mirrors/tabby
1
0
Fork 0
mirror of https://github.com/TabbyML/tabby synced 2024-11-22 08:21:59 +00:00
Code Issues Packages Projects Releases Wiki Activity

chore(scheduler): ignore katana request without status code (#2481)

Browse Source
This commit is contained in:
Meng Zhang 2024-06-23 13:57:30 +08:00 committed by GitHub
parent cb9091289f
commit 141c20966a
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 6 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
crates/tabby-scheduler/src/crawl/mod.rs
View File

@ -46,7 +46,7 @@ async fn crawl_url(start_url: &str) -> anyhow::Result<impl Stream<Item = KatanaR
};
// Skip if the status code is not 200
if data.response.status_code != 200 {
if data.response.status_code != Some(200) {
continue;
}
@ -76,7 +76,7 @@ fn to_document(data: KatanaRequestResponse) -> Option<CrawledDocument> {
let (html, metadata) = {
let (node, metadata) = Readability::new()
.base_url(Url::parse(&data.request.endpoint).ok()?)
.parse(&data.response.body);
.parse(&data.response.body?);
let mut html_bytes = vec![];
node.serialize(&mut html_bytes).ok()?;
@ -134,9 +134,9 @@ mod tests {
raw: "GET / HTTP/1.1\nHost: example.com\n".to_owned(),
},
response: types::KatanaResponse {
status_code: 200,
status_code: Some(200),
headers,
body: "<p>Hello, World!</p>".to_owned(),
body: Some("<p>Hello, World!</p>".to_owned()),
technologies: Default::default(),
raw: "HTTP/1.1 200 OK\nContent-Type: text/html\n".to_owned(),
},

4
crates/tabby-scheduler/src/crawl/types.rs
View File

@ -18,9 +18,9 @@ pub struct KatanaRequest {
#[derive(Deserialize, Debug)]
pub struct KatanaResponse {
pub status_code: u16,
pub status_code: Option<u16>,
pub headers: HashMap<String, String>,
pub body: String,
pub body: Option<String>,
pub technologies: Vec<String>,
pub raw: String,
}