fix: test

Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>
Co-authored-by: -LAN- <laipz8200@outlook.com>

.github/workflows/build-push.yml

@@ -8,6 +8,7 @@ on:
       - "deploy/enterprise"
       - "build/**"
       - "release/e-*"
+      - "hotfix/**"
     tags:
       - "*"
 

api/.env.example

@@ -158,7 +158,7 @@ WEB_API_CORS_ALLOW_ORIGINS=http://localhost:3000,*
 CONSOLE_CORS_ALLOW_ORIGINS=http://localhost:3000,*
 
 # Vector database configuration
-# Supported values are `weaviate`, `qdrant`, `milvus`, `myscale`, `relyt`, `pgvector`, `pgvecto-rs`, `chroma`, `opensearch`, `oracle`, `tencent`, `elasticsearch`, `elasticsearch-ja`, `analyticdb`, `couchbase`, `vikingdb`, `oceanbase`, `opengauss`, `tablestore`,`vastbase`,`tidb`,`tidb_on_qdrant`,`baidu`,`lindorm`,`huawei_cloud`,`upstash`, `matrixone`.
+# Supported values are `weaviate`, `qdrant`, `milvus`, `myscale`, `relyt`, `pgvector`, `pgvecto-rs`, `chroma`, `opensearch`, `oracle`, `tencent`, `elasticsearch`, `elasticsearch-ja`, `analyticdb`, `couchbase`, `vikingdb`, `oceanbase`, `opengauss`, `tablestore`,`vastbase`,`tidb`,`tidb_on_qdrant`,`baidu`,`lindorm`,`huawei_cloud`,`upstash`, `matrixone`, `pinecone`.
 VECTOR_STORE=weaviate
 # Prefix used to create collection name in vector database
 VECTOR_INDEX_NAME_PREFIX=Vector_index
@@ -365,6 +365,16 @@ PROMPT_GENERATION_MAX_TOKENS=512
 CODE_GENERATION_MAX_TOKENS=1024
 PLUGIN_BASED_TOKEN_COUNTING_ENABLED=false
 
+# Pinecone configuration, only available when VECTOR_STORE is `pinecone`
+PINECONE_API_KEY=your-pinecone-api-key
+PINECONE_ENVIRONMENT=your-pinecone-environment
+PINECONE_INDEX_NAME=dify-index
+PINECONE_CLIENT_TIMEOUT=30
+PINECONE_BATCH_SIZE=100
+PINECONE_METRIC=cosine
+PINECONE_PODS=1
+PINECONE_POD_TYPE=s1
+
 # Mail configuration, support: resend, smtp, sendgrid
 MAIL_TYPE=
 # If using SendGrid, use the 'from' field for authentication if necessary.

api/commands.py

@@ -10,6 +10,7 @@ from flask import current_app
 from pydantic import TypeAdapter
 from sqlalchemy import select
 from sqlalchemy.exc import SQLAlchemyError
+from sqlalchemy.orm import sessionmaker
 
 from configs import dify_config
 from constants.languages import languages
@@ -61,31 +62,30 @@ def reset_password(email, new_password, password_confirm):
     if str(new_password).strip() != str(password_confirm).strip():
         click.echo(click.style("Passwords do not match.", fg="red"))
         return
+    with sessionmaker(db.engine, expire_on_commit=False).begin() as session:
+        account = session.query(Account).where(Account.email == email).one_or_none()
 
-    account = db.session.query(Account).where(Account.email == email).one_or_none()
+        if not account:
+            click.echo(click.style(f"Account not found for email: {email}", fg="red"))
+            return
 
-    if not account:
-        click.echo(click.style(f"Account not found for email: {email}", fg="red"))
-        return
+        try:
+            valid_password(new_password)
+        except:
+            click.echo(click.style(f"Invalid password. Must match {password_pattern}", fg="red"))
+            return
 
-    try:
-        valid_password(new_password)
-    except:
-        click.echo(click.style(f"Invalid password. Must match {password_pattern}", fg="red"))
-        return
+        # generate password salt
+        salt = secrets.token_bytes(16)
+        base64_salt = base64.b64encode(salt).decode()
 
-    # generate password salt
-    salt = secrets.token_bytes(16)
-    base64_salt = base64.b64encode(salt).decode()
-
-    # encrypt password with salt
-    password_hashed = hash_password(new_password, salt)
-    base64_password_hashed = base64.b64encode(password_hashed).decode()
-    account.password = base64_password_hashed
-    account.password_salt = base64_salt
-    db.session.commit()
-    AccountService.reset_login_error_rate_limit(email)
-    click.echo(click.style("Password reset successfully.", fg="green"))
+        # encrypt password with salt
+        password_hashed = hash_password(new_password, salt)
+        base64_password_hashed = base64.b64encode(password_hashed).decode()
+        account.password = base64_password_hashed
+        account.password_salt = base64_salt
+        AccountService.reset_login_error_rate_limit(email)
+        click.echo(click.style("Password reset successfully.", fg="green"))
 
 
 @click.command("reset-email", help="Reset the account email.")
@@ -100,22 +100,21 @@ def reset_email(email, new_email, email_confirm):
     if str(new_email).strip() != str(email_confirm).strip():
         click.echo(click.style("New emails do not match.", fg="red"))
         return
+    with sessionmaker(db.engine, expire_on_commit=False).begin() as session:
+        account = session.query(Account).where(Account.email == email).one_or_none()
 
-    account = db.session.query(Account).where(Account.email == email).one_or_none()
+        if not account:
+            click.echo(click.style(f"Account not found for email: {email}", fg="red"))
+            return
 
-    if not account:
-        click.echo(click.style(f"Account not found for email: {email}", fg="red"))
-        return
+        try:
+            email_validate(new_email)
+        except:
+            click.echo(click.style(f"Invalid email: {new_email}", fg="red"))
+            return
 
-    try:
-        email_validate(new_email)
-    except:
-        click.echo(click.style(f"Invalid email: {new_email}", fg="red"))
-        return
-
-    account.email = new_email
-    db.session.commit()
-    click.echo(click.style("Email updated successfully.", fg="green"))
+        account.email = new_email
+        click.echo(click.style("Email updated successfully.", fg="green"))
 
 
 @click.command(
@@ -139,25 +138,24 @@ def reset_encrypt_key_pair():
     if dify_config.EDITION != "SELF_HOSTED":
         click.echo(click.style("This command is only for SELF_HOSTED installations.", fg="red"))
         return
+    with sessionmaker(db.engine, expire_on_commit=False).begin() as session:
+        tenants = session.query(Tenant).all()
+        for tenant in tenants:
+            if not tenant:
+                click.echo(click.style("No workspaces found. Run /install first.", fg="red"))
+                return
 
-    tenants = db.session.query(Tenant).all()
-    for tenant in tenants:
-        if not tenant:
-            click.echo(click.style("No workspaces found. Run /install first.", fg="red"))
-            return
+            tenant.encrypt_public_key = generate_key_pair(tenant.id)
 
-        tenant.encrypt_public_key = generate_key_pair(tenant.id)
+            session.query(Provider).where(Provider.provider_type == "custom", Provider.tenant_id == tenant.id).delete()
+            session.query(ProviderModel).where(ProviderModel.tenant_id == tenant.id).delete()
 
-        db.session.query(Provider).where(Provider.provider_type == "custom", Provider.tenant_id == tenant.id).delete()
-        db.session.query(ProviderModel).where(ProviderModel.tenant_id == tenant.id).delete()
-        db.session.commit()
-
-        click.echo(
-            click.style(
-                f"Congratulations! The asymmetric key pair of workspace {tenant.id} has been reset.",
-                fg="green",
+            click.echo(
+                click.style(
+                    f"Congratulations! The asymmetric key pair of workspace {tenant.id} has been reset.",
+                    fg="green",
+                )
             )
-        )
 
 
 @click.command("vdb-migrate", help="Migrate vector db.")
@@ -182,14 +180,15 @@ def migrate_annotation_vector_database():
         try:
             # get apps info
             per_page = 50
-            apps = (
-                db.session.query(App)
-                .where(App.status == "normal")
-                .order_by(App.created_at.desc())
-                .limit(per_page)
-                .offset((page - 1) * per_page)
-                .all()
-            )
+            with sessionmaker(db.engine, expire_on_commit=False).begin() as session:
+                apps = (
+                    session.query(App)
+                    .where(App.status == "normal")
+                    .order_by(App.created_at.desc())
+                    .limit(per_page)
+                    .offset((page - 1) * per_page)
+                    .all()
+                )
             if not apps:
                 break
         except SQLAlchemyError:
@@ -203,26 +202,27 @@ def migrate_annotation_vector_database():
             )
             try:
                 click.echo(f"Creating app annotation index: {app.id}")
-                app_annotation_setting = (
-                    db.session.query(AppAnnotationSetting).where(AppAnnotationSetting.app_id == app.id).first()
-                )
+                with sessionmaker(db.engine, expire_on_commit=False).begin() as session:
+                    app_annotation_setting = (
+                        session.query(AppAnnotationSetting).where(AppAnnotationSetting.app_id == app.id).first()
+                    )
 
-                if not app_annotation_setting:
-                    skipped_count = skipped_count + 1
-                    click.echo(f"App annotation setting disabled: {app.id}")
-                    continue
-                # get dataset_collection_binding info
-                dataset_collection_binding = (
-                    db.session.query(DatasetCollectionBinding)
-                    .where(DatasetCollectionBinding.id == app_annotation_setting.collection_binding_id)
-                    .first()
-                )
-                if not dataset_collection_binding:
-                    click.echo(f"App annotation collection binding not found: {app.id}")
-                    continue
-                annotations = db.session.scalars(
-                    select(MessageAnnotation).where(MessageAnnotation.app_id == app.id)
-                ).all()
+                    if not app_annotation_setting:
+                        skipped_count = skipped_count + 1
+                        click.echo(f"App annotation setting disabled: {app.id}")
+                        continue
+                    # get dataset_collection_binding info
+                    dataset_collection_binding = (
+                        session.query(DatasetCollectionBinding)
+                        .where(DatasetCollectionBinding.id == app_annotation_setting.collection_binding_id)
+                        .first()
+                    )
+                    if not dataset_collection_binding:
+                        click.echo(f"App annotation collection binding not found: {app.id}")
+                        continue
+                    annotations = session.scalars(
+                        select(MessageAnnotation).where(MessageAnnotation.app_id == app.id)
+                    ).all()
                 dataset = Dataset(
                     id=app.id,
                     tenant_id=app.tenant_id,

api/configs/middleware/__init__.py

@@ -35,6 +35,7 @@ from .vdb.opensearch_config import OpenSearchConfig
 from .vdb.oracle_config import OracleConfig
 from .vdb.pgvector_config import PGVectorConfig
 from .vdb.pgvectors_config import PGVectoRSConfig
+from .vdb.pinecone_config import PineconeConfig
 from .vdb.qdrant_config import QdrantConfig
 from .vdb.relyt_config import RelytConfig
 from .vdb.tablestore_config import TableStoreConfig
@@ -336,6 +337,7 @@ class MiddlewareConfig(
     PGVectorConfig,
     VastbaseVectorConfig,
     PGVectoRSConfig,
+    PineconeConfig,
     QdrantConfig,
     RelytConfig,
     TencentVectorDBConfig,

api/configs/middleware/vdb/oceanbase_config.py

@@ -40,8 +40,12 @@ class OceanBaseVectorConfig(BaseSettings):
 
     OCEANBASE_FULLTEXT_PARSER: str | None = Field(
         description=(
-            "Fulltext parser to use for text indexing. Options: 'japanese_ftparser' (Japanese), "
-            "'thai_ftparser' (Thai), 'ik' (Chinese). Default is 'ik'"
+            "Fulltext parser to use for text indexing. "
+            "Built-in options: 'ngram' (N-gram tokenizer for English/numbers), "
+            "'beng' (Basic English tokenizer), 'space' (Space-based tokenizer), "
+            "'ngram2' (Improved N-gram tokenizer), 'ik' (Chinese tokenizer). "
+            "External plugins (require installation): 'japanese_ftparser' (Japanese tokenizer), "
+            "'thai_ftparser' (Thai tokenizer). Default is 'ik'"
         ),
         default="ik",
     )

api/configs/middleware/vdb/pinecone_config.py

@@ -0,0 +1,38 @@
+from pydantic import Field, PositiveInt
+from pydantic_settings import BaseSettings
+
+
+class PineconeConfig(BaseSettings):
+    """
+    Configuration settings for Pinecone vector database
+    """
+
+    PINECONE_API_KEY: str | None = Field(
+        description="API key for authenticating with Pinecone service",
+        default=None,
+    )
+
+    PINECONE_ENVIRONMENT: str | None = Field(
+        description="Pinecone environment (e.g., 'us-west1-gcp', 'us-east-1-aws')",
+        default=None,
+    )
+
+    PINECONE_INDEX_NAME: str | None = Field(
+        description="Default Pinecone index name",
+        default=None,
+    )
+
+    PINECONE_CLIENT_TIMEOUT: PositiveInt = Field(
+        description="Timeout in seconds for Pinecone client operations (default is 30 seconds)",
+        default=30,
+    )
+
+    PINECONE_BATCH_SIZE: PositiveInt = Field(
+        description="Batch size for Pinecone operations (default is 100)",
+        default=100,
+    )
+
+    PINECONE_METRIC: str = Field(
+        description="Distance metric for Pinecone index (cosine, euclidean, dotproduct)",
+        default="cosine",
+    )

api/controllers/console/app/conversation.py

@@ -1,6 +1,7 @@
 from datetime import datetime
 
 import pytz  # pip install pytz
+import sqlalchemy as sa
 from flask_login import current_user
 from flask_restx import Resource, marshal_with, reqparse
 from flask_restx.inputs import int_range
@@ -70,7 +71,7 @@ class CompletionConversationApi(Resource):
         parser.add_argument("limit", type=int_range(1, 100), default=20, location="args")
         args = parser.parse_args()
 
-        query = db.select(Conversation).where(
+        query = sa.select(Conversation).where(
             Conversation.app_id == app_model.id, Conversation.mode == "completion", Conversation.is_deleted.is_(False)
         )
 
@@ -236,7 +237,7 @@ class ChatConversationApi(Resource):
             .subquery()
         )
 
-        query = db.select(Conversation).where(Conversation.app_id == app_model.id, Conversation.is_deleted.is_(False))
+        query = sa.select(Conversation).where(Conversation.app_id == app_model.id, Conversation.is_deleted.is_(False))
 
         if args["keyword"]:
             keyword_filter = f"%{args['keyword']}%"

api/controllers/console/datasets/datasets.py

@@ -784,6 +784,7 @@ class DatasetRetrievalSettingApi(Resource):
                 | VectorType.PGVECTO_RS
                 | VectorType.VIKINGDB
                 | VectorType.UPSTASH
+                | VectorType.PINECONE
             ):
                 return {"retrieval_method": [RetrievalMethod.SEMANTIC_SEARCH.value]}
             case (
@@ -840,6 +841,7 @@ class DatasetRetrievalSettingMockApi(Resource):
                 | VectorType.PGVECTO_RS
                 | VectorType.VIKINGDB
                 | VectorType.UPSTASH
+                | VectorType.PINECONE
             ):
                 return {"retrieval_method": [RetrievalMethod.SEMANTIC_SEARCH.value]}
             case (

api/controllers/console/datasets/datasets_document.py

@@ -4,6 +4,7 @@ from argparse import ArgumentTypeError
 from collections.abc import Sequence
 from typing import Literal, cast
 
+import sqlalchemy as sa
 from flask import request
 from flask_login import current_user
 from flask_restx import Resource, fields, marshal, marshal_with, reqparse
@@ -211,13 +212,13 @@ class DatasetDocumentListApi(Resource):
 
         if sort == "hit_count":
             sub_query = (
-                db.select(DocumentSegment.document_id, db.func.sum(DocumentSegment.hit_count).label("total_hit_count"))
+                sa.select(DocumentSegment.document_id, sa.func.sum(DocumentSegment.hit_count).label("total_hit_count"))
                 .group_by(DocumentSegment.document_id)
                 .subquery()
             )
 
             query = query.outerjoin(sub_query, sub_query.c.document_id == Document.id).order_by(
-                sort_logic(db.func.coalesce(sub_query.c.total_hit_count, 0)),
+                sort_logic(sa.func.coalesce(sub_query.c.total_hit_count, 0)),
                 sort_logic(Document.position),
             )
         elif sort == "created_at":

api/core/rag/datasource/vdb/oceanbase/oceanbase_vector.py

@@ -123,7 +123,7 @@ class OceanBaseVector(BaseVector):
                 # Get parser from config or use default ik parser
                 parser_name = dify_config.OCEANBASE_FULLTEXT_PARSER or "ik"
 
-                allowed_parsers = ["ik", "japanese_ftparser", "thai_ftparser"]
+                allowed_parsers = ["ngram", "beng", "space", "ngram2", "ik", "japanese_ftparser", "thai_ftparser"]
                 if parser_name not in allowed_parsers:
                     raise ValueError(
                         f"Invalid OceanBase full-text parser: {parser_name}. "

api/core/rag/datasource/vdb/pinecone/pinecone_vector.py

@@ -0,0 +1,338 @@
+import json
+import time
+from typing import Any
+
+from pinecone import Pinecone, ServerlessSpec
+from pydantic import BaseModel
+
+from configs import dify_config
+from core.rag.datasource.vdb.field import Field
+from core.rag.datasource.vdb.vector_base import BaseVector
+from core.rag.datasource.vdb.vector_factory import AbstractVectorFactory
+from core.rag.datasource.vdb.vector_type import VectorType
+from core.rag.embedding.embedding_base import Embeddings
+from core.rag.models.document import Document
+from extensions.ext_database import db
+from extensions.ext_redis import redis_client
+from models.dataset import Dataset, DatasetCollectionBinding
+
+
+class PineconeConfig(BaseModel):
+    """Pinecone configuration class"""
+
+    api_key: str
+    environment: str
+    index_name: str | None = None
+    timeout: float = 30
+    batch_size: int = 100
+    metric: str = "cosine"
+
+
+class PineconeVector(BaseVector):
+    """Pinecone vector database concrete implementation class"""
+
+    def __init__(self, collection_name: str, group_id: str, config: PineconeConfig):
+        super().__init__(collection_name)
+        self._client_config = config
+        self._group_id = group_id
+
+        # Initialize Pinecone client with SSL configuration
+        try:
+            self._pc = Pinecone(
+                api_key=config.api_key,
+                # Configure SSL to handle connection issues
+                ssl_ca_certs=None,  # Use system default CA certificates
+            )
+        except Exception as e:
+            # Fallback to basic initialization if SSL config fails
+            self._pc = Pinecone(api_key=config.api_key)
+
+        # Normalize index name: lowercase, only a-z0-9- and <=45 chars
+        import hashlib
+        import re
+
+        base_name = collection_name.lower()
+        base_name = re.sub(r"[^a-z0-9-]+", "-", base_name)  # replace invalid chars with '-'
+        base_name = re.sub(r"-+", "-", base_name).strip("-")
+        # Use longer secure suffix to reduce collision risk
+        suffix_len = 24  # 24 hex digits (96-bit entropy)
+        if len(base_name) > 45:
+            hash_suffix = hashlib.sha256(base_name.encode()).hexdigest()[:suffix_len]
+            truncated_name = base_name[: 45 - (suffix_len + 1)].rstrip("-")
+            self._index_name = f"{truncated_name}-{hash_suffix}"
+        else:
+            self._index_name = base_name
+        # Guard empty name
+        if not self._index_name:
+            self._index_name = f"index-{hashlib.sha256(collection_name.encode()).hexdigest()[:suffix_len]}"
+        # Pinecone index handle, lazily initialized
+        self._index: Any | None = None
+
+    def get_type(self) -> str:
+        """Return vector database type identifier."""
+        return VectorType.PINECONE
+
+    def _ensure_index_initialized(self) -> None:
+        """Ensure that self._index is attached to an existing Pinecone index."""
+        if self._index is not None:
+            return
+        try:
+            existing_indexes = self._pc.list_indexes().names()
+            if self._index_name in existing_indexes:
+                self._index = self._pc.Index(self._index_name)
+            else:
+                raise ValueError("Index not initialized. Please ingest documents to create index.")
+        except Exception:
+            raise
+
+    def to_index_struct(self) -> dict:
+        """Generate index structure dictionary"""
+        return {"type": self.get_type(), "vector_store": {"class_prefix": self._collection_name}}
+
+    def create(self, texts: list[Document], embeddings: list[list[float]], **kwargs):
+        """Create vector index"""
+        if texts:
+            # Get vector dimension
+            vector_size = len(embeddings[0])
+
+            # Create Pinecone index
+            self.create_index(vector_size)
+
+            # Add vector data
+            self.add_texts(texts, embeddings, **kwargs)
+
+    def create_index(self, dimension: int):
+        """Create Pinecone index"""
+        lock_name = f"vector_indexing_lock_{self._index_name}"
+
+        with redis_client.lock(lock_name, timeout=30):
+            # Check Redis cache
+            index_exist_cache_key = f"vector_indexing_{self._index_name}"
+            if redis_client.get(index_exist_cache_key):
+                self._index = self._pc.Index(self._index_name)
+                return
+
+            # Check if index already exists
+            existing_indexes = self._pc.list_indexes().names()
+
+            if self._index_name not in existing_indexes:
+                # Create new index using ServerlessSpec
+                self._pc.create_index(
+                    name=self._index_name,
+                    dimension=dimension,
+                    metric=self._client_config.metric,
+                    spec=ServerlessSpec(cloud="aws", region=self._client_config.environment),
+                )
+
+                # Wait for index creation to complete
+                while not self._pc.describe_index(self._index_name).status["ready"]:
+                    time.sleep(1)
+            else:
+                # Get index instance
+                self._index = self._pc.Index(self._index_name)
+
+            # Set cache
+            redis_client.set(index_exist_cache_key, 1, ex=3600)
+
+    def add_texts(self, documents: list[Document], embeddings: list[list[float]], **kwargs):
+        """Batch add document vectors"""
+        if not self._index:
+            raise ValueError("Index not initialized. Call create() first.")
+
+        total_docs = len(documents)
+
+        uuids = self._get_uuids(documents)
+        batch_size = self._client_config.batch_size
+        added_ids = []
+
+        # Batch processing
+        total_batches = (total_docs + batch_size - 1) // batch_size  # Ceiling division
+        for batch_idx, i in enumerate(range(0, len(documents), batch_size), 1):
+            batch_documents = documents[i : i + batch_size]
+            batch_embeddings = embeddings[i : i + batch_size]
+            batch_uuids = uuids[i : i + batch_size]
+            batch_size_actual = len(batch_documents)
+
+            # Build Pinecone vector data (metadata must be primitives or list[str])
+            vectors_to_upsert = []
+            for doc, embedding, doc_id in zip(batch_documents, batch_embeddings, batch_uuids):
+                raw_meta = doc.metadata or {}
+                safe_meta: dict[str, Any] = {}
+                # lift common identifiers to top-level fields for filtering
+                for k, v in raw_meta.items():
+                    if isinstance(v, (str, int, float, bool)) or (
+                        isinstance(v, list) and all(isinstance(x, str) for x in v)
+                    ):
+                        safe_meta[k] = v
+                    else:
+                        safe_meta[k] = json.dumps(v, ensure_ascii=False)
+
+                # keep content as string metadata if needed
+                safe_meta[Field.CONTENT_KEY.value] = doc.page_content
+                # group id as string
+                safe_meta[Field.GROUP_KEY.value] = str(self._group_id)
+
+                vectors_to_upsert.append({"id": doc_id, "values": embedding, "metadata": safe_meta})
+
+            # Batch insert to Pinecone
+            try:
+                self._index.upsert(vectors=vectors_to_upsert)
+                added_ids.extend(batch_uuids)
+            except Exception as e:
+                raise
+
+        return added_ids
+
+    def search_by_vector(self, query_vector: list[float], **kwargs) -> list[Document]:
+        """Vector similarity search"""
+        # Lazily attach to an existing index if needed
+        self._ensure_index_initialized()
+
+        top_k = kwargs.get("top_k", 4)
+        score_threshold = float(kwargs.get("score_threshold", 0.0))
+
+        # Build filter conditions
+        filter_dict = {Field.GROUP_KEY.value: {"$eq": str(self._group_id)}}
+
+        # Document scope filtering
+        document_ids_filter = kwargs.get("document_ids_filter")
+        if document_ids_filter:
+            filter_dict["document_id"] = {"$in": document_ids_filter}
+
+        # Execute search
+        try:
+            index = self._index
+            assert index is not None
+            response = index.query(vector=query_vector, top_k=top_k, include_metadata=True, filter=filter_dict)
+        except Exception as e:
+            raise
+
+        # Convert results
+        docs = []
+        filtered_count = 0
+        for match in response.matches:
+            if match.score >= score_threshold:
+                page_content = match.metadata.get(Field.CONTENT_KEY.value, "")
+                metadata = dict(match.metadata or {})
+                metadata.pop(Field.CONTENT_KEY.value, None)
+                metadata.pop(Field.GROUP_KEY.value, None)
+                metadata["score"] = match.score
+
+                doc = Document(page_content=page_content, metadata=metadata)
+                docs.append(doc)
+            else:
+                filtered_count += 1
+
+        # Sort by similarity score in descending order
+        docs.sort(key=lambda x: x.metadata.get("score", 0), reverse=True)
+
+        return docs
+
+    def search_by_full_text(self, query: str, **kwargs) -> list[Document]:
+        """Full-text search - Pinecone does not natively support it, returns empty list"""
+        return []
+
+    def delete_by_metadata_field(self, key: str, value: str):
+        """Delete by metadata field"""
+        self._ensure_index_initialized()
+
+        try:
+            # Build filter conditions
+            filter_dict = {
+                Field.GROUP_KEY.value: {"$eq": self._group_id},
+                f"{Field.METADATA_KEY.value}.{key}": {"$eq": value},
+            }
+
+            # Pinecone delete operation
+            index = self._index
+            assert index is not None
+            index.delete(filter=filter_dict)
+        except Exception as e:
+            # Ignore delete errors
+            pass
+
+    def delete_by_ids(self, ids: list[str]) -> None:
+        """Batch delete by ID list"""
+        self._ensure_index_initialized()
+
+        try:
+            # Pinecone delete by ID
+            index = self._index
+            assert index is not None
+            index.delete(ids=ids)
+        except Exception as e:
+            raise
+
+    def delete(self) -> None:
+        """Delete all vector data for the entire dataset"""
+        self._ensure_index_initialized()
+
+        try:
+            # Delete all vectors by group_id
+            filter_dict = {Field.GROUP_KEY.value: {"$eq": self._group_id}}
+            index = self._index
+            assert index is not None
+            index.delete(filter=filter_dict)
+        except Exception as e:
+            raise
+
+    def text_exists(self, id: str) -> bool:
+        """Check if document exists"""
+        try:
+            self._ensure_index_initialized()
+        except Exception:
+            return False
+
+        try:
+            # Check if vector exists through query
+            index = self._index
+            assert index is not None
+            response = index.fetch(ids=[id])
+            exists = id in response.vectors
+            return exists
+        except Exception as e:
+            return False
+
+
+class PineconeVectorFactory(AbstractVectorFactory):
+    """Pinecone vector database factory class"""
+
+    def init_vector(self, dataset: Dataset, attributes: list, embeddings: Embeddings) -> PineconeVector:
+        """Create PineconeVector instance"""
+
+        # Determine index name
+        if dataset.collection_binding_id:
+            dataset_collection_binding = (
+                db.session.query(DatasetCollectionBinding)
+                .where(DatasetCollectionBinding.id == dataset.collection_binding_id)
+                .one_or_none()
+            )
+            if dataset_collection_binding:
+                collection_name = dataset_collection_binding.collection_name
+            else:
+                raise ValueError("Dataset Collection Bindings does not exist!")
+        else:
+            if dataset.index_struct_dict:
+                class_prefix: str = dataset.index_struct_dict["vector_store"]["class_prefix"]
+                collection_name = class_prefix
+            else:
+                dataset_id = dataset.id
+                collection_name = Dataset.gen_collection_name_by_id(dataset_id)
+
+        # Set index structure
+        if not dataset.index_struct_dict:
+            dataset.index_struct = json.dumps(self.gen_index_struct_dict(VectorType.PINECONE, collection_name))
+
+        # Create PineconeVector instance
+        return PineconeVector(
+            collection_name=collection_name,
+            group_id=dataset.id,
+            config=PineconeConfig(
+                api_key=dify_config.PINECONE_API_KEY or "",
+                environment=dify_config.PINECONE_ENVIRONMENT or "",
+                index_name=dify_config.PINECONE_INDEX_NAME,
+                timeout=dify_config.PINECONE_CLIENT_TIMEOUT,
+                batch_size=dify_config.PINECONE_BATCH_SIZE,
+                metric=dify_config.PINECONE_METRIC,
+            ),
+        )

api/core/rag/datasource/vdb/vector_factory.py

@@ -87,6 +87,10 @@ class Vector:
                 from core.rag.datasource.vdb.pgvecto_rs.pgvecto_rs import PGVectoRSFactory
 
                 return PGVectoRSFactory
+            case VectorType.PINECONE:
+                from core.rag.datasource.vdb.pinecone.pinecone_vector import PineconeVectorFactory
+
+                return PineconeVectorFactory
             case VectorType.QDRANT:
                 from core.rag.datasource.vdb.qdrant.qdrant_vector import QdrantVectorFactory
 

api/core/rag/datasource/vdb/vector_type.py

@@ -31,3 +31,4 @@ class VectorType(StrEnum):
     HUAWEI_CLOUD = "huawei_cloud"
     MATRIXONE = "matrixone"
     CLICKZETTA = "clickzetta"
+    PINECONE = "pinecone"

api/core/workflow/graph_engine/layers/debug_logging.py

@@ -146,6 +146,12 @@ class DebugLoggingLayer(GraphEngineLayer):
                 self.logger.info("  Partial outputs: %s", self._format_dict(event.outputs))
 
         # Node-level events
+        # Retry before Started because Retry subclasses Started;
+        elif isinstance(event, NodeRunRetryEvent):
+            self.retry_count += 1
+            self.logger.warning("🔄 Node retry: %s (attempt %s)", event.node_id, event.retry_index)
+            self.logger.warning("  Previous error: %s", event.error)
+
         elif isinstance(event, NodeRunStartedEvent):
             self.node_count += 1
             self.logger.info('▶️ Node started: %s - "%s" (type: %s)', event.node_id, event.node_title, event.node_type)
@@ -175,11 +181,6 @@ class DebugLoggingLayer(GraphEngineLayer):
             self.logger.warning("⚠️ Node exception handled: %s", event.node_id)
             self.logger.warning("  Error: %s", event.error)
 
-        elif isinstance(event, NodeRunRetryEvent):
-            self.retry_count += 1
-            self.logger.warning("🔄 Node retry: %s (attempt %s)", event.node_id, event.retry_index)
-            self.logger.warning("  Previous error: %s", event.error)
-
         elif isinstance(event, NodeRunStreamChunkEvent):
             # Log stream chunks at debug level to avoid spam
             final_indicator = " (FINAL)" if event.is_final else ""

api/core/workflow/nodes/iteration/iteration_node.py

@@ -1,9 +1,11 @@
+import contextvars
 import logging
 from collections.abc import Generator, Mapping, Sequence
 from concurrent.futures import Future, ThreadPoolExecutor, as_completed
 from datetime import UTC, datetime
 from typing import TYPE_CHECKING, Any, NewType, cast
 
+from flask import Flask, current_app
 from typing_extensions import TypeIs
 
 from core.variables import IntegerVariable, NoneSegment
@@ -35,6 +37,7 @@ from core.workflow.nodes.base.entities import BaseNodeData, RetryConfig
 from core.workflow.nodes.base.node import Node
 from core.workflow.nodes.iteration.entities import ErrorHandleMode, IterationNodeData
 from libs.datetime_utils import naive_utc_now
+from libs.flask_utils import preserve_flask_contexts
 
 from .exc import (
     InvalidIteratorValueError,
@@ -239,6 +242,8 @@ class IterationNode(Node):
                     self._execute_single_iteration_parallel,
                     index=index,
                     item=item,
+                    flask_app=current_app._get_current_object(),  # type: ignore
+                    context_vars=contextvars.copy_context(),
                 )
                 future_to_index[future] = index
 
@@ -281,26 +286,29 @@ class IterationNode(Node):
         self,
         index: int,
         item: object,
+        flask_app: Flask,
+        context_vars: contextvars.Context,
     ) -> tuple[datetime, list[GraphNodeEventBase], object | None, int]:
         """Execute a single iteration in parallel mode and return results."""
-        iter_start_at = datetime.now(UTC).replace(tzinfo=None)
-        events: list[GraphNodeEventBase] = []
-        outputs_temp: list[object] = []
+        with preserve_flask_contexts(flask_app=flask_app, context_vars=context_vars):
+            iter_start_at = datetime.now(UTC).replace(tzinfo=None)
+            events: list[GraphNodeEventBase] = []
+            outputs_temp: list[object] = []
 
-        graph_engine = self._create_graph_engine(index, item)
+            graph_engine = self._create_graph_engine(index, item)
 
-        # Collect events instead of yielding them directly
-        for event in self._run_single_iter(
-            variable_pool=graph_engine.graph_runtime_state.variable_pool,
-            outputs=outputs_temp,
-            graph_engine=graph_engine,
-        ):
-            events.append(event)
+            # Collect events instead of yielding them directly
+            for event in self._run_single_iter(
+                variable_pool=graph_engine.graph_runtime_state.variable_pool,
+                outputs=outputs_temp,
+                graph_engine=graph_engine,
+            ):
+                events.append(event)
 
-        # Get the output value from the temporary outputs list
-        output_value = outputs_temp[0] if outputs_temp else None
+            # Get the output value from the temporary outputs list
+            output_value = outputs_temp[0] if outputs_temp else None
 
-        return iter_start_at, events, output_value, graph_engine.graph_runtime_state.total_tokens
+            return iter_start_at, events, output_value, graph_engine.graph_runtime_state.total_tokens
 
     def _handle_iteration_success(
         self,

api/core/workflow/nodes/knowledge_index/entities.py

@@ -63,7 +63,7 @@ class RetrievalSetting(BaseModel):
     Retrieval Setting.
     """
 
-    search_method: Literal["semantic_search", "keyword_search", "fulltext_search", "hybrid_search"]
+    search_method: Literal["semantic_search", "keyword_search", "full_text_search", "hybrid_search"]
     top_k: int
     score_threshold: float | None = 0.5
     score_threshold_enabled: bool = False

api/models/dataset.py

@@ -910,7 +910,7 @@ class AppDatasetJoin(Base):
     id = mapped_column(StringUUID, primary_key=True, nullable=False, server_default=sa.text("uuid_generate_v4()"))
     app_id = mapped_column(StringUUID, nullable=False)
     dataset_id = mapped_column(StringUUID, nullable=False)
-    created_at: Mapped[datetime] = mapped_column(DateTime, nullable=False, server_default=db.func.current_timestamp())
+    created_at: Mapped[datetime] = mapped_column(DateTime, nullable=False, server_default=sa.func.current_timestamp())
 
     @property
     def app(self):
@@ -931,7 +931,7 @@ class DatasetQuery(Base):
     source_app_id = mapped_column(StringUUID, nullable=True)
     created_by_role = mapped_column(String, nullable=False)
     created_by = mapped_column(StringUUID, nullable=False)
-    created_at: Mapped[datetime] = mapped_column(DateTime, nullable=False, server_default=db.func.current_timestamp())
+    created_at: Mapped[datetime] = mapped_column(DateTime, nullable=False, server_default=sa.func.current_timestamp())
 
 
 class DatasetKeywordTable(Base):

api/models/model.py

@@ -1731,7 +1731,7 @@ class MessageChain(Base):
     type: Mapped[str] = mapped_column(String(255), nullable=False)
     input = mapped_column(sa.Text, nullable=True)
     output = mapped_column(sa.Text, nullable=True)
-    created_at = mapped_column(sa.DateTime, nullable=False, server_default=db.func.current_timestamp())
+    created_at = mapped_column(sa.DateTime, nullable=False, server_default=sa.func.current_timestamp())
 
 
 class MessageAgentThought(Base):
@@ -1769,7 +1769,7 @@ class MessageAgentThought(Base):
     latency: Mapped[float | None] = mapped_column(sa.Float, nullable=True)
     created_by_role = mapped_column(String, nullable=False)
     created_by = mapped_column(StringUUID, nullable=False)
-    created_at = mapped_column(sa.DateTime, nullable=False, server_default=db.func.current_timestamp())
+    created_at = mapped_column(sa.DateTime, nullable=False, server_default=sa.func.current_timestamp())
 
     @property
     def files(self) -> list[Any]:
@@ -1872,7 +1872,7 @@ class DatasetRetrieverResource(Base):
     index_node_hash = mapped_column(sa.Text, nullable=True)
     retriever_from = mapped_column(sa.Text, nullable=False)
     created_by = mapped_column(StringUUID, nullable=False)
-    created_at = mapped_column(sa.DateTime, nullable=False, server_default=db.func.current_timestamp())
+    created_at = mapped_column(sa.DateTime, nullable=False, server_default=sa.func.current_timestamp())
 
 
 class Tag(Base):

api/pyproject.toml

@@ -88,7 +88,8 @@ dependencies = [
     "httpx-sse~=0.4.0",
     "sendgrid~=6.12.3",
     "flask-restx~=1.3.0",
-    "packaging~=23.2",
+    "packaging>=24.2,<25.0",
+    "pinecone>=7.3.0",
 ]
 # Before adding new dependency, consider place it in
 # alphabet order (a-z) and suitable group.

api/services/app_service.py

@@ -2,6 +2,7 @@ import json
 import logging
 from typing import TypedDict, cast
 
+import sqlalchemy as sa
 from flask_sqlalchemy.pagination import Pagination
 
 from configs import dify_config
@@ -65,7 +66,7 @@ class AppService:
                 return None
 
         app_models = db.paginate(
-            db.select(App).where(*filters).order_by(App.created_at.desc()),
+            sa.select(App).where(*filters).order_by(App.created_at.desc()),
             page=args["page"],
             per_page=args["limit"],
             error_out=False,

api/services/dataset_service.py

@@ -115,12 +115,12 @@ class DatasetService:
                     # Check if permitted_dataset_ids is not empty to avoid WHERE false condition
                     if permitted_dataset_ids and len(permitted_dataset_ids) > 0:
                         query = query.where(
-                            db.or_(
+                            sa.or_(
                                 Dataset.permission == DatasetPermissionEnum.ALL_TEAM,
-                                db.and_(
+                                sa.and_(
                                     Dataset.permission == DatasetPermissionEnum.ONLY_ME, Dataset.created_by == user.id
                                 ),
-                                db.and_(
+                                sa.and_(
                                     Dataset.permission == DatasetPermissionEnum.PARTIAL_TEAM,
                                     Dataset.id.in_(permitted_dataset_ids),
                                 ),
@@ -128,9 +128,9 @@ class DatasetService:
                         )
                     else:
                         query = query.where(
-                            db.or_(
+                            sa.or_(
                                 Dataset.permission == DatasetPermissionEnum.ALL_TEAM,
-                                db.and_(
+                                sa.and_(
                                     Dataset.permission == DatasetPermissionEnum.ONLY_ME, Dataset.created_by == user.id
                                 ),
                             )
@@ -1879,7 +1879,7 @@ class DocumentService:
     #                 for notion_info in notion_info_list:
     #                     workspace_id = notion_info.workspace_id
     #                     data_source_binding = DataSourceOauthBinding.query.filter(
-    #                         db.and_(
+    #                         sa.and_(
     #                             DataSourceOauthBinding.tenant_id == current_user.current_tenant_id,
     #                             DataSourceOauthBinding.provider == "notion",
     #                             DataSourceOauthBinding.disabled == False,

api/services/entities/knowledge_entities/rag_pipeline_entities.py

@@ -83,7 +83,7 @@ class RetrievalSetting(BaseModel):
     Retrieval Setting.
     """
 
-    search_method: Literal["semantic_search", "fulltext_search", "keyword_search", "hybrid_search"]
+    search_method: Literal["semantic_search", "full_text_search", "keyword_search", "hybrid_search"]
     top_k: int
     score_threshold: float | None = 0.5
     score_threshold_enabled: bool = False

api/services/plugin/plugin_migration.py

@@ -471,7 +471,7 @@ class PluginMigration:
         total_failed_tenant = 0
         while True:
             # paginate
-            tenants = db.paginate(db.select(Tenant).order_by(Tenant.created_at.desc()), page=page, per_page=100)
+            tenants = db.paginate(sa.select(Tenant).order_by(Tenant.created_at.desc()), page=page, per_page=100)
             if tenants.items is None or len(tenants.items) == 0:
                 break
 

api/services/tag_service.py

@@ -1,5 +1,6 @@
 import uuid
 
+import sqlalchemy as sa
 from flask_login import current_user
 from sqlalchemy import func, select
 from werkzeug.exceptions import NotFound
@@ -18,7 +19,7 @@ class TagService:
             .where(Tag.type == tag_type, Tag.tenant_id == current_tenant_id)
         )
         if keyword:
-            query = query.where(db.and_(Tag.name.ilike(f"%{keyword}%")))
+            query = query.where(sa.and_(Tag.name.ilike(f"%{keyword}%")))
         query = query.group_by(Tag.id, Tag.type, Tag.name, Tag.created_at)
         results: list = query.order_by(Tag.created_at.desc()).all()
         return results

api/services/variable_truncator.py

@@ -262,6 +262,14 @@ class VariableTruncator:
         target_length = self._array_element_limit
 
         for i, item in enumerate(value):
+            # Dirty fix:
+            # The output of `Start` node may contain list of `File` elements,
+            # causing `AssertionError` while invoking `_truncate_json_primitives`.
+            #
+            # This check ensures that `list[File]` are handled separately
+            if isinstance(item, File):
+                truncated_value.append(item)
+                continue
             if i >= target_length:
                 return _PartResult(truncated_value, used_size, True)
             if i > 0:

api/tasks/document_indexing_sync_task.py

@@ -2,6 +2,7 @@ import logging
 import time
 
 import click
+import sqlalchemy as sa
 from celery import shared_task
 from sqlalchemy import select
 
@@ -51,7 +52,7 @@ def document_indexing_sync_task(dataset_id: str, document_id: str):
         data_source_binding = (
             db.session.query(DataSourceOauthBinding)
             .where(
-                db.and_(
+                sa.and_(
                     DataSourceOauthBinding.tenant_id == document.tenant_id,
                     DataSourceOauthBinding.provider == "notion",
                     DataSourceOauthBinding.disabled == False,

api/tests/integration_tests/vdb/pinecone/test_pinecone.py

@@ -0,0 +1,27 @@
+from core.rag.datasource.vdb.pinecone.pinecone_vector import PineconeConfig, PineconeVector
+from tests.integration_tests.vdb.test_vector_store import (
+    AbstractVectorTest,
+    setup_mock_redis,
+)
+
+
+class PineconeVectorTest(AbstractVectorTest):
+    def __init__(self):
+        super().__init__()
+        self.attributes = ["doc_id", "dataset_id", "document_id", "doc_hash"]
+        self.vector = PineconeVector(
+            collection_name=self.collection_name,
+            group_id=self.dataset_id,
+            config=PineconeConfig(
+                api_key="test_api_key",
+                environment="test_environment",
+                index_name="test_index",
+            ),
+        )
+
+    def search_by_vector(self):
+        super().search_by_vector()
+
+
+def test_pinecone_vector():
+    PineconeVectorTest().run_all_tests()

api/tests/unit_tests/services/test_variable_truncator.py

@@ -588,3 +588,11 @@ class TestIntegrationScenarios:
             if isinstance(result.result, ObjectSegment):
                 result_size = truncator.calculate_json_size(result.result.value)
                 assert result_size <= original_size
+
+    def test_file_and_array_file_variable_mapping(self, file):
+        truncator = VariableTruncator(string_length_limit=30, array_element_limit=3, max_size_bytes=300)
+
+        mapping = {"array_file": [file]}
+        truncated_mapping, truncated = truncator.truncate_variable_mapping(mapping)
+        assert truncated is False
+        assert truncated_mapping == mapping

api/uv.lock

@@ -1334,6 +1334,7 @@ dependencies = [
     { name = "packaging" },
     { name = "pandas", extra = ["excel", "output-formatting", "performance"] },
     { name = "pandoc" },
+    { name = "pinecone" },
     { name = "psycogreen" },
     { name = "psycopg2-binary" },
     { name = "pycryptodome" },
@@ -1525,9 +1526,10 @@ requires-dist = [
     { name = "opentelemetry-semantic-conventions", specifier = "==0.48b0" },
     { name = "opentelemetry-util-http", specifier = "==0.48b0" },
     { name = "opik", specifier = "~=1.7.25" },
-    { name = "packaging", specifier = "~=23.2" },
+    { name = "packaging", specifier = ">=24.2,<25.0" },
     { name = "pandas", extras = ["excel", "output-formatting", "performance"], specifier = "~=2.2.2" },
     { name = "pandoc", specifier = "~=2.4" },
+    { name = "pinecone", specifier = ">=7.3.0" },
     { name = "psycogreen", specifier = "~=1.0.2" },
     { name = "psycopg2-binary", specifier = "~=2.9.6" },
     { name = "pycryptodome", specifier = "==3.19.1" },
@@ -4161,11 +4163,11 @@ wheels = [
 
 [[package]]
 name = "packaging"
-version = "23.2"
+version = "24.2"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/fb/2b/9b9c33ffed44ee921d0967086d653047286054117d584f1b1a7c22ceaf7b/packaging-23.2.tar.gz", hash = "sha256:048fb0e9405036518eaaf48a55953c750c11e1a1b68e0dd1a9d62ed0c092cfc5", size = 146714, upload-time = "2023-10-01T13:50:05.279Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/d0/63/68dbb6eb2de9cb10ee4c9c14a0148804425e13c4fb20d61cce69f53106da/packaging-24.2.tar.gz", hash = "sha256:c228a6dc5e932d346bc5739379109d49e8853dd8223571c7c5b55260edc0b97f", size = 163950, upload-time = "2024-11-08T09:47:47.202Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/ec/1a/610693ac4ee14fcdf2d9bf3c493370e4f2ef7ae2e19217d7a237ff42367d/packaging-23.2-py3-none-any.whl", hash = "sha256:8c491190033a9af7e1d931d0b5dacc2ef47509b34dd0de67ed209b5203fc88c7", size = 53011, upload-time = "2023-10-01T13:50:03.745Z" },
+    { url = "https://files.pythonhosted.org/packages/88/ef/eb23f262cca3c0c4eb7ab1933c3b1f03d021f2c48f54763065b6f0e321be/packaging-24.2-py3-none-any.whl", hash = "sha256:09abb1bccd265c01f4a3aa3f7a7db064b36514d2cba19a2f694fe6150451a759", size = 65451, upload-time = "2024-11-08T09:47:44.722Z" },
 ]
 
 [[package]]
@@ -4326,6 +4328,45 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/34/e7/ae39f538fd6844e982063c3a5e4598b8ced43b9633baa3a85ef33af8c05c/pillow-11.3.0-pp311-pypy311_pp73-win_amd64.whl", hash = "sha256:c84d689db21a1c397d001aa08241044aa2069e7587b398c8cc63020390b1c1b8", size = 6984598, upload-time = "2025-07-01T09:16:27.732Z" },
 ]
 
+[[package]]
+name = "pinecone"
+version = "7.3.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "certifi" },
+    { name = "pinecone-plugin-assistant" },
+    { name = "pinecone-plugin-interface" },
+    { name = "python-dateutil" },
+    { name = "typing-extensions" },
+    { name = "urllib3" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/fa/38/12731d4af470851b4963eba616605868a8599ef4df51c7b6c928e5f3166d/pinecone-7.3.0.tar.gz", hash = "sha256:307edc155621d487c20dc71b76c3ad5d6f799569ba42064190d03917954f9a7b", size = 235256, upload-time = "2025-06-27T20:03:51.498Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/b7/a6/c5d54a5fb1de3983a8739c1a1660e7a7074db2cbadfa875b823fcf29b629/pinecone-7.3.0-py3-none-any.whl", hash = "sha256:315b8fef20320bef723ecbb695dec0aafa75d8434d86e01e5a0e85933e1009a8", size = 587563, upload-time = "2025-06-27T20:03:50.249Z" },
+]
+
+[[package]]
+name = "pinecone-plugin-assistant"
+version = "1.8.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "packaging" },
+    { name = "requests" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/b2/01/65c4c3a81732fa379f8e7f78a8c18aa57a1139f5b79d58b93a69f2fc8cb0/pinecone_plugin_assistant-1.8.0.tar.gz", hash = "sha256:8e8682cff30f9bae9243b384021aba71c91f4e6ef1650e9d63ee64aab83cba87", size = 150435, upload-time = "2025-08-31T14:31:18.046Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/dd/49/62ab8e2f9098bf8593e36bbe6e729fcc0500bafca7d88be7b62eac66c8b0/pinecone_plugin_assistant-1.8.0-py3-none-any.whl", hash = "sha256:71ae42c3b4478d23138cbc4fe3505db561319a826f5aff4ef2e306a25ac56686", size = 259281, upload-time = "2025-08-31T14:31:16.587Z" },
+]
+
+[[package]]
+name = "pinecone-plugin-interface"
+version = "0.0.7"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/f4/fb/e8a4063264953ead9e2b24d9b390152c60f042c951c47f4592e9996e57ff/pinecone_plugin_interface-0.0.7.tar.gz", hash = "sha256:b8e6675e41847333aa13923cc44daa3f85676d7157324682dc1640588a982846", size = 3370, upload-time = "2024-06-05T01:57:52.093Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/3b/1d/a21fdfcd6d022cb64cef5c2a29ee6691c6c103c4566b41646b080b7536a5/pinecone_plugin_interface-0.0.7-py3-none-any.whl", hash = "sha256:875857ad9c9fc8bbc074dbe780d187a2afd21f5bfe0f3b08601924a61ef1bba8", size = 6249, upload-time = "2024-06-05T01:57:50.583Z" },
+]
+
 [[package]]
 name = "platformdirs"
 version = "4.4.0"

docker/.env.example

@@ -655,6 +655,8 @@ LINDORM_USING_UGC=True
 LINDORM_QUERY_TIMEOUT=1
 
 # OceanBase Vector configuration, only available when VECTOR_STORE is `oceanbase`
+# Built-in fulltext parsers are `ngram`, `beng`, `space`, `ngram2`, `ik`
+# External fulltext parsers (require plugin installation) are `japanese_ftparser`, `thai_ftparser`
 OCEANBASE_VECTOR_HOST=oceanbase
 OCEANBASE_VECTOR_PORT=2881
 OCEANBASE_VECTOR_USER=root@test
@@ -706,6 +708,16 @@ CLICKZETTA_ANALYZER_TYPE=chinese
 CLICKZETTA_ANALYZER_MODE=smart
 CLICKZETTA_VECTOR_DISTANCE_FUNCTION=cosine_distance
 
+# Pinecone configuration, only available when VECTOR_STORE is `pinecone`
+PINECONE_API_KEY=your-pinecone-api-key
+PINECONE_ENVIRONMENT=your-pinecone-environment
+PINECONE_INDEX_NAME=dify-index
+PINECONE_CLIENT_TIMEOUT=30
+PINECONE_BATCH_SIZE=100
+PINECONE_METRIC=cosine
+PINECONE_PODS=1
+PINECONE_POD_TYPE=s1
+
 # ------------------------------
 # Knowledge Configuration
 # ------------------------------

docker/docker-compose.yaml

@@ -339,6 +339,14 @@ x-shared-env: &shared-api-worker-env
   CLICKZETTA_ANALYZER_TYPE: ${CLICKZETTA_ANALYZER_TYPE:-chinese}
   CLICKZETTA_ANALYZER_MODE: ${CLICKZETTA_ANALYZER_MODE:-smart}
   CLICKZETTA_VECTOR_DISTANCE_FUNCTION: ${CLICKZETTA_VECTOR_DISTANCE_FUNCTION:-cosine_distance}
+  PINECONE_API_KEY: ${PINECONE_API_KEY:-your-pinecone-api-key}
+  PINECONE_ENVIRONMENT: ${PINECONE_ENVIRONMENT:-your-pinecone-environment}
+  PINECONE_INDEX_NAME: ${PINECONE_INDEX_NAME:-dify-index}
+  PINECONE_CLIENT_TIMEOUT: ${PINECONE_CLIENT_TIMEOUT:-30}
+  PINECONE_BATCH_SIZE: ${PINECONE_BATCH_SIZE:-100}
+  PINECONE_METRIC: ${PINECONE_METRIC:-cosine}
+  PINECONE_PODS: ${PINECONE_PODS:-1}
+  PINECONE_POD_TYPE: ${PINECONE_POD_TYPE:-s1}
   UPLOAD_FILE_SIZE_LIMIT: ${UPLOAD_FILE_SIZE_LIMIT:-15}
   UPLOAD_FILE_BATCH_LIMIT: ${UPLOAD_FILE_BATCH_LIMIT:-5}
   ETL_TYPE: ${ETL_TYPE:-dify}

web/app/components/app/log/list.tsx

@@ -35,7 +35,7 @@ import { useStore as useAppStore } from '@/app/components/app/store'
 import { useAppContext } from '@/context/app-context'
 import useTimestamp from '@/hooks/use-timestamp'
 import Tooltip from '@/app/components/base/tooltip'
-import { CopyIcon } from '@/app/components/base/copy-icon'
+import CopyIcon from '@/app/components/base/copy-icon'
 import { buildChatItemTree, getThreadMessages } from '@/app/components/base/chat/utils'
 import { getProcessedFilesFromResponse } from '@/app/components/base/file-uploader/utils'
 import cn from '@/utils/classnames'

web/app/components/base/copy-icon/index.tsx

@@ -15,7 +15,7 @@ type Props = {
 
 const prefixEmbedded = 'appOverview.overview.appInfo.embedded'
 
-export const CopyIcon = ({ content }: Props) => {
+const CopyIcon = ({ content }: Props) => {
   const { t } = useTranslation()
   const [isCopied, setIsCopied] = useState<boolean>(false)
 

web/app/components/base/markdown-blocks/think-block.tsx

@@ -63,7 +63,7 @@ const useThinkTimer = (children: any) => {
   return { elapsedTime, isComplete }
 }
 
-export const ThinkBlock = ({ children, ...props }: any) => {
+const ThinkBlock = ({ children, ...props }: React.ComponentProps<'details'>) => {
   const { elapsedTime, isComplete } = useThinkTimer(children)
   const displayContent = removeEndThink(children)
   const { t } = useTranslation()

web/app/components/base/search-input/index.tsx

@@ -53,7 +53,7 @@ const SearchInput: FC<SearchInputProps> = ({
         }}
         onCompositionEnd={(e) => {
           isComposing.current = false
-          onChange(e.data)
+          onChange(e.currentTarget.value)
         }}
         onFocus={() => setFocus(true)}
         onBlur={() => setFocus(false)}

web/app/components/base/svg-gallery/index.tsx

@@ -3,7 +3,7 @@ import { SVG } from '@svgdotjs/svg.js'
 import DOMPurify from 'dompurify'
 import ImagePreview from '@/app/components/base/image-uploader/image-preview'
 
-export const SVGRenderer = ({ content }: { content: string }) => {
+const SVGRenderer = ({ content }: { content: string }) => {
   const svgRef = useRef<HTMLDivElement>(null)
   const [imagePreview, setImagePreview] = useState('')
   const [windowSize, setWindowSize] = useState({

web/app/components/datasets/create-from-pipeline/list/built-in-pipeline-list.tsx

@@ -1,9 +1,18 @@
 import { usePipelineTemplateList } from '@/service/use-pipeline'
 import TemplateCard from './template-card'
 import CreateCard from './create-card'
+import { useI18N } from '@/context/i18n'
+import { useMemo } from 'react'
+import { LanguagesSupported } from '@/i18n-config/language'
 
 const BuiltInPipelineList = () => {
-  const { data: pipelineList, isLoading } = usePipelineTemplateList({ type: 'built-in' })
+  const { locale } = useI18N()
+  const language = useMemo(() => {
+    if (['zh-Hans', 'ja-JP'].includes(locale))
+      return locale
+    return LanguagesSupported[0]
+  }, [locale])
+  const { data: pipelineList, isLoading } = usePipelineTemplateList({ type: 'built-in', language })
   const list = pipelineList?.pipeline_templates || []
 
   return (

web/app/components/plugins/plugin-item/index.tsx

@@ -146,7 +146,6 @@ const PluginItem: FC<Props> = ({
         {/* Organization & Name */}
         <div className='flex grow items-center overflow-hidden'>
           <OrgInfo
-            className='mt-0.5'
             orgName={orgName}
             packageName={name}
             packageNameClassName='w-auto max-w-[150px]'
@@ -154,8 +153,8 @@ const PluginItem: FC<Props> = ({
           {category === PluginType.extension && (
             <>
               <div className='system-xs-regular mx-2 text-text-quaternary'>·</div>
-              <div className='system-xs-regular flex space-x-1 overflow-hidden text-text-tertiary'>
-                <RiLoginCircleLine className='h-4 w-4 shrink-0' />
+              <div className='system-xs-regular flex items-center gap-x-1 overflow-hidden text-text-tertiary'>
+                <RiLoginCircleLine className='size-3 shrink-0' />
                 <span
                   className='truncate'
                   title={t('plugin.endpointsEnabled', { num: endpoints_active })}
@@ -184,7 +183,7 @@ const PluginItem: FC<Props> = ({
             && <>
               <a href={getMarketplaceUrl(`/plugins/${author}/${name}`, { theme })} target='_blank' className='flex items-center gap-0.5'>
                 <div className='system-2xs-medium-uppercase text-text-tertiary'>{t('plugin.from')} <span className='text-text-secondary'>marketplace</span></div>
-                <RiArrowRightUpLine className='h-3 w-3 text-text-tertiary' />
+                <RiArrowRightUpLine className='h-3 w-3 text-text-secondary' />
               </a>
             </>
           }

web/app/components/tools/provider/detail.tsx

@@ -244,9 +244,8 @@ const ProviderDetail = ({
               <div className="flex h-5 items-center">
                 <Title title={collection.label[language]} />
               </div>
-              <div className='mb-1 flex h-4 items-center justify-between'>
+              <div className='mb-1 mt-0.5 flex h-4 items-center justify-between'>
                 <OrgInfo
-                  className="mt-0.5"
                   packageNameClassName='w-auto'
                   orgName={collection.author}
                   packageName={collection.name}

web/app/components/workflow/nodes/llm/components/json-schema-config-modal/json-schema-generator/index.tsx

@@ -30,7 +30,7 @@ enum GeneratorView {
   result = 'result',
 }
 
-export const JsonSchemaGenerator: FC<JsonSchemaGeneratorProps> = ({
+const JsonSchemaGenerator: FC<JsonSchemaGeneratorProps> = ({
   onApply,
   crossAxisOffset,
 }) => {

web/app/styles/globals.css

@@ -1,12 +1,18 @@
 @import "preflight.css";
-@tailwind base;
-@tailwind components;
+
 
 @import '../../themes/light.css';
 @import '../../themes/dark.css';
 @import "../../themes/manual-light.css";
 @import "../../themes/manual-dark.css";
 
+@import "../components/base/button/index.css";
+@import "../components/base/action-button/index.css";
+@import "../components/base/modal/index.css";
+
+@tailwind base;
+@tailwind components;
+
 html {
   color-scheme: light;
 }
@@ -680,10 +686,6 @@ button:focus-within {
   display: none;
 }
 
-@import "../components/base/button/index.css";
-@import "../components/base/action-button/index.css";
-@import "../components/base/modal/index.css";
-
 @tailwind utilities;
 
 @layer utilities {

web/models/pipeline.ts

@@ -17,6 +17,7 @@ export enum DatasourceType {
 
 export type PipelineTemplateListParams = {
   type: 'built-in' | 'customized'
+  language?: string
 }
 
 export type PipelineTemplate = {

web/next.config.js

@@ -91,12 +91,10 @@ const remoteImageURLs = [hasSetWebPrefix ? new URL(`${process.env.NEXT_PUBLIC_WE
 /** @type {import('next').NextConfig} */
 const nextConfig = {
   basePath: process.env.NEXT_PUBLIC_BASE_PATH || '',
-  webpack: (config, { dev, isServer }) => {
-    if (dev) {
-      config.plugins.push(codeInspectorPlugin({ bundler: 'webpack' }))
-    }
-
-    return config
+  turbopack: {
+    rules: codeInspectorPlugin({
+      bundler: 'turbopack'
+    })
   },
   productionBrowserSourceMaps: false, // enable browser source map generation during the production build
   // Configure pageExtensions to include md and mdx
@@ -112,6 +110,10 @@ const nextConfig = {
     })),
   },
   experimental: {
+    optimizePackageImports: [
+      '@remixicon/react',
+      '@heroicons/react'
+    ],
   },
   // fix all before production. Now it slow the develop speed.
   eslint: {

web/package.json

@@ -19,7 +19,7 @@
     "and_qq >= 14.9"
   ],
   "scripts": {
-    "dev": "cross-env NODE_OPTIONS='--inspect' next dev",
+    "dev": "cross-env NODE_OPTIONS='--inspect' next dev --turbopack",
     "build": "next build",
     "build:docker": "next build && node scripts/optimize-standalone.js",
     "start": "cp -r .next/static .next/standalone/.next/static && cp -r public .next/standalone/public && cross-env PORT=$npm_config_port HOSTNAME=$npm_config_host node .next/standalone/server.js",
@@ -203,7 +203,7 @@
     "autoprefixer": "^10.4.20",
     "babel-loader": "^10.0.0",
     "bing-translate-api": "^4.0.2",
-    "code-inspector-plugin": "^0.18.1",
+    "code-inspector-plugin": "1.2.9",
     "cross-env": "^7.0.3",
     "eslint": "^9.35.0",
     "eslint-config-next": "15.5.0",

web/pnpm-lock.yaml

@@ -519,8 +519,8 @@ importers:
         specifier: ^4.0.2
         version: 4.1.0
       code-inspector-plugin:
-        specifier: ^0.18.1
-        version: 0.18.3
+        specifier: 1.2.9
+        version: 1.2.9
       cross-env:
         specifier: ^7.0.3
         version: 7.0.3
@@ -1372,6 +1372,24 @@ packages:
   '@clack/prompts@0.11.0':
     resolution: {integrity: sha512-pMN5FcrEw9hUkZA4f+zLlzivQSeQf5dRGJjSUbvVYDLvpKCdQx5OaknvKzgbtXOizhP+SJJJjqEbOe55uKKfAw==}
 
+  '@code-inspector/core@1.2.9':
+    resolution: {integrity: sha512-A1w+G73HlTB6S8X6sA6tT+ziWHTAcTyH+7FZ1Sgd3ZLXF/E/jT+hgRbKposjXMwxcbodRc6hBG6UyiV+VxwE6Q==}
+
+  '@code-inspector/esbuild@1.2.9':
+    resolution: {integrity: sha512-DuyfxGupV43CN8YElIqynAniBtE86i037+3OVJYrm3jlJscXzbV98/kOzvu+VJQQvElcDgpgD6C/aGmPvFEiUg==}
+
+  '@code-inspector/mako@1.2.9':
+    resolution: {integrity: sha512-8N+MHdr64AnthLB4v+YGe8/9bgog3BnkxIW/fqX5iVS0X06mF7X1pxfZOD2bABVtv1tW25lRtNs5AgvYJs0vpg==}
+
+  '@code-inspector/turbopack@1.2.9':
+    resolution: {integrity: sha512-UVOUbqU6rpi5eOkrFamKrdeSWb0/OFFJQBaxbgs1RK5V5f4/iVwC5KjO2wkjv8cOGU4EppLfBVSBI1ysOo8S5A==}
+
+  '@code-inspector/vite@1.2.9':
+    resolution: {integrity: sha512-saIokJ3o3SdrHEgTEg1fbbowbKfh7J4mYtu0i1mVfah1b1UfdCF/iFHTEJ6SADMiY47TeNZTg0TQWTlU1AWPww==}
+
+  '@code-inspector/webpack@1.2.9':
+    resolution: {integrity: sha512-9YEykVrOIc0zMV7pyTyZhCprjScjn6gPPmxb4/OQXKCrP2fAm+NB188rg0s95e4sM7U3qRUpPA4NUH5F7Ogo+g==}
+
   '@cspotcode/source-map-support@0.8.1':
     resolution: {integrity: sha512-IchNf6dN4tHoMFIn/7OE8LWZ19Y6q/67Bmf6vnGREv8RSbBVb9LPJxEcnwrcwX6ixSvaiGoomAUvu4YSxXrVgw==}
     engines: {node: '>=12'}
@@ -4425,11 +4443,8 @@ packages:
     resolution: {integrity: sha512-QVb0dM5HvG+uaxitm8wONl7jltx8dqhfU33DcqtOZcLSVIKSDDLDi7+0LbAKiyI8hD9u42m2YxXSkMGWThaecQ==}
     engines: {iojs: '>= 1.0.0', node: '>= 0.12.0'}
 
-  code-inspector-core@0.18.3:
-    resolution: {integrity: sha512-60pT2cPoguMTUYdN1MMpjoPUnuF0ud/u7M2y+Vqit/bniLEit9dySEWAVxLU/Ukc5ILrDeLKEttc6fCMl9RUrA==}
-
-  code-inspector-plugin@0.18.3:
-    resolution: {integrity: sha512-d9oJXZUsnvfTaQDwFmDNA2F+AR/TXIxWg1rr8KGcEskltR2prbZsfuu1z70EAn4khpx0smfi/PvIIwNJQ7FAMw==}
+  code-inspector-plugin@1.2.9:
+    resolution: {integrity: sha512-PGp/AQ03vaajimG9rn5+eQHGifrym5CSNLCViPtwzot7FM3MqEkGNqcvimH0FVuv3wDOcP5KvETAUSLf1BE3HA==}
 
   collapse-white-space@2.1.0:
     resolution: {integrity: sha512-loKTxY1zCOuG4j9f6EPnuyyYkf58RnhhWTvRoZEokgB+WbdXehfjFviyOVYkqzEWz1Q5kRiZdBYS5SwxbQYwzw==}
@@ -5055,9 +5070,6 @@ packages:
   esast-util-from-js@2.0.1:
     resolution: {integrity: sha512-8Ja+rNJ0Lt56Pcf3TAmpBZjmx8ZcK5Ts4cAzIOjsjevg9oSXJnl6SUQ2EevU8tv3h6ZLWmoKL5H4fgWvdvfETw==}
 
-  esbuild-code-inspector-plugin@0.18.3:
-    resolution: {integrity: sha512-FaPt5eFMtW1oXMWqAcqfAJByNagP1V/R9dwDDLQO29JmryMF35+frskTqy+G53whmTaVi19+TCrFqhNbMZH5ZQ==}
-
   esbuild-register@3.6.0:
     resolution: {integrity: sha512-H2/S7Pm8a9CL1uhp9OvjwrBh5Pvx0H8qVOxNu8Wed9Y7qv56MPtq+GGM8RJpq6glYJn9Wspr8uw7l55uyinNeg==}
     peerDependencies:
@@ -6413,8 +6425,8 @@ packages:
     resolution: {integrity: sha512-MbjN408fEndfiQXbFQ1vnd+1NoLDsnQW41410oQBXiyXDMYH5z505juWa4KUE1LqxRC7DgOgZDbKLxHIwm27hA==}
     engines: {node: '>=0.10'}
 
-  launch-ide@1.0.1:
-    resolution: {integrity: sha512-U7qBxSNk774PxWq4XbmRe0ThiIstPoa4sMH/OGSYxrFVvg8x3biXcF1fsH6wasDpEmEXMdINUrQhBdwsSgKyMg==}
+  launch-ide@1.2.0:
+    resolution: {integrity: sha512-7nXSPQOt3b2JT52Ge8jp4miFcY+nrUEZxNLWBzrEfjmByDTb9b5ytqMSwGhsNwY6Cntwop+6n7rWIFN0+S8PTw==}
 
   layout-base@1.0.2:
     resolution: {integrity: sha512-8h2oVEZNktL4BH2JCOI90iD1yXwL6iNW7KcCKT2QZgQJR2vbqDsldCTPRU9NifTCqHZci57XvQQ15YTu+sTYPg==}
@@ -8693,9 +8705,6 @@ packages:
   vfile@6.0.3:
     resolution: {integrity: sha512-KzIbH/9tXat2u30jf+smMwFCsno4wHVdNmzFyL+T/L3UGqqk6JKfVqOFOZEpZSHADH1k40ab6NUIXZq422ov3Q==}
 
-  vite-code-inspector-plugin@0.18.3:
-    resolution: {integrity: sha512-178H73vbDUHE+JpvfAfioUHlUr7qXCYIEa2YNXtzenFQGOjtae59P1jjcxGfa6pPHEnOoaitb13K+0qxwhi/WA==}
-
   vm-browserify@1.1.2:
     resolution: {integrity: sha512-2ham8XPWTONajOR0ohOKOHXkm3+gaBmGut3SRuu75xLd/RRaY6vqgh8NBYYk7+RW3u5AtzPQZG8F10LHkl0lAQ==}
 
@@ -8754,9 +8763,6 @@ packages:
     engines: {node: '>= 10.13.0'}
     hasBin: true
 
-  webpack-code-inspector-plugin@0.18.3:
-    resolution: {integrity: sha512-3782rsJhBnRiw0IpR6EqnyGDQoiSq0CcGeLJ52rZXlszYCe8igXtcujq7OhI0byaivWQ1LW7sXKyMEoVpBhq0w==}
-
   webpack-dev-middleware@6.1.3:
     resolution: {integrity: sha512-A4ChP0Qj8oGociTs6UdlRUGANIGrCDL3y+pmQMc+dSsraXHCatFpmMey4mYELA+juqwUqwQsUgJJISXl1KWmiw==}
     engines: {node: '>= 14.15.0'}
@@ -9993,6 +9999,48 @@ snapshots:
       picocolors: 1.1.1
       sisteransi: 1.0.5
 
+  '@code-inspector/core@1.2.9':
+    dependencies:
+      '@vue/compiler-dom': 3.5.17
+      chalk: 4.1.2
+      dotenv: 16.6.1
+      launch-ide: 1.2.0
+      portfinder: 1.0.37
+    transitivePeerDependencies:
+      - supports-color
+
+  '@code-inspector/esbuild@1.2.9':
+    dependencies:
+      '@code-inspector/core': 1.2.9
+    transitivePeerDependencies:
+      - supports-color
+
+  '@code-inspector/mako@1.2.9':
+    dependencies:
+      '@code-inspector/core': 1.2.9
+    transitivePeerDependencies:
+      - supports-color
+
+  '@code-inspector/turbopack@1.2.9':
+    dependencies:
+      '@code-inspector/core': 1.2.9
+      '@code-inspector/webpack': 1.2.9
+    transitivePeerDependencies:
+      - supports-color
+
+  '@code-inspector/vite@1.2.9':
+    dependencies:
+      '@code-inspector/core': 1.2.9
+      chalk: 4.1.1
+    transitivePeerDependencies:
+      - supports-color
+
+  '@code-inspector/webpack@1.2.9':
+    dependencies:
+      '@code-inspector/core': 1.2.9
+    transitivePeerDependencies:
+      - supports-color
+
   '@cspotcode/source-map-support@0.8.1':
     dependencies:
       '@jridgewell/trace-mapping': 0.3.9
@@ -12799,7 +12847,7 @@ snapshots:
 
   '@vue/compiler-core@3.5.17':
     dependencies:
-      '@babel/parser': 7.28.0
+      '@babel/parser': 7.28.4
       '@vue/shared': 3.5.17
       entities: 4.5.0
       estree-walker: 2.0.2
@@ -13503,24 +13551,15 @@ snapshots:
 
   co@4.6.0: {}
 
-  code-inspector-core@0.18.3:
+  code-inspector-plugin@1.2.9:
     dependencies:
-      '@vue/compiler-dom': 3.5.17
+      '@code-inspector/core': 1.2.9
+      '@code-inspector/esbuild': 1.2.9
+      '@code-inspector/mako': 1.2.9
+      '@code-inspector/turbopack': 1.2.9
+      '@code-inspector/vite': 1.2.9
+      '@code-inspector/webpack': 1.2.9
       chalk: 4.1.1
-      dotenv: 16.6.1
-      launch-ide: 1.0.1
-      portfinder: 1.0.37
-    transitivePeerDependencies:
-      - supports-color
-
-  code-inspector-plugin@0.18.3:
-    dependencies:
-      chalk: 4.1.1
-      code-inspector-core: 0.18.3
-      dotenv: 16.6.1
-      esbuild-code-inspector-plugin: 0.18.3
-      vite-code-inspector-plugin: 0.18.3
-      webpack-code-inspector-plugin: 0.18.3
     transitivePeerDependencies:
       - supports-color
 
@@ -14160,12 +14199,6 @@ snapshots:
       esast-util-from-estree: 2.0.0
       vfile-message: 4.0.2
 
-  esbuild-code-inspector-plugin@0.18.3:
-    dependencies:
-      code-inspector-core: 0.18.3
-    transitivePeerDependencies:
-      - supports-color
-
   esbuild-register@3.6.0(esbuild@0.25.0):
     dependencies:
       debug: 4.4.1
@@ -16020,7 +16053,7 @@ snapshots:
     dependencies:
       language-subtag-registry: 0.3.23
 
-  launch-ide@1.0.1:
+  launch-ide@1.2.0:
     dependencies:
       chalk: 4.1.2
       dotenv: 16.6.1
@@ -18779,12 +18812,6 @@ snapshots:
       '@types/unist': 3.0.3
       vfile-message: 4.0.2
 
-  vite-code-inspector-plugin@0.18.3:
-    dependencies:
-      code-inspector-core: 0.18.3
-    transitivePeerDependencies:
-      - supports-color
-
   vm-browserify@1.1.2: {}
 
   void-elements@3.1.0: {}
@@ -18855,12 +18882,6 @@ snapshots:
       - bufferutil
       - utf-8-validate
 
-  webpack-code-inspector-plugin@0.18.3:
-    dependencies:
-      code-inspector-core: 0.18.3
-    transitivePeerDependencies:
-      - supports-color
-
   webpack-dev-middleware@6.1.3(webpack@5.100.2(esbuild@0.25.0)(uglify-js@3.19.3)):
     dependencies:
       colorette: 2.0.20

web/service/use-pipeline.ts

@@ -40,8 +40,9 @@ const NAME_SPACE = 'pipeline'
 
 export const PipelineTemplateListQueryKeyPrefix = [NAME_SPACE, 'template-list']
 export const usePipelineTemplateList = (params: PipelineTemplateListParams) => {
+  const { type, language } = params
   return useQuery<PipelineTemplateListResponse>({
-    queryKey: [...PipelineTemplateListQueryKeyPrefix, params.type],
+    queryKey: [...PipelineTemplateListQueryKeyPrefix, type, language],
     queryFn: () => {
       return get<PipelineTemplateListResponse>('/rag/pipeline/templates', { params })
     },
@@ -55,7 +56,7 @@ export const useInvalidCustomizedTemplateList = () => {
 export const usePipelineTemplateById = (params: PipelineTemplateByIdRequest, enabled: boolean) => {
   const { template_id, type } = params
   return useQuery<PipelineTemplateByIdResponse>({
-    queryKey: [NAME_SPACE, 'template', template_id],
+    queryKey: [NAME_SPACE, 'template', type, template_id],
     queryFn: () => {
       return get<PipelineTemplateByIdResponse>(`/rag/pipeline/templates/${template_id}`, {
         params: {
@@ -64,6 +65,7 @@ export const usePipelineTemplateById = (params: PipelineTemplateByIdRequest, ena
       })
     },
     enabled,
+    staleTime: 0,
   })
 }
 

web/tsconfig.json

@@ -26,6 +26,9 @@
     "paths": {
       "@/*": [
         "./*"
+      ],
+      "~@/*": [
+        "./*"
       ]
     }
   },