searchengines-web/MeiliSearch
searchengines-web/MeiliSearch
1
0
Fork 0
mirror of https://github.com/meilisearch/MeiliSearch synced 2025-07-03 20:07:09 +02:00
Code Issues Releases Wiki Activity

add the retrieveVectors parameter to the get and fetch documents route

Browse source
This commit is contained in:
Tamo 2024-06-05 23:40:29 +02:00
parent ea61e5cbec
commit 6607875f49
6 changed files with 325 additions and 87 deletions
Download patch file Download diff file Expand all files Collapse all files

37
meilisearch/tests/common/index.rs
View file

@ -182,14 +182,10 @@ impl Index<'_> {
self.service.get(url).await
}
pub async fn get_document(
&self,
id: u64,
options: Option<GetDocumentOptions>,
) -> (Value, StatusCode) {
pub async fn get_document(&self, id: u64, options: Option<Value>) -> (Value, StatusCode) {
let mut url = format!("/indexes/{}/documents/{}", urlencode(self.uid.as_ref()), id);
if let Some(fields) = options.and_then(|o| o.fields) {
let _ = write!(url, "?fields={}", fields.join(","));
if let Some(options) = options {
write!(url, "?{}", yaup::to_string(&options).unwrap()).unwrap();
}
self.service.get(url).await
}
@ -205,18 +201,11 @@ impl Index<'_> {
}
pub async fn get_all_documents(&self, options: GetAllDocumentsOptions) -> (Value, StatusCode) {
let mut url = format!("/indexes/{}/documents?", urlencode(self.uid.as_ref()));
if let Some(limit) = options.limit {
let _ = write!(url, "limit={}&", limit);
}
if let Some(offset) = options.offset {
let _ = write!(url, "offset={}&", offset);
}
if let Some(attributes_to_retrieve) = options.attributes_to_retrieve {
let _ = write!(url, "fields={}&", attributes_to_retrieve.join(","));
}
let url = format!(
"/indexes/{}/documents?{}",
urlencode(self.uid.as_ref()),
yaup::to_string(&options).unwrap()
);
self.service.get(url).await
}
@ -435,13 +424,11 @@ impl Index<'_> {
}
}
pub struct GetDocumentOptions {
pub fields: Option<Vec<&'static str>>,
}
#[derive(Debug, Default)]
#[derive(Debug, Default, serde::Serialize)]
#[serde(rename_all = "camelCase")]
pub struct GetAllDocumentsOptions {
pub limit: Option<usize>,
pub offset: Option<usize>,
pub attributes_to_retrieve: Option<Vec<&'static str>>,
pub fields: Option<Vec<&'static str>>,
pub retrieve_vectors: bool,
}

2
meilisearch/tests/common/mod.rs
View file

@ -6,7 +6,7 @@ pub mod service;
use std::fmt::{self, Display};
#[allow(unused)]
pub use index::{GetAllDocumentsOptions, GetDocumentOptions};
pub use index::GetAllDocumentsOptions;
use meili_snap::json_string;
use serde::{Deserialize, Serialize};
#[allow(unused)]

24
meilisearch/tests/documents/errors.rs
View file

@ -795,3 +795,27 @@ async fn fetch_document_by_filter() {
}
"###);
}
#[actix_rt::test]
async fn retrieve_vectors() {
let server = Server::new().await;
let index = server.index("doggo");
let (response, _code) = index.get_all_documents_raw("?retrieveVectors=tamo").await;
snapshot!(json_string!(response), @r###"
{
"message": "Invalid value in parameter `retrieveVectors`: could not parse `tamo` as a boolean, expected either `true` or `false`",
"code": "invalid_document_retrieve_vectors",
"type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_document_retrieve_vectors"
}
"###);
let (response, _code) = index.get_document(0, Some(json!({"retrieveVectors": "tamo"}))).await;
snapshot!(json_string!(response), @r###"
{
"message": "Invalid value in parameter `retrieveVectors`: could not parse `tamo` as a boolean, expected either `true` or `false`",
"code": "invalid_document_retrieve_vectors",
"type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_document_retrieve_vectors"
}
"###);
}

268
meilisearch/tests/documents/get_documents.rs
View file

@ -4,7 +4,7 @@ use meili_snap::*;
use urlencoding::encode as urlencode;
use crate::common::encoder::Encoder;
use crate::common::{GetAllDocumentsOptions, GetDocumentOptions, Server, Value};
use crate::common::{GetAllDocumentsOptions, Server, Value};
use crate::json;
// TODO: partial test since we are testing error, amd error is not yet fully implemented in
@ -59,8 +59,7 @@ async fn get_document() {
})
);
let (response, code) =
index.get_document(0, Some(GetDocumentOptions { fields: Some(vec!["id"]) })).await;
let (response, code) = index.get_document(0, Some(json!({ "fields": ["id"] }))).await;
assert_eq!(code, 200);
assert_eq!(
response,
@ -69,9 +68,8 @@ async fn get_document() {
})
);
let (response, code) = index
.get_document(0, Some(GetDocumentOptions { fields: Some(vec!["nested.content"]) }))
.await;
let (response, code) =
index.get_document(0, Some(json!({ "fields": ["nested.content"] }))).await;
assert_eq!(code, 200);
assert_eq!(
response,
@ -211,7 +209,7 @@ async fn test_get_all_documents_attributes_to_retrieve() {
let (response, code) = index
.get_all_documents(GetAllDocumentsOptions {
attributes_to_retrieve: Some(vec!["name"]),
fields: Some(vec!["name"]),
..Default::default()
})
.await;
@ -225,9 +223,21 @@ async fn test_get_all_documents_attributes_to_retrieve() {
assert_eq!(response["limit"], json!(20));
assert_eq!(response["total"], json!(77));
let (response, code) = index
.get_all_documents(GetAllDocumentsOptions { fields: Some(vec![]), ..Default::default() })
.await;
assert_eq!(code, 200);
assert_eq!(response["results"].as_array().unwrap().len(), 20);
for results in response["results"].as_array().unwrap() {
assert_eq!(results.as_object().unwrap().keys().count(), 0);
}
assert_eq!(response["offset"], json!(0));
assert_eq!(response["limit"], json!(20));
assert_eq!(response["total"], json!(77));
let (response, code) = index
.get_all_documents(GetAllDocumentsOptions {
attributes_to_retrieve: Some(vec![]),
fields: Some(vec!["wrong"]),
..Default::default()
})
.await;
@ -242,22 +252,7 @@ async fn test_get_all_documents_attributes_to_retrieve() {
let (response, code) = index
.get_all_documents(GetAllDocumentsOptions {
attributes_to_retrieve: Some(vec!["wrong"]),
..Default::default()
})
.await;
assert_eq!(code, 200);
assert_eq!(response["results"].as_array().unwrap().len(), 20);
for results in response["results"].as_array().unwrap() {
assert_eq!(results.as_object().unwrap().keys().count(), 0);
}
assert_eq!(response["offset"], json!(0));
assert_eq!(response["limit"], json!(20));
assert_eq!(response["total"], json!(77));
let (response, code) = index
.get_all_documents(GetAllDocumentsOptions {
attributes_to_retrieve: Some(vec!["name", "tags"]),
fields: Some(vec!["name", "tags"]),
..Default::default()
})
.await;
@ -270,10 +265,7 @@ async fn test_get_all_documents_attributes_to_retrieve() {
}
let (response, code) = index
.get_all_documents(GetAllDocumentsOptions {
attributes_to_retrieve: Some(vec!["*"]),
..Default::default()
})
.get_all_documents(GetAllDocumentsOptions { fields: Some(vec!["*"]), ..Default::default() })
.await;
assert_eq!(code, 200);
assert_eq!(response["results"].as_array().unwrap().len(), 20);
@ -283,7 +275,7 @@ async fn test_get_all_documents_attributes_to_retrieve() {
let (response, code) = index
.get_all_documents(GetAllDocumentsOptions {
attributes_to_retrieve: Some(vec!["*", "wrong"]),
fields: Some(vec!["*", "wrong"]),
..Default::default()
})
.await;
@ -316,12 +308,10 @@ async fn get_document_s_nested_attributes_to_retrieve() {
assert_eq!(code, 202);
index.wait_task(1).await;
let (response, code) =
index.get_document(0, Some(GetDocumentOptions { fields: Some(vec!["content"]) })).await;
let (response, code) = index.get_document(0, Some(json!({ "fields": ["content"] }))).await;
assert_eq!(code, 200);
assert_eq!(response, json!({}));
let (response, code) =
index.get_document(1, Some(GetDocumentOptions { fields: Some(vec!["content"]) })).await;
let (response, code) = index.get_document(1, Some(json!({ "fields": ["content"] }))).await;
assert_eq!(code, 200);
assert_eq!(
response,
@ -333,9 +323,7 @@ async fn get_document_s_nested_attributes_to_retrieve() {
})
);
let (response, code) = index
.get_document(0, Some(GetDocumentOptions { fields: Some(vec!["content.truc"]) }))
.await;
let (response, code) = index.get_document(0, Some(json!({ "fields": ["content.truc"] }))).await;
assert_eq!(code, 200);
assert_eq!(
response,
@ -343,9 +331,7 @@ async fn get_document_s_nested_attributes_to_retrieve() {
"content.truc": "foobar",
})
);
let (response, code) = index
.get_document(1, Some(GetDocumentOptions { fields: Some(vec!["content.truc"]) }))
.await;
let (response, code) = index.get_document(1, Some(json!({ "fields": ["content.truc"] }))).await;
assert_eq!(code, 200);
assert_eq!(
response,
@ -540,3 +526,207 @@ async fn get_document_by_filter() {
}
"###);
}
#[actix_rt::test]
async fn get_document_with_vectors() {
let server = Server::new().await;
let index = server.index("doggo");
let (value, code) = server.set_features(json!({"vectorStore": true})).await;
snapshot!(code, @"200 OK");
snapshot!(value, @r###"
{
"vectorStore": true,
"metrics": false,
"logsRoute": false
}
"###);
let (response, code) = index
.update_settings(json!({
"embedders": {
"manual": {
"source": "userProvided",
"dimensions": 3,
}
},
}))
.await;
snapshot!(code, @"202 Accepted");
server.wait_task(response.uid()).await;
let documents = json!([
{"id": 0, "name": "kefir", "_vectors": { "manual": [0, 0, 0] }},
{"id": 1, "name": "echo", "_vectors": { "manual": null }},
]);
let (value, code) = index.add_documents(documents, None).await;
snapshot!(code, @"202 Accepted");
index.wait_task(value.uid()).await;
// by default you shouldn't see the `_vectors` object
let (documents, _code) = index.get_all_documents(Default::default()).await;
snapshot!(json_string!(documents), @r###"
{
"results": [
{
"id": 0,
"name": "kefir"
},
{
"id": 1,
"name": "echo"
}
],
"offset": 0,
"limit": 20,
"total": 2
}
"###);
let (documents, _code) = index.get_document(0, None).await;
snapshot!(json_string!(documents), @r###"
{
"id": 0,
"name": "kefir"
}
"###);
// if we try to retrieve the vectors with the `fields` parameter they
// still shouldn't be displayed
let (documents, _code) = index
.get_all_documents(GetAllDocumentsOptions {
fields: Some(vec!["name", "_vectors"]),
..Default::default()
})
.await;
snapshot!(json_string!(documents), @r###"
{
"results": [
{
"name": "kefir"
},
{
"name": "echo"
}
],
"offset": 0,
"limit": 20,
"total": 2
}
"###);
let (documents, _code) =
index.get_document(0, Some(json!({"fields": ["name", "_vectors"]}))).await;
snapshot!(json_string!(documents), @r###"
{
"name": "kefir"
}
"###);
// If we specify the retrieve vectors boolean and nothing else we should get the vectors
let (documents, _code) = index
.get_all_documents(GetAllDocumentsOptions { retrieve_vectors: true, ..Default::default() })
.await;
snapshot!(json_string!(documents), @r###"
{
"results": [
{
"id": 0,
"name": "kefir",
"_vectors": {
"manual": {
"embeddings": [
[
0.0,
0.0,
0.0
]
],
"userProvided": true
}
}
},
{
"id": 1,
"name": "echo",
"_vectors": {}
}
],
"offset": 0,
"limit": 20,
"total": 2
}
"###);
let (documents, _code) = index.get_document(0, Some(json!({"retrieveVectors": true}))).await;
snapshot!(json_string!(documents), @r###"
{
"id": 0,
"name": "kefir",
"_vectors": {
"manual": {
"embeddings": [
[
0.0,
0.0,
0.0
]
],
"userProvided": true
}
}
}
"###);
// If we specify the retrieve vectors boolean and exclude vectors form the `fields` we should still get the vectors
let (documents, _code) = index
.get_all_documents(GetAllDocumentsOptions {
retrieve_vectors: true,
fields: Some(vec!["name"]),
..Default::default()
})
.await;
snapshot!(json_string!(documents), @r###"
{
"results": [
{
"name": "kefir",
"_vectors": {
"manual": {
"embeddings": [
[
0.0,
0.0,
0.0
]
],
"userProvided": true
}
}
},
{
"name": "echo",
"_vectors": {}
}
],
"offset": 0,
"limit": 20,
"total": 2
}
"###);
let (documents, _code) =
index.get_document(0, Some(json!({"retrieveVectors": true, "fields": ["name"]}))).await;
snapshot!(json_string!(documents), @r###"
{
"name": "kefir",
"_vectors": {
"manual": {
"embeddings": [
[
0.0,
0.0,
0.0
]
],
"userProvided": true
}
}
}
"###);
}