From b4825166a7c5890407b45f8247090a521e1c8744 Mon Sep 17 00:00:00 2001
From: Haruko386 <tryeverypossible@163.com>
Date: Thu, 2 Jul 2026 12:06:53 +0800
Subject: [PATCH] fix: JSONMap scan in dataset index chunking config (#16489)

### Summary

As title

This PR fixes dataset index task creation failing with unsupported data
type: entity.JSONMap when loading document chunking config.

#### issues:
```
2026/06/30 15:19:40 /home/infiniflow/Documents/development/ragflow/internal/dao/document.go:162
[error] unsupported data type: ragflow/internal/entity.JSONMap
```

#### Changes:
+ Adds the missing GORM type:longtext tag to ParserConfig in
DocumentDAO.GetChunkingConfig.
+ Adds a DAO regression test covering GetChunkingConfig joins across
document, knowledgebase, and tenant while scanning parser_config.
---
 internal/dao/document.go      |  2 +-
 internal/dao/document_test.go | 60 +++++++++++++++++++++++++++++++++++
 2 files changed, 61 insertions(+), 1 deletion(-)

diff --git a/internal/dao/document.go b/internal/dao/document.go
index 3d0adefc8b..281affd230 100644
--- a/internal/dao/document.go
+++ b/internal/dao/document.go
@@ -130,7 +130,7 @@ func (dao *DocumentDAO) GetChunkingConfig(docID string) (map[string]interface{},
 		ID           string         `gorm:"column:id"`
 		KbID         string         `gorm:"column:kb_id"`
 		ParserID     string         `gorm:"column:parser_id"`
-		ParserConfig entity.JSONMap `gorm:"column:parser_config"`
+		ParserConfig entity.JSONMap `gorm:"column:parser_config;type:longtext"`
 		Size         int64          `gorm:"column:size"`
 		ContentHash  *string        `gorm:"column:content_hash"`
 		Language     *string        `gorm:"column:language"`
diff --git a/internal/dao/document_test.go b/internal/dao/document_test.go
index 31e8f36a99..484a94f3f0 100644
--- a/internal/dao/document_test.go
+++ b/internal/dao/document_test.go
@@ -35,6 +35,8 @@ func setupDocumentTestDB(t *testing.T) *gorm.DB {
 	}
 	if err := db.AutoMigrate(
 		&entity.Document{},
+		&entity.Knowledgebase{},
+		&entity.Tenant{},
 	); err != nil {
 		t.Fatalf("failed to migrate: %v", err)
 	}
@@ -164,4 +166,62 @@ func TestDocumentGetByDocumentIDAndDatasetIDUsesKBID(t *testing.T) {
 	}
 }
 
+func TestDocumentGetChunkingConfigScansParserConfig(t *testing.T) {
+	db := setupDocumentTestDB(t)
+	pushDocDB(t, db)
+
+	if err := db.Create(&entity.Tenant{
+		ID:        "tenant1",
+		LLMID:     "llm1",
+		EmbdID:    "embd1",
+		ASRID:     "asr1",
+		Img2TxtID: "img2txt1",
+		RerankID:  "rerank1",
+		ParserIDs: "naive",
+	}).Error; err != nil {
+		t.Fatalf("create tenant: %v", err)
+	}
+	if err := db.Create(&entity.Knowledgebase{
+		ID:           "kb1",
+		TenantID:     "tenant1",
+		Name:         "Dataset 1",
+		Language:     sp("English"),
+		EmbdID:       "kb-embd1",
+		Permission:   "me",
+		CreatedBy:    "user1",
+		ParserID:     "naive",
+		ParserConfig: entity.JSONMap{},
+	}).Error; err != nil {
+		t.Fatalf("create knowledgebase: %v", err)
+	}
+	if err := db.Create(&entity.Document{
+		ID:           "doc1",
+		KbID:         "kb1",
+		ParserID:     "naive",
+		ParserConfig: entity.JSONMap{"chunk_token_num": float64(128), "delimiter": "\\n"},
+		SourceType:   "local",
+		Type:         "doc",
+		CreatedBy:    "user1",
+		Size:         42,
+		Suffix:       ".txt",
+	}).Error; err != nil {
+		t.Fatalf("create document: %v", err)
+	}
+
+	config, err := NewDocumentDAO().GetChunkingConfig("doc1")
+	if err != nil {
+		t.Fatalf("GetChunkingConfig failed: %v", err)
+	}
+	parserConfig, ok := config["parser_config"].(entity.JSONMap)
+	if !ok {
+		t.Fatalf("parser_config type = %T, want entity.JSONMap", config["parser_config"])
+	}
+	if parserConfig["chunk_token_num"] != float64(128) || parserConfig["delimiter"] != "\\n" {
+		t.Fatalf("unexpected parser_config: %#v", parserConfig)
+	}
+	if config["tenant_id"] != "tenant1" || config["embd_id"] != "kb-embd1" {
+		t.Fatalf("unexpected joined config: %#v", config)
+	}
+}
+
 func sp(s string) *string { return &s }