Files
ragflow/internal/server/config/doc_engine_config.go

204 lines
5.5 KiB
Go
Raw Normal View History

//
// Copyright 2026 The InfiniFlow Authors. All Rights Reserved.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
//
package config
import (
"github.com/spf13/viper"
)
type DocEngineConfig struct {
ES ElasticsearchConfig `mapstructure:"es"`
Infinity InfinityConfig `mapstructure:"infinity"`
feat(serenedb): add SereneDB doc-store engine (Go + Python connectors) (#17375) ## What Adds [**SereneDB**](https://serenedb.com) as a selectable doc-store engine on **both** RAGFlow paths: - the **Go** `DocEngine` (`internal/engine/serenedb`), alongside Elasticsearch and Infinity; - the **Python** `DocStoreConnection` (`rag/utils/serenedb_conn.py`) + `DOC_ENGINE=serenedb` registration. SereneDB is a PostgreSQL-wire engine (DuckDB execution) whose single inverted index carries **both** a scored text column (`@@`, BM25) and an IVF vector column (`<#>`, inner product), so hybrid search is one SQL statement. The Go engine connects with `database/sql` + `lib/pq` (already a dependency, no new module); the Python connector uses psycopg2 (already a dependency). ## Storage model One table per tenant with `kb_id` as a filter column - the **Elasticsearch / OceanBase** model, not Infinity's per-dataset tables. This keeps BM25 statistics (IDF, avgdl) computed over the whole tenant corpus (global IDF). Both connectors use this identical layout, so they are storage- and retrieval-compatible: `hybrid` proxy routing and Python↔Go switching are safe. On the Python side the connector is wired as OceanBase's plain-SQL sibling (chunk_data JSON metadata, inline chunk vectors, verbatim ES field names); the ES tokenizer path is unchanged. Metadata stays one table per tenant (`ragflow_doc_meta_<tenant>`). The query shapes mirror the Python connector, including the five empirically-found landmines: the scored dictionary needs `frequency + norm` (else `BM25()` silently returns 0.0), the `@@` query is the tokenized query, the scored lexical branch matches one column, vectors use an L2-normalized shadow column with `ip`/`sq8`, and the similarity threshold goes directly in the ANN scan's `WHERE`. **Minimum engine version: SereneDB 26.07.4.** --------- Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
2026-08-04 08:16:39 +02:00
SereneDB SereneDBConfig `mapstructure:"serenedb"`
}
// ElasticsearchConfig Elasticsearch configuration
type ElasticsearchConfig struct {
Hosts string `mapstructure:"hosts"`
Username string `mapstructure:"username"`
Password string `mapstructure:"password"`
}
// InfinityConfig Infinity configuration
type InfinityConfig struct {
URI string `mapstructure:"uri"`
PostgresPort int `mapstructure:"postgres_port"`
DBName string `mapstructure:"db_name"`
MappingFileName string `mapstructure:"mapping_file_name"`
DocMetaMappingFileName string `mapstructure:"doc_meta_mapping_file_name"`
}
feat(serenedb): add SereneDB doc-store engine (Go + Python connectors) (#17375) ## What Adds [**SereneDB**](https://serenedb.com) as a selectable doc-store engine on **both** RAGFlow paths: - the **Go** `DocEngine` (`internal/engine/serenedb`), alongside Elasticsearch and Infinity; - the **Python** `DocStoreConnection` (`rag/utils/serenedb_conn.py`) + `DOC_ENGINE=serenedb` registration. SereneDB is a PostgreSQL-wire engine (DuckDB execution) whose single inverted index carries **both** a scored text column (`@@`, BM25) and an IVF vector column (`<#>`, inner product), so hybrid search is one SQL statement. The Go engine connects with `database/sql` + `lib/pq` (already a dependency, no new module); the Python connector uses psycopg2 (already a dependency). ## Storage model One table per tenant with `kb_id` as a filter column - the **Elasticsearch / OceanBase** model, not Infinity's per-dataset tables. This keeps BM25 statistics (IDF, avgdl) computed over the whole tenant corpus (global IDF). Both connectors use this identical layout, so they are storage- and retrieval-compatible: `hybrid` proxy routing and Python↔Go switching are safe. On the Python side the connector is wired as OceanBase's plain-SQL sibling (chunk_data JSON metadata, inline chunk vectors, verbatim ES field names); the ES tokenizer path is unchanged. Metadata stays one table per tenant (`ragflow_doc_meta_<tenant>`). The query shapes mirror the Python connector, including the five empirically-found landmines: the scored dictionary needs `frequency + norm` (else `BM25()` silently returns 0.0), the `@@` query is the tokenized query, the scored lexical branch matches one column, vectors use an L2-normalized shadow column with `ip`/`sq8`, and the similarity threshold goes directly in the ANN scan's `WHERE`. **Minimum engine version: SereneDB 26.07.4.** --------- Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
2026-08-04 08:16:39 +02:00
// SereneDBConfig SereneDB configuration. SereneDB speaks the PostgreSQL wire
// protocol, so the engine connects with database/sql + lib/pq.
type SereneDBConfig struct {
Host string `mapstructure:"host"`
Port int `mapstructure:"port"`
User string `mapstructure:"user"`
Password string `mapstructure:"password"`
DBName string `mapstructure:"db_name"`
// SSLMode is the lib/pq sslmode; empty defaults to "disable" for a trusted
// local deployment. Set it (e.g. "require") to encrypt the connection.
SSLMode string `mapstructure:"ssl_mode"`
}
func (c *Config) ParseDocEngineConfig(v *viper.Viper) error {
c.parseInfinityConfig(v)
c.parseElasticsearchConfig(v)
feat(serenedb): add SereneDB doc-store engine (Go + Python connectors) (#17375) ## What Adds [**SereneDB**](https://serenedb.com) as a selectable doc-store engine on **both** RAGFlow paths: - the **Go** `DocEngine` (`internal/engine/serenedb`), alongside Elasticsearch and Infinity; - the **Python** `DocStoreConnection` (`rag/utils/serenedb_conn.py`) + `DOC_ENGINE=serenedb` registration. SereneDB is a PostgreSQL-wire engine (DuckDB execution) whose single inverted index carries **both** a scored text column (`@@`, BM25) and an IVF vector column (`<#>`, inner product), so hybrid search is one SQL statement. The Go engine connects with `database/sql` + `lib/pq` (already a dependency, no new module); the Python connector uses psycopg2 (already a dependency). ## Storage model One table per tenant with `kb_id` as a filter column - the **Elasticsearch / OceanBase** model, not Infinity's per-dataset tables. This keeps BM25 statistics (IDF, avgdl) computed over the whole tenant corpus (global IDF). Both connectors use this identical layout, so they are storage- and retrieval-compatible: `hybrid` proxy routing and Python↔Go switching are safe. On the Python side the connector is wired as OceanBase's plain-SQL sibling (chunk_data JSON metadata, inline chunk vectors, verbatim ES field names); the ES tokenizer path is unchanged. Metadata stays one table per tenant (`ragflow_doc_meta_<tenant>`). The query shapes mirror the Python connector, including the five empirically-found landmines: the scored dictionary needs `frequency + norm` (else `BM25()` silently returns 0.0), the `@@` query is the tokenized query, the scored lexical branch matches one column, vectors use an L2-normalized shadow column with `ip`/`sq8`, and the similarity threshold goes directly in the ANN scan's `WHERE`. **Minimum engine version: SereneDB 26.07.4.** --------- Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
2026-08-04 08:16:39 +02:00
c.parseSereneDBConfig(v)
return nil
}
feat(serenedb): add SereneDB doc-store engine (Go + Python connectors) (#17375) ## What Adds [**SereneDB**](https://serenedb.com) as a selectable doc-store engine on **both** RAGFlow paths: - the **Go** `DocEngine` (`internal/engine/serenedb`), alongside Elasticsearch and Infinity; - the **Python** `DocStoreConnection` (`rag/utils/serenedb_conn.py`) + `DOC_ENGINE=serenedb` registration. SereneDB is a PostgreSQL-wire engine (DuckDB execution) whose single inverted index carries **both** a scored text column (`@@`, BM25) and an IVF vector column (`<#>`, inner product), so hybrid search is one SQL statement. The Go engine connects with `database/sql` + `lib/pq` (already a dependency, no new module); the Python connector uses psycopg2 (already a dependency). ## Storage model One table per tenant with `kb_id` as a filter column - the **Elasticsearch / OceanBase** model, not Infinity's per-dataset tables. This keeps BM25 statistics (IDF, avgdl) computed over the whole tenant corpus (global IDF). Both connectors use this identical layout, so they are storage- and retrieval-compatible: `hybrid` proxy routing and Python↔Go switching are safe. On the Python side the connector is wired as OceanBase's plain-SQL sibling (chunk_data JSON metadata, inline chunk vectors, verbatim ES field names); the ES tokenizer path is unchanged. Metadata stays one table per tenant (`ragflow_doc_meta_<tenant>`). The query shapes mirror the Python connector, including the five empirically-found landmines: the scored dictionary needs `frequency + norm` (else `BM25()` silently returns 0.0), the `@@` query is the tokenized query, the scored lexical branch matches one column, vectors use an L2-normalized shadow column with `ip`/`sq8`, and the similarity threshold goes directly in the ANN scan's `WHERE`. **Minimum engine version: SereneDB 26.07.4.** --------- Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
2026-08-04 08:16:39 +02:00
func (c *Config) parseSereneDBConfig(v *viper.Viper) {
// Default SereneDB config
c.docEngine.SereneDB.Host = "localhost"
c.docEngine.SereneDB.Port = 5432
c.docEngine.SereneDB.User = "postgres"
c.docEngine.SereneDB.DBName = "default_db"
if !v.IsSet("serenedb") {
return
}
sub := v.Sub("serenedb")
if sub == nil {
return
}
if sub.IsSet("host") {
c.docEngine.SereneDB.Host = sub.GetString("host")
}
if sub.IsSet("port") {
c.docEngine.SereneDB.Port = sub.GetInt("port")
}
if sub.IsSet("user") {
c.docEngine.SereneDB.User = sub.GetString("user")
}
if sub.IsSet("password") {
c.docEngine.SereneDB.Password = sub.GetString("password")
}
if sub.IsSet("db_name") {
c.docEngine.SereneDB.DBName = sub.GetString("db_name")
}
if sub.IsSet("ssl_mode") {
c.docEngine.SereneDB.SSLMode = sub.GetString("ssl_mode")
}
}
func (c *Config) parseInfinityConfig(v *viper.Viper) {
// Default Infinity config
c.docEngine.Infinity.URI = "localhost:23817"
c.docEngine.Infinity.PostgresPort = 5432
c.docEngine.Infinity.DBName = "default_db"
c.docEngine.Infinity.MappingFileName = "infinity_mapping.json"
c.docEngine.Infinity.DocMetaMappingFileName = "doc_meta_infinity_mapping.json"
if !v.IsSet("infinity") {
return
}
sub := v.Sub("infinity")
if sub == nil {
return
}
if sub.IsSet("uri") {
c.docEngine.Infinity.URI = sub.GetString("uri")
}
if sub.IsSet("postgres_port") {
c.docEngine.Infinity.PostgresPort = sub.GetInt("postgres_port")
}
if sub.IsSet("db_name") {
c.docEngine.Infinity.DBName = sub.GetString("db_name")
}
if sub.IsSet("mapping_file_name") {
c.docEngine.Infinity.MappingFileName = sub.GetString("mapping_file_name")
}
if sub.IsSet("doc_meta_mapping_file_name") {
c.docEngine.Infinity.DocMetaMappingFileName = sub.GetString("doc_meta_mapping_file_name")
}
}
func (c *Config) parseElasticsearchConfig(v *viper.Viper) {
// Default Elasticsearch config
c.docEngine.ES.Hosts = "http://localhost:1200"
c.docEngine.ES.Username = "elastic"
c.docEngine.ES.Password = "infini_rag_flow"
if !v.IsSet("es") {
return
}
sub := v.Sub("es")
if sub == nil {
return
}
if sub.IsSet("hosts") {
c.docEngine.ES.Hosts = sub.GetString("hosts")
}
if sub.IsSet("username") {
c.docEngine.ES.Username = sub.GetString("username")
}
if sub.IsSet("password") {
c.docEngine.ES.Password = sub.GetString("password")
}
}
func (c *Config) GetElasticsearchConfig() ElasticsearchConfig {
return c.docEngine.ES
}
func (e ElasticsearchConfig) ExportConfigs() map[string]interface{} {
var esConfigs map[string]interface{}
esConfigs = make(map[string]interface{})
esConfigs["hosts"] = e.Hosts
esConfigs["username"] = e.Username
esConfigs["password"] = e.Password
return esConfigs
}
func (c *Config) IsElasticConfigured() bool {
return c.docEngine.ES.Hosts != ""
}
func (c *Config) GetInfinityConfig() InfinityConfig {
return c.docEngine.Infinity
}
func (i InfinityConfig) ExportConfigs() map[string]interface{} {
var infinityConfigs map[string]interface{}
infinityConfigs = make(map[string]interface{})
infinityConfigs["uri"] = i.URI
infinityConfigs["postgres_port"] = i.PostgresPort
infinityConfigs["db_name"] = i.DBName
infinityConfigs["mapping_file_name"] = i.MappingFileName
infinityConfigs["doc_meta_mapping_file_name"] = i.DocMetaMappingFileName
return infinityConfigs
}
feat(serenedb): add SereneDB doc-store engine (Go + Python connectors) (#17375) ## What Adds [**SereneDB**](https://serenedb.com) as a selectable doc-store engine on **both** RAGFlow paths: - the **Go** `DocEngine` (`internal/engine/serenedb`), alongside Elasticsearch and Infinity; - the **Python** `DocStoreConnection` (`rag/utils/serenedb_conn.py`) + `DOC_ENGINE=serenedb` registration. SereneDB is a PostgreSQL-wire engine (DuckDB execution) whose single inverted index carries **both** a scored text column (`@@`, BM25) and an IVF vector column (`<#>`, inner product), so hybrid search is one SQL statement. The Go engine connects with `database/sql` + `lib/pq` (already a dependency, no new module); the Python connector uses psycopg2 (already a dependency). ## Storage model One table per tenant with `kb_id` as a filter column - the **Elasticsearch / OceanBase** model, not Infinity's per-dataset tables. This keeps BM25 statistics (IDF, avgdl) computed over the whole tenant corpus (global IDF). Both connectors use this identical layout, so they are storage- and retrieval-compatible: `hybrid` proxy routing and Python↔Go switching are safe. On the Python side the connector is wired as OceanBase's plain-SQL sibling (chunk_data JSON metadata, inline chunk vectors, verbatim ES field names); the ES tokenizer path is unchanged. Metadata stays one table per tenant (`ragflow_doc_meta_<tenant>`). The query shapes mirror the Python connector, including the five empirically-found landmines: the scored dictionary needs `frequency + norm` (else `BM25()` silently returns 0.0), the `@@` query is the tokenized query, the scored lexical branch matches one column, vectors use an L2-normalized shadow column with `ip`/`sq8`, and the similarity threshold goes directly in the ANN scan's `WHERE`. **Minimum engine version: SereneDB 26.07.4.** --------- Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
2026-08-04 08:16:39 +02:00
func (c *Config) GetSereneDBConfig() SereneDBConfig {
return c.docEngine.SereneDB
}