reduce number of allocations (#2257)

* reduce number of allocations

Explanation makes up around 50% of all allocations (numbers not perf).
It's created during serialization but not called.

- Make Explanation optional in BM25
- Avoid allocations when using Explanation

* use Cow

Author: PSeitz

src/postings/serializer.rs

@@ -355,7 +355,7 @@ impl<W: Write> PostingsSerializer<W> {
             return;
         }
 
-        self.bm25_weight = Some(Bm25Weight::for_one_term(
+        self.bm25_weight = Some(Bm25Weight::for_one_term_without_explain(
             term_doc_freq as u64,
             num_docs_in_segment,
             self.avg_fieldnorm,

src/query/bm25.rs

@@ -77,7 +77,7 @@ pub struct Bm25Params {
 /// A struct used for computing BM25 scores.
 #[derive(Clone)]
 pub struct Bm25Weight {
-    idf_explain: Explanation,
+    idf_explain: Option<Explanation>,
     weight: Score,
     cache: [Score; 256],
     average_fieldnorm: Score,
@@ -147,11 +147,30 @@ impl Bm25Weight {
         idf_explain.add_const("N, total number of docs", total_num_docs as Score);
         Bm25Weight::new(idf_explain, avg_fieldnorm)
     }
+    /// Construct a [Bm25Weight] for a single term.
+    /// This method does not carry the [Explanation] for the idf.
+    pub fn for_one_term_without_explain(
+        term_doc_freq: u64,
+        total_num_docs: u64,
+        avg_fieldnorm: Score,
+    ) -> Bm25Weight {
+        let idf = idf(term_doc_freq, total_num_docs);
+        Bm25Weight::new_without_explain(idf, avg_fieldnorm)
+    }
 
     pub(crate) fn new(idf_explain: Explanation, average_fieldnorm: Score) -> Bm25Weight {
         let weight = idf_explain.value() * (1.0 + K1);
         Bm25Weight {
-            idf_explain,
+            idf_explain: Some(idf_explain),
+            weight,
+            cache: compute_tf_cache(average_fieldnorm),
+            average_fieldnorm,
+        }
+    }
+    pub(crate) fn new_without_explain(idf: f32, average_fieldnorm: Score) -> Bm25Weight {
+        let weight = idf * (1.0 + K1);
+        Bm25Weight {
+            idf_explain: None,
             weight,
             cache: compute_tf_cache(average_fieldnorm),
             average_fieldnorm,
@@ -202,7 +221,9 @@ impl Bm25Weight {
 
         let mut explanation = Explanation::new("TermQuery, product of...", score);
         explanation.add_detail(Explanation::new("(K1+1)", K1 + 1.0));
-        explanation.add_detail(self.idf_explain.clone());
+        if let Some(idf_explain) = &self.idf_explain {
+            explanation.add_detail(idf_explain.clone());
+        }
         explanation.add_detail(tf_explanation);
         explanation
     }

src/query/boost_query.rs

@@ -74,7 +74,8 @@ impl Weight for BoostWeight {
     fn explain(&self, reader: &SegmentReader, doc: u32) -> crate::Result<Explanation> {
         let underlying_explanation = self.weight.explain(reader, doc)?;
         let score = underlying_explanation.value() * self.boost;
-        let mut explanation = Explanation::new(format!("Boost x{} of ...", self.boost), score);
+        let mut explanation =
+            Explanation::new_with_string(format!("Boost x{} of ...", self.boost), score);
         explanation.add_detail(underlying_explanation);
         Ok(explanation)
     }
@@ -151,7 +152,7 @@ mod tests {
         let explanation = query.explain(&searcher, DocAddress::new(0, 0u32)).unwrap();
         assert_eq!(
             explanation.to_pretty_json(),
-            "{\n  \"value\": 0.2,\n  \"description\": \"Boost x0.2 of ...\",\n  \"details\": [\n    {\n      \"value\": 1.0,\n      \"description\": \"AllQuery\",\n      \"context\": []\n    }\n  ],\n  \"context\": []\n}"
+            "{\n  \"value\": 0.2,\n  \"description\": \"Boost x0.2 of ...\",\n  \"details\": [\n    {\n      \"value\": 1.0,\n      \"description\": \"AllQuery\"\n    }\n  ]\n}"
         );
         Ok(())
     }

src/query/const_score_query.rs

@@ -164,11 +164,9 @@ mod tests {
   "details": [
     {
       "value": 1.0,
-      "description": "AllQuery",
-      "context": []
+      "description": "AllQuery"
     }
-  ],
-  "context": []
+  ]
 }"#
         );
         Ok(())

src/query/explanation.rs

@@ -1,3 +1,4 @@
+use std::borrow::Cow;
 use std::fmt;
 
 use serde::Serialize;
@@ -16,12 +17,12 @@ pub(crate) fn does_not_match(doc: DocId) -> TantivyError {
 #[derive(Clone, Serialize)]
 pub struct Explanation {
     value: Score,
-    description: String,
-    #[serde(skip_serializing_if = "Vec::is_empty")]
-    details: Vec<Explanation>,
-    context: Vec<String>,
+    description: Cow<'static, str>,
+    #[serde(skip_serializing_if = "Option::is_none")]
+    details: Option<Vec<Explanation>>,
+    #[serde(skip_serializing_if = "Option::is_none")]
+    context: Option<Vec<String>>,
 }
-
 impl fmt::Debug for Explanation {
     fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
         write!(f, "Explanation({})", self.to_pretty_json())
@@ -30,12 +31,21 @@ impl fmt::Debug for Explanation {
 
 impl Explanation {
     /// Creates a new explanation object.
-    pub fn new<T: ToString>(description: T, value: Score) -> Explanation {
+    pub fn new_with_string(description: String, value: Score) -> Explanation {
+        Explanation {
+            value,
+            description: Cow::Owned(description),
+            details: None,
+            context: None,
+        }
+    }
+    /// Creates a new explanation object.
+    pub fn new(description: &'static str, value: Score) -> Explanation {
         Explanation {
             value,
-            description: description.to_string(),
-            details: vec![],
-            context: vec![],
+            description: Cow::Borrowed(description),
+            details: None,
+            context: None,
         }
     }
 
@@ -48,17 +58,21 @@ impl Explanation {
     ///
     /// Details are treated as child of the current node.
     pub fn add_detail(&mut self, child_explanation: Explanation) {
-        self.details.push(child_explanation);
+        self.details
+            .get_or_insert_with(Vec::new)
+            .push(child_explanation);
     }
 
     /// Adds some extra context to the explanation.
     pub fn add_context(&mut self, context: String) {
-        self.context.push(context);
+        self.context.get_or_insert_with(Vec::new).push(context);
     }
 
     /// Shortcut for `self.details.push(Explanation::new(name, value));`
-    pub fn add_const<T: ToString>(&mut self, name: T, value: Score) {
-        self.details.push(Explanation::new(name, value));
+    pub fn add_const(&mut self, name: &'static str, value: Score) {
+        self.details
+            .get_or_insert_with(Vec::new)
+            .push(Explanation::new(name, value));
     }
 
     /// Returns an indented json representation of the explanation tree for debug usage.

src/query/term_query/term_query.rs

@@ -101,7 +101,7 @@ impl TermQuery {
                 ..
             } => Bm25Weight::for_terms(statistics_provider, &[self.term.clone()])?,
             EnableScoring::Disabled { .. } => {
-                Bm25Weight::new(Explanation::new("<no score>".to_string(), 1.0f32), 1.0f32)
+                Bm25Weight::new(Explanation::new("<no score>", 1.0f32), 1.0f32)
             }
         };
         let scoring_enabled = enable_scoring.is_scoring_enabled();