From 0b061f1e7090e62d1f7b52bd434ac389831ecf7e Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Cl=C3=A9ment=20Renault?= <clement@meilisearch.com>
Date: Wed, 4 Sep 2024 17:40:24 +0200
Subject: [PATCH] Introduce the FieldIdFacetIsEmptyDocidsExtractor

---
 milli/src/update/new/extract/faceted/mod.rs | 32 ++++++++++++++++++++-
 milli/src/update/new/indexer/mod.rs         |  2 ++
 2 files changed, 33 insertions(+), 1 deletion(-)

diff --git a/milli/src/update/new/extract/faceted/mod.rs b/milli/src/update/new/extract/faceted/mod.rs
index b230549c1..e3c89b0e4 100644
--- a/milli/src/update/new/extract/faceted/mod.rs
+++ b/milli/src/update/new/extract/faceted/mod.rs
@@ -186,7 +186,6 @@ impl FacetedExtractor for FieldIdFacetNumberDocidsExtractor {
     }
 }
 
-/// TODO It doesn't keep the original string in the value
 pub struct FieldIdFacetStringDocidsExtractor;
 impl FacetedExtractor for FieldIdFacetStringDocidsExtractor {
     fn attributes_to_extract<'a>(rtxn: &'a RoTxn, index: &'a Index) -> Result<HashSet<String>> {
@@ -211,6 +210,37 @@ impl FacetedExtractor for FieldIdFacetStringDocidsExtractor {
     }
 }
 
+// Extract fieldid facet isempty docids
+// Extract fieldid facet isnull docids
+// Extract fieldid facet exists docids
+
+pub struct FieldIdFacetIsEmptyDocidsExtractor;
+impl FacetedExtractor for FieldIdFacetIsEmptyDocidsExtractor {
+    fn attributes_to_extract<'a>(rtxn: &'a RoTxn, index: &'a Index) -> Result<HashSet<String>> {
+        index.user_defined_faceted_fields(rtxn)
+    }
+
+    fn build_key<'b>(
+        field_id: FieldId,
+        value: &Value,
+        output: &'b mut Vec<u8>,
+    ) -> Option<&'b [u8]> {
+        let is_empty = match value {
+            Value::Null | Value::Bool(_) | Value::Number(_) => false,
+            Value::String(s) => s.is_empty(),
+            Value::Array(a) => a.is_empty(),
+            Value::Object(o) => o.is_empty(),
+        };
+
+        if is_empty {
+            output.extend_from_slice(&field_id.to_be_bytes());
+            Some(&*output)
+        } else {
+            None
+        }
+    }
+}
+
 pub fn extract_document_facets(
     attributes_to_extract: &[&str],
     obkv: &KvReaderFieldId,
diff --git a/milli/src/update/new/indexer/mod.rs b/milli/src/update/new/indexer/mod.rs
index 21e28fc84..ed42c03b1 100644
--- a/milli/src/update/new/indexer/mod.rs
+++ b/milli/src/update/new/indexer/mod.rs
@@ -131,7 +131,9 @@ where
                 // Extract fieldid docid facet number
                 // Extract fieldid docid facet string
                 // Extract facetid string fst
+                // Extract facetid normalized string strings
 
+                // TODO Inverted Indexes again
                 // Extract fieldid facet isempty docids
                 // Extract fieldid facet isnull docids
                 // Extract fieldid facet exists docids