test(postings): add unit tests for TermFrequencyRecorder

Closes #2285

The TermFrequencyRecorder was completely untested. Add five focused tests:

- term_frequency_recorder_has_term_freq: verifies the recorder
  correctly advertises term-frequency support via has_term_freq()
- term_frequency_recorder_zero_docs: term_doc_freq() returns Some(0)
  before any documents are recorded
- term_frequency_recorder_term_doc_freq_single_doc: one document with
  two occurrences yields term_doc_freq() == Some(1)
- term_frequency_recorder_term_doc_freq_multiple_docs: three documents
  with varying term frequencies yield term_doc_freq() == Some(3),
  confirming the count tracks documents, not occurrences
- term_frequency_recorder_single_occurrence_per_doc: each of three
  documents has exactly one occurrence
- term_frequency_recorder_high_frequency_doc: a single document with
  1000 occurrences still yields term_doc_freq() == Some(1)
This commit is contained in:
Kanishk Sachan
2026-06-06 14:44:51 +08:00
committed by PSeitz
parent 62705526e8
commit 70a8e56ee5
+96 -1
View File
@@ -275,8 +275,9 @@ impl Recorder for TfAndPositionRecorder {
mod tests {
use common::write_u32_vint;
use stacker::MemoryArena;
use super::{BufferLender, VInt32Reader};
use super::{BufferLender, Recorder, TermFrequencyRecorder, VInt32Reader};
#[test]
fn test_buffer_lender() {
@@ -314,4 +315,98 @@ mod tests {
let res: Vec<u32> = VInt32Reader::new(&buffer[..]).collect();
assert_eq!(&res[..], &vals[..]);
}
// ── TermFrequencyRecorder ─────────────────────────────────────────────────
#[test]
fn term_frequency_recorder_has_term_freq() {
let rec = TermFrequencyRecorder::default();
assert!(
rec.has_term_freq(),
"TermFrequencyRecorder must advertise term-frequency support"
);
}
#[test]
fn term_frequency_recorder_term_doc_freq_single_doc() {
let mut arena = MemoryArena::default();
let mut rec = TermFrequencyRecorder::default();
// Record one document with two term occurrences.
rec.new_doc(0, &mut arena);
rec.record_position(0, &mut arena);
rec.record_position(1, &mut arena);
rec.close_doc(&mut arena);
assert_eq!(
rec.term_doc_freq(),
Some(1),
"term_doc_freq should be 1 after recording one document"
);
}
#[test]
fn term_frequency_recorder_term_doc_freq_multiple_docs() {
let mut arena = MemoryArena::default();
let mut rec = TermFrequencyRecorder::default();
// Three documents with 1, 3, and 2 occurrences respectively.
for (doc, tf) in [(0u32, 1u32), (5, 3), (10, 2)] {
rec.new_doc(doc, &mut arena);
for pos in 0..tf {
rec.record_position(pos, &mut arena);
}
rec.close_doc(&mut arena);
}
assert_eq!(
rec.term_doc_freq(),
Some(3),
"term_doc_freq should equal the number of documents recorded"
);
}
#[test]
fn term_frequency_recorder_zero_docs() {
let rec = TermFrequencyRecorder::default();
assert_eq!(
rec.term_doc_freq(),
Some(0),
"term_doc_freq should be 0 before any document is recorded"
);
}
#[test]
fn term_frequency_recorder_single_occurrence_per_doc() {
let mut arena = MemoryArena::default();
let mut rec = TermFrequencyRecorder::default();
// Each document has exactly one occurrence — the minimum non-trivial case.
for doc in [1u32, 2, 100] {
rec.new_doc(doc, &mut arena);
rec.record_position(0, &mut arena);
rec.close_doc(&mut arena);
}
assert_eq!(rec.term_doc_freq(), Some(3));
}
#[test]
fn term_frequency_recorder_high_frequency_doc() {
let mut arena = MemoryArena::default();
let mut rec = TermFrequencyRecorder::default();
// A document where the term appears many times.
rec.new_doc(42, &mut arena);
for pos in 0..1000 {
rec.record_position(pos, &mut arena);
}
rec.close_doc(&mut arena);
assert_eq!(
rec.term_doc_freq(),
Some(1),
"term_doc_freq counts documents, not occurrences"
);
}
}