From 28478cdc4147da511c87a4dce60b37afb9299858 Mon Sep 17 00:00:00 2001
From: kenwoodjw <blackxin55+@gmail.com>
Date: Wed, 30 Jul 2025 16:13:45 +0800
Subject: [PATCH] feat: support metadata condition filter string array (#23111)

Signed-off-by: kenwoodjw <blackxin55+@gmail.com>
---
 api/core/app/app_config/entities.py           |  2 ++
 api/core/rag/entities/metadata_entities.py    |  2 ++
 .../nodes/knowledge_retrieval/entities.py     |  2 ++
 .../knowledge_retrieval_node.py               | 22 +++++++++++++++++++
 .../metadata/condition-list/utils.ts          |  2 ++
 web/i18n/zh-Hans/workflow.ts                  |  4 ++--
 6 files changed, 32 insertions(+), 2 deletions(-)

diff --git a/api/core/app/app_config/entities.py b/api/core/app/app_config/entities.py
index 75bd2f677..0df0aa59b 100644
--- a/api/core/app/app_config/entities.py
+++ b/api/core/app/app_config/entities.py
@@ -148,6 +148,8 @@ SupportedComparisonOperator = Literal[
     "is not",
     "empty",
     "not empty",
+    "in",
+    "not in",
     # for number
     "=",
     "≠",
diff --git a/api/core/rag/entities/metadata_entities.py b/api/core/rag/entities/metadata_entities.py
index 6ef932ad2..1f054bccd 100644
--- a/api/core/rag/entities/metadata_entities.py
+++ b/api/core/rag/entities/metadata_entities.py
@@ -13,6 +13,8 @@ SupportedComparisonOperator = Literal[
     "is not",
     "empty",
     "not empty",
+    "in",
+    "not in",
     # for number
     "=",
     "≠",
diff --git a/api/core/workflow/nodes/knowledge_retrieval/entities.py b/api/core/workflow/nodes/knowledge_retrieval/entities.py
index f1767bdf9..b71271abe 100644
--- a/api/core/workflow/nodes/knowledge_retrieval/entities.py
+++ b/api/core/workflow/nodes/knowledge_retrieval/entities.py
@@ -74,6 +74,8 @@ SupportedComparisonOperator = Literal[
     "is not",
     "empty",
     "not empty",
+    "in",
+    "not in",
     # for number
     "=",
     "≠",
diff --git a/api/core/workflow/nodes/knowledge_retrieval/knowledge_retrieval_node.py b/api/core/workflow/nodes/knowledge_retrieval/knowledge_retrieval_node.py
index e041e217c..7303b6850 100644
--- a/api/core/workflow/nodes/knowledge_retrieval/knowledge_retrieval_node.py
+++ b/api/core/workflow/nodes/knowledge_retrieval/knowledge_retrieval_node.py
@@ -602,6 +602,28 @@ class KnowledgeRetrievalNode(BaseNode):
                         **{key: metadata_name, key_value: f"%{value}"}
                     )
                 )
+            case "in":
+                if isinstance(value, str):
+                    escaped_values = [v.strip().replace("'", "''") for v in str(value).split(",")]
+                    escaped_value_str = ",".join(escaped_values)
+                else:
+                    escaped_value_str = str(value)
+                filters.append(
+                    (text(f"documents.doc_metadata ->> :{key} = any(string_to_array(:{key_value},','))")).params(
+                        **{key: metadata_name, key_value: escaped_value_str}
+                    )
+                )
+            case "not in":
+                if isinstance(value, str):
+                    escaped_values = [v.strip().replace("'", "''") for v in str(value).split(",")]
+                    escaped_value_str = ",".join(escaped_values)
+                else:
+                    escaped_value_str = str(value)
+                filters.append(
+                    (text(f"documents.doc_metadata ->> :{key} != all(string_to_array(:{key_value},','))")).params(
+                        **{key: metadata_name, key_value: escaped_value_str}
+                    )
+                )
             case "=" | "is":
                 if isinstance(value, str):
                     filters.append(Document.doc_metadata[metadata_name] == f'"{value}"')
diff --git a/web/app/components/workflow/nodes/knowledge-retrieval/components/metadata/condition-list/utils.ts b/web/app/components/workflow/nodes/knowledge-retrieval/components/metadata/condition-list/utils.ts
index 639702399..10ee1aff1 100644
--- a/web/app/components/workflow/nodes/knowledge-retrieval/components/metadata/condition-list/utils.ts
+++ b/web/app/components/workflow/nodes/knowledge-retrieval/components/metadata/condition-list/utils.ts
@@ -32,6 +32,8 @@ export const getOperators = (type?: MetadataFilteringVariableType) => {
         ComparisonOperator.endWith,
         ComparisonOperator.empty,
         ComparisonOperator.notEmpty,
+        ComparisonOperator.in,
+        ComparisonOperator.notIn,
       ]
     case MetadataFilteringVariableType.number:
       return [
diff --git a/web/i18n/zh-Hans/workflow.ts b/web/i18n/zh-Hans/workflow.ts
index 1f0300ae2..b1c28c466 100644
--- a/web/i18n/zh-Hans/workflow.ts
+++ b/web/i18n/zh-Hans/workflow.ts
@@ -591,8 +591,8 @@ const translation = {
         'not empty': '不为空',
         'null': '空',
         'not null': '不为空',
-        'in': '是',
-        'not in': '不是',
+        'in': '在',
+        'not in': '不在',
         'all of': '全部是',
         'exists': '存在',
         'not exists': '不存在',