Oliver/internet_mapper
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

work off content-type header

Browse Source
This commit is contained in:
Rushmore75 2025-04-17 09:35:57 -06:00
parent 7fab961d76
commit 647c4cd324
1 changed files with 67 additions and 52 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

119
src/main.rs
View File

@ -19,7 +19,7 @@ use metrics::{counter, gauge};
use metrics_exporter_prometheus::PrometheusBuilder;
use serde::Deserialize;
use surrealdb::{engine::remote::ws::Client, Surreal};
use tokio::{io::AsyncWriteExt, task::JoinSet};
use tokio::{io::{AsyncWriteExt, BufWriter}, task::JoinSet};
use tracing::{debug, error, info, instrument, level_filters::LevelFilter, trace, trace_span, warn};
use tracing_subscriber::{fmt, layer::SubscriberExt, EnvFilter, Layer, Registry};
@ -175,63 +175,78 @@ async fn process(mut site: Website, db: Surreal<Client>, reqwest: reqwest::Clien
if let Ok(response) = request_builder.send().await {
// Get body from response
let headers = response.headers();
let ct = headers.get("Content-Type");
if let Some(ct) = ct {
let path = filesystem::as_path(&site.site, ct);
#[allow(non_snake_case)]
let CT = headers.get("Content-Type");
let ct = headers.get("content-type");
// make sure that the file is good to go
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();
if CT.is_none() && ct.is_none() {
}
let ct = match (CT,ct) {
(None, None) => {
warn!("Server did not respond with Content-Type header. Url: {} Headers: ({:?})", site.site.to_string(), headers);
return
},
(None, Some(a)) => a,
(Some(a), None) => a,
(Some(a), Some(_)) => a,
};
// 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)
},
}
let path = filesystem::as_path(&site.site, ct);
// make sure that the file is good to go
if let Some(file) = filesystem::init(&path).await {
let should_parse = path.to_string_lossy().ends_with(".html");
let mut buf: Vec<u8> = Vec::new();
let mut writer = BufWriter::new(file);
// stream the response onto the disk
let mut stream = response.bytes_stream();
info!("Writing at: {:?}", path);
while let Some(data) = stream.next().await {
match data {
Ok(data) => {
let _ = writer.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 {
// Parse document and get relationships
let sites = parser::parse(&site, &buf).await;
// De-duplicate this list
let prev_len = sites.len();
let set = sites.into_iter().fold(HashSet::new(), |mut set, item| {
set.insert(item);
set
});
let de_dupe_sites: Vec<Website> = set.into_iter().collect();
let diff = prev_len - de_dupe_sites.len();
trace!("Saved {diff} from being entered into the db by de-duping");
// Store all the other sites so that we can link to them.
let _ = Website::store_all(de_dupe_sites, &db).await;
}
// METRICS
g.decrement(1);
counter!(GET_METRIC).increment(1);
// update self in db
site.set_crawled();
Website::store_all(vec![site], &db).await;
} else {
error!("File failed to cooperate: {:?}", path);
}
let _ = writer.flush();
if should_parse {
// Parse document and get relationships
let sites = parser::parse(&site, &buf).await;
// De-duplicate this list
let prev_len = sites.len();
let set = sites.into_iter().fold(HashSet::new(), |mut set, item| {
set.insert(item);
set
});
let de_dupe_sites: Vec<Website> = set.into_iter().collect();
let diff = prev_len - de_dupe_sites.len();
trace!("Saved {diff} from being entered into the db by de-duping");
// Store all the other sites so that we can link to them.
let _ = Website::store_all(de_dupe_sites, &db).await;
}
// METRICS
g.decrement(1);
counter!(GET_METRIC).increment(1);
// update self in db
site.set_crawled();
Website::store_all(vec![site], &db).await;
} else {
warn!("Server did not respond with Content-Type header: {}", site.site.to_string());
error!("File failed to cooperate: {:?}", path);
}
} else {
error!("Failed to get: {}", &site.site);