fmt

Signed-off-by: Kushal Kumar <kushalkumargupta4@gmail.com>

.github/workflows/bench-pr.yml

@@ -55,7 +55,7 @@ jobs:
               reaction-type: "rocket"
               repo-token: ${{ env.GH_TOKEN }}
 
-        - uses: xt0rted/pull-request-comment-branch@v2
+        - uses: xt0rted/pull-request-comment-branch@v3
           id: comment-branch
           with:
             repo_token: ${{ env.GH_TOKEN }}

.github/workflows/benchmarks-pr.yml

@@ -56,7 +56,7 @@ jobs:
           reaction-type: "eyes"
           repo-token: ${{ env.GH_TOKEN }}
 
-      - uses: xt0rted/pull-request-comment-branch@v2
+      - uses: xt0rted/pull-request-comment-branch@v3
         id: comment-branch
         with:
           repo_token: ${{ env.GH_TOKEN }}

BENCHMARKS.md

@@ -48,6 +48,27 @@ cargo xtask bench --no-dashboard -- workloads/my_workload_1.json workloads/my_wo
 
 For processing the results, look at [Looking at benchmark results/Without dashboard](#without-dashboard).
 
+#### Sending a workload by hand
+
+Sometimes you want to visualize the metrics of a worlkoad that comes from a custom report.
+It is not quite easy to trick the benchboard in thinking that your report is legitimate but here are the commands you can run to upload your firefox report on a running benchboard.
+
+```bash
+# Name this hostname whatever you want
+echo '{ "hostname": "the-best-place" }' | xh PUT 'http://127.0.0.1:9001/api/v1/machine'
+
+# You'll receive an UUID from this command that we will call $invocation_uuid
+echo '{ "commit": { "sha1": "1234567", "commit_date": "2024-09-05 12:00:12.0 +00:00:00", "message": "A cool message" }, "machine_hostname": "the-best-place", "max_workloads": 1 }' | xh PUT 'http://127.0.0.1:9001/api/v1/invocation'
+
+# Just use UUID from the previous command
+# and you'll receive another UUID that we will call $workload_uuid
+echo '{ "invocation_uuid": "$invocation_uuid", "name": "toto", "max_runs": 1 }' | xh PUT 'http://127.0.0.1:9001/api/v1/workload'
+
+# And now use your $workload_uuid and the content of your firefox report
+# but don't forget to convert your firefox report from JSONLines into an object
+echo '{ "workload_uuid": "$workload_uuid", "data": $REPORT_JSON_DATA }' | xh PUT 'http://127.0.0.1:9001/api/v1/run'
+```
+
 ### In CI
 
 We have dedicated runners to run workloads on CI. Currently, there are three ways of running the CI:

crates/meilisearch/src/routes/indexes/mod.rs

@@ -29,6 +29,8 @@ pub mod documents;
 pub mod facet_search;
 pub mod search;
 mod search_analytics;
+#[cfg(test)]
+mod search_test;
 pub mod settings;
 mod settings_analytics;
 pub mod similar;

crates/meilisearch/src/routes/indexes/search.rs

@@ -198,7 +198,7 @@ impl TryFrom<SearchQueryGet> for SearchQuery {
 // TODO: TAMO: split on :asc, and :desc, instead of doing some weird things
 
 /// Transform the sort query parameter into something that matches the post expected format.
-fn fix_sort_query_parameters(sort_query: &str) -> Vec<String> {
+pub fn fix_sort_query_parameters(sort_query: &str) -> Vec<String> {
     let mut sort_parameters = Vec::new();
     let mut merge = false;
     for current_sort in sort_query.trim_matches('"').split(',').map(|s| s.trim()) {
@@ -356,30 +356,3 @@ pub fn search_kind(
         (_, None, Some(_)) => Err(MeilisearchHttpError::MissingSearchHybrid.into()),
     }
 }
-
-#[cfg(test)]
-mod test {
-    use super::*;
-
-    #[test]
-    fn test_fix_sort_query_parameters() {
-        let sort = fix_sort_query_parameters("_geoPoint(12, 13):asc");
-        assert_eq!(sort, vec!["_geoPoint(12,13):asc".to_string()]);
-        let sort = fix_sort_query_parameters("doggo:asc,_geoPoint(12.45,13.56):desc");
-        assert_eq!(sort, vec!["doggo:asc".to_string(), "_geoPoint(12.45,13.56):desc".to_string(),]);
-        let sort = fix_sort_query_parameters(
-            "doggo:asc , _geoPoint(12.45, 13.56, 2590352):desc , catto:desc",
-        );
-        assert_eq!(
-            sort,
-            vec![
-                "doggo:asc".to_string(),
-                "_geoPoint(12.45,13.56,2590352):desc".to_string(),
-                "catto:desc".to_string(),
-            ]
-        );
-        let sort = fix_sort_query_parameters("doggo:asc , _geoPoint(1, 2), catto:desc");
-        // This is ugly but eh, I don't want to write a full parser just for this unused route
-        assert_eq!(sort, vec!["doggo:asc".to_string(), "_geoPoint(1,2),catto:desc".to_string(),]);
-    }
-}

crates/meilisearch/src/routes/indexes/search_test.rs

@@ -0,0 +1,22 @@
+use crate::routes::indexes::search::fix_sort_query_parameters;
+
+#[test]
+fn test_fix_sort_query_parameters() {
+    let sort = fix_sort_query_parameters("_geoPoint(12, 13):asc");
+    assert_eq!(sort, vec!["_geoPoint(12,13):asc".to_string()]);
+    let sort = fix_sort_query_parameters("doggo:asc,_geoPoint(12.45,13.56):desc");
+    assert_eq!(sort, vec!["doggo:asc".to_string(), "_geoPoint(12.45,13.56):desc".to_string(),]);
+    let sort =
+        fix_sort_query_parameters("doggo:asc , _geoPoint(12.45, 13.56, 2590352):desc , catto:desc");
+    assert_eq!(
+        sort,
+        vec![
+            "doggo:asc".to_string(),
+            "_geoPoint(12.45,13.56,2590352):desc".to_string(),
+            "catto:desc".to_string(),
+        ]
+    );
+    let sort = fix_sort_query_parameters("doggo:asc , _geoPoint(1, 2), catto:desc");
+    // This is ugly but eh, I don't want to write a full parser just for this unused route
+    assert_eq!(sort, vec!["doggo:asc".to_string(), "_geoPoint(1,2),catto:desc".to_string(),]);
+}

crates/meilisearch/src/search/mod.rs

@@ -30,6 +30,8 @@ use milli::{
 use regex::Regex;
 use serde::Serialize;
 use serde_json::{json, Value};
+#[cfg(test)]
+mod mod_test;
 
 use crate::error::MeilisearchHttpError;
 
@@ -1557,7 +1559,7 @@ pub fn perform_similar(
     Ok(result)
 }
 
-fn insert_geo_distance(sorts: &[String], document: &mut Document) {
+pub fn insert_geo_distance(sorts: &[String], document: &mut Document) {
     lazy_static::lazy_static! {
         static ref GEO_REGEX: Regex =
             Regex::new(r"_geoPoint\(\s*([[:digit:].\-]+)\s*,\s*([[:digit:].\-]+)\s*\)").unwrap();

crates/meilisearch/src/search/mod_test.rs

@@ -0,0 +1,114 @@
+use meilisearch_types::Document;
+use serde_json::json;
+
+use crate::search::insert_geo_distance;
+
+#[test]
+fn test_insert_geo_distance() {
+    let value: Document = serde_json::from_str(
+        r#"{
+          "_geo": {
+            "lat": 50.629973371633746,
+            "lng": 3.0569447399419567
+          },
+          "city": "Lille",
+          "id": "1"
+        }"#,
+    )
+    .unwrap();
+
+    let sorters = &["_geoPoint(50.629973371633746,3.0569447399419567):desc".to_string()];
+    let mut document = value.clone();
+    insert_geo_distance(sorters, &mut document);
+    assert_eq!(document.get("_geoDistance"), Some(&json!(0)));
+
+    let sorters = &["_geoPoint(50.629973371633746, 3.0569447399419567):asc".to_string()];
+    let mut document = value.clone();
+    insert_geo_distance(sorters, &mut document);
+    assert_eq!(document.get("_geoDistance"), Some(&json!(0)));
+
+    let sorters = &["_geoPoint(   50.629973371633746   ,  3.0569447399419567   ):desc".to_string()];
+    let mut document = value.clone();
+    insert_geo_distance(sorters, &mut document);
+    assert_eq!(document.get("_geoDistance"), Some(&json!(0)));
+
+    let sorters = &[
+        "prix:asc",
+        "villeneuve:desc",
+        "_geoPoint(50.629973371633746, 3.0569447399419567):asc",
+        "ubu:asc",
+    ]
+    .map(|s| s.to_string());
+    let mut document = value.clone();
+    insert_geo_distance(sorters, &mut document);
+    assert_eq!(document.get("_geoDistance"), Some(&json!(0)));
+
+    // only the first geoPoint is used to compute the distance
+    let sorters = &[
+        "chien:desc",
+        "_geoPoint(50.629973371633746, 3.0569447399419567):asc",
+        "pangolin:desc",
+        "_geoPoint(100.0, -80.0):asc",
+        "chat:asc",
+    ]
+    .map(|s| s.to_string());
+    let mut document = value.clone();
+    insert_geo_distance(sorters, &mut document);
+    assert_eq!(document.get("_geoDistance"), Some(&json!(0)));
+
+    // there was no _geoPoint so nothing is inserted in the document
+    let sorters = &["chien:asc".to_string()];
+    let mut document = value;
+    insert_geo_distance(sorters, &mut document);
+    assert_eq!(document.get("_geoDistance"), None);
+}
+
+#[test]
+fn test_insert_geo_distance_with_coords_as_string() {
+    let value: Document = serde_json::from_str(
+        r#"{
+          "_geo": {
+            "lat": "50",
+            "lng": 3
+          }
+        }"#,
+    )
+    .unwrap();
+
+    let sorters = &["_geoPoint(50,3):desc".to_string()];
+    let mut document = value.clone();
+    insert_geo_distance(sorters, &mut document);
+    assert_eq!(document.get("_geoDistance"), Some(&json!(0)));
+
+    let value: Document = serde_json::from_str(
+        r#"{
+          "_geo": {
+            "lat": "50",
+            "lng": "3"
+          },
+          "id": "1"
+        }"#,
+    )
+    .unwrap();
+
+    let sorters = &["_geoPoint(50,3):desc".to_string()];
+    let mut document = value.clone();
+    insert_geo_distance(sorters, &mut document);
+    assert_eq!(document.get("_geoDistance"), Some(&json!(0)));
+
+    let value: Document = serde_json::from_str(
+        r#"{
+          "_geo": {
+            "lat": 50,
+            "lng": "3"
+          },
+          "id": "1"
+        }"#,
+    )
+    .unwrap();
+
+    let sorters = &["_geoPoint(50,3):desc".to_string()];
+    let mut document = value.clone();
+    insert_geo_distance(sorters, &mut document);
+    assert_eq!(document.get("_geoDistance"), Some(&json!(0)));
+}