garage-unix-socket/src/api/s3_list.rs

288 lines
7.5 KiB
Rust
Raw Normal View History

2021-02-19 15:44:06 +00:00
use std::collections::{BTreeMap, BTreeSet, HashMap};
2020-04-24 18:47:11 +00:00
use std::fmt::Write;
use std::sync::Arc;
2020-04-24 18:56:00 +00:00
use chrono::{DateTime, NaiveDateTime, SecondsFormat, Utc};
use hyper::{Body, Response};
2020-04-24 18:47:11 +00:00
2021-02-19 15:44:06 +00:00
use garage_util::error::Error as GarageError;
2020-04-24 18:47:11 +00:00
2020-07-07 11:59:22 +00:00
use garage_model::garage::Garage;
2020-07-08 15:33:24 +00:00
use garage_model::object_table::*;
2020-04-24 18:47:11 +00:00
use garage_table::DeletedFilter;
2020-04-28 10:18:14 +00:00
use crate::encoding::*;
2021-02-19 15:44:06 +00:00
use crate::error::*;
#[derive(Debug)]
pub struct ListObjectsQuery {
pub is_v2: bool,
pub bucket: String,
pub delimiter: Option<String>,
pub max_keys: usize,
pub prefix: String,
pub marker: Option<String>,
pub continuation_token: Option<String>,
pub start_after: Option<String>,
pub urlencode_resp: bool,
}
2020-04-24 18:47:11 +00:00
#[derive(Debug)]
struct ListResultInfo {
last_modified: u64,
size: u64,
etag: String,
2020-04-24 18:47:11 +00:00
}
2021-02-19 15:44:06 +00:00
pub fn parse_list_objects_query(
bucket: &str,
params: &HashMap<String, String>,
) -> Result<ListObjectsQuery, Error> {
Ok(ListObjectsQuery {
is_v2: params.get("list-type").map(|x| x == "2").unwrap_or(false),
bucket: bucket.to_string(),
2021-03-15 13:46:37 +00:00
delimiter: params.get("delimiter").filter(|x| !x.is_empty()).cloned(),
2021-02-19 15:44:06 +00:00
max_keys: params
.get("max-keys")
.map(|x| {
x.parse::<usize>()
.ok_or_bad_request("Invalid value for max-keys")
})
.unwrap_or(Ok(1000))?,
prefix: params.get("prefix").cloned().unwrap_or(String::new()),
marker: params.get("marker").cloned(),
continuation_token: params.get("continuation-token").cloned(),
start_after: params.get("start-after").cloned(),
urlencode_resp: params
.get("encoding-type")
.map(|x| x == "url")
.unwrap_or(false),
})
}
2020-04-24 18:47:11 +00:00
pub async fn handle_list(
garage: Arc<Garage>,
2021-02-19 15:44:06 +00:00
query: &ListObjectsQuery,
) -> Result<Response<Body>, Error> {
let mut result_keys = BTreeMap::<String, ListResultInfo>::new();
let mut result_common_prefixes = BTreeSet::<String>::new();
2021-02-19 15:44:06 +00:00
let mut next_chunk_start = if query.is_v2 {
if let Some(ct) = &query.continuation_token {
String::from_utf8(base64::decode(ct.as_bytes())?)?
} else {
query.start_after.clone().unwrap_or(query.prefix.clone())
}
} else {
query.marker.clone().unwrap_or(query.prefix.clone())
};
2020-04-24 18:47:11 +00:00
2021-02-19 15:44:06 +00:00
debug!(
"List request: `{:?}` {} `{}`",
query.delimiter, query.max_keys, query.prefix
);
2020-04-24 18:47:11 +00:00
2020-05-01 15:52:35 +00:00
let truncated;
'query_loop: loop {
2020-04-24 18:47:11 +00:00
let objects = garage
.object_table
.get_range(
2021-02-19 15:44:06 +00:00
&query.bucket,
2020-04-24 18:47:11 +00:00
Some(next_chunk_start.clone()),
Some(DeletedFilter::NotDeleted),
2021-02-19 15:44:06 +00:00
query.max_keys + 1,
2020-04-24 18:47:11 +00:00
)
.await?;
debug!(
"List: get range {} (max {}), results: {}",
next_chunk_start,
2021-02-19 15:44:06 +00:00
query.max_keys + 1,
objects.len()
);
2020-04-24 18:47:11 +00:00
for object in objects.iter() {
2021-02-19 15:44:06 +00:00
if !object.key.starts_with(&query.prefix) {
truncated = None;
2020-05-01 15:52:35 +00:00
break 'query_loop;
}
2021-02-19 15:44:06 +00:00
if query.is_v2 && query.start_after.as_ref() == Some(&object.key) {
continue;
}
2020-04-26 18:55:13 +00:00
if let Some(version) = object.versions().iter().find(|x| x.is_data()) {
2021-02-19 15:44:06 +00:00
if result_keys.len() + result_common_prefixes.len() >= query.max_keys {
truncated = Some(object.key.to_string());
2020-05-04 13:09:23 +00:00
break 'query_loop;
}
2021-02-19 15:44:06 +00:00
let common_prefix = if let Some(delimiter) = &query.delimiter {
let relative_key = &object.key[query.prefix.len()..];
2020-11-11 15:12:42 +00:00
relative_key
.find(delimiter)
2021-02-19 15:44:06 +00:00
.map(|i| &object.key[..query.prefix.len() + i + delimiter.len()])
} else {
None
2020-04-24 18:47:11 +00:00
};
if let Some(pfx) = common_prefix {
result_common_prefixes.insert(pfx.to_string());
} else {
let meta = match &version.state {
ObjectVersionState::Complete(ObjectVersionData::Inline(meta, _)) => meta,
2020-07-08 15:34:37 +00:00
ObjectVersionState::Complete(ObjectVersionData::FirstBlock(meta, _)) => {
meta
2020-07-08 15:34:37 +00:00
}
_ => unreachable!(),
};
let info = match result_keys.get(&object.key) {
None => ListResultInfo {
last_modified: version.timestamp,
size: meta.size,
etag: meta.etag.to_string(),
},
2020-04-26 18:55:13 +00:00
Some(_lri) => {
2021-02-19 15:44:06 +00:00
return Err(Error::InternalError(GarageError::Message(format!(
"Duplicate key?? {}",
object.key
))))
2020-04-26 18:55:13 +00:00
}
};
result_keys.insert(object.key.clone(), info);
2020-04-24 18:47:11 +00:00
};
}
}
2021-02-19 15:44:06 +00:00
if objects.len() < query.max_keys + 1 {
truncated = None;
2020-05-04 13:09:23 +00:00
break 'query_loop;
2020-04-24 18:47:11 +00:00
}
if objects.len() > 0 {
next_chunk_start = objects[objects.len() - 1].key.clone();
}
}
let mut xml = String::new();
writeln!(&mut xml, r#"<?xml version="1.0" encoding="UTF-8"?>"#).unwrap();
writeln!(
&mut xml,
r#"<ListBucketResult xmlns="http://s3.amazonaws.com/doc/2006-03-01/">"#
)
.unwrap();
2021-02-19 15:44:06 +00:00
writeln!(&mut xml, "\t<Name>{}</Name>", query.bucket).unwrap();
// TODO: in V1, is this supposed to be urlencoded when encoding-type is URL??
writeln!(
&mut xml,
"\t<Prefix>{}</Prefix>",
xml_encode_key(&query.prefix, query.urlencode_resp),
)
.unwrap();
if let Some(delim) = &query.delimiter {
// TODO: in V1, is this supposed to be urlencoded when encoding-type is URL??
writeln!(
&mut xml,
"\t<Delimiter>{}</Delimiter>",
xml_encode_key(delim, query.urlencode_resp),
)
.unwrap();
}
writeln!(&mut xml, "\t<MaxKeys>{}</MaxKeys>", query.max_keys).unwrap();
if query.urlencode_resp {
writeln!(&mut xml, "\t<EncodingType>url</EncodingType>").unwrap();
}
2021-02-19 15:44:06 +00:00
writeln!(
&mut xml,
"\t<KeyCount>{}</KeyCount>",
result_keys.len() + result_common_prefixes.len()
)
.unwrap();
writeln!(
&mut xml,
"\t<IsTruncated>{}</IsTruncated>",
truncated.is_some()
)
.unwrap();
2021-02-19 15:44:06 +00:00
if query.is_v2 {
if let Some(ct) = &query.continuation_token {
writeln!(&mut xml, "\t<ContinuationToken>{}</ContinuationToken>", ct).unwrap();
}
if let Some(sa) = &query.start_after {
writeln!(
&mut xml,
"\t<StartAfter>{}</StartAfter>",
xml_encode_key(sa, query.urlencode_resp)
)
.unwrap();
}
if let Some(nct) = truncated {
writeln!(
&mut xml,
"\t<NextContinuationToken>{}</NextContinuationToken>",
base64::encode(nct.as_bytes())
)
.unwrap();
}
} else {
// TODO: are these supposed to be urlencoded when encoding-type is URL??
if let Some(mkr) = &query.marker {
2021-02-23 17:46:25 +00:00
writeln!(
&mut xml,
"\t<Marker>{}</Marker>",
xml_encode_key(mkr, query.urlencode_resp)
)
.unwrap();
2021-02-19 15:44:06 +00:00
}
if let Some(next_marker) = truncated {
2021-02-23 17:46:25 +00:00
writeln!(
&mut xml,
"\t<NextMarker>{}</NextMarker>",
xml_encode_key(&next_marker, query.urlencode_resp)
)
.unwrap();
2021-02-19 15:44:06 +00:00
}
}
2021-02-19 15:44:06 +00:00
for (key, info) in result_keys.iter() {
2020-04-24 18:47:11 +00:00
let last_modif = NaiveDateTime::from_timestamp(info.last_modified as i64 / 1000, 0);
let last_modif = DateTime::<Utc>::from_utc(last_modif, Utc);
let last_modif = last_modif.to_rfc3339_opts(SecondsFormat::Millis, true);
writeln!(&mut xml, "\t<Contents>").unwrap();
2020-04-28 10:18:14 +00:00
writeln!(
&mut xml,
"\t\t<Key>{}</Key>",
2021-02-19 15:44:06 +00:00
xml_encode_key(key, query.urlencode_resp),
2020-04-28 10:18:14 +00:00
)
.unwrap();
2020-04-24 18:47:11 +00:00
writeln!(&mut xml, "\t\t<LastModified>{}</LastModified>", last_modif).unwrap();
writeln!(&mut xml, "\t\t<Size>{}</Size>", info.size).unwrap();
if !info.etag.is_empty() {
writeln!(&mut xml, "\t\t<ETag>\"{}\"</ETag>", info.etag).unwrap();
}
2020-04-24 18:47:11 +00:00
writeln!(&mut xml, "\t\t<StorageClass>STANDARD</StorageClass>").unwrap();
writeln!(&mut xml, "\t</Contents>").unwrap();
}
2021-02-19 15:44:06 +00:00
for pfx in result_common_prefixes.iter() {
writeln!(&mut xml, "\t<CommonPrefixes>").unwrap();
//TODO: in V1, are these urlencoded when urlencode_resp is true ?? (proably)
writeln!(
&mut xml,
"\t\t<Prefix>{}</Prefix>",
xml_encode_key(pfx, query.urlencode_resp),
)
.unwrap();
writeln!(&mut xml, "\t</CommonPrefixes>").unwrap();
}
2021-02-19 15:44:06 +00:00
2020-04-24 18:47:11 +00:00
writeln!(&mut xml, "</ListBucketResult>").unwrap();
debug!("{}", xml);
2020-04-24 18:47:11 +00:00
Ok(Response::builder()
2021-02-23 17:46:25 +00:00
.header("Content-Type", "application/xml")
.body(Body::from(xml.into_bytes()))?)
2020-04-26 18:55:13 +00:00
}