throw an error on unknown fields specified in the _geo field

Author: irevoire

filter-parser/src/lib.rs

@@ -591,8 +591,8 @@ pub mod tests {
         "###);
 
         insta::assert_display_snapshot!(p("colour NOT EXIST"), @r###"
-        Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, or `_geoRadius` at `colour NOT EXIST`.
-        1:17 colour NOT EXIST
+        Found unexpected characters at the end of the filter: `ORdog != \'bernese mountain\'`. You probably forgot an `OR` or an `AND` rule.
+        17:44 channel = ponce ORdog != 'bernese mountain'
         "###);
 
         insta::assert_display_snapshot!(p("subscribers 100 TO1000"), @r###"

milli/src/error.rs

@@ -154,6 +154,8 @@ only composed of alphanumeric characters (a-z A-Z 0-9), hyphens (-) and undersco
 pub enum GeoError {
     #[error("The `_geo` field in the document with the id: `{document_id}` is not an object. Was expecting an object with the `_geo.lat` and `_geo.lng` fields but instead got `{value}`.")]
     NotAnObject { document_id: Value, value: Value },
+    #[error("The `_geo` field in the document with the id: `{document_id}` contains the following unexpected fields: `{value}`.")]
+    UnexpectedExtraFields { document_id: Value, value: Value },
     #[error("Could not find latitude nor longitude in the document with the id: `{document_id}`. Was expecting `_geo.lat` and `_geo.lng` fields.")]
     MissingLatitudeAndLongitude { document_id: Value },
     #[error("Could not find latitude in the document with the id: `{document_id}`. Was expecting a `_geo.lat` field.")]

milli/src/index.rs

@@ -2326,4 +2326,35 @@ pub(crate) mod tests {
 
         db_snap!(index, geo_faceted_documents_ids); // ensure that no more document was inserted
     }
+
+    #[test]
+    fn unexpected_extra_fields_in_geo_field() {
+        let index = TempIndex::new();
+
+        index
+            .update_settings(|settings| {
+                settings.set_primary_key("id".to_string());
+                settings.set_filterable_fields(HashSet::from(["_geo".to_string()]));
+            })
+            .unwrap();
+
+        let err = index
+            .add_documents(
+                documents!({ "id" : "doggo", "_geo": { "lat": 1, "lng": 2, "doggo": "are the best" }}),
+            )
+            .unwrap_err();
+        insta::assert_display_snapshot!(err, @r###"The `_geo` field in the document with the id: `"\"doggo\""` contains the following unexpected fields: `{"doggo":"are the best"}`."###);
+
+        db_snap!(index, geo_faceted_documents_ids); // ensure that no documents were inserted
+
+        // multiple fields and complex values
+        let err = index
+            .add_documents(
+                documents!({ "id" : "doggo", "_geo": { "lat": 1, "lng": 2, "doggo": "are the best", "and": { "all": ["cats", { "are": "beautiful" } ] } } }),
+            )
+            .unwrap_err();
+        insta::assert_display_snapshot!(err, @r###"The `_geo` field in the document with the id: `"\"doggo\""` contains the following unexpected fields: `{"and":{"all":["cats",{"are":"beautiful"}]},"doggo":"are the best"}`."###);
+
+        db_snap!(index, geo_faceted_documents_ids); // ensure that no documents were inserted
+    }
 }

milli/src/snapshots/index.rs/unexpected_extra_fields_in_geo_field/geo_faceted_documents_ids.snap

@@ -0,0 +1,4 @@
+---
+source: milli/src/index.rs
+---
+[]

milli/src/update/index_documents/enrich.rs

@@ -379,6 +379,10 @@ pub fn validate_geo_from_json(id: &DocumentId, bytes: &[u8]) -> Result<StdResult
         Value::Object(mut object) => match (object.remove("lat"), object.remove("lng")) {
             (Some(lat), Some(lng)) => {
                 match (extract_finite_float_from_value(lat), extract_finite_float_from_value(lng)) {
+                    (Ok(_), Ok(_)) if !object.is_empty() => Ok(Err(UnexpectedExtraFields {
+                        document_id: debug_id(),
+                        value: object.into(),
+                    })),
                     (Ok(_), Ok(_)) => Ok(Ok(())),
                     (Err(value), Ok(_)) => Ok(Err(BadLatitude { document_id: debug_id(), value })),
                     (Ok(_), Err(value)) => Ok(Err(BadLongitude { document_id: debug_id(), value })),