Duration::from_secs(10);
pub const DEFAULT_START_INSTANCES: &str = "lemmy.ml";
pub const DEFAULT_MAX_CRAWL_DEPTH: &str = "1";
pub const EXCLUDE_INSTANCES: &'static [&str] = &[
"ds9.lemmy.ml",
"enterprise.lemmy.ml",
"voyager.lemmy.ml",
"test.lemmy.ml",
];
pub const EXCLUDE_INSTANCES: &str =
"ds9.lemmy.ml, enterprise.lemmy.ml, voyager.lemmy.ml, test.lemmy.ml";

View file

@ -1,7 +1,7 @@
use anyhow::Error;
use clap::{App, Arg};
use lemmy_stats_crawler::crawl::{crawl, InstanceDetails};
use lemmy_stats_crawler::{DEFAULT_MAX_CRAWL_DEPTH, DEFAULT_START_INSTANCES};
use lemmy_stats_crawler::{DEFAULT_MAX_CRAWL_DEPTH, DEFAULT_START_INSTANCES, EXCLUDE_INSTANCES};
use serde::Serialize;
#[tokio::main]
@ -12,6 +12,7 @@ pub async fn main() -> Result<(), Error> {
.long("start-instances")
.takes_value(true),
)
.arg(Arg::with_name("exclude").long("exclude").takes_value(true))
.arg(
Arg::with_name("max-crawl-depth")
.long("max-crawl-depth")
@ -22,7 +23,13 @@ pub async fn main() -> Result<(), Error> {
.value_of("start-instances")
.unwrap_or(DEFAULT_START_INSTANCES)
.split(',')
.map(|s| s.to_string())
.map(|s| s.trim().to_string())
.collect();
let exclude: Vec<String> = matches
.value_of("exclude")
.unwrap_or(EXCLUDE_INSTANCES)
.split(',')
.map(|s| s.trim().to_string())
.collect();
let max_crawl_depth: i32 = matches
.value_of("max-crawl-depth")
@ -30,7 +37,8 @@ pub async fn main() -> Result<(), Error> {
.parse()?;
eprintln!("Crawling...");
let (instance_details, failed_instances) = crawl(start_instances, max_crawl_depth).await?;
let (instance_details, failed_instances) =
crawl(start_instances, exclude, max_crawl_depth).await?;
let total_stats = aggregate(instance_details, failed_instances);
println!("{}", serde_json::to_string_pretty(&total_stats)?);