fix json_path_writer add sep logic

common/src/json_path_writer.rs

@@ -53,7 +53,7 @@ impl JsonPathWriter {
     pub fn push(&mut self, segment: &str) {
         let len_path = self.path.len();
         self.indices.push(len_path);
-        if !self.path.is_empty() {
+        if self.indices.len() > 1 {
             self.path.push(JSON_PATH_SEGMENT_SEP as char);
         }
         self.path.push_str(segment);

src/aggregation/metric/top_hits.rs

@@ -2,6 +2,7 @@ use std::collections::HashMap;
 use std::net::Ipv6Addr;
 
 use columnar::{ColumnarReader, DynamicColumn};
+use common::json_path_writer::JSON_PATH_SEGMENT_SEP_STR;
 use common::DateTime;
 use regex::Regex;
 use serde::ser::SerializeMap;
@@ -15,7 +16,6 @@ use crate::aggregation::intermediate_agg_result::{
 use crate::aggregation::segment_agg_result::SegmentAggregationCollector;
 use crate::aggregation::AggregationError;
 use crate::collector::TopNComputer;
-use crate::schema::term::JSON_PATH_SEGMENT_SEP_STR;
 use crate::schema::OwnedValue;
 use crate::{DocAddress, DocId, SegmentOrdinal};
 

src/aggregation/mod.rs

@@ -159,10 +159,6 @@ use itertools::Itertools;
 use serde::de::{self, Visitor};
 use serde::{Deserialize, Deserializer, Serialize};
 
-pub(crate) fn invalid_agg_request(message: String) -> crate::TantivyError {
-    crate::TantivyError::AggregationError(AggregationError::InvalidRequest(message))
-}
-
 fn parse_str_into_f64<E: de::Error>(value: &str) -> Result<f64, E> {
     let parsed = value.parse::<f64>().map_err(|_err| {
         de::Error::custom(format!("Failed to parse f64 from string: {:?}", value))

src/core/json_utils.rs

@@ -1,4 +1,5 @@
-use common::JsonPathWriter;
+use common::json_path_writer::JSON_PATH_SEGMENT_SEP;
+use common::{replace_in_place, JsonPathWriter};
 use rustc_hash::FxHashMap;
 
 use crate::postings::{IndexingContext, IndexingPosition, PostingsWriter};

src/core/tests.rs

@@ -421,10 +421,7 @@ fn test_non_text_json_term_freq() {
     term.append_type_and_fast_value(75u64);
 
     let postings = inv_idx
-        .read_postings(
-            &json_term_writer.term(),
-            IndexRecordOption::WithFreqsAndPositions,
-        )
+        .read_postings(&term, IndexRecordOption::WithFreqsAndPositions)
         .unwrap()
         .unwrap();
     assert_eq!(postings.doc(), 0);
@@ -458,10 +455,7 @@ fn test_non_text_json_term_freq_bitpacked() {
     term.append_type_and_fast_value(75u64);
 
     let mut postings = inv_idx
-        .read_postings(
-            &json_term_writer.term(),
-            IndexRecordOption::WithFreqsAndPositions,
-        )
+        .read_postings(&term, IndexRecordOption::WithFreqsAndPositions)
         .unwrap()
         .unwrap();
     assert_eq!(postings.doc(), 0);

src/functional_test.rs

@@ -4,6 +4,7 @@ use rand::{thread_rng, Rng};
 
 use crate::indexer::index_writer::MEMORY_BUDGET_NUM_BYTES_MIN;
 use crate::schema::*;
+#[allow(deprecated)] // Remove with index sorting
 use crate::{doc, schema, Index, IndexSettings, IndexSortByField, IndexWriter, Order, Searcher};
 
 fn check_index_content(searcher: &Searcher, vals: &[u64]) -> crate::Result<()> {

src/indexer/mod.rs

@@ -156,16 +156,24 @@ mod tests_mmap {
     }
     #[test]
     fn test_json_field_1byte() {
-        // Test when field name contains a 1 byte, which has special meaning in tantivy.
+        // Test when field name contains a '1' byte, which has special meaning in tantivy.
+        // The 1 byte can be addressed as '1' byte or '.'.
         let field_name_in = "\u{0001}";
         let field_name_out = "\u{0001}";
         test_json_field_name(field_name_in, field_name_out);
 
-        // Test when field name contains a 1 byte, which has special meaning in tantivy.
+        // Test when field name contains a '1' byte, which has special meaning in tantivy.
         let field_name_in = "\u{0001}";
         let field_name_out = ".";
         test_json_field_name(field_name_in, field_name_out);
     }
+    #[test]
+    fn test_json_field_dot() {
+        // Test when field name contains a '.'
+        let field_name_in = ".";
+        let field_name_out = ".";
+        test_json_field_name(field_name_in, field_name_out);
+    }
     fn test_json_field_name(field_name_in: &str, field_name_out: &str) {
         let mut schema_builder = Schema::builder();
 
@@ -205,10 +213,10 @@ mod tests_mmap {
         let reader = index.reader().unwrap();
         let searcher = reader.searcher();
         let parse_query = QueryParser::for_index(&index, Vec::new());
-        let test_query = |field_name: &str| {
-            let query = parse_query.parse_query(field_name).unwrap();
+        let test_query = |query_str: &str| {
+            let query = parse_query.parse_query(query_str).unwrap();
             let num_docs = searcher.search(&query, &Count).unwrap();
-            assert_eq!(num_docs, 1);
+            assert_eq!(num_docs, 1, "{}", query_str);
         };
         test_query(format!("json.{field_name_out}:test1").as_str());
         test_query(format!("json.a{field_name_out}:test2").as_str());

src/query/query_parser/query_parser.rs

@@ -10,11 +10,10 @@ use query_grammar::{UserInputAst, UserInputBound, UserInputLeaf, UserInputLitera
 use rustc_hash::FxHashMap;
 
 use super::logical_ast::*;
-use crate::core::Index;
+use crate::index::Index;
 use crate::json_utils::{
     convert_to_fast_value_and_append_to_json_term, split_json_path, term_from_json_paths,
 };
-use crate::core::Index;
 use crate::query::range_query::{is_type_valid_for_fastfield_range_query, RangeQuery};
 use crate::query::{
     AllQuery, BooleanQuery, BoostQuery, EmptyQuery, FuzzyTermQuery, Occur, PhrasePrefixQuery,