1
0
mirror of https://github.com/redis/go-redis.git synced 2025-11-26 06:23:09 +03:00
Files
go-redis/search_commands_parse_test.go
Nedyalko Dyakov 5fdc947723 feat(ftinfo): add vector specific attributes in ftinfo (#3596)
* add vector specific attributes in ftinfo

* safer parsing

* try to fix tests

* oops, should test agains resp2 only

* fix asserts

* skip test for RE
2025-11-12 13:29:26 +02:00

371 lines
12 KiB
Go

package redis
import (
"testing"
)
// TestParseFTInfo tests the parseFTInfo function with a comprehensive FT.INFO response
// This test uses the actual response structure from Redis with vector fields
func TestParseFTInfo(t *testing.T) {
// This is the data structure that would be returned by Redis for FT.INFO
// Based on the redis-cli output provided by the user
data := map[string]interface{}{
"index_name": "rand:42d1f2820b3048b6bc783d4dcdb9094a",
"index_options": []interface{}{},
"index_definition": []interface{}{
"key_type", "HASH",
"prefixes", []interface{}{"rand:42d1f2820b3048b6bc783d4dcdb9094a"},
"default_score", "1",
"indexes_all", false,
},
"attributes": []interface{}{
// prompt field (TEXT)
[]interface{}{
"identifier", "prompt",
"attribute", "prompt",
"type", "TEXT",
"WEIGHT", "1",
},
// response field (TEXT)
[]interface{}{
"identifier", "response",
"attribute", "response",
"type", "TEXT",
"WEIGHT", "1",
},
// exact_digest field (TAG)
[]interface{}{
"identifier", "exact_digest",
"attribute", "exact_digest",
"type", "TAG",
"SEPARATOR", ",",
},
// prompt_vector field (VECTOR)
[]interface{}{
"identifier", "prompt_vector",
"attribute", "prompt_vector",
"type", "VECTOR",
"algorithm", "HNSW",
"data_type", "FLOAT32",
"dim", int64(1536),
"distance_metric", "COSINE",
"M", int64(16),
"ef_construction", int64(64),
},
},
"num_docs": int64(0),
"max_doc_id": int64(0),
"num_terms": int64(0),
"num_records": int64(0),
"inverted_sz_mb": "0",
"vector_index_sz_mb": "0",
"total_inverted_index_blocks": int64(0),
"offset_vectors_sz_mb": "0",
"doc_table_size_mb": "0.01532745361328125",
"sortable_values_size_mb": "0",
"key_table_size_mb": "2.288818359375e-5",
"tag_overhead_sz_mb": "0",
"text_overhead_sz_mb": "0",
"total_index_memory_sz_mb": "0.015350341796875",
"geoshapes_sz_mb": "0",
"records_per_doc_avg": "nan",
"bytes_per_record_avg": "nan",
"offsets_per_term_avg": "nan",
"offset_bits_per_record_avg": "nan",
"hash_indexing_failures": int64(0),
"total_indexing_time": "0",
"indexing": int64(0),
"percent_indexed": "1",
"number_of_uses": int64(2),
"cleaning": int64(0),
"gc_stats": []interface{}{
"bytes_collected", "0",
"total_ms_run", "0",
"total_cycles", "0",
"average_cycle_time_ms", "nan",
"last_run_time_ms", "0",
"gc_numeric_trees_missed", "0",
"gc_blocks_denied", "0",
},
"cursor_stats": []interface{}{
"global_idle", int64(0),
"global_total", int64(0),
"index_capacity", int64(128),
"index_total", int64(0),
},
"dialect_stats": []interface{}{
"dialect_1", int64(0),
"dialect_2", int64(0),
"dialect_3", int64(0),
"dialect_4", int64(0),
},
"Index Errors": []interface{}{
"indexing failures", int64(0),
"last indexing error", "N/A",
"last indexing error key", "N/A",
"background indexing status", "OK",
},
"field statistics": []interface{}{
[]interface{}{
"identifier", "prompt",
"attribute", "prompt",
"Index Errors", []interface{}{
"indexing failures", int64(0),
"last indexing error", "N/A",
"last indexing error key", "N/A",
},
},
[]interface{}{
"identifier", "response",
"attribute", "response",
"Index Errors", []interface{}{
"indexing failures", int64(0),
"last indexing error", "N/A",
"last indexing error key", "N/A",
},
},
[]interface{}{
"identifier", "exact_digest",
"attribute", "exact_digest",
"Index Errors", []interface{}{
"indexing failures", int64(0),
"last indexing error", "N/A",
"last indexing error key", "N/A",
},
},
[]interface{}{
"identifier", "prompt_vector",
"attribute", "prompt_vector",
"Index Errors", []interface{}{
"indexing failures", int64(0),
"last indexing error", "N/A",
"last indexing error key", "N/A",
},
"memory", int64(0),
"marked_deleted", int64(0),
},
},
}
// Parse the data
result, err := parseFTInfo(data)
if err != nil {
t.Fatalf("parseFTInfo failed: %v", err)
}
// Validate index name
if result.IndexName != "rand:42d1f2820b3048b6bc783d4dcdb9094a" {
t.Errorf("IndexName = %v, want %v", result.IndexName, "rand:42d1f2820b3048b6bc783d4dcdb9094a")
}
// Validate index definition
if result.IndexDefinition.KeyType != "HASH" {
t.Errorf("IndexDefinition.KeyType = %v, want HASH", result.IndexDefinition.KeyType)
}
if len(result.IndexDefinition.Prefixes) != 1 || result.IndexDefinition.Prefixes[0] != "rand:42d1f2820b3048b6bc783d4dcdb9094a" {
t.Errorf("IndexDefinition.Prefixes = %v, want [rand:42d1f2820b3048b6bc783d4dcdb9094a]", result.IndexDefinition.Prefixes)
}
if result.IndexDefinition.DefaultScore != 1.0 {
t.Errorf("IndexDefinition.DefaultScore = %v, want 1.0", result.IndexDefinition.DefaultScore)
}
// Validate attributes
if len(result.Attributes) != 4 {
t.Fatalf("len(Attributes) = %v, want 4", len(result.Attributes))
}
// Check prompt field (TEXT)
promptAttr := result.Attributes[0]
if promptAttr.Identifier != "prompt" {
t.Errorf("Attributes[0].Identifier = %v, want prompt", promptAttr.Identifier)
}
if promptAttr.Attribute != "prompt" {
t.Errorf("Attributes[0].Attribute = %v, want prompt", promptAttr.Attribute)
}
if promptAttr.Type != "TEXT" {
t.Errorf("Attributes[0].Type = %v, want TEXT", promptAttr.Type)
}
if promptAttr.Weight != 1.0 {
t.Errorf("Attributes[0].Weight = %v, want 1.0", promptAttr.Weight)
}
// Check response field (TEXT)
responseAttr := result.Attributes[1]
if responseAttr.Identifier != "response" {
t.Errorf("Attributes[1].Identifier = %v, want response", responseAttr.Identifier)
}
if responseAttr.Attribute != "response" {
t.Errorf("Attributes[1].Attribute = %v, want response", responseAttr.Attribute)
}
if responseAttr.Type != "TEXT" {
t.Errorf("Attributes[1].Type = %v, want TEXT", responseAttr.Type)
}
// Check exact_digest field (TAG)
tagAttr := result.Attributes[2]
if tagAttr.Identifier != "exact_digest" {
t.Errorf("Attributes[2].Identifier = %v, want exact_digest", tagAttr.Identifier)
}
if tagAttr.Attribute != "exact_digest" {
t.Errorf("Attributes[2].Attribute = %v, want exact_digest", tagAttr.Attribute)
}
if tagAttr.Type != "TAG" {
t.Errorf("Attributes[2].Type = %v, want TAG", tagAttr.Type)
}
// Check prompt_vector field (VECTOR)
vectorAttr := result.Attributes[3]
if vectorAttr.Identifier != "prompt_vector" {
t.Errorf("Attributes[3].Identifier = %v, want prompt_vector", vectorAttr.Identifier)
}
if vectorAttr.Attribute != "prompt_vector" {
t.Errorf("Attributes[3].Attribute = %v, want prompt_vector", vectorAttr.Attribute)
}
if vectorAttr.Type != "VECTOR" {
t.Errorf("Attributes[3].Type = %v, want VECTOR", vectorAttr.Type)
}
if vectorAttr.Algorithm != "HNSW" {
t.Errorf("Attributes[3].Algorithm = %v, want HNSW", vectorAttr.Algorithm)
}
if vectorAttr.DataType != "FLOAT32" {
t.Errorf("Attributes[3].DataType = %v, want FLOAT32", vectorAttr.DataType)
}
if vectorAttr.Dim != 1536 {
t.Errorf("Attributes[3].Dim = %v, want 1536", vectorAttr.Dim)
}
if vectorAttr.DistanceMetric != "COSINE" {
t.Errorf("Attributes[3].DistanceMetric = %v, want COSINE", vectorAttr.DistanceMetric)
}
if vectorAttr.M != 16 {
t.Errorf("Attributes[3].M = %v, want 16", vectorAttr.M)
}
if vectorAttr.EFConstruction != 64 {
t.Errorf("Attributes[3].EFConstruction = %v, want 64", vectorAttr.EFConstruction)
}
// Validate numeric fields
if result.NumDocs != 0 {
t.Errorf("NumDocs = %v, want 0", result.NumDocs)
}
if result.MaxDocID != 0 {
t.Errorf("MaxDocID = %v, want 0", result.MaxDocID)
}
if result.NumTerms != 0 {
t.Errorf("NumTerms = %v, want 0", result.NumTerms)
}
if result.NumRecords != 0 {
t.Errorf("NumRecords = %v, want 0", result.NumRecords)
}
if result.Indexing != 0 {
t.Errorf("Indexing = %v, want 0", result.Indexing)
}
if result.PercentIndexed != 1.0 {
t.Errorf("PercentIndexed = %v, want 1.0", result.PercentIndexed)
}
if result.HashIndexingFailures != 0 {
t.Errorf("HashIndexingFailures = %v, want 0", result.HashIndexingFailures)
}
if result.Cleaning != 0 {
t.Errorf("Cleaning = %v, want 0", result.Cleaning)
}
if result.NumberOfUses != 2 {
t.Errorf("NumberOfUses = %v, want 2", result.NumberOfUses)
}
// Validate average stats (should be "nan" for empty index)
if result.RecordsPerDocAvg != "nan" {
t.Errorf("RecordsPerDocAvg = %v, want nan", result.RecordsPerDocAvg)
}
if result.BytesPerRecordAvg != "nan" {
t.Errorf("BytesPerRecordAvg = %v, want nan", result.BytesPerRecordAvg)
}
if result.OffsetsPerTermAvg != "nan" {
t.Errorf("OffsetsPerTermAvg = %v, want nan", result.OffsetsPerTermAvg)
}
if result.OffsetBitsPerRecordAvg != "nan" {
t.Errorf("OffsetBitsPerRecordAvg = %v, want nan", result.OffsetBitsPerRecordAvg)
}
// Validate cursor stats
if result.CursorStats.GlobalIdle != 0 {
t.Errorf("CursorStats.GlobalIdle = %v, want 0", result.CursorStats.GlobalIdle)
}
if result.CursorStats.GlobalTotal != 0 {
t.Errorf("CursorStats.GlobalTotal = %v, want 0", result.CursorStats.GlobalTotal)
}
if result.CursorStats.IndexCapacity != 128 {
t.Errorf("CursorStats.IndexCapacity = %v, want 128", result.CursorStats.IndexCapacity)
}
if result.CursorStats.IndexTotal != 0 {
t.Errorf("CursorStats.IndexTotal = %v, want 0", result.CursorStats.IndexTotal)
}
// Validate dialect stats
if result.DialectStats["dialect_1"] != 0 {
t.Errorf("DialectStats[dialect_1] = %v, want 0", result.DialectStats["dialect_1"])
}
if result.DialectStats["dialect_2"] != 0 {
t.Errorf("DialectStats[dialect_2] = %v, want 0", result.DialectStats["dialect_2"])
}
if result.DialectStats["dialect_3"] != 0 {
t.Errorf("DialectStats[dialect_3] = %v, want 0", result.DialectStats["dialect_3"])
}
if result.DialectStats["dialect_4"] != 0 {
t.Errorf("DialectStats[dialect_4] = %v, want 0", result.DialectStats["dialect_4"])
}
// Validate GC stats
if result.GCStats.BytesCollected != 0 {
t.Errorf("GCStats.BytesCollected = %v, want 0", result.GCStats.BytesCollected)
}
if result.GCStats.TotalMsRun != 0 {
t.Errorf("GCStats.TotalMsRun = %v, want 0", result.GCStats.TotalMsRun)
}
if result.GCStats.TotalCycles != 0 {
t.Errorf("GCStats.TotalCycles = %v, want 0", result.GCStats.TotalCycles)
}
if result.GCStats.AverageCycleTimeMs != "nan" {
t.Errorf("GCStats.AverageCycleTimeMs = %v, want nan", result.GCStats.AverageCycleTimeMs)
}
// Validate Index Errors
if result.IndexErrors.IndexingFailures != 0 {
t.Errorf("IndexErrors.IndexingFailures = %v, want 0", result.IndexErrors.IndexingFailures)
}
if result.IndexErrors.LastIndexingError != "N/A" {
t.Errorf("IndexErrors.LastIndexingError = %v, want N/A", result.IndexErrors.LastIndexingError)
}
if result.IndexErrors.LastIndexingErrorKey != "N/A" {
t.Errorf("IndexErrors.LastIndexingErrorKey = %v, want N/A", result.IndexErrors.LastIndexingErrorKey)
}
// Validate field statistics
if len(result.FieldStatistics) != 4 {
t.Fatalf("len(FieldStatistics) = %v, want 4", len(result.FieldStatistics))
}
expectedIdentifiers := map[string]bool{
"prompt": true,
"response": true,
"exact_digest": true,
"prompt_vector": true,
}
for _, fieldStat := range result.FieldStatistics {
if !expectedIdentifiers[fieldStat.Identifier] {
t.Errorf("Unexpected field statistic identifier: %v", fieldStat.Identifier)
}
if fieldStat.IndexErrors.IndexingFailures != 0 {
t.Errorf("FieldStatistic[%s].IndexErrors.IndexingFailures = %v, want 0", fieldStat.Identifier, fieldStat.IndexErrors.IndexingFailures)
}
if fieldStat.IndexErrors.LastIndexingError != "N/A" {
t.Errorf("FieldStatistic[%s].IndexErrors.LastIndexingError = %v, want N/A", fieldStat.Identifier, fieldStat.IndexErrors.LastIndexingError)
}
if fieldStat.IndexErrors.LastIndexingErrorKey != "N/A" {
t.Errorf("FieldStatistic[%s].IndexErrors.LastIndexingErrorKey = %v, want N/A", fieldStat.Identifier, fieldStat.IndexErrors.LastIndexingErrorKey)
}
}
t.Logf("Successfully parsed FT.INFO response with %d attributes", len(result.Attributes))
}