mirror of
https://github.com/langgenius/dify.git
synced 2024-11-16 03:32:23 +08:00
Merge branch 'main' into tag_redis
This commit is contained in:
liuhaoran
GitHub
commit
aaee1824d9
3
.github/workflows/api-tests.yml
vendored
3
.github/workflows/api-tests.yml
vendored
|
|
@ -78,7 +78,7 @@ jobs:
|
|||
- name: Run Workflow
|
||||
run: poetry run -C api bash dev/pytest/pytest_workflow.sh
|
||||
|
||||
- name: Set up Vector Stores (Weaviate, Qdrant, PGVector, Milvus, PgVecto-RS, Chroma, MyScale, ElasticSearch)
|
||||
- name: Set up Vector Stores (Weaviate, Qdrant, PGVector, Milvus, PgVecto-RS, Chroma, MyScale, ElasticSearch, Couchbase)
|
||||
uses: hoverkraft-tech/compose-action@v2.0.0
|
||||
with:
|
||||
compose-file: |
|
||||
|
|
@ -86,6 +86,7 @@ jobs:
|
|||
services: |
|
||||
weaviate
|
||||
qdrant
|
||||
couchbase-server
|
||||
etcd
|
||||
minio
|
||||
milvus-standalone
|
||||
|
|
|
|||
4
.github/workflows/expose_service_ports.sh
vendored
4
.github/workflows/expose_service_ports.sh
vendored
|
|
@ -7,5 +7,7 @@ yq eval '.services["milvus-standalone"].ports += ["19530:19530"]' -i docker/dock
|
|||
yq eval '.services.pgvector.ports += ["5433:5432"]' -i docker/docker-compose.yaml
|
||||
yq eval '.services["pgvecto-rs"].ports += ["5431:5432"]' -i docker/docker-compose.yaml
|
||||
yq eval '.services["elasticsearch"].ports += ["9200:9200"]' -i docker/docker-compose.yaml
|
||||
yq eval '.services.couchbase-server.ports += ["8091-8096:8091-8096"]' -i docker/docker-compose.yaml
|
||||
yq eval '.services.couchbase-server.ports += ["11210:11210"]' -i docker/docker-compose.yaml
|
||||
|
||||
echo "Ports exposed for sandbox, weaviate, qdrant, chroma, milvus, pgvector, pgvecto-rs, elasticsearch"
|
||||
echo "Ports exposed for sandbox, weaviate, qdrant, chroma, milvus, pgvector, pgvecto-rs, elasticsearch, couchbase"
|
||||
|
|
|
|||
3
.gitignore
vendored
3
.gitignore
vendored
|
|
@ -173,6 +173,7 @@ docker/volumes/myscale/log/*
|
|||
docker/volumes/unstructured/*
|
||||
docker/volumes/pgvector/data/*
|
||||
docker/volumes/pgvecto_rs/data/*
|
||||
docker/volumes/couchbase/*
|
||||
|
||||
docker/nginx/conf.d/default.conf
|
||||
docker/nginx/ssl/*
|
||||
|
|
@ -189,4 +190,4 @@ pyrightconfig.json
|
|||
api/.vscode
|
||||
|
||||
.idea/
|
||||
.vscode
|
||||
.vscode
|
||||
|
|
|
|||
|
|
@ -1,5 +1,9 @@
|
|||
|
||||
|
||||
<p align="center">
|
||||
📌 <a href="https://dify.ai/blog/introducing-dify-workflow-file-upload-a-demo-on-ai-podcast">Introducing Dify Workflow File Upload: Recreate Google NotebookLM Podcast</a>
|
||||
</p>
|
||||
|
||||
<p align="center">
|
||||
<a href="https://cloud.dify.ai">Dify Cloud</a> ·
|
||||
<a href="https://docs.dify.ai/getting-started/install-self-hosted">Self-hosting</a> ·
|
||||
|
|
|
|||
|
|
@ -154,7 +154,7 @@ Dify 是一个开源的 LLM 应用开发平台。其直观的界面结合了 AI
|
|||
我们提供[ Dify 云服务](https://dify.ai)，任何人都可以零设置尝试。它提供了自部署版本的所有功能，并在沙盒计划中包含 200 次免费的 GPT-4 调用。
|
||||
|
||||
- **自托管 Dify 社区版</br>**
|
||||
使用这个[入门指南](#quick-start)快速在您的环境中运行 Dify。
|
||||
使用这个[入门指南](#快速启动)快速在您的环境中运行 Dify。
|
||||
使用我们的[文档](https://docs.dify.ai)进行进一步的参考和更深入的说明。
|
||||
|
||||
- **面向企业/组织的 Dify</br>**
|
||||
|
|
|
|||
|
|
@ -31,8 +31,17 @@ REDIS_HOST=localhost
|
|||
REDIS_PORT=6379
|
||||
REDIS_USERNAME=
|
||||
REDIS_PASSWORD=difyai123456
|
||||
REDIS_USE_SSL=false
|
||||
REDIS_DB=0
|
||||
|
||||
# redis Sentinel configuration.
|
||||
REDIS_USE_SENTINEL=false
|
||||
REDIS_SENTINELS=
|
||||
REDIS_SENTINEL_SERVICE_NAME=
|
||||
REDIS_SENTINEL_USERNAME=
|
||||
REDIS_SENTINEL_PASSWORD=
|
||||
REDIS_SENTINEL_SOCKET_TIMEOUT=0.1
|
||||
|
||||
# PostgreSQL database configuration
|
||||
DB_USERNAME=postgres
|
||||
DB_PASSWORD=difyai123456
|
||||
|
|
@ -111,7 +120,7 @@ SUPABASE_URL=your-server-url
|
|||
WEB_API_CORS_ALLOW_ORIGINS=http://127.0.0.1:3000,*
|
||||
CONSOLE_CORS_ALLOW_ORIGINS=http://127.0.0.1:3000,*
|
||||
|
||||
# Vector database configuration, support: weaviate, qdrant, milvus, myscale, relyt, pgvecto_rs, pgvector, pgvector, chroma, opensearch, tidb_vector, vikingdb, upstash
|
||||
# Vector database configuration, support: weaviate, qdrant, milvus, myscale, relyt, pgvecto_rs, pgvector, pgvector, chroma, opensearch, tidb_vector, couchbase, vikingdb, upstash
|
||||
VECTOR_STORE=weaviate
|
||||
|
||||
# Weaviate configuration
|
||||
|
|
@ -127,6 +136,13 @@ QDRANT_CLIENT_TIMEOUT=20
|
|||
QDRANT_GRPC_ENABLED=false
|
||||
QDRANT_GRPC_PORT=6334
|
||||
|
||||
#Couchbase configuration
|
||||
COUCHBASE_CONNECTION_STRING=127.0.0.1
|
||||
COUCHBASE_USER=Administrator
|
||||
COUCHBASE_PASSWORD=password
|
||||
COUCHBASE_BUCKET_NAME=Embeddings
|
||||
COUCHBASE_SCOPE_NAME=_default
|
||||
|
||||
# Milvus configuration
|
||||
MILVUS_URI=http://127.0.0.1:19530
|
||||
MILVUS_TOKEN=
|
||||
|
|
|
|||
|
|
@ -55,7 +55,9 @@ RUN apt-get update \
|
|||
&& echo "deb http://deb.debian.org/debian testing main" > /etc/apt/sources.list \
|
||||
&& apt-get update \
|
||||
# For Security
|
||||
&& apt-get install -y --no-install-recommends zlib1g=1:1.3.dfsg+really1.3.1-1 expat=2.6.3-1 libldap-2.5-0=2.5.18+dfsg-3 perl=5.38.2-5 libsqlite3-0=3.46.1-1 \
|
||||
&& apt-get install -y --no-install-recommends zlib1g=1:1.3.dfsg+really1.3.1-1 expat=2.6.3-1 libldap-2.5-0=2.5.18+dfsg-3+b1 perl=5.40.0-6 libsqlite3-0=3.46.1-1 \
|
||||
# install a chinese font to support the use of tools like matplotlib
|
||||
&& apt-get install -y fonts-noto-cjk \
|
||||
&& apt-get autoremove -y \
|
||||
&& rm -rf /var/lib/apt/lists/*
|
||||
|
||||
|
|
|
|||
|
|
@ -278,6 +278,7 @@ def migrate_knowledge_vector_database():
|
|||
VectorType.BAIDU,
|
||||
VectorType.VIKINGDB,
|
||||
VectorType.UPSTASH,
|
||||
VectorType.COUCHBASE,
|
||||
}
|
||||
page = 1
|
||||
while True:
|
||||
|
|
|
|||
|
|
@ -571,6 +571,11 @@ class DataSetConfig(BaseSettings):
|
|||
default=False,
|
||||
)
|
||||
|
||||
TIDB_SERVERLESS_NUMBER: PositiveInt = Field(
|
||||
description="number of tidb serverless cluster",
|
||||
default=500,
|
||||
)
|
||||
|
||||
|
||||
class WorkspaceConfig(BaseSettings):
|
||||
"""
|
||||
|
|
|
|||
|
|
@ -17,6 +17,7 @@ from configs.middleware.storage.tencent_cos_storage_config import TencentCloudCO
|
|||
from configs.middleware.storage.volcengine_tos_storage_config import VolcengineTOSStorageConfig
|
||||
from configs.middleware.vdb.analyticdb_config import AnalyticdbConfig
|
||||
from configs.middleware.vdb.chroma_config import ChromaConfig
|
||||
from configs.middleware.vdb.couchbase_config import CouchbaseConfig
|
||||
from configs.middleware.vdb.elasticsearch_config import ElasticsearchConfig
|
||||
from configs.middleware.vdb.milvus_config import MilvusConfig
|
||||
from configs.middleware.vdb.myscale_config import MyScaleConfig
|
||||
|
|
@ -27,6 +28,7 @@ from configs.middleware.vdb.pgvectors_config import PGVectoRSConfig
|
|||
from configs.middleware.vdb.qdrant_config import QdrantConfig
|
||||
from configs.middleware.vdb.relyt_config import RelytConfig
|
||||
from configs.middleware.vdb.tencent_vector_config import TencentVectorDBConfig
|
||||
from configs.middleware.vdb.tidb_on_qdrant_config import TidbOnQdrantConfig
|
||||
from configs.middleware.vdb.tidb_vector_config import TiDBVectorConfig
|
||||
from configs.middleware.vdb.upstash_config import UpstashConfig
|
||||
from configs.middleware.vdb.vikingdb_config import VikingDBConfig
|
||||
|
|
@ -54,6 +56,11 @@ class VectorStoreConfig(BaseSettings):
|
|||
default=None,
|
||||
)
|
||||
|
||||
VECTOR_STORE_WHITELIST_ENABLE: Optional[bool] = Field(
|
||||
description="Enable whitelist for vector store.",
|
||||
default=False,
|
||||
)
|
||||
|
||||
|
||||
class KeywordStoreConfig(BaseSettings):
|
||||
KEYWORD_STORE: str = Field(
|
||||
|
|
@ -245,8 +252,10 @@ class MiddlewareConfig(
|
|||
TiDBVectorConfig,
|
||||
WeaviateConfig,
|
||||
ElasticsearchConfig,
|
||||
CouchbaseConfig,
|
||||
InternalTestConfig,
|
||||
VikingDBConfig,
|
||||
UpstashConfig,
|
||||
TidbOnQdrantConfig,
|
||||
):
|
||||
pass
|
||||
|
|
|
|||
34
api/configs/middleware/vdb/couchbase_config.py
Normal file
34
api/configs/middleware/vdb/couchbase_config.py
Normal file
|
|
@ -0,0 +1,34 @@
|
|||
from typing import Optional
|
||||
|
||||
from pydantic import BaseModel, Field
|
||||
|
||||
|
||||
class CouchbaseConfig(BaseModel):
|
||||
"""
|
||||
Couchbase configs
|
||||
"""
|
||||
|
||||
COUCHBASE_CONNECTION_STRING: Optional[str] = Field(
|
||||
description="COUCHBASE connection string",
|
||||
default=None,
|
||||
)
|
||||
|
||||
COUCHBASE_USER: Optional[str] = Field(
|
||||
description="COUCHBASE user",
|
||||
default=None,
|
||||
)
|
||||
|
||||
COUCHBASE_PASSWORD: Optional[str] = Field(
|
||||
description="COUCHBASE password",
|
||||
default=None,
|
||||
)
|
||||
|
||||
COUCHBASE_BUCKET_NAME: Optional[str] = Field(
|
||||
description="COUCHBASE bucket name",
|
||||
default=None,
|
||||
)
|
||||
|
||||
COUCHBASE_SCOPE_NAME: Optional[str] = Field(
|
||||
description="COUCHBASE scope name",
|
||||
default=None,
|
||||
)
|
||||
65
api/configs/middleware/vdb/tidb_on_qdrant_config.py
Normal file
65
api/configs/middleware/vdb/tidb_on_qdrant_config.py
Normal file
|
|
@ -0,0 +1,65 @@
|
|||
from typing import Optional
|
||||
|
||||
from pydantic import Field, NonNegativeInt, PositiveInt
|
||||
from pydantic_settings import BaseSettings
|
||||
|
||||
|
||||
class TidbOnQdrantConfig(BaseSettings):
|
||||
"""
|
||||
Tidb on Qdrant configs
|
||||
"""
|
||||
|
||||
TIDB_ON_QDRANT_URL: Optional[str] = Field(
|
||||
description="Tidb on Qdrant url",
|
||||
default=None,
|
||||
)
|
||||
|
||||
TIDB_ON_QDRANT_API_KEY: Optional[str] = Field(
|
||||
description="Tidb on Qdrant api key",
|
||||
default=None,
|
||||
)
|
||||
|
||||
TIDB_ON_QDRANT_CLIENT_TIMEOUT: NonNegativeInt = Field(
|
||||
description="Tidb on Qdrant client timeout in seconds",
|
||||
default=20,
|
||||
)
|
||||
|
||||
TIDB_ON_QDRANT_GRPC_ENABLED: bool = Field(
|
||||
description="whether enable grpc support for Tidb on Qdrant connection",
|
||||
default=False,
|
||||
)
|
||||
|
||||
TIDB_ON_QDRANT_GRPC_PORT: PositiveInt = Field(
|
||||
description="Tidb on Qdrant grpc port",
|
||||
default=6334,
|
||||
)
|
||||
|
||||
TIDB_PUBLIC_KEY: Optional[str] = Field(
|
||||
description="Tidb account public key",
|
||||
default=None,
|
||||
)
|
||||
|
||||
TIDB_PRIVATE_KEY: Optional[str] = Field(
|
||||
description="Tidb account private key",
|
||||
default=None,
|
||||
)
|
||||
|
||||
TIDB_API_URL: Optional[str] = Field(
|
||||
description="Tidb API url",
|
||||
default=None,
|
||||
)
|
||||
|
||||
TIDB_IAM_API_URL: Optional[str] = Field(
|
||||
description="Tidb IAM API url",
|
||||
default=None,
|
||||
)
|
||||
|
||||
TIDB_REGION: Optional[str] = Field(
|
||||
description="Tidb serverless region",
|
||||
default="regions/aws-us-east-1",
|
||||
)
|
||||
|
||||
TIDB_PROJECT_ID: Optional[str] = Field(
|
||||
description="Tidb project id",
|
||||
default=None,
|
||||
)
|
||||
|
|
@ -9,7 +9,7 @@ class PackagingInfo(BaseSettings):
|
|||
|
||||
CURRENT_VERSION: str = Field(
|
||||
description="Dify version",
|
||||
default="0.10.1",
|
||||
default="0.10.2",
|
||||
)
|
||||
|
||||
COMMIT_SHA: str = Field(
|
||||
|
|
|
|||
|
|
@ -105,6 +105,8 @@ class ChatMessageListApi(Resource):
|
|||
if rest_count > 0:
|
||||
has_more = True
|
||||
|
||||
history_messages = list(reversed(history_messages))
|
||||
|
||||
return InfiniteScrollPagination(data=history_messages, limit=args["limit"], has_more=has_more)
|
||||
|
||||
|
||||
|
|
|
|||
|
|
@ -102,6 +102,13 @@ class DatasetListApi(Resource):
|
|||
help="type is required. Name must be between 1 to 40 characters.",
|
||||
type=_validate_name,
|
||||
)
|
||||
parser.add_argument(
|
||||
"description",
|
||||
type=str,
|
||||
nullable=True,
|
||||
required=False,
|
||||
default="",
|
||||
)
|
||||
parser.add_argument(
|
||||
"indexing_technique",
|
||||
type=str,
|
||||
|
|
@ -140,6 +147,7 @@ class DatasetListApi(Resource):
|
|||
dataset = DatasetService.create_empty_dataset(
|
||||
tenant_id=current_user.current_tenant_id,
|
||||
name=args["name"],
|
||||
description=args["description"],
|
||||
indexing_technique=args["indexing_technique"],
|
||||
account=current_user,
|
||||
permission=DatasetPermissionEnum.ONLY_ME,
|
||||
|
|
@ -631,6 +639,8 @@ class DatasetRetrievalSettingApi(Resource):
|
|||
| VectorType.ORACLE
|
||||
| VectorType.ELASTICSEARCH
|
||||
| VectorType.PGVECTOR
|
||||
| VectorType.TIDB_ON_QDRANT
|
||||
| VectorType.COUCHBASE
|
||||
):
|
||||
return {
|
||||
"retrieval_method": [
|
||||
|
|
@ -669,6 +679,7 @@ class DatasetRetrievalSettingMockApi(Resource):
|
|||
| VectorType.MYSCALE
|
||||
| VectorType.ORACLE
|
||||
| VectorType.ELASTICSEARCH
|
||||
| VectorType.COUCHBASE
|
||||
| VectorType.PGVECTOR
|
||||
):
|
||||
return {
|
||||
|
|
|
|||
|
|
@ -21,7 +21,12 @@ class AppParameterApi(InstalledAppResource):
|
|||
"options": fields.List(fields.String),
|
||||
}
|
||||
|
||||
system_parameters_fields = {"image_file_size_limit": fields.String}
|
||||
system_parameters_fields = {
|
||||
"image_file_size_limit": fields.Integer,
|
||||
"video_file_size_limit": fields.Integer,
|
||||
"audio_file_size_limit": fields.Integer,
|
||||
"file_size_limit": fields.Integer,
|
||||
}
|
||||
|
||||
parameters_fields = {
|
||||
"opening_statement": fields.String,
|
||||
|
|
@ -82,7 +87,12 @@ class AppParameterApi(InstalledAppResource):
|
|||
}
|
||||
},
|
||||
),
|
||||
"system_parameters": {"image_file_size_limit": dify_config.UPLOAD_IMAGE_FILE_SIZE_LIMIT},
|
||||
"system_parameters": {
|
||||
"image_file_size_limit": dify_config.UPLOAD_IMAGE_FILE_SIZE_LIMIT,
|
||||
"video_file_size_limit": dify_config.UPLOAD_VIDEO_FILE_SIZE_LIMIT,
|
||||
"audio_file_size_limit": dify_config.UPLOAD_AUDIO_FILE_SIZE_LIMIT,
|
||||
"file_size_limit": dify_config.UPLOAD_FILE_SIZE_LIMIT,
|
||||
},
|
||||
}
|
||||
|
||||
|
||||
|
|
|
|||
|
|
@ -21,7 +21,7 @@ class EnterpriseWorkspace(Resource):
|
|||
if account is None:
|
||||
return {"message": "owner account not found."}, 404
|
||||
|
||||
tenant = TenantService.create_tenant(args["name"])
|
||||
tenant = TenantService.create_tenant(args["name"], is_from_dashboard=True)
|
||||
TenantService.create_tenant_member(tenant, account, role="owner")
|
||||
|
||||
tenant_was_created.send(tenant)
|
||||
|
|
|
|||
|
|
@ -21,7 +21,12 @@ class AppParameterApi(Resource):
|
|||
"options": fields.List(fields.String),
|
||||
}
|
||||
|
||||
system_parameters_fields = {"image_file_size_limit": fields.String}
|
||||
system_parameters_fields = {
|
||||
"image_file_size_limit": fields.Integer,
|
||||
"video_file_size_limit": fields.Integer,
|
||||
"audio_file_size_limit": fields.Integer,
|
||||
"file_size_limit": fields.Integer,
|
||||
}
|
||||
|
||||
parameters_fields = {
|
||||
"opening_statement": fields.String,
|
||||
|
|
@ -81,7 +86,12 @@ class AppParameterApi(Resource):
|
|||
}
|
||||
},
|
||||
),
|
||||
"system_parameters": {"image_file_size_limit": dify_config.UPLOAD_IMAGE_FILE_SIZE_LIMIT},
|
||||
"system_parameters": {
|
||||
"image_file_size_limit": dify_config.UPLOAD_IMAGE_FILE_SIZE_LIMIT,
|
||||
"video_file_size_limit": dify_config.UPLOAD_VIDEO_FILE_SIZE_LIMIT,
|
||||
"audio_file_size_limit": dify_config.UPLOAD_AUDIO_FILE_SIZE_LIMIT,
|
||||
"file_size_limit": dify_config.UPLOAD_FILE_SIZE_LIMIT,
|
||||
},
|
||||
}
|
||||
|
||||
|
||||
|
|
|
|||
|
|
@ -66,6 +66,13 @@ class DatasetListApi(DatasetApiResource):
|
|||
help="type is required. Name must be between 1 to 40 characters.",
|
||||
type=_validate_name,
|
||||
)
|
||||
parser.add_argument(
|
||||
"description",
|
||||
type=str,
|
||||
nullable=True,
|
||||
required=False,
|
||||
default="",
|
||||
)
|
||||
parser.add_argument(
|
||||
"indexing_technique",
|
||||
type=str,
|
||||
|
|
@ -108,6 +115,7 @@ class DatasetListApi(DatasetApiResource):
|
|||
dataset = DatasetService.create_empty_dataset(
|
||||
tenant_id=tenant_id,
|
||||
name=args["name"],
|
||||
description=args["description"],
|
||||
indexing_technique=args["indexing_technique"],
|
||||
account=current_user,
|
||||
permission=args["permission"],
|
||||
|
|
|
|||
|
|
@ -21,7 +21,12 @@ class AppParameterApi(WebApiResource):
|
|||
"options": fields.List(fields.String),
|
||||
}
|
||||
|
||||
system_parameters_fields = {"image_file_size_limit": fields.String}
|
||||
system_parameters_fields = {
|
||||
"image_file_size_limit": fields.Integer,
|
||||
"video_file_size_limit": fields.Integer,
|
||||
"audio_file_size_limit": fields.Integer,
|
||||
"file_size_limit": fields.Integer,
|
||||
}
|
||||
|
||||
parameters_fields = {
|
||||
"opening_statement": fields.String,
|
||||
|
|
@ -80,7 +85,12 @@ class AppParameterApi(WebApiResource):
|
|||
}
|
||||
},
|
||||
),
|
||||
"system_parameters": {"image_file_size_limit": dify_config.UPLOAD_IMAGE_FILE_SIZE_LIMIT},
|
||||
"system_parameters": {
|
||||
"image_file_size_limit": dify_config.UPLOAD_IMAGE_FILE_SIZE_LIMIT,
|
||||
"video_file_size_limit": dify_config.UPLOAD_VIDEO_FILE_SIZE_LIMIT,
|
||||
"audio_file_size_limit": dify_config.UPLOAD_AUDIO_FILE_SIZE_LIMIT,
|
||||
"file_size_limit": dify_config.UPLOAD_FILE_SIZE_LIMIT,
|
||||
},
|
||||
}
|
||||
|
||||
|
||||
|
|
|
|||
|
|
@ -165,6 +165,12 @@ class BaseAgentRunner(AppRunner):
|
|||
continue
|
||||
|
||||
parameter_type = parameter.type.as_normal_type()
|
||||
if parameter.type in {
|
||||
ToolParameter.ToolParameterType.SYSTEM_FILES,
|
||||
ToolParameter.ToolParameterType.FILE,
|
||||
ToolParameter.ToolParameterType.FILES,
|
||||
}:
|
||||
continue
|
||||
enum = []
|
||||
if parameter.type == ToolParameter.ToolParameterType.SELECT:
|
||||
enum = [option.value for option in parameter.options]
|
||||
|
|
@ -250,6 +256,12 @@ class BaseAgentRunner(AppRunner):
|
|||
continue
|
||||
|
||||
parameter_type = parameter.type.as_normal_type()
|
||||
if parameter.type in {
|
||||
ToolParameter.ToolParameterType.SYSTEM_FILES,
|
||||
ToolParameter.ToolParameterType.FILE,
|
||||
ToolParameter.ToolParameterType.FILES,
|
||||
}:
|
||||
continue
|
||||
enum = []
|
||||
if parameter.type == ToolParameter.ToolParameterType.SELECT:
|
||||
enum = [option.value for option in parameter.options]
|
||||
|
|
|
|||
|
|
@ -76,8 +76,16 @@ def to_prompt_message_content(f: File, /):
|
|||
|
||||
|
||||
def download(f: File, /):
|
||||
upload_file = file_repository.get_upload_file(session=db.session(), file=f)
|
||||
return _download_file_content(upload_file.key)
|
||||
if f.transfer_method == FileTransferMethod.TOOL_FILE:
|
||||
tool_file = file_repository.get_tool_file(session=db.session(), file=f)
|
||||
return _download_file_content(tool_file.file_key)
|
||||
elif f.transfer_method == FileTransferMethod.LOCAL_FILE:
|
||||
upload_file = file_repository.get_upload_file(session=db.session(), file=f)
|
||||
return _download_file_content(upload_file.key)
|
||||
# remote file
|
||||
response = ssrf_proxy.get(f.remote_url, follow_redirects=True)
|
||||
response.raise_for_status()
|
||||
return response.content
|
||||
|
||||
|
||||
def _download_file_content(path: str, /):
|
||||
|
|
|
|||
|
|
@ -105,6 +105,7 @@ class LLMResult(BaseModel):
|
|||
Model class for llm result.
|
||||
"""
|
||||
|
||||
id: Optional[str] = None
|
||||
model: str
|
||||
prompt_messages: list[PromptMessage]
|
||||
message: AssistantPromptMessage
|
||||
|
|
|
|||
|
|
@ -53,6 +53,9 @@ model_credential_schema:
|
|||
type: select
|
||||
required: true
|
||||
options:
|
||||
- label:
|
||||
en_US: 2024-10-01-preview
|
||||
value: 2024-10-01-preview
|
||||
- label:
|
||||
en_US: 2024-09-01-preview
|
||||
value: 2024-09-01-preview
|
||||
|
|
|
|||
|
|
@ -45,9 +45,7 @@ class AzureOpenAILargeLanguageModel(_CommonAzureOpenAI, LargeLanguageModel):
|
|||
stream: bool = True,
|
||||
user: Optional[str] = None,
|
||||
) -> Union[LLMResult, Generator]:
|
||||
base_model_name = credentials.get("base_model_name")
|
||||
if not base_model_name:
|
||||
raise ValueError("Base Model Name is required")
|
||||
base_model_name = self._get_base_model_name(credentials)
|
||||
ai_model_entity = self._get_ai_model_entity(base_model_name=base_model_name, model=model)
|
||||
|
||||
if ai_model_entity and ai_model_entity.entity.model_properties.get(ModelPropertyKey.MODE) == LLMMode.CHAT.value:
|
||||
|
|
@ -81,9 +79,7 @@ class AzureOpenAILargeLanguageModel(_CommonAzureOpenAI, LargeLanguageModel):
|
|||
prompt_messages: list[PromptMessage],
|
||||
tools: Optional[list[PromptMessageTool]] = None,
|
||||
) -> int:
|
||||
base_model_name = credentials.get("base_model_name")
|
||||
if not base_model_name:
|
||||
raise ValueError("Base Model Name is required")
|
||||
base_model_name = self._get_base_model_name(credentials)
|
||||
model_entity = self._get_ai_model_entity(base_model_name=base_model_name, model=model)
|
||||
if not model_entity:
|
||||
raise ValueError(f"Base Model Name {base_model_name} is invalid")
|
||||
|
|
@ -108,9 +104,7 @@ class AzureOpenAILargeLanguageModel(_CommonAzureOpenAI, LargeLanguageModel):
|
|||
if "base_model_name" not in credentials:
|
||||
raise CredentialsValidateFailedError("Base Model Name is required")
|
||||
|
||||
base_model_name = credentials.get("base_model_name")
|
||||
if not base_model_name:
|
||||
raise CredentialsValidateFailedError("Base Model Name is required")
|
||||
base_model_name = self._get_base_model_name(credentials)
|
||||
ai_model_entity = self._get_ai_model_entity(base_model_name=base_model_name, model=model)
|
||||
|
||||
if not ai_model_entity:
|
||||
|
|
@ -149,9 +143,7 @@ class AzureOpenAILargeLanguageModel(_CommonAzureOpenAI, LargeLanguageModel):
|
|||
raise CredentialsValidateFailedError(str(ex))
|
||||
|
||||
def get_customizable_model_schema(self, model: str, credentials: dict) -> Optional[AIModelEntity]:
|
||||
base_model_name = credentials.get("base_model_name")
|
||||
if not base_model_name:
|
||||
raise ValueError("Base Model Name is required")
|
||||
base_model_name = self._get_base_model_name(credentials)
|
||||
ai_model_entity = self._get_ai_model_entity(base_model_name=base_model_name, model=model)
|
||||
return ai_model_entity.entity if ai_model_entity else None
|
||||
|
||||
|
|
@ -308,11 +300,6 @@ class AzureOpenAILargeLanguageModel(_CommonAzureOpenAI, LargeLanguageModel):
|
|||
|
||||
if tools:
|
||||
extra_model_kwargs["tools"] = [helper.dump_model(PromptMessageFunction(function=tool)) for tool in tools]
|
||||
# extra_model_kwargs['functions'] = [{
|
||||
# "name": tool.name,
|
||||
# "description": tool.description,
|
||||
# "parameters": tool.parameters
|
||||
# } for tool in tools]
|
||||
|
||||
if stop:
|
||||
extra_model_kwargs["stop"] = stop
|
||||
|
|
@ -769,3 +756,9 @@ class AzureOpenAILargeLanguageModel(_CommonAzureOpenAI, LargeLanguageModel):
|
|||
ai_model_entity_copy.entity.label.en_US = model
|
||||
ai_model_entity_copy.entity.label.zh_Hans = model
|
||||
return ai_model_entity_copy
|
||||
|
||||
def _get_base_model_name(self, credentials: dict) -> str:
|
||||
base_model_name = credentials.get("base_model_name")
|
||||
if not base_model_name:
|
||||
raise ValueError("Base Model Name is required")
|
||||
return base_model_name
|
||||
|
|
|
|||
File diff suppressed because one or more lines are too long
|
After Width: | Height: | Size: 9.8 KiB |
|
|
@ -0,0 +1,3 @@
|
|||
<svg width="40" height="40" viewBox="0 0 40 40" fill="none" xmlns="http://www.w3.org/2000/svg">
|
||||
<path fill-rule="evenodd" clip-rule="evenodd" d="M25.132 24.3947C25.497 25.7527 25.8984 27.1413 26.3334 28.5834C26.7302 29.8992 25.5459 30.4167 25.0752 29.1758C24.571 27.8466 24.0885 26.523 23.6347 25.1729C21.065 26.4654 18.5025 27.5424 15.5961 28.7541C16.7581 33.0256 17.8309 36.5984 19.4952 39.9935C19.4953 39.9936 19.4953 39.9937 19.4954 39.9938C19.6631 39.9979 19.8313 40 20 40C31.0457 40 40 31.0457 40 20C40 16.0335 38.8453 12.3366 36.8537 9.22729C31.6585 9.69534 27.0513 10.4562 22.8185 11.406C22.8882 12.252 22.9677 13.0739 23.0555 13.855C23.3824 16.7604 23.9112 19.5281 24.6137 22.3836C27.0581 21.2848 29.084 20.3225 30.6816 19.522C32.2154 18.7535 33.6943 18.7062 31.2018 20.6594C29.0388 22.1602 27.0644 23.3566 25.132 24.3947ZM36.1559 8.20846C33.0001 3.89184 28.1561 0.887462 22.5955 0.166882C22.4257 2.86234 22.4785 6.26344 22.681 9.50447C26.7473 8.88859 31.1721 8.46032 36.1559 8.20846ZM19.9369 9.73661e-05C19.7594 2.92694 19.8384 6.65663 20.19 9.91293C17.3748 10.4109 14.7225 11.0064 12.1592 11.7038C12.0486 10.4257 11.9927 9.25764 11.9927 8.24178C11.9927 7.5054 11.3957 6.90844 10.6593 6.90844C9.92296 6.90844 9.32601 7.5054 9.32601 8.24178C9.32601 9.47868 9.42873 10.898 9.61402 12.438C8.33567 12.8278 7.07397 13.2443 5.81918 13.688C5.12493 13.9336 4.76118 14.6954 5.0067 15.3896C5.25223 16.0839 6.01406 16.4476 6.7083 16.2021C7.7931 15.8185 8.88482 15.4388 9.98927 15.0659C10.5222 18.3344 11.3344 21.9428 12.2703 25.4156C12.4336 26.0218 12.6062 26.6262 12.7863 27.2263C9.34168 28.4135 5.82612 29.3782 2.61128 29.8879C0.949407 26.9716 0 23.5967 0 20C0 8.97534 8.92023 0.0341108 19.9369 9.73661e-05ZM4.19152 32.2527C7.45069 36.4516 12.3458 39.3173 17.9204 39.8932C16.5916 37.455 14.9338 33.717 13.5405 29.5901C10.4404 30.7762 7.25883 31.6027 4.19152 32.2527ZM22.9735 23.1135C22.1479 20.41 21.4462 17.5441 20.9225 14.277C20.746 13.5841 20.5918 12.8035 20.4593 11.9636C17.6508 12.6606 14.9992 13.4372 12.4356 14.2598C12.8479 17.4766 13.5448 21.1334 14.5118 24.7218C14.662 25.2792 14.8081 25.8248 14.9514 26.3594L14.9516 26.3603L14.9524 26.3634L14.9526 26.3639L14.973 26.4401C16.1833 25.9872 17.3746 25.5123 18.53 25.0259C20.1235 24.3552 21.6051 23.7165 22.9735 23.1135Z" fill="#141519"/>
|
||||
</svg>
|
||||
|
After Width: | Height: | Size: 2.2 KiB |
47
api/core/model_runtime/model_providers/gitee_ai/_common.py
Normal file
47
api/core/model_runtime/model_providers/gitee_ai/_common.py
Normal file
|
|
@ -0,0 +1,47 @@
|
|||
from dashscope.common.error import (
|
||||
AuthenticationError,
|
||||
InvalidParameter,
|
||||
RequestFailure,
|
||||
ServiceUnavailableError,
|
||||
UnsupportedHTTPMethod,
|
||||
UnsupportedModel,
|
||||
)
|
||||
|
||||
from core.model_runtime.errors.invoke import (
|
||||
InvokeAuthorizationError,
|
||||
InvokeBadRequestError,
|
||||
InvokeConnectionError,
|
||||
InvokeError,
|
||||
InvokeRateLimitError,
|
||||
InvokeServerUnavailableError,
|
||||
)
|
||||
|
||||
|
||||
class _CommonGiteeAI:
|
||||
@property
|
||||
def _invoke_error_mapping(self) -> dict[type[InvokeError], list[type[Exception]]]:
|
||||
"""
|
||||
Map model invoke error to unified error
|
||||
The key is the error type thrown to the caller
|
||||
The value is the error type thrown by the model,
|
||||
which needs to be converted into a unified error type for the caller.
|
||||
|
||||
:return: Invoke error mapping
|
||||
"""
|
||||
return {
|
||||
InvokeConnectionError: [
|
||||
RequestFailure,
|
||||
],
|
||||
InvokeServerUnavailableError: [
|
||||
ServiceUnavailableError,
|
||||
],
|
||||
InvokeRateLimitError: [],
|
||||
InvokeAuthorizationError: [
|
||||
AuthenticationError,
|
||||
],
|
||||
InvokeBadRequestError: [
|
||||
InvalidParameter,
|
||||
UnsupportedModel,
|
||||
UnsupportedHTTPMethod,
|
||||
],
|
||||
}
|
||||
25
api/core/model_runtime/model_providers/gitee_ai/gitee_ai.py
Normal file
25
api/core/model_runtime/model_providers/gitee_ai/gitee_ai.py
Normal file
|
|
@ -0,0 +1,25 @@
|
|||
import logging
|
||||
|
||||
from core.model_runtime.entities.model_entities import ModelType
|
||||
from core.model_runtime.errors.validate import CredentialsValidateFailedError
|
||||
from core.model_runtime.model_providers.__base.model_provider import ModelProvider
|
||||
|
||||
logger = logging.getLogger(__name__)
|
||||
|
||||
|
||||
class GiteeAIProvider(ModelProvider):
|
||||
def validate_provider_credentials(self, credentials: dict) -> None:
|
||||
"""
|
||||
Validate provider credentials
|
||||
if validate failed, raise exception
|
||||
|
||||
:param credentials: provider credentials, credentials form defined in `provider_credential_schema`.
|
||||
"""
|
||||
try:
|
||||
model_instance = self.get_model_instance(ModelType.LLM)
|
||||
model_instance.validate_credentials(model="Qwen2-7B-Instruct", credentials=credentials)
|
||||
except CredentialsValidateFailedError as ex:
|
||||
raise ex
|
||||
except Exception as ex:
|
||||
logger.exception(f"{self.get_provider_schema().provider} credentials validate failed")
|
||||
raise ex
|
||||
|
|
@ -0,0 +1,35 @@
|
|||
provider: gitee_ai
|
||||
label:
|
||||
en_US: Gitee AI
|
||||
zh_Hans: Gitee AI
|
||||
description:
|
||||
en_US: 快速体验大模型，领先探索 AI 开源世界
|
||||
zh_Hans: 快速体验大模型，领先探索 AI 开源世界
|
||||
icon_small:
|
||||
en_US: Gitee-AI-Logo.svg
|
||||
icon_large:
|
||||
en_US: Gitee-AI-Logo-full.svg
|
||||
help:
|
||||
title:
|
||||
en_US: Get your token from Gitee AI
|
||||
zh_Hans: 从 Gitee AI 获取 token
|
||||
url:
|
||||
en_US: https://ai.gitee.com/dashboard/settings/tokens
|
||||
supported_model_types:
|
||||
- llm
|
||||
- text-embedding
|
||||
- rerank
|
||||
- speech2text
|
||||
- tts
|
||||
configurate_methods:
|
||||
- predefined-model
|
||||
provider_credential_schema:
|
||||
credential_form_schemas:
|
||||
- variable: api_key
|
||||
label:
|
||||
en_US: API Key
|
||||
type: secret-input
|
||||
required: true
|
||||
placeholder:
|
||||
zh_Hans: 在此输入您的 API Key
|
||||
en_US: Enter your API Key
|
||||
|
|
@ -0,0 +1,105 @@
|
|||
model: Qwen2-72B-Instruct
|
||||
label:
|
||||
zh_Hans: Qwen2-72B-Instruct
|
||||
en_US: Qwen2-72B-Instruct
|
||||
model_type: llm
|
||||
features:
|
||||
- agent-thought
|
||||
model_properties:
|
||||
mode: chat
|
||||
context_size: 6400
|
||||
parameter_rules:
|
||||
- name: stream
|
||||
use_template: boolean
|
||||
label:
|
||||
en_US: "Stream"
|
||||
zh_Hans: "流式"
|
||||
type: boolean
|
||||
default: true
|
||||
required: true
|
||||
help:
|
||||
en_US: "Whether to return the results in batches through streaming. If set to true, the generated text will be pushed to the user in real time during the generation process."
|
||||
zh_Hans: "是否通过流式分批返回结果。如果设置为 true，生成过程中实时地向用户推送每一部分生成的文本。"
|
||||
|
||||
- name: max_tokens
|
||||
use_template: max_tokens
|
||||
label:
|
||||
en_US: "Max Tokens"
|
||||
zh_Hans: "最大Token数"
|
||||
type: int
|
||||
default: 512
|
||||
min: 1
|
||||
required: true
|
||||
help:
|
||||
en_US: "The maximum number of tokens that can be generated by the model varies depending on the model."
|
||||
zh_Hans: "模型可生成的最大 token 个数，不同模型上限不同。"
|
||||
|
||||
- name: temperature
|
||||
use_template: temperature
|
||||
label:
|
||||
en_US: "Temperature"
|
||||
zh_Hans: "采样温度"
|
||||
type: float
|
||||
default: 0.7
|
||||
min: 0.0
|
||||
max: 1.0
|
||||
precision: 1
|
||||
required: true
|
||||
help:
|
||||
en_US: "The randomness of the sampling temperature control output. The temperature value is within the range of [0.0, 1.0]. The higher the value, the more random and creative the output; the lower the value, the more stable it is. It is recommended to adjust either top_p or temperature parameters according to your needs to avoid adjusting both at the same time."
|
||||
zh_Hans: "采样温度控制输出的随机性。温度值在 [0.0, 1.0] 范围内，值越高，输出越随机和创造性；值越低，输出越稳定。建议根据需求调整 top_p 或 temperature 参数，避免同时调整两者。"
|
||||
|
||||
- name: top_p
|
||||
use_template: top_p
|
||||
label:
|
||||
en_US: "Top P"
|
||||
zh_Hans: "Top P"
|
||||
type: float
|
||||
default: 0.7
|
||||
min: 0.0
|
||||
max: 1.0
|
||||
precision: 1
|
||||
required: true
|
||||
help:
|
||||
en_US: "The value range of the sampling method is [0.0, 1.0]. The top_p value determines that the model selects tokens from the top p% of candidate words with the highest probability; when top_p is 0, this parameter is invalid. It is recommended to adjust either top_p or temperature parameters according to your needs to avoid adjusting both at the same time."
|
||||
zh_Hans: "采样方法的取值范围为 [0.0,1.0]。top_p 值确定模型从概率最高的前p%的候选词中选取 tokens；当 top_p 为 0 时，此参数无效。建议根据需求调整 top_p 或 temperature 参数，避免同时调整两者。"
|
||||
|
||||
- name: top_k
|
||||
use_template: top_k
|
||||
label:
|
||||
en_US: "Top K"
|
||||
zh_Hans: "Top K"
|
||||
type: int
|
||||
default: 50
|
||||
min: 0
|
||||
max: 100
|
||||
required: true
|
||||
help:
|
||||
en_US: "The value range is [0,100], which limits the model to only select from the top k words with the highest probability when choosing the next word at each step. The larger the value, the more diverse text generation will be."
|
||||
zh_Hans: "取值范围为 [0,100]，限制模型在每一步选择下一个词时，只从概率最高的前 k 个词中选取。数值越大，文本生成越多样。"
|
||||
|
||||
- name: frequency_penalty
|
||||
use_template: frequency_penalty
|
||||
label:
|
||||
en_US: "Frequency Penalty"
|
||||
zh_Hans: "频率惩罚"
|
||||
type: float
|
||||
default: 0
|
||||
min: -1.0
|
||||
max: 1.0
|
||||
precision: 1
|
||||
required: false
|
||||
help:
|
||||
en_US: "Used to adjust the frequency of repeated content in automatically generated text. Positive numbers reduce repetition, while negative numbers increase repetition. After setting this parameter, if a word has already appeared in the text, the model will decrease the probability of choosing that word for subsequent generation."
|
||||
zh_Hans: "用于调整自动生成文本中重复内容的频率。正数减少重复，负数增加重复。设置此参数后，如果一个词在文本中已经出现过，模型在后续生成中选择该词的概率会降低。"
|
||||
|
||||
- name: user
|
||||
use_template: text
|
||||
label:
|
||||
en_US: "User"
|
||||
zh_Hans: "用户"
|
||||
type: string
|
||||
required: false
|
||||
help:
|
||||
en_US: "Used to track and differentiate conversation requests from different users."
|
||||
zh_Hans: "用于追踪和区分不同用户的对话请求。"
|
||||
|
|
@ -0,0 +1,105 @@
|
|||
model: Qwen2-7B-Instruct
|
||||
label:
|
||||
zh_Hans: Qwen2-7B-Instruct
|
||||
en_US: Qwen2-7B-Instruct
|
||||
model_type: llm
|
||||
features:
|
||||
- agent-thought
|
||||
model_properties:
|
||||
mode: chat
|
||||
context_size: 32768
|
||||
parameter_rules:
|
||||
- name: stream
|
||||
use_template: boolean
|
||||
label:
|
||||
en_US: "Stream"
|
||||
zh_Hans: "流式"
|
||||
type: boolean
|
||||
default: true
|
||||
required: true
|
||||
help:
|
||||
en_US: "Whether to return the results in batches through streaming. If set to true, the generated text will be pushed to the user in real time during the generation process."
|
||||
zh_Hans: "是否通过流式分批返回结果。如果设置为 true，生成过程中实时地向用户推送每一部分生成的文本。"
|
||||
|
||||
- name: max_tokens
|
||||
use_template: max_tokens
|
||||
label:
|
||||
en_US: "Max Tokens"
|
||||
zh_Hans: "最大Token数"
|
||||
type: int
|
||||
default: 512
|
||||
min: 1
|
||||
required: true
|
||||
help:
|
||||
en_US: "The maximum number of tokens that can be generated by the model varies depending on the model."
|
||||
zh_Hans: "模型可生成的最大 token 个数，不同模型上限不同。"
|
||||
|
||||
- name: temperature
|
||||
use_template: temperature
|
||||
label:
|
||||
en_US: "Temperature"
|
||||
zh_Hans: "采样温度"
|
||||
type: float
|
||||
default: 0.7
|
||||
min: 0.0
|
||||
max: 1.0
|
||||
precision: 1
|
||||
required: true
|
||||
help:
|
||||
en_US: "The randomness of the sampling temperature control output. The temperature value is within the range of [0.0, 1.0]. The higher the value, the more random and creative the output; the lower the value, the more stable it is. It is recommended to adjust either top_p or temperature parameters according to your needs to avoid adjusting both at the same time."
|
||||
zh_Hans: "采样温度控制输出的随机性。温度值在 [0.0, 1.0] 范围内，值越高，输出越随机和创造性；值越低，输出越稳定。建议根据需求调整 top_p 或 temperature 参数，避免同时调整两者。"
|
||||
|
||||
- name: top_p
|
||||
use_template: top_p
|
||||
label:
|
||||
en_US: "Top P"
|
||||
zh_Hans: "Top P"
|
||||
type: float
|
||||
default: 0.7
|
||||
min: 0.0
|
||||
max: 1.0
|
||||
precision: 1
|
||||
required: true
|
||||
help:
|
||||
en_US: "The value range of the sampling method is [0.0, 1.0]. The top_p value determines that the model selects tokens from the top p% of candidate words with the highest probability; when top_p is 0, this parameter is invalid. It is recommended to adjust either top_p or temperature parameters according to your needs to avoid adjusting both at the same time."
|
||||
zh_Hans: "采样方法的取值范围为 [0.0,1.0]。top_p 值确定模型从概率最高的前p%的候选词中选取 tokens；当 top_p 为 0 时，此参数无效。建议根据需求调整 top_p 或 temperature 参数，避免同时调整两者。"
|
||||
|
||||
- name: top_k
|
||||
use_template: top_k
|
||||
label:
|
||||
en_US: "Top K"
|
||||
zh_Hans: "Top K"
|
||||
type: int
|
||||
default: 50
|
||||
min: 0
|
||||
max: 100
|
||||
required: true
|
||||
help:
|
||||
en_US: "The value range is [0,100], which limits the model to only select from the top k words with the highest probability when choosing the next word at each step. The larger the value, the more diverse text generation will be."
|
||||
zh_Hans: "取值范围为 [0,100]，限制模型在每一步选择下一个词时，只从概率最高的前 k 个词中选取。数值越大，文本生成越多样。"
|
||||
|
||||
- name: frequency_penalty
|
||||
use_template: frequency_penalty
|
||||
label:
|
||||
en_US: "Frequency Penalty"
|
||||
zh_Hans: "频率惩罚"
|
||||
type: float
|
||||
default: 0
|
||||
min: -1.0
|
||||
max: 1.0
|
||||
precision: 1
|
||||
required: false
|
||||
help:
|
||||
en_US: "Used to adjust the frequency of repeated content in automatically generated text. Positive numbers reduce repetition, while negative numbers increase repetition. After setting this parameter, if a word has already appeared in the text, the model will decrease the probability of choosing that word for subsequent generation."
|
||||
zh_Hans: "用于调整自动生成文本中重复内容的频率。正数减少重复，负数增加重复。设置此参数后，如果一个词在文本中已经出现过，模型在后续生成中选择该词的概率会降低。"
|
||||
|
||||
- name: user
|
||||
use_template: text
|
||||
label:
|
||||
en_US: "User"
|
||||
zh_Hans: "用户"
|
||||
type: string
|
||||
required: false
|
||||
help:
|
||||
en_US: "Used to track and differentiate conversation requests from different users."
|
||||
zh_Hans: "用于追踪和区分不同用户的对话请求。"
|
||||
|
|
@ -0,0 +1,105 @@
|
|||
model: Yi-1.5-34B-Chat
|
||||
label:
|
||||
zh_Hans: Yi-1.5-34B-Chat
|
||||
en_US: Yi-1.5-34B-Chat
|
||||
model_type: llm
|
||||
features:
|
||||
- agent-thought
|
||||
model_properties:
|
||||
mode: chat
|
||||
context_size: 4096
|
||||
parameter_rules:
|
||||
- name: stream
|
||||
use_template: boolean
|
||||
label:
|
||||
en_US: "Stream"
|
||||
zh_Hans: "流式"
|
||||
type: boolean
|
||||
default: true
|
||||
required: true
|
||||
help:
|
||||
en_US: "Whether to return the results in batches through streaming. If set to true, the generated text will be pushed to the user in real time during the generation process."
|
||||
zh_Hans: "是否通过流式分批返回结果。如果设置为 true，生成过程中实时地向用户推送每一部分生成的文本。"
|
||||
|
||||
- name: max_tokens
|
||||
use_template: max_tokens
|
||||
label:
|
||||
en_US: "Max Tokens"
|
||||
zh_Hans: "最大Token数"
|
||||
type: int
|
||||
default: 512
|
||||
min: 1
|
||||
required: true
|
||||
help:
|
||||
en_US: "The maximum number of tokens that can be generated by the model varies depending on the model."
|
||||
zh_Hans: "模型可生成的最大 token 个数，不同模型上限不同。"
|
||||
|
||||
- name: temperature
|
||||
use_template: temperature
|
||||
label:
|
||||
en_US: "Temperature"
|
||||
zh_Hans: "采样温度"
|
||||
type: float
|
||||
default: 0.7
|
||||
min: 0.0
|
||||
max: 1.0
|
||||
precision: 1
|
||||
required: true
|
||||
help:
|
||||
en_US: "The randomness of the sampling temperature control output. The temperature value is within the range of [0.0, 1.0]. The higher the value, the more random and creative the output; the lower the value, the more stable it is. It is recommended to adjust either top_p or temperature parameters according to your needs to avoid adjusting both at the same time."
|
||||
zh_Hans: "采样温度控制输出的随机性。温度值在 [0.0, 1.0] 范围内，值越高，输出越随机和创造性；值越低，输出越稳定。建议根据需求调整 top_p 或 temperature 参数，避免同时调整两者。"
|
||||
|
||||
- name: top_p
|
||||
use_template: top_p
|
||||
label:
|
||||
en_US: "Top P"
|
||||
zh_Hans: "Top P"
|
||||
type: float
|
||||
default: 0.7
|
||||
min: 0.0
|
||||
max: 1.0
|
||||
precision: 1
|
||||
required: true
|
||||
help:
|
||||
en_US: "The value range of the sampling method is [0.0, 1.0]. The top_p value determines that the model selects tokens from the top p% of candidate words with the highest probability; when top_p is 0, this parameter is invalid. It is recommended to adjust either top_p or temperature parameters according to your needs to avoid adjusting both at the same time."
|
||||
zh_Hans: "采样方法的取值范围为 [0.0,1.0]。top_p 值确定模型从概率最高的前p%的候选词中选取 tokens；当 top_p 为 0 时，此参数无效。建议根据需求调整 top_p 或 temperature 参数，避免同时调整两者。"
|
||||
|
||||
- name: top_k
|
||||
use_template: top_k
|
||||
label:
|
||||
en_US: "Top K"
|
||||
zh_Hans: "Top K"
|
||||
type: int
|
||||
default: 50
|
||||
min: 0
|
||||
max: 100
|
||||
required: true
|
||||
help:
|
||||
en_US: "The value range is [0,100], which limits the model to only select from the top k words with the highest probability when choosing the next word at each step. The larger the value, the more diverse text generation will be."
|
||||
zh_Hans: "取值范围为 [0,100]，限制模型在每一步选择下一个词时，只从概率最高的前 k 个词中选取。数值越大，文本生成越多样。"
|
||||
|
||||
- name: frequency_penalty
|
||||
use_template: frequency_penalty
|
||||
label:
|
||||
en_US: "Frequency Penalty"
|
||||
zh_Hans: "频率惩罚"
|
||||
type: float
|
||||
default: 0
|
||||
min: -1.0
|
||||
max: 1.0
|
||||
precision: 1
|
||||
required: false
|
||||
help:
|
||||
en_US: "Used to adjust the frequency of repeated content in automatically generated text. Positive numbers reduce repetition, while negative numbers increase repetition. After setting this parameter, if a word has already appeared in the text, the model will decrease the probability of choosing that word for subsequent generation."
|
||||
zh_Hans: "用于调整自动生成文本中重复内容的频率。正数减少重复，负数增加重复。设置此参数后，如果一个词在文本中已经出现过，模型在后续生成中选择该词的概率会降低。"
|
||||
|
||||
- name: user
|
||||
use_template: text
|
||||
label:
|
||||
en_US: "User"
|
||||
zh_Hans: "用户"
|
||||
type: string
|
||||
required: false
|
||||
help:
|
||||
en_US: "Used to track and differentiate conversation requests from different users."
|
||||
zh_Hans: "用于追踪和区分不同用户的对话请求。"
|
||||
|
|
@ -0,0 +1,7 @@
|
|||
- Qwen2-7B-Instruct
|
||||
- Qwen2-72B-Instruct
|
||||
- Yi-1.5-34B-Chat
|
||||
- glm-4-9b-chat
|
||||
- deepseek-coder-33B-instruct-chat
|
||||
- deepseek-coder-33B-instruct-completions
|
||||
- codegeex4-all-9b
|
||||
|
|
@ -0,0 +1,105 @@
|
|||
model: codegeex4-all-9b
|
||||
label:
|
||||
zh_Hans: codegeex4-all-9b
|
||||
en_US: codegeex4-all-9b
|
||||
model_type: llm
|
||||
features:
|
||||
- agent-thought
|
||||
model_properties:
|
||||
mode: chat
|
||||
context_size: 40960
|
||||
parameter_rules:
|
||||
- name: stream
|
||||
use_template: boolean
|
||||
label:
|
||||
en_US: "Stream"
|
||||
zh_Hans: "流式"
|
||||
type: boolean
|
||||
default: true
|
||||
required: true
|
||||
help:
|
||||
en_US: "Whether to return the results in batches through streaming. If set to true, the generated text will be pushed to the user in real time during the generation process."
|
||||
zh_Hans: "是否通过流式分批返回结果。如果设置为 true，生成过程中实时地向用户推送每一部分生成的文本。"
|
||||
|
||||
- name: max_tokens
|
||||
use_template: max_tokens
|
||||
label:
|
||||
en_US: "Max Tokens"
|
||||
zh_Hans: "最大Token数"
|
||||
type: int
|
||||
default: 512
|
||||
min: 1
|
||||
required: true
|
||||
help:
|
||||
en_US: "The maximum number of tokens that can be generated by the model varies depending on the model."
|
||||
zh_Hans: "模型可生成的最大 token 个数，不同模型上限不同。"
|
||||
|
||||
- name: temperature
|
||||
use_template: temperature
|
||||
label:
|
||||
en_US: "Temperature"
|
||||
zh_Hans: "采样温度"
|
||||
type: float
|
||||
default: 0.7
|
||||
min: 0.0
|
||||
max: 1.0
|
||||
precision: 1
|
||||
required: true
|
||||
help:
|
||||
en_US: "The randomness of the sampling temperature control output. The temperature value is within the range of [0.0, 1.0]. The higher the value, the more random and creative the output; the lower the value, the more stable it is. It is recommended to adjust either top_p or temperature parameters according to your needs to avoid adjusting both at the same time."
|
||||
zh_Hans: "采样温度控制输出的随机性。温度值在 [0.0, 1.0] 范围内，值越高，输出越随机和创造性；值越低，输出越稳定。建议根据需求调整 top_p 或 temperature 参数，避免同时调整两者。"
|
||||
|
||||
- name: top_p
|
||||
use_template: top_p
|
||||
label:
|
||||
en_US: "Top P"
|
||||
zh_Hans: "Top P"
|
||||
type: float
|
||||
default: 0.7
|
||||
min: 0.0
|
||||
max: 1.0
|
||||
precision: 1
|
||||
required: true
|
||||
help:
|
||||
en_US: "The value range of the sampling method is [0.0, 1.0]. The top_p value determines that the model selects tokens from the top p% of candidate words with the highest probability; when top_p is 0, this parameter is invalid. It is recommended to adjust either top_p or temperature parameters according to your needs to avoid adjusting both at the same time."
|
||||
zh_Hans: "采样方法的取值范围为 [0.0,1.0]。top_p 值确定模型从概率最高的前p%的候选词中选取 tokens；当 top_p 为 0 时，此参数无效。建议根据需求调整 top_p 或 temperature 参数，避免同时调整两者。"
|
||||
|
||||
- name: top_k
|
||||
use_template: top_k
|
||||
label:
|
||||
en_US: "Top K"
|
||||
zh_Hans: "Top K"
|
||||
type: int
|
||||
default: 50
|
||||
min: 0
|
||||
max: 100
|
||||
required: true
|
||||
help:
|
||||
en_US: "The value range is [0,100], which limits the model to only select from the top k words with the highest probability when choosing the next word at each step. The larger the value, the more diverse text generation will be."
|
||||
zh_Hans: "取值范围为 [0,100]，限制模型在每一步选择下一个词时，只从概率最高的前 k 个词中选取。数值越大，文本生成越多样。"
|
||||
|
||||
- name: frequency_penalty
|
||||
use_template: frequency_penalty
|
||||
label:
|
||||
en_US: "Frequency Penalty"
|
||||
zh_Hans: "频率惩罚"
|
||||
type: float
|
||||
default: 0
|
||||
min: -1.0
|
||||
max: 1.0
|
||||
precision: 1
|
||||
required: false
|
||||
help:
|
||||
en_US: "Used to adjust the frequency of repeated content in automatically generated text. Positive numbers reduce repetition, while negative numbers increase repetition. After setting this parameter, if a word has already appeared in the text, the model will decrease the probability of choosing that word for subsequent generation."
|
||||
zh_Hans: "用于调整自动生成文本中重复内容的频率。正数减少重复，负数增加重复。设置此参数后，如果一个词在文本中已经出现过，模型在后续生成中选择该词的概率会降低。"
|
||||
|
||||
- name: user
|
||||
use_template: text
|
||||
label:
|
||||
en_US: "User"
|
||||
zh_Hans: "用户"
|
||||
type: string
|
||||
required: false
|
||||
help:
|
||||
en_US: "Used to track and differentiate conversation requests from different users."
|
||||
zh_Hans: "用于追踪和区分不同用户的对话请求。"
|
||||
|
|
@ -0,0 +1,105 @@
|
|||
model: deepseek-coder-33B-instruct-chat
|
||||
label:
|
||||
zh_Hans: deepseek-coder-33B-instruct-chat
|
||||
en_US: deepseek-coder-33B-instruct-chat
|
||||
model_type: llm
|
||||
features:
|
||||
- agent-thought
|
||||
model_properties:
|
||||
mode: chat
|
||||
context_size: 9000
|
||||
parameter_rules:
|
||||
- name: stream
|
||||
use_template: boolean
|
||||
label:
|
||||
en_US: "Stream"
|
||||
zh_Hans: "流式"
|
||||
type: boolean
|
||||
default: true
|
||||
required: true
|
||||
help:
|
||||
en_US: "Whether to return the results in batches through streaming. If set to true, the generated text will be pushed to the user in real time during the generation process."
|
||||
zh_Hans: "是否通过流式分批返回结果。如果设置为 true，生成过程中实时地向用户推送每一部分生成的文本。"
|
||||
|
||||
- name: max_tokens
|
||||
use_template: max_tokens
|
||||
label:
|
||||
en_US: "Max Tokens"
|
||||
zh_Hans: "最大Token数"
|
||||
type: int
|
||||
default: 512
|
||||
min: 1
|
||||
required: true
|
||||
help:
|
||||
en_US: "The maximum number of tokens that can be generated by the model varies depending on the model."
|
||||
zh_Hans: "模型可生成的最大 token 个数，不同模型上限不同。"
|
||||
|
||||
- name: temperature
|
||||
use_template: temperature
|
||||
label:
|
||||
en_US: "Temperature"
|
||||
zh_Hans: "采样温度"
|
||||
type: float
|
||||
default: 0.7
|
||||
min: 0.0
|
||||
max: 1.0
|
||||
precision: 1
|
||||
required: true
|
||||
help:
|
||||
en_US: "The randomness of the sampling temperature control output. The temperature value is within the range of [0.0, 1.0]. The higher the value, the more random and creative the output; the lower the value, the more stable it is. It is recommended to adjust either top_p or temperature parameters according to your needs to avoid adjusting both at the same time."
|
||||
zh_Hans: "采样温度控制输出的随机性。温度值在 [0.0, 1.0] 范围内，值越高，输出越随机和创造性；值越低，输出越稳定。建议根据需求调整 top_p 或 temperature 参数，避免同时调整两者。"
|
||||
|
||||
- name: top_p
|
||||
use_template: top_p
|
||||
label:
|
||||
en_US: "Top P"
|
||||
zh_Hans: "Top P"
|
||||
type: float
|
||||
default: 0.7
|
||||
min: 0.0
|
||||
max: 1.0
|
||||
precision: 1
|
||||
required: true
|
||||
help:
|
||||
en_US: "The value range of the sampling method is [0.0, 1.0]. The top_p value determines that the model selects tokens from the top p% of candidate words with the highest probability; when top_p is 0, this parameter is invalid. It is recommended to adjust either top_p or temperature parameters according to your needs to avoid adjusting both at the same time."
|
||||
zh_Hans: "采样方法的取值范围为 [0.0,1.0]。top_p 值确定模型从概率最高的前p%的候选词中选取 tokens；当 top_p 为 0 时，此参数无效。建议根据需求调整 top_p 或 temperature 参数，避免同时调整两者。"
|
||||
|
||||
- name: top_k
|
||||
use_template: top_k
|
||||
label:
|
||||
en_US: "Top K"
|
||||
zh_Hans: "Top K"
|
||||
type: int
|
||||
default: 50
|
||||
min: 0
|
||||
max: 100
|
||||
required: true
|
||||
help:
|
||||
en_US: "The value range is [0,100], which limits the model to only select from the top k words with the highest probability when choosing the next word at each step. The larger the value, the more diverse text generation will be."
|
||||
zh_Hans: "取值范围为 [0,100]，限制模型在每一步选择下一个词时，只从概率最高的前 k 个词中选取。数值越大，文本生成越多样。"
|
||||
|
||||
- name: frequency_penalty
|
||||
use_template: frequency_penalty
|
||||
label:
|
||||
en_US: "Frequency Penalty"
|
||||
zh_Hans: "频率惩罚"
|
||||
type: float
|
||||
default: 0
|
||||
min: -1.0
|
||||
max: 1.0
|
||||
precision: 1
|
||||
required: false
|
||||
help:
|
||||
en_US: "Used to adjust the frequency of repeated content in automatically generated text. Positive numbers reduce repetition, while negative numbers increase repetition. After setting this parameter, if a word has already appeared in the text, the model will decrease the probability of choosing that word for subsequent generation."
|
||||
zh_Hans: "用于调整自动生成文本中重复内容的频率。正数减少重复，负数增加重复。设置此参数后，如果一个词在文本中已经出现过，模型在后续生成中选择该词的概率会降低。"
|
||||
|
||||
- name: user
|
||||
use_template: text
|
||||
label:
|
||||
en_US: "User"
|
||||
zh_Hans: "用户"
|
||||
type: string
|
||||
required: false
|
||||
help:
|
||||
en_US: "Used to track and differentiate conversation requests from different users."
|
||||
zh_Hans: "用于追踪和区分不同用户的对话请求。"
|
||||
|
|
@ -0,0 +1,91 @@
|
|||
model: deepseek-coder-33B-instruct-completions
|
||||
label:
|
||||
zh_Hans: deepseek-coder-33B-instruct-completions
|
||||
en_US: deepseek-coder-33B-instruct-completions
|
||||
model_type: llm
|
||||
features:
|
||||
- agent-thought
|
||||
model_properties:
|
||||
mode: completion
|
||||
context_size: 9000
|
||||
parameter_rules:
|
||||
- name: stream
|
||||
use_template: boolean
|
||||
label:
|
||||
en_US: "Stream"
|
||||
zh_Hans: "流式"
|
||||
type: boolean
|
||||
default: true
|
||||
required: true
|
||||
help:
|
||||
en_US: "Whether to return the results in batches through streaming. If set to true, the generated text will be pushed to the user in real time during the generation process."
|
||||
zh_Hans: "是否通过流式分批返回结果。如果设置为 true，生成过程中实时地向用户推送每一部分生成的文本。"
|
||||
|
||||
- name: max_tokens
|
||||
use_template: max_tokens
|
||||
label:
|
||||
en_US: "Max Tokens"
|
||||
zh_Hans: "最大Token数"
|
||||
type: int
|
||||
default: 512
|
||||
min: 1
|
||||
required: true
|
||||
help:
|
||||
en_US: "The maximum number of tokens that can be generated by the model varies depending on the model."
|
||||
zh_Hans: "模型可生成的最大 token 个数，不同模型上限不同。"
|
||||
|
||||
- name: temperature
|
||||
use_template: temperature
|
||||
label:
|
||||
en_US: "Temperature"
|
||||
zh_Hans: "采样温度"
|
||||
type: float
|
||||
default: 0.7
|
||||
min: 0.0
|
||||
max: 1.0
|
||||
precision: 1
|
||||
required: true
|
||||
help:
|
||||
en_US: "The randomness of the sampling temperature control output. The temperature value is within the range of [0.0, 1.0]. The higher the value, the more random and creative the output; the lower the value, the more stable it is. It is recommended to adjust either top_p or temperature parameters according to your needs to avoid adjusting both at the same time."
|
||||
zh_Hans: "采样温度控制输出的随机性。温度值在 [0.0, 1.0] 范围内，值越高，输出越随机和创造性；值越低，输出越稳定。建议根据需求调整 top_p 或 temperature 参数，避免同时调整两者。"
|
||||
|
||||
- name: top_p
|
||||
use_template: top_p
|
||||
label:
|
||||
en_US: "Top P"
|
||||
zh_Hans: "Top P"
|
||||
type: float
|
||||
default: 0.7
|
||||
min: 0.0
|
||||
max: 1.0
|
||||
precision: 1
|
||||
required: true
|
||||
help:
|
||||
en_US: "The value range of the sampling method is [0.0, 1.0]. The top_p value determines that the model selects tokens from the top p% of candidate words with the highest probability; when top_p is 0, this parameter is invalid. It is recommended to adjust either top_p or temperature parameters according to your needs to avoid adjusting both at the same time."
|
||||
zh_Hans: "采样方法的取值范围为 [0.0,1.0]。top_p 值确定模型从概率最高的前p%的候选词中选取 tokens；当 top_p 为 0 时，此参数无效。建议根据需求调整 top_p 或 temperature 参数，避免同时调整两者。"
|
||||
|
||||
- name: frequency_penalty
|
||||
use_template: frequency_penalty
|
||||
label:
|
||||
en_US: "Frequency Penalty"
|
||||
zh_Hans: "频率惩罚"
|
||||
type: float
|
||||
default: 0
|
||||
min: -1.0
|
||||
max: 1.0
|
||||
precision: 1
|
||||
required: false
|
||||
help:
|
||||
en_US: "Used to adjust the frequency of repeated content in automatically generated text. Positive numbers reduce repetition, while negative numbers increase repetition. After setting this parameter, if a word has already appeared in the text, the model will decrease the probability of choosing that word for subsequent generation."
|
||||
zh_Hans: "用于调整自动生成文本中重复内容的频率。正数减少重复，负数增加重复。设置此参数后，如果一个词在文本中已经出现过，模型在后续生成中选择该词的概率会降低。"
|
||||
|
||||
- name: user
|
||||
use_template: text
|
||||
label:
|
||||
en_US: "User"
|
||||
zh_Hans: "用户"
|
||||
type: string
|
||||
required: false
|
||||
help:
|
||||
en_US: "Used to track and differentiate conversation requests from different users."
|
||||
zh_Hans: "用于追踪和区分不同用户的对话请求。"
|
||||
|
|
@ -0,0 +1,105 @@
|
|||
model: glm-4-9b-chat
|
||||
label:
|
||||
zh_Hans: glm-4-9b-chat
|
||||
en_US: glm-4-9b-chat
|
||||
model_type: llm
|
||||
features:
|
||||
- agent-thought
|
||||
model_properties:
|
||||
mode: chat
|
||||
context_size: 32768
|
||||
parameter_rules:
|
||||
- name: stream
|
||||
use_template: boolean
|
||||
label:
|
||||
en_US: "Stream"
|
||||
zh_Hans: "流式"
|
||||
type: boolean
|
||||
default: true
|
||||
required: true
|
||||
help:
|
||||
en_US: "Whether to return the results in batches through streaming. If set to true, the generated text will be pushed to the user in real time during the generation process."
|
||||
zh_Hans: "是否通过流式分批返回结果。如果设置为 true，生成过程中实时地向用户推送每一部分生成的文本。"
|
||||
|
||||
- name: max_tokens
|
||||
use_template: max_tokens
|
||||
label:
|
||||
en_US: "Max Tokens"
|
||||
zh_Hans: "最大Token数"
|
||||
type: int
|
||||
default: 512
|
||||
min: 1
|
||||
required: true
|
||||
help:
|
||||
en_US: "The maximum number of tokens that can be generated by the model varies depending on the model."
|
||||
zh_Hans: "模型可生成的最大 token 个数，不同模型上限不同。"
|
||||
|
||||
- name: temperature
|
||||
use_template: temperature
|
||||
label:
|
||||
en_US: "Temperature"
|
||||
zh_Hans: "采样温度"
|
||||
type: float
|
||||
default: 0.7
|
||||
min: 0.0
|
||||
max: 1.0
|
||||
precision: 1
|
||||
required: true
|
||||
help:
|
||||
en_US: "The randomness of the sampling temperature control output. The temperature value is within the range of [0.0, 1.0]. The higher the value, the more random and creative the output; the lower the value, the more stable it is. It is recommended to adjust either top_p or temperature parameters according to your needs to avoid adjusting both at the same time."
|
||||
zh_Hans: "采样温度控制输出的随机性。温度值在 [0.0, 1.0] 范围内，值越高，输出越随机和创造性；值越低，输出越稳定。建议根据需求调整 top_p 或 temperature 参数，避免同时调整两者。"
|
||||
|
||||
- name: top_p
|
||||
use_template: top_p
|
||||
label:
|
||||
en_US: "Top P"
|
||||
zh_Hans: "Top P"
|
||||
type: float
|
||||
default: 0.7
|
||||
min: 0.0
|
||||
max: 1.0
|
||||
precision: 1
|
||||
required: true
|
||||
help:
|
||||
en_US: "The value range of the sampling method is [0.0, 1.0]. The top_p value determines that the model selects tokens from the top p% of candidate words with the highest probability; when top_p is 0, this parameter is invalid. It is recommended to adjust either top_p or temperature parameters according to your needs to avoid adjusting both at the same time."
|
||||
zh_Hans: "采样方法的取值范围为 [0.0,1.0]。top_p 值确定模型从概率最高的前p%的候选词中选取 tokens；当 top_p 为 0 时，此参数无效。建议根据需求调整 top_p 或 temperature 参数，避免同时调整两者。"
|
||||
|
||||
- name: top_k
|
||||
use_template: top_k
|
||||
label:
|
||||
en_US: "Top K"
|
||||
zh_Hans: "Top K"
|
||||
type: int
|
||||
default: 50
|
||||
min: 0
|
||||
max: 100
|
||||
required: true
|
||||
help:
|
||||
en_US: "The value range is [0,100], which limits the model to only select from the top k words with the highest probability when choosing the next word at each step. The larger the value, the more diverse text generation will be."
|
||||
zh_Hans: "取值范围为 [0,100]，限制模型在每一步选择下一个词时，只从概率最高的前 k 个词中选取。数值越大，文本生成越多样。"
|
||||
|
||||
- name: frequency_penalty
|
||||
use_template: frequency_penalty
|
||||
label:
|
||||
en_US: "Frequency Penalty"
|
||||
zh_Hans: "频率惩罚"
|
||||
type: float
|
||||
default: 0
|
||||
min: -1.0
|
||||
max: 1.0
|
||||
precision: 1
|
||||
required: false
|
||||
help:
|
||||
en_US: "Used to adjust the frequency of repeated content in automatically generated text. Positive numbers reduce repetition, while negative numbers increase repetition. After setting this parameter, if a word has already appeared in the text, the model will decrease the probability of choosing that word for subsequent generation."
|
||||
zh_Hans: "用于调整自动生成文本中重复内容的频率。正数减少重复，负数增加重复。设置此参数后，如果一个词在文本中已经出现过，模型在后续生成中选择该词的概率会降低。"
|
||||
|
||||
- name: user
|
||||
use_template: text
|
||||
label:
|
||||
en_US: "User"
|
||||
zh_Hans: "用户"
|
||||
type: string
|
||||
required: false
|
||||
help:
|
||||
en_US: "Used to track and differentiate conversation requests from different users."
|
||||
zh_Hans: "用于追踪和区分不同用户的对话请求。"
|
||||
47
api/core/model_runtime/model_providers/gitee_ai/llm/llm.py
Normal file
47
api/core/model_runtime/model_providers/gitee_ai/llm/llm.py
Normal file
|
|
@ -0,0 +1,47 @@
|
|||
from collections.abc import Generator
|
||||
from typing import Optional, Union
|
||||
|
||||
from core.model_runtime.entities.llm_entities import LLMMode, LLMResult
|
||||
from core.model_runtime.entities.message_entities import (
|
||||
PromptMessage,
|
||||
PromptMessageTool,
|
||||
)
|
||||
from core.model_runtime.model_providers.openai_api_compatible.llm.llm import OAIAPICompatLargeLanguageModel
|
||||
|
||||
|
||||
class GiteeAILargeLanguageModel(OAIAPICompatLargeLanguageModel):
|
||||
MODEL_TO_IDENTITY: dict[str, str] = {
|
||||
"Yi-1.5-34B-Chat": "Yi-34B-Chat",
|
||||
"deepseek-coder-33B-instruct-completions": "deepseek-coder-33B-instruct",
|
||||
"deepseek-coder-33B-instruct-chat": "deepseek-coder-33B-instruct",
|
||||
}
|
||||
|
||||
def _invoke(
|
||||
self,
|
||||
model: str,
|
||||
credentials: dict,
|
||||
prompt_messages: list[PromptMessage],
|
||||
model_parameters: dict,
|
||||
tools: Optional[list[PromptMessageTool]] = None,
|
||||
stop: Optional[list[str]] = None,
|
||||
stream: bool = True,
|
||||
user: Optional[str] = None,
|
||||
) -> Union[LLMResult, Generator]:
|
||||
self._add_custom_parameters(credentials, model, model_parameters)
|
||||
return super()._invoke(model, credentials, prompt_messages, model_parameters, tools, stop, stream)
|
||||
|
||||
def validate_credentials(self, model: str, credentials: dict) -> None:
|
||||
self._add_custom_parameters(credentials, model, None)
|
||||
super().validate_credentials(model, credentials)
|
||||
|
||||
@staticmethod
|
||||
def _add_custom_parameters(credentials: dict, model: str, model_parameters: dict) -> None:
|
||||
if model is None:
|
||||
model = "bge-large-zh-v1.5"
|
||||
|
||||
model_identity = GiteeAILargeLanguageModel.MODEL_TO_IDENTITY.get(model, model)
|
||||
credentials["endpoint_url"] = f"https://ai.gitee.com/api/serverless/{model_identity}/"
|
||||
if model.endswith("completions"):
|
||||
credentials["mode"] = LLMMode.COMPLETION.value
|
||||
else:
|
||||
credentials["mode"] = LLMMode.CHAT.value
|
||||
|
|
@ -0,0 +1 @@
|
|||
- bge-reranker-v2-m3
|
||||
|
|
@ -0,0 +1,4 @@
|
|||
model: bge-reranker-v2-m3
|
||||
model_type: rerank
|
||||
model_properties:
|
||||
context_size: 1024
|
||||
128
api/core/model_runtime/model_providers/gitee_ai/rerank/rerank.py
Normal file
128
api/core/model_runtime/model_providers/gitee_ai/rerank/rerank.py
Normal file
|
|
@ -0,0 +1,128 @@
|
|||
from typing import Optional
|
||||
|
||||
import httpx
|
||||
|
||||
from core.model_runtime.entities.common_entities import I18nObject
|
||||
from core.model_runtime.entities.model_entities import AIModelEntity, FetchFrom, ModelPropertyKey, ModelType
|
||||
from core.model_runtime.entities.rerank_entities import RerankDocument, RerankResult
|
||||
from core.model_runtime.errors.invoke import (
|
||||
InvokeAuthorizationError,
|
||||
InvokeBadRequestError,
|
||||
InvokeConnectionError,
|
||||
InvokeError,
|
||||
InvokeRateLimitError,
|
||||
InvokeServerUnavailableError,
|
||||
)
|
||||
from core.model_runtime.errors.validate import CredentialsValidateFailedError
|
||||
from core.model_runtime.model_providers.__base.rerank_model import RerankModel
|
||||
|
||||
|
||||
class GiteeAIRerankModel(RerankModel):
|
||||
"""
|
||||
Model class for rerank model.
|
||||
"""
|
||||
|
||||
def _invoke(
|
||||
self,
|
||||
model: str,
|
||||
credentials: dict,
|
||||
query: str,
|
||||
docs: list[str],
|
||||
score_threshold: Optional[float] = None,
|
||||
top_n: Optional[int] = None,
|
||||
user: Optional[str] = None,
|
||||
) -> RerankResult:
|
||||
"""
|
||||
Invoke rerank model
|
||||
|
||||
:param model: model name
|
||||
:param credentials: model credentials
|
||||
:param query: search query
|
||||
:param docs: docs for reranking
|
||||
:param score_threshold: score threshold
|
||||
:param top_n: top n documents to return
|
||||
:param user: unique user id
|
||||
:return: rerank result
|
||||
"""
|
||||
if len(docs) == 0:
|
||||
return RerankResult(model=model, docs=[])
|
||||
|
||||
base_url = credentials.get("base_url", "https://ai.gitee.com/api/serverless")
|
||||
base_url = base_url.removesuffix("/")
|
||||
|
||||
try:
|
||||
body = {"model": model, "query": query, "documents": docs}
|
||||
if top_n is not None:
|
||||
body["top_n"] = top_n
|
||||
response = httpx.post(
|
||||
f"{base_url}/{model}/rerank",
|
||||
json=body,
|
||||
headers={"Authorization": f"Bearer {credentials.get('api_key')}"},
|
||||
)
|
||||
|
||||
response.raise_for_status()
|
||||
results = response.json()
|
||||
|
||||
rerank_documents = []
|
||||
for result in results["results"]:
|
||||
rerank_document = RerankDocument(
|
||||
index=result["index"],
|
||||
text=result["document"]["text"],
|
||||
score=result["relevance_score"],
|
||||
)
|
||||
if score_threshold is None or result["relevance_score"] >= score_threshold:
|
||||
rerank_documents.append(rerank_document)
|
||||
return RerankResult(model=model, docs=rerank_documents)
|
||||
except httpx.HTTPStatusError as e:
|
||||
raise InvokeServerUnavailableError(str(e))
|
||||
|
||||
def validate_credentials(self, model: str, credentials: dict) -> None:
|
||||
"""
|
||||
Validate model credentials
|
||||
|
||||
:param model: model name
|
||||
:param credentials: model credentials
|
||||
:return:
|
||||
"""
|
||||
try:
|
||||
self._invoke(
|
||||
model=model,
|
||||
credentials=credentials,
|
||||
query="What is the capital of the United States?",
|
||||
docs=[
|
||||
"Carson City is the capital city of the American state of Nevada. At the 2010 United States "
|
||||
"Census, Carson City had a population of 55,274.",
|
||||
"The Commonwealth of the Northern Mariana Islands is a group of islands in the Pacific Ocean that "
|
||||
"are a political division controlled by the United States. Its capital is Saipan.",
|
||||
],
|
||||
score_threshold=0.01,
|
||||
)
|
||||
except Exception as ex:
|
||||
raise CredentialsValidateFailedError(str(ex))
|
||||
|
||||
@property
|
||||
def _invoke_error_mapping(self) -> dict[type[InvokeError], list[type[Exception]]]:
|
||||
"""
|
||||
Map model invoke error to unified error
|
||||
"""
|
||||
return {
|
||||
InvokeConnectionError: [httpx.ConnectError],
|
||||
InvokeServerUnavailableError: [httpx.RemoteProtocolError],
|
||||
InvokeRateLimitError: [],
|
||||
InvokeAuthorizationError: [httpx.HTTPStatusError],
|
||||
InvokeBadRequestError: [httpx.RequestError],
|
||||
}
|
||||
|
||||
def get_customizable_model_schema(self, model: str, credentials: dict) -> AIModelEntity:
|
||||
"""
|
||||
generate custom model entities from credentials
|
||||
"""
|
||||
entity = AIModelEntity(
|
||||
model=model,
|
||||
label=I18nObject(en_US=model),
|
||||
model_type=ModelType.RERANK,
|
||||
fetch_from=FetchFrom.CUSTOMIZABLE_MODEL,
|
||||
model_properties={ModelPropertyKey.CONTEXT_SIZE: int(credentials.get("context_size"))},
|
||||
)
|
||||
|
||||
return entity
|
||||
|
|
@ -0,0 +1,2 @@
|
|||
- whisper-base
|
||||
- whisper-large
|
||||
|
|
@ -0,0 +1,53 @@
|
|||
import os
|
||||
from typing import IO, Optional
|
||||
|
||||
import requests
|
||||
|
||||
from core.model_runtime.errors.invoke import InvokeBadRequestError
|
||||
from core.model_runtime.errors.validate import CredentialsValidateFailedError
|
||||
from core.model_runtime.model_providers.__base.speech2text_model import Speech2TextModel
|
||||
from core.model_runtime.model_providers.gitee_ai._common import _CommonGiteeAI
|
||||
|
||||
|
||||
class GiteeAISpeech2TextModel(_CommonGiteeAI, Speech2TextModel):
|
||||
"""
|
||||
Model class for OpenAI Compatible Speech to text model.
|
||||
"""
|
||||
|
||||
def _invoke(self, model: str, credentials: dict, file: IO[bytes], user: Optional[str] = None) -> str:
|
||||
"""
|
||||
Invoke speech2text model
|
||||
|
||||
:param model: model name
|
||||
:param credentials: model credentials
|
||||
:param file: audio file
|
||||
:param user: unique user id
|
||||
:return: text for given audio file
|
||||
"""
|
||||
# doc: https://ai.gitee.com/docs/openapi/serverless#tag/serverless/POST/{service}/speech-to-text
|
||||
|
||||
endpoint_url = f"https://ai.gitee.com/api/serverless/{model}/speech-to-text"
|
||||
files = [("file", file)]
|
||||
_, file_ext = os.path.splitext(file.name)
|
||||
headers = {"Content-Type": f"audio/{file_ext}", "Authorization": f"Bearer {credentials.get('api_key')}"}
|
||||
response = requests.post(endpoint_url, headers=headers, files=files)
|
||||
if response.status_code != 200:
|
||||
raise InvokeBadRequestError(response.text)
|
||||
response_data = response.json()
|
||||
return response_data["text"]
|
||||
|
||||
def validate_credentials(self, model: str, credentials: dict) -> None:
|
||||
"""
|
||||
Validate model credentials
|
||||
|
||||
:param model: model name
|
||||
:param credentials: model credentials
|
||||
:return:
|
||||
"""
|
||||
try:
|
||||
audio_file_path = self._get_demo_file_path()
|
||||
|
||||
with open(audio_file_path, "rb") as audio_file:
|
||||
self._invoke(model, credentials, audio_file)
|
||||
except Exception as ex:
|
||||
raise CredentialsValidateFailedError(str(ex))
|
||||
|
|
@ -0,0 +1,5 @@
|
|||
model: whisper-base
|
||||
model_type: speech2text
|
||||
model_properties:
|
||||
file_upload_limit: 1
|
||||
supported_file_extensions: flac,mp3,mp4,mpeg,mpga,m4a,ogg,wav,webm
|
||||
|
|
@ -0,0 +1,5 @@
|
|||
model: whisper-large
|
||||
model_type: speech2text
|
||||
model_properties:
|
||||
file_upload_limit: 1
|
||||
supported_file_extensions: flac,mp3,mp4,mpeg,mpga,m4a,ogg,wav,webm
|
||||
|
|
@ -0,0 +1,3 @@
|
|||
- bge-large-zh-v1.5
|
||||
- bge-small-zh-v1.5
|
||||
- bge-m3
|
||||
|
|
@ -0,0 +1,8 @@
|
|||
model: bge-large-zh-v1.5
|
||||
label:
|
||||
zh_Hans: bge-large-zh-v1.5
|
||||
en_US: bge-large-zh-v1.5
|
||||
model_type: text-embedding
|
||||
model_properties:
|
||||
context_size: 200000
|
||||
max_chunks: 20
|
||||
|
|
@ -0,0 +1,8 @@
|
|||
model: bge-m3
|
||||
label:
|
||||
zh_Hans: bge-m3
|
||||
en_US: bge-m3
|
||||
model_type: text-embedding
|
||||
model_properties:
|
||||
context_size: 200000
|
||||
max_chunks: 20
|
||||
|
|
@ -0,0 +1,8 @@
|
|||
model: bge-small-zh-v1.5
|
||||
label:
|
||||
zh_Hans: bge-small-zh-v1.5
|
||||
en_US: bge-small-zh-v1.5
|
||||
model_type: text-embedding
|
||||
model_properties:
|
||||
context_size: 200000
|
||||
max_chunks: 20
|
||||
|
|
@ -0,0 +1,31 @@
|
|||
from typing import Optional
|
||||
|
||||
from core.entities.embedding_type import EmbeddingInputType
|
||||
from core.model_runtime.entities.text_embedding_entities import TextEmbeddingResult
|
||||
from core.model_runtime.model_providers.openai_api_compatible.text_embedding.text_embedding import (
|
||||
OAICompatEmbeddingModel,
|
||||
)
|
||||
|
||||
|
||||
class GiteeAIEmbeddingModel(OAICompatEmbeddingModel):
|
||||
def _invoke(
|
||||
self,
|
||||
model: str,
|
||||
credentials: dict,
|
||||
texts: list[str],
|
||||
user: Optional[str] = None,
|
||||
input_type: EmbeddingInputType = EmbeddingInputType.DOCUMENT,
|
||||
) -> TextEmbeddingResult:
|
||||
self._add_custom_parameters(credentials, model)
|
||||
return super()._invoke(model, credentials, texts, user, input_type)
|
||||
|
||||
def validate_credentials(self, model: str, credentials: dict) -> None:
|
||||
self._add_custom_parameters(credentials, None)
|
||||
super().validate_credentials(model, credentials)
|
||||
|
||||
@staticmethod
|
||||
def _add_custom_parameters(credentials: dict, model: str) -> None:
|
||||
if model is None:
|
||||
model = "bge-m3"
|
||||
|
||||
credentials["endpoint_url"] = f"https://ai.gitee.com/api/serverless/{model}/v1/"
|
||||
|
|
@ -0,0 +1,11 @@
|
|||
model: ChatTTS
|
||||
model_type: tts
|
||||
model_properties:
|
||||
default_voice: 'default'
|
||||
voices:
|
||||
- mode: 'default'
|
||||
name: 'Default'
|
||||
language: [ 'zh-Hans', 'en-US', 'de-DE', 'fr-FR', 'es-ES', 'it-IT', 'th-TH', 'id-ID' ]
|
||||
word_limit: 3500
|
||||
audio_type: 'mp3'
|
||||
max_workers: 5
|
||||
|
|
@ -0,0 +1,11 @@
|
|||
model: FunAudioLLM-CosyVoice-300M
|
||||
model_type: tts
|
||||
model_properties:
|
||||
default_voice: 'default'
|
||||
voices:
|
||||
- mode: 'default'
|
||||
name: 'Default'
|
||||
language: [ 'zh-Hans', 'en-US', 'de-DE', 'fr-FR', 'es-ES', 'it-IT', 'th-TH', 'id-ID' ]
|
||||
word_limit: 3500
|
||||
audio_type: 'mp3'
|
||||
max_workers: 5
|
||||
|
|
@ -0,0 +1,4 @@
|
|||
- speecht5_tts
|
||||
- ChatTTS
|
||||
- fish-speech-1.2-sft
|
||||
- FunAudioLLM-CosyVoice-300M
|
||||
|
|
@ -0,0 +1,11 @@
|
|||
model: fish-speech-1.2-sft
|
||||
model_type: tts
|
||||
model_properties:
|
||||
default_voice: 'default'
|
||||
voices:
|
||||
- mode: 'default'
|
||||
name: 'Default'
|
||||
language: [ 'zh-Hans', 'en-US', 'de-DE', 'fr-FR', 'es-ES', 'it-IT', 'th-TH', 'id-ID' ]
|
||||
word_limit: 3500
|
||||
audio_type: 'mp3'
|
||||
max_workers: 5
|
||||
|
|
@ -0,0 +1,11 @@
|
|||
model: speecht5_tts
|
||||
model_type: tts
|
||||
model_properties:
|
||||
default_voice: 'default'
|
||||
voices:
|
||||
- mode: 'default'
|
||||
name: 'Default'
|
||||
language: [ 'zh-Hans', 'en-US', 'de-DE', 'fr-FR', 'es-ES', 'it-IT', 'th-TH', 'id-ID' ]
|
||||
word_limit: 3500
|
||||
audio_type: 'mp3'
|
||||
max_workers: 5
|
||||
79
api/core/model_runtime/model_providers/gitee_ai/tts/tts.py
Normal file
79
api/core/model_runtime/model_providers/gitee_ai/tts/tts.py
Normal file
|
|
@ -0,0 +1,79 @@
|
|||
from typing import Optional
|
||||
|
||||
import requests
|
||||
|
||||
from core.model_runtime.errors.invoke import InvokeBadRequestError
|
||||
from core.model_runtime.errors.validate import CredentialsValidateFailedError
|
||||
from core.model_runtime.model_providers.__base.tts_model import TTSModel
|
||||
from core.model_runtime.model_providers.gitee_ai._common import _CommonGiteeAI
|
||||
|
||||
|
||||
class GiteeAIText2SpeechModel(_CommonGiteeAI, TTSModel):
|
||||
"""
|
||||
Model class for OpenAI Speech to text model.
|
||||
"""
|
||||
|
||||
def _invoke(
|
||||
self, model: str, tenant_id: str, credentials: dict, content_text: str, voice: str, user: Optional[str] = None
|
||||
) -> any:
|
||||
"""
|
||||
_invoke text2speech model
|
||||
|
||||
:param model: model name
|
||||
:param tenant_id: user tenant id
|
||||
:param credentials: model credentials
|
||||
:param content_text: text content to be translated
|
||||
:param voice: model timbre
|
||||
:param user: unique user id
|
||||
:return: text translated to audio file
|
||||
"""
|
||||
return self._tts_invoke_streaming(model=model, credentials=credentials, content_text=content_text, voice=voice)
|
||||
|
||||
def validate_credentials(self, model: str, credentials: dict) -> None:
|
||||
"""
|
||||
validate credentials text2speech model
|
||||
|
||||
:param model: model name
|
||||
:param credentials: model credentials
|
||||
:return: text translated to audio file
|
||||
"""
|
||||
try:
|
||||
self._tts_invoke_streaming(
|
||||
model=model,
|
||||
credentials=credentials,
|
||||
content_text="Hello Dify!",
|
||||
voice=self._get_model_default_voice(model, credentials),
|
||||
)
|
||||
except Exception as ex:
|
||||
raise CredentialsValidateFailedError(str(ex))
|
||||
|
||||
def _tts_invoke_streaming(self, model: str, credentials: dict, content_text: str, voice: str) -> any:
|
||||
"""
|
||||
_tts_invoke_streaming text2speech model
|
||||
:param model: model name
|
||||
:param credentials: model credentials
|
||||
:param content_text: text content to be translated
|
||||
:param voice: model timbre
|
||||
:return: text translated to audio file
|
||||
"""
|
||||
try:
|
||||
# doc: https://ai.gitee.com/docs/openapi/serverless#tag/serverless/POST/{service}/text-to-speech
|
||||
endpoint_url = "https://ai.gitee.com/api/serverless/" + model + "/text-to-speech"
|
||||
|
||||
headers = {"Content-Type": "application/json"}
|
||||
api_key = credentials.get("api_key")
|
||||
if api_key:
|
||||
headers["Authorization"] = f"Bearer {api_key}"
|
||||
|
||||
payload = {"inputs": content_text}
|
||||
response = requests.post(endpoint_url, headers=headers, json=payload)
|
||||
|
||||
if response.status_code != 200:
|
||||
raise InvokeBadRequestError(response.text)
|
||||
|
||||
data = response.content
|
||||
|
||||
for i in range(0, len(data), 1024):
|
||||
yield data[i : i + 1024]
|
||||
except Exception as ex:
|
||||
raise InvokeBadRequestError(str(ex))
|
||||
|
|
@ -116,26 +116,33 @@ class GoogleLargeLanguageModel(LargeLanguageModel):
|
|||
:param tools: tool messages
|
||||
:return: glm tools
|
||||
"""
|
||||
return glm.Tool(
|
||||
function_declarations=[
|
||||
glm.FunctionDeclaration(
|
||||
name=tool.name,
|
||||
parameters=glm.Schema(
|
||||
type=glm.Type.OBJECT,
|
||||
properties={
|
||||
key: {
|
||||
"type_": value.get("type", "string").upper(),
|
||||
"description": value.get("description", ""),
|
||||
"enum": value.get("enum", []),
|
||||
}
|
||||
for key, value in tool.parameters.get("properties", {}).items()
|
||||
},
|
||||
required=tool.parameters.get("required", []),
|
||||
),
|
||||
function_declarations = []
|
||||
for tool in tools:
|
||||
properties = {}
|
||||
for key, value in tool.parameters.get("properties", {}).items():
|
||||
properties[key] = {
|
||||
"type_": glm.Type.STRING,
|
||||
"description": value.get("description", ""),
|
||||
"enum": value.get("enum", []),
|
||||
}
|
||||
|
||||
if properties:
|
||||
parameters = glm.Schema(
|
||||
type=glm.Type.OBJECT,
|
||||
properties=properties,
|
||||
required=tool.parameters.get("required", []),
|
||||
)
|
||||
for tool in tools
|
||||
]
|
||||
)
|
||||
else:
|
||||
parameters = None
|
||||
|
||||
function_declaration = glm.FunctionDeclaration(
|
||||
name=tool.name,
|
||||
parameters=parameters,
|
||||
description=tool.description,
|
||||
)
|
||||
function_declarations.append(function_declaration)
|
||||
|
||||
return glm.Tool(function_declarations=function_declarations)
|
||||
|
||||
def validate_credentials(self, model: str, credentials: dict) -> None:
|
||||
"""
|
||||
|
|
|
|||
|
|
@ -44,6 +44,9 @@ class MoonshotLargeLanguageModel(OAIAPICompatLargeLanguageModel):
|
|||
self._add_custom_parameters(credentials)
|
||||
self._add_function_call(model, credentials)
|
||||
user = user[:32] if user else None
|
||||
# {"response_format": "json_object"} need convert to {"response_format": {"type": "json_object"}}
|
||||
if "response_format" in model_parameters:
|
||||
model_parameters["response_format"] = {"type": model_parameters.get("response_format")}
|
||||
return super()._invoke(model, credentials, prompt_messages, model_parameters, tools, stop, stream, user)
|
||||
|
||||
def validate_credentials(self, model: str, credentials: dict) -> None:
|
||||
|
|
|
|||
|
|
@ -397,16 +397,21 @@ class OAIAPICompatLargeLanguageModel(_CommonOaiApiCompat, LargeLanguageModel):
|
|||
chunk_index = 0
|
||||
|
||||
def create_final_llm_result_chunk(
|
||||
index: int, message: AssistantPromptMessage, finish_reason: str
|
||||
id: Optional[str], index: int, message: AssistantPromptMessage, finish_reason: str, usage: dict
|
||||
) -> LLMResultChunk:
|
||||
# calculate num tokens
|
||||
prompt_tokens = self._num_tokens_from_string(model, prompt_messages[0].content)
|
||||
completion_tokens = self._num_tokens_from_string(model, full_assistant_content)
|
||||
prompt_tokens = usage and usage.get("prompt_tokens")
|
||||
if prompt_tokens is None:
|
||||
prompt_tokens = self._num_tokens_from_string(model, prompt_messages[0].content)
|
||||
completion_tokens = usage and usage.get("completion_tokens")
|
||||
if completion_tokens is None:
|
||||
completion_tokens = self._num_tokens_from_string(model, full_assistant_content)
|
||||
|
||||
# transform usage
|
||||
usage = self._calc_response_usage(model, credentials, prompt_tokens, completion_tokens)
|
||||
|
||||
return LLMResultChunk(
|
||||
id=id,
|
||||
model=model,
|
||||
prompt_messages=prompt_messages,
|
||||
delta=LLMResultChunkDelta(index=index, message=message, finish_reason=finish_reason, usage=usage),
|
||||
|
|
@ -450,7 +455,7 @@ class OAIAPICompatLargeLanguageModel(_CommonOaiApiCompat, LargeLanguageModel):
|
|||
tool_call.function.arguments += new_tool_call.function.arguments
|
||||
|
||||
finish_reason = None # The default value of finish_reason is None
|
||||
|
||||
message_id, usage = None, None
|
||||
for chunk in response.iter_lines(decode_unicode=True, delimiter=delimiter):
|
||||
chunk = chunk.strip()
|
||||
if chunk:
|
||||
|
|
@ -462,20 +467,26 @@ class OAIAPICompatLargeLanguageModel(_CommonOaiApiCompat, LargeLanguageModel):
|
|||
continue
|
||||
|
||||
try:
|
||||
chunk_json = json.loads(decoded_chunk)
|
||||
chunk_json: dict = json.loads(decoded_chunk)
|
||||
# stream ended
|
||||
except json.JSONDecodeError as e:
|
||||
yield create_final_llm_result_chunk(
|
||||
id=message_id,
|
||||
index=chunk_index + 1,
|
||||
message=AssistantPromptMessage(content=""),
|
||||
finish_reason="Non-JSON encountered.",
|
||||
usage=usage,
|
||||
)
|
||||
break
|
||||
if chunk_json:
|
||||
if u := chunk_json.get("usage"):
|
||||
usage = u
|
||||
if not chunk_json or len(chunk_json["choices"]) == 0:
|
||||
continue
|
||||
|
||||
choice = chunk_json["choices"][0]
|
||||
finish_reason = chunk_json["choices"][0].get("finish_reason")
|
||||
message_id = chunk_json.get("id")
|
||||
chunk_index += 1
|
||||
|
||||
if "delta" in choice:
|
||||
|
|
@ -524,6 +535,7 @@ class OAIAPICompatLargeLanguageModel(_CommonOaiApiCompat, LargeLanguageModel):
|
|||
continue
|
||||
|
||||
yield LLMResultChunk(
|
||||
id=message_id,
|
||||
model=model,
|
||||
prompt_messages=prompt_messages,
|
||||
delta=LLMResultChunkDelta(
|
||||
|
|
@ -536,6 +548,7 @@ class OAIAPICompatLargeLanguageModel(_CommonOaiApiCompat, LargeLanguageModel):
|
|||
|
||||
if tools_calls:
|
||||
yield LLMResultChunk(
|
||||
id=message_id,
|
||||
model=model,
|
||||
prompt_messages=prompt_messages,
|
||||
delta=LLMResultChunkDelta(
|
||||
|
|
@ -545,17 +558,22 @@ class OAIAPICompatLargeLanguageModel(_CommonOaiApiCompat, LargeLanguageModel):
|
|||
)
|
||||
|
||||
yield create_final_llm_result_chunk(
|
||||
index=chunk_index, message=AssistantPromptMessage(content=""), finish_reason=finish_reason
|
||||
id=message_id,
|
||||
index=chunk_index,
|
||||
message=AssistantPromptMessage(content=""),
|
||||
finish_reason=finish_reason,
|
||||
usage=usage,
|
||||
)
|
||||
|
||||
def _handle_generate_response(
|
||||
self, model: str, credentials: dict, response: requests.Response, prompt_messages: list[PromptMessage]
|
||||
) -> LLMResult:
|
||||
response_json = response.json()
|
||||
response_json: dict = response.json()
|
||||
|
||||
completion_type = LLMMode.value_of(credentials["mode"])
|
||||
|
||||
output = response_json["choices"][0]
|
||||
message_id = response_json.get("id")
|
||||
|
||||
response_content = ""
|
||||
tool_calls = None
|
||||
|
|
@ -593,6 +611,7 @@ class OAIAPICompatLargeLanguageModel(_CommonOaiApiCompat, LargeLanguageModel):
|
|||
|
||||
# transform response
|
||||
result = LLMResult(
|
||||
id=message_id,
|
||||
model=response_json["model"],
|
||||
prompt_messages=prompt_messages,
|
||||
message=assistant_message,
|
||||
|
|
|
|||
0
api/core/rag/datasource/vdb/couchbase/__init__.py
Normal file
0
api/core/rag/datasource/vdb/couchbase/__init__.py
Normal file
378
api/core/rag/datasource/vdb/couchbase/couchbase_vector.py
Normal file
378
api/core/rag/datasource/vdb/couchbase/couchbase_vector.py
Normal file
|
|
@ -0,0 +1,378 @@
|
|||
import json
|
||||
import logging
|
||||
import time
|
||||
import uuid
|
||||
from datetime import timedelta
|
||||
from typing import Any
|
||||
|
||||
from couchbase import search
|
||||
from couchbase.auth import PasswordAuthenticator
|
||||
from couchbase.cluster import Cluster
|
||||
from couchbase.management.search import SearchIndex
|
||||
|
||||
# needed for options -- cluster, timeout, SQL++ (N1QL) query, etc.
|
||||
from couchbase.options import ClusterOptions, SearchOptions
|
||||
from couchbase.vector_search import VectorQuery, VectorSearch
|
||||
from flask import current_app
|
||||
from pydantic import BaseModel, model_validator
|
||||
|
||||
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_redis import redis_client
|
||||
from models.dataset import Dataset
|
||||
|
||||
logger = logging.getLogger(__name__)
|
||||
|
||||
|
||||
class CouchbaseConfig(BaseModel):
|
||||
connection_string: str
|
||||
user: str
|
||||
password: str
|
||||
bucket_name: str
|
||||
scope_name: str
|
||||
|
||||
@model_validator(mode="before")
|
||||
@classmethod
|
||||
def validate_config(cls, values: dict) -> dict:
|
||||
if not values.get("connection_string"):
|
||||
raise ValueError("config COUCHBASE_CONNECTION_STRING is required")
|
||||
if not values.get("user"):
|
||||
raise ValueError("config COUCHBASE_USER is required")
|
||||
if not values.get("password"):
|
||||
raise ValueError("config COUCHBASE_PASSWORD is required")
|
||||
if not values.get("bucket_name"):
|
||||
raise ValueError("config COUCHBASE_PASSWORD is required")
|
||||
if not values.get("scope_name"):
|
||||
raise ValueError("config COUCHBASE_SCOPE_NAME is required")
|
||||
return values
|
||||
|
||||
|
||||
class CouchbaseVector(BaseVector):
|
||||
def __init__(self, collection_name: str, config: CouchbaseConfig):
|
||||
super().__init__(collection_name)
|
||||
self._client_config = config
|
||||
|
||||
"""Connect to couchbase"""
|
||||
|
||||
auth = PasswordAuthenticator(config.user, config.password)
|
||||
options = ClusterOptions(auth)
|
||||
self._cluster = Cluster(config.connection_string, options)
|
||||
self._bucket = self._cluster.bucket(config.bucket_name)
|
||||
self._scope = self._bucket.scope(config.scope_name)
|
||||
self._bucket_name = config.bucket_name
|
||||
self._scope_name = config.scope_name
|
||||
|
||||
# Wait until the cluster is ready for use.
|
||||
self._cluster.wait_until_ready(timedelta(seconds=5))
|
||||
|
||||
def create(self, texts: list[Document], embeddings: list[list[float]], **kwargs):
|
||||
index_id = str(uuid.uuid4()).replace("-", "")
|
||||
self._create_collection(uuid=index_id, vector_length=len(embeddings[0]))
|
||||
self.add_texts(texts, embeddings)
|
||||
|
||||
def _create_collection(self, vector_length: int, uuid: str):
|
||||
lock_name = "vector_indexing_lock_{}".format(self._collection_name)
|
||||
with redis_client.lock(lock_name, timeout=20):
|
||||
collection_exist_cache_key = "vector_indexing_{}".format(self._collection_name)
|
||||
if redis_client.get(collection_exist_cache_key):
|
||||
return
|
||||
if self._collection_exists(self._collection_name):
|
||||
return
|
||||
manager = self._bucket.collections()
|
||||
manager.create_collection(self._client_config.scope_name, self._collection_name)
|
||||
|
||||
index_manager = self._scope.search_indexes()
|
||||
|
||||
index_definition = json.loads("""
|
||||
{
|
||||
"type": "fulltext-index",
|
||||
"name": "Embeddings._default.Vector_Search",
|
||||
"uuid": "26d4db528e78b716",
|
||||
"sourceType": "gocbcore",
|
||||
"sourceName": "Embeddings",
|
||||
"sourceUUID": "2242e4a25b4decd6650c9c7b3afa1dbf",
|
||||
"planParams": {
|
||||
"maxPartitionsPerPIndex": 1024,
|
||||
"indexPartitions": 1
|
||||
},
|
||||
"params": {
|
||||
"doc_config": {
|
||||
"docid_prefix_delim": "",
|
||||
"docid_regexp": "",
|
||||
"mode": "scope.collection.type_field",
|
||||
"type_field": "type"
|
||||
},
|
||||
"mapping": {
|
||||
"analysis": { },
|
||||
"default_analyzer": "standard",
|
||||
"default_datetime_parser": "dateTimeOptional",
|
||||
"default_field": "_all",
|
||||
"default_mapping": {
|
||||
"dynamic": true,
|
||||
"enabled": true
|
||||
},
|
||||
"default_type": "_default",
|
||||
"docvalues_dynamic": false,
|
||||
"index_dynamic": true,
|
||||
"store_dynamic": true,
|
||||
"type_field": "_type",
|
||||
"types": {
|
||||
"collection_name": {
|
||||
"dynamic": true,
|
||||
"enabled": true,
|
||||
"properties": {
|
||||
"embedding": {
|
||||
"dynamic": false,
|
||||
"enabled": true,
|
||||
"fields": [
|
||||
{
|
||||
"dims": 1536,
|
||||
"index": true,
|
||||
"name": "embedding",
|
||||
"similarity": "dot_product",
|
||||
"type": "vector",
|
||||
"vector_index_optimized_for": "recall"
|
||||
}
|
||||
]
|
||||
},
|
||||
"metadata": {
|
||||
"dynamic": true,
|
||||
"enabled": true
|
||||
},
|
||||
"text": {
|
||||
"dynamic": false,
|
||||
"enabled": true,
|
||||
"fields": [
|
||||
{
|
||||
"index": true,
|
||||
"name": "text",
|
||||
"store": true,
|
||||
"type": "text"
|
||||
}
|
||||
]
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
},
|
||||
"store": {
|
||||
"indexType": "scorch",
|
||||
"segmentVersion": 16
|
||||
}
|
||||
},
|
||||
"sourceParams": { }
|
||||
}
|
||||
""")
|
||||
index_definition["name"] = self._collection_name + "_search"
|
||||
index_definition["uuid"] = uuid
|
||||
index_definition["params"]["mapping"]["types"]["collection_name"]["properties"]["embedding"]["fields"][0][
|
||||
"dims"
|
||||
] = vector_length
|
||||
index_definition["params"]["mapping"]["types"][self._scope_name + "." + self._collection_name] = (
|
||||
index_definition["params"]["mapping"]["types"].pop("collection_name")
|
||||
)
|
||||
time.sleep(2)
|
||||
index_manager.upsert_index(
|
||||
SearchIndex(
|
||||
index_definition["name"],
|
||||
params=index_definition["params"],
|
||||
source_name=self._bucket_name,
|
||||
),
|
||||
)
|
||||
time.sleep(1)
|
||||
|
||||
redis_client.set(collection_exist_cache_key, 1, ex=3600)
|
||||
|
||||
def _collection_exists(self, name: str):
|
||||
scope_collection_map: dict[str, Any] = {}
|
||||
|
||||
# Get a list of all scopes in the bucket
|
||||
for scope in self._bucket.collections().get_all_scopes():
|
||||
scope_collection_map[scope.name] = []
|
||||
|
||||
# Get a list of all the collections in the scope
|
||||
for collection in scope.collections:
|
||||
scope_collection_map[scope.name].append(collection.name)
|
||||
|
||||
# Check if the collection exists in the scope
|
||||
return self._collection_name in scope_collection_map[self._scope_name]
|
||||
|
||||
def get_type(self) -> str:
|
||||
return VectorType.COUCHBASE
|
||||
|
||||
def add_texts(self, documents: list[Document], embeddings: list[list[float]], **kwargs):
|
||||
uuids = self._get_uuids(documents)
|
||||
texts = [d.page_content for d in documents]
|
||||
metadatas = [d.metadata for d in documents]
|
||||
|
||||
doc_ids = []
|
||||
|
||||
documents_to_insert = [
|
||||
{"text": text, "embedding": vector, "metadata": metadata}
|
||||
for id, text, vector, metadata in zip(uuids, texts, embeddings, metadatas)
|
||||
]
|
||||
for doc, id in zip(documents_to_insert, uuids):
|
||||
result = self._scope.collection(self._collection_name).upsert(id, doc)
|
||||
|
||||
doc_ids.extend(uuids)
|
||||
|
||||
return doc_ids
|
||||
|
||||
def text_exists(self, id: str) -> bool:
|
||||
# Use a parameterized query for safety and correctness
|
||||
query = f"""
|
||||
SELECT COUNT(1) AS count FROM
|
||||
`{self._client_config.bucket_name}`.{self._client_config.scope_name}.{self._collection_name}
|
||||
WHERE META().id = $doc_id
|
||||
"""
|
||||
# Pass the id as a parameter to the query
|
||||
result = self._cluster.query(query, named_parameters={"doc_id": id}).execute()
|
||||
for row in result:
|
||||
return row["count"] > 0
|
||||
return False # Return False if no rows are returned
|
||||
|
||||
def delete_by_ids(self, ids: list[str]) -> None:
|
||||
query = f"""
|
||||
DELETE FROM `{self._bucket_name}`.{self._client_config.scope_name}.{self._collection_name}
|
||||
WHERE META().id IN $doc_ids;
|
||||
"""
|
||||
try:
|
||||
self._cluster.query(query, named_parameters={"doc_ids": ids}).execute()
|
||||
except Exception as e:
|
||||
logger.error(e)
|
||||
|
||||
def delete_by_document_id(self, document_id: str):
|
||||
query = f"""
|
||||
DELETE FROM
|
||||
`{self._client_config.bucket_name}`.{self._client_config.scope_name}.{self._collection_name}
|
||||
WHERE META().id = $doc_id;
|
||||
"""
|
||||
self._cluster.query(query, named_parameters={"doc_id": document_id}).execute()
|
||||
|
||||
# def get_ids_by_metadata_field(self, key: str, value: str):
|
||||
# query = f"""
|
||||
# SELECT id FROM
|
||||
# `{self._client_config.bucket_name}`.{self._client_config.scope_name}.{self._collection_name}
|
||||
# WHERE `metadata.{key}` = $value;
|
||||
# """
|
||||
# result = self._cluster.query(query, named_parameters={'value':value})
|
||||
# return [row['id'] for row in result.rows()]
|
||||
|
||||
def delete_by_metadata_field(self, key: str, value: str) -> None:
|
||||
query = f"""
|
||||
DELETE FROM `{self._client_config.bucket_name}`.{self._client_config.scope_name}.{self._collection_name}
|
||||
WHERE metadata.{key} = $value;
|
||||
"""
|
||||
self._cluster.query(query, named_parameters={"value": value}).execute()
|
||||
|
||||
def search_by_vector(self, query_vector: list[float], **kwargs: Any) -> list[Document]:
|
||||
top_k = kwargs.get("top_k", 5)
|
||||
score_threshold = kwargs.get("score_threshold") or 0.0
|
||||
|
||||
search_req = search.SearchRequest.create(
|
||||
VectorSearch.from_vector_query(
|
||||
VectorQuery(
|
||||
"embedding",
|
||||
query_vector,
|
||||
top_k,
|
||||
)
|
||||
)
|
||||
)
|
||||
try:
|
||||
search_iter = self._scope.search(
|
||||
self._collection_name + "_search",
|
||||
search_req,
|
||||
SearchOptions(limit=top_k, collections=[self._collection_name], fields=["*"]),
|
||||
)
|
||||
|
||||
docs = []
|
||||
# Parse the results
|
||||
for row in search_iter.rows():
|
||||
text = row.fields.pop("text")
|
||||
metadata = self._format_metadata(row.fields)
|
||||
score = row.score
|
||||
metadata["score"] = score
|
||||
doc = Document(page_content=text, metadata=metadata)
|
||||
if score >= score_threshold:
|
||||
docs.append(doc)
|
||||
except Exception as e:
|
||||
raise ValueError(f"Search failed with error: {e}")
|
||||
|
||||
return docs
|
||||
|
||||
def search_by_full_text(self, query: str, **kwargs: Any) -> list[Document]:
|
||||
top_k = kwargs.get("top_k", 2)
|
||||
try:
|
||||
CBrequest = search.SearchRequest.create(search.QueryStringQuery("text:" + query))
|
||||
search_iter = self._scope.search(
|
||||
self._collection_name + "_search", CBrequest, SearchOptions(limit=top_k, fields=["*"])
|
||||
)
|
||||
|
||||
docs = []
|
||||
for row in search_iter.rows():
|
||||
text = row.fields.pop("text")
|
||||
metadata = self._format_metadata(row.fields)
|
||||
score = row.score
|
||||
metadata["score"] = score
|
||||
doc = Document(page_content=text, metadata=metadata)
|
||||
docs.append(doc)
|
||||
|
||||
except Exception as e:
|
||||
raise ValueError(f"Search failed with error: {e}")
|
||||
|
||||
return docs
|
||||
|
||||
def delete(self):
|
||||
manager = self._bucket.collections()
|
||||
scopes = manager.get_all_scopes()
|
||||
|
||||
for scope in scopes:
|
||||
for collection in scope.collections:
|
||||
if collection.name == self._collection_name:
|
||||
manager.drop_collection("_default", self._collection_name)
|
||||
|
||||
def _format_metadata(self, row_fields: dict[str, Any]) -> dict[str, Any]:
|
||||
"""Helper method to format the metadata from the Couchbase Search API.
|
||||
Args:
|
||||
row_fields (Dict[str, Any]): The fields to format.
|
||||
|
||||
Returns:
|
||||
Dict[str, Any]: The formatted metadata.
|
||||
"""
|
||||
metadata = {}
|
||||
for key, value in row_fields.items():
|
||||
# Couchbase Search returns the metadata key with a prefix
|
||||
# `metadata.` We remove it to get the original metadata key
|
||||
if key.startswith("metadata"):
|
||||
new_key = key.split("metadata" + ".")[-1]
|
||||
metadata[new_key] = value
|
||||
else:
|
||||
metadata[key] = value
|
||||
|
||||
return metadata
|
||||
|
||||
|
||||
class CouchbaseVectorFactory(AbstractVectorFactory):
|
||||
def init_vector(self, dataset: Dataset, attributes: list, embeddings: Embeddings) -> CouchbaseVector:
|
||||
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)
|
||||
dataset.index_struct = json.dumps(self.gen_index_struct_dict(VectorType.COUCHBASE, collection_name))
|
||||
|
||||
config = current_app.config
|
||||
return CouchbaseVector(
|
||||
collection_name=collection_name,
|
||||
config=CouchbaseConfig(
|
||||
connection_string=config.get("COUCHBASE_CONNECTION_STRING"),
|
||||
user=config.get("COUCHBASE_USER"),
|
||||
password=config.get("COUCHBASE_PASSWORD"),
|
||||
bucket_name=config.get("COUCHBASE_BUCKET_NAME"),
|
||||
scope_name=config.get("COUCHBASE_SCOPE_NAME"),
|
||||
),
|
||||
)
|
||||
|
|
@ -142,7 +142,7 @@ class ElasticSearchVector(BaseVector):
|
|||
|
||||
def search_by_full_text(self, query: str, **kwargs: Any) -> list[Document]:
|
||||
query_str = {"match": {Field.CONTENT_KEY.value: query}}
|
||||
results = self._client.search(index=self._collection_name, query=query_str)
|
||||
results = self._client.search(index=self._collection_name, query=query_str, size=kwargs.get("top_k", 4))
|
||||
docs = []
|
||||
for hit in results["hits"]["hits"]:
|
||||
docs.append(
|
||||
|
|
|
|||
17
api/core/rag/datasource/vdb/tidb_on_qdrant/tidb_entities.py
Normal file
17
api/core/rag/datasource/vdb/tidb_on_qdrant/tidb_entities.py
Normal file
|
|
@ -0,0 +1,17 @@
|
|||
from typing import Optional
|
||||
|
||||
from pydantic import BaseModel
|
||||
|
||||
|
||||
class ClusterEntity(BaseModel):
|
||||
"""
|
||||
Model Config Entity.
|
||||
"""
|
||||
|
||||
name: str
|
||||
cluster_id: str
|
||||
displayName: str
|
||||
region: str
|
||||
spendingLimit: Optional[int] = 1000
|
||||
version: str
|
||||
createdBy: str
|
||||
|
|
@ -0,0 +1,526 @@
|
|||
import json
|
||||
import os
|
||||
import uuid
|
||||
from collections.abc import Generator, Iterable, Sequence
|
||||
from itertools import islice
|
||||
from typing import TYPE_CHECKING, Any, Optional, Union, cast
|
||||
|
||||
import qdrant_client
|
||||
import requests
|
||||
from flask import current_app
|
||||
from pydantic import BaseModel
|
||||
from qdrant_client.http import models as rest
|
||||
from qdrant_client.http.models import (
|
||||
FilterSelector,
|
||||
HnswConfigDiff,
|
||||
PayloadSchemaType,
|
||||
TextIndexParams,
|
||||
TextIndexType,
|
||||
TokenizerType,
|
||||
)
|
||||
from qdrant_client.local.qdrant_local import QdrantLocal
|
||||
from requests.auth import HTTPDigestAuth
|
||||
|
||||
from configs import dify_config
|
||||
from core.rag.datasource.vdb.field import Field
|
||||
from core.rag.datasource.vdb.tidb_on_qdrant.tidb_service import TidbService
|
||||
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, TidbAuthBinding
|
||||
|
||||
if TYPE_CHECKING:
|
||||
from qdrant_client import grpc # noqa
|
||||
from qdrant_client.conversions import common_types
|
||||
from qdrant_client.http import models as rest
|
||||
|
||||
DictFilter = dict[str, Union[str, int, bool, dict, list]]
|
||||
MetadataFilter = Union[DictFilter, common_types.Filter]
|
||||
|
||||
|
||||
class TidbOnQdrantConfig(BaseModel):
|
||||
endpoint: str
|
||||
api_key: Optional[str] = None
|
||||
timeout: float = 20
|
||||
root_path: Optional[str] = None
|
||||
grpc_port: int = 6334
|
||||
prefer_grpc: bool = False
|
||||
|
||||
def to_qdrant_params(self):
|
||||
if self.endpoint and self.endpoint.startswith("path:"):
|
||||
path = self.endpoint.replace("path:", "")
|
||||
if not os.path.isabs(path):
|
||||
path = os.path.join(self.root_path, path)
|
||||
|
||||
return {"path": path}
|
||||
else:
|
||||
return {
|
||||
"url": self.endpoint,
|
||||
"api_key": self.api_key,
|
||||
"timeout": self.timeout,
|
||||
"verify": False,
|
||||
"grpc_port": self.grpc_port,
|
||||
"prefer_grpc": self.prefer_grpc,
|
||||
}
|
||||
|
||||
|
||||
class TidbConfig(BaseModel):
|
||||
api_url: str
|
||||
public_key: str
|
||||
private_key: str
|
||||
|
||||
|
||||
class TidbOnQdrantVector(BaseVector):
|
||||
def __init__(self, collection_name: str, group_id: str, config: TidbOnQdrantConfig, distance_func: str = "Cosine"):
|
||||
super().__init__(collection_name)
|
||||
self._client_config = config
|
||||
self._client = qdrant_client.QdrantClient(**self._client_config.to_qdrant_params())
|
||||
self._distance_func = distance_func.upper()
|
||||
self._group_id = group_id
|
||||
|
||||
def get_type(self) -> str:
|
||||
return VectorType.TIDB_ON_QDRANT
|
||||
|
||||
def to_index_struct(self) -> dict:
|
||||
return {"type": self.get_type(), "vector_store": {"class_prefix": self._collection_name}}
|
||||
|
||||
def create(self, texts: list[Document], embeddings: list[list[float]], **kwargs):
|
||||
if texts:
|
||||
# get embedding vector size
|
||||
vector_size = len(embeddings[0])
|
||||
# get collection name
|
||||
collection_name = self._collection_name
|
||||
# create collection
|
||||
self.create_collection(collection_name, vector_size)
|
||||
|
||||
self.add_texts(texts, embeddings, **kwargs)
|
||||
|
||||
def create_collection(self, collection_name: str, vector_size: int):
|
||||
lock_name = "vector_indexing_lock_{}".format(collection_name)
|
||||
with redis_client.lock(lock_name, timeout=20):
|
||||
collection_exist_cache_key = "vector_indexing_{}".format(self._collection_name)
|
||||
if redis_client.get(collection_exist_cache_key):
|
||||
return
|
||||
collection_name = collection_name or uuid.uuid4().hex
|
||||
all_collection_name = []
|
||||
collections_response = self._client.get_collections()
|
||||
collection_list = collections_response.collections
|
||||
for collection in collection_list:
|
||||
all_collection_name.append(collection.name)
|
||||
if collection_name not in all_collection_name:
|
||||
from qdrant_client.http import models as rest
|
||||
|
||||
vectors_config = rest.VectorParams(
|
||||
size=vector_size,
|
||||
distance=rest.Distance[self._distance_func],
|
||||
)
|
||||
hnsw_config = HnswConfigDiff(
|
||||
m=0,
|
||||
payload_m=16,
|
||||
ef_construct=100,
|
||||
full_scan_threshold=10000,
|
||||
max_indexing_threads=0,
|
||||
on_disk=False,
|
||||
)
|
||||
self._client.recreate_collection(
|
||||
collection_name=collection_name,
|
||||
vectors_config=vectors_config,
|
||||
hnsw_config=hnsw_config,
|
||||
timeout=int(self._client_config.timeout),
|
||||
)
|
||||
|
||||
# create group_id payload index
|
||||
self._client.create_payload_index(
|
||||
collection_name, Field.GROUP_KEY.value, field_schema=PayloadSchemaType.KEYWORD
|
||||
)
|
||||
# create doc_id payload index
|
||||
self._client.create_payload_index(
|
||||
collection_name, Field.DOC_ID.value, field_schema=PayloadSchemaType.KEYWORD
|
||||
)
|
||||
# create full text index
|
||||
text_index_params = TextIndexParams(
|
||||
type=TextIndexType.TEXT,
|
||||
tokenizer=TokenizerType.MULTILINGUAL,
|
||||
min_token_len=2,
|
||||
max_token_len=20,
|
||||
lowercase=True,
|
||||
)
|
||||
self._client.create_payload_index(
|
||||
collection_name, Field.CONTENT_KEY.value, field_schema=text_index_params
|
||||
)
|
||||
redis_client.set(collection_exist_cache_key, 1, ex=3600)
|
||||
|
||||
def add_texts(self, documents: list[Document], embeddings: list[list[float]], **kwargs):
|
||||
uuids = self._get_uuids(documents)
|
||||
texts = [d.page_content for d in documents]
|
||||
metadatas = [d.metadata for d in documents]
|
||||
|
||||
added_ids = []
|
||||
for batch_ids, points in self._generate_rest_batches(texts, embeddings, metadatas, uuids, 64, self._group_id):
|
||||
self._client.upsert(collection_name=self._collection_name, points=points)
|
||||
added_ids.extend(batch_ids)
|
||||
|
||||
return added_ids
|
||||
|
||||
def _generate_rest_batches(
|
||||
self,
|
||||
texts: Iterable[str],
|
||||
embeddings: list[list[float]],
|
||||
metadatas: Optional[list[dict]] = None,
|
||||
ids: Optional[Sequence[str]] = None,
|
||||
batch_size: int = 64,
|
||||
group_id: Optional[str] = None,
|
||||
) -> Generator[tuple[list[str], list[rest.PointStruct]], None, None]:
|
||||
from qdrant_client.http import models as rest
|
||||
|
||||
texts_iterator = iter(texts)
|
||||
embeddings_iterator = iter(embeddings)
|
||||
metadatas_iterator = iter(metadatas or [])
|
||||
ids_iterator = iter(ids or [uuid.uuid4().hex for _ in iter(texts)])
|
||||
while batch_texts := list(islice(texts_iterator, batch_size)):
|
||||
# Take the corresponding metadata and id for each text in a batch
|
||||
batch_metadatas = list(islice(metadatas_iterator, batch_size)) or None
|
||||
batch_ids = list(islice(ids_iterator, batch_size))
|
||||
|
||||
# Generate the embeddings for all the texts in a batch
|
||||
batch_embeddings = list(islice(embeddings_iterator, batch_size))
|
||||
|
||||
points = [
|
||||
rest.PointStruct(
|
||||
id=point_id,
|
||||
vector=vector,
|
||||
payload=payload,
|
||||
)
|
||||
for point_id, vector, payload in zip(
|
||||
batch_ids,
|
||||
batch_embeddings,
|
||||
self._build_payloads(
|
||||
batch_texts,
|
||||
batch_metadatas,
|
||||
Field.CONTENT_KEY.value,
|
||||
Field.METADATA_KEY.value,
|
||||
group_id,
|
||||
Field.GROUP_KEY.value,
|
||||
),
|
||||
)
|
||||
]
|
||||
|
||||
yield batch_ids, points
|
||||
|
||||
@classmethod
|
||||
def _build_payloads(
|
||||
cls,
|
||||
texts: Iterable[str],
|
||||
metadatas: Optional[list[dict]],
|
||||
content_payload_key: str,
|
||||
metadata_payload_key: str,
|
||||
group_id: str,
|
||||
group_payload_key: str,
|
||||
) -> list[dict]:
|
||||
payloads = []
|
||||
for i, text in enumerate(texts):
|
||||
if text is None:
|
||||
raise ValueError(
|
||||
"At least one of the texts is None. Please remove it before "
|
||||
"calling .from_texts or .add_texts on Qdrant instance."
|
||||
)
|
||||
metadata = metadatas[i] if metadatas is not None else None
|
||||
payloads.append({content_payload_key: text, metadata_payload_key: metadata, group_payload_key: group_id})
|
||||
|
||||
return payloads
|
||||
|
||||
def delete_by_metadata_field(self, key: str, value: str):
|
||||
from qdrant_client.http import models
|
||||
from qdrant_client.http.exceptions import UnexpectedResponse
|
||||
|
||||
try:
|
||||
filter = models.Filter(
|
||||
must=[
|
||||
models.FieldCondition(
|
||||
key=f"metadata.{key}",
|
||||
match=models.MatchValue(value=value),
|
||||
),
|
||||
],
|
||||
)
|
||||
|
||||
self._reload_if_needed()
|
||||
|
||||
self._client.delete(
|
||||
collection_name=self._collection_name,
|
||||
points_selector=FilterSelector(filter=filter),
|
||||
)
|
||||
except UnexpectedResponse as e:
|
||||
# Collection does not exist, so return
|
||||
if e.status_code == 404:
|
||||
return
|
||||
# Some other error occurred, so re-raise the exception
|
||||
else:
|
||||
raise e
|
||||
|
||||
def delete(self):
|
||||
from qdrant_client.http.exceptions import UnexpectedResponse
|
||||
|
||||
try:
|
||||
self._client.delete_collection(collection_name=self._collection_name)
|
||||
except UnexpectedResponse as e:
|
||||
# Collection does not exist, so return
|
||||
if e.status_code == 404:
|
||||
return
|
||||
# Some other error occurred, so re-raise the exception
|
||||
else:
|
||||
raise e
|
||||
|
||||
def delete_by_ids(self, ids: list[str]) -> None:
|
||||
from qdrant_client.http import models
|
||||
from qdrant_client.http.exceptions import UnexpectedResponse
|
||||
|
||||
for node_id in ids:
|
||||
try:
|
||||
filter = models.Filter(
|
||||
must=[
|
||||
models.FieldCondition(
|
||||
key="metadata.doc_id",
|
||||
match=models.MatchValue(value=node_id),
|
||||
),
|
||||
],
|
||||
)
|
||||
self._client.delete(
|
||||
collection_name=self._collection_name,
|
||||
points_selector=FilterSelector(filter=filter),
|
||||
)
|
||||
except UnexpectedResponse as e:
|
||||
# Collection does not exist, so return
|
||||
if e.status_code == 404:
|
||||
return
|
||||
# Some other error occurred, so re-raise the exception
|
||||
else:
|
||||
raise e
|
||||
|
||||
def text_exists(self, id: str) -> bool:
|
||||
all_collection_name = []
|
||||
collections_response = self._client.get_collections()
|
||||
collection_list = collections_response.collections
|
||||
for collection in collection_list:
|
||||
all_collection_name.append(collection.name)
|
||||
if self._collection_name not in all_collection_name:
|
||||
return False
|
||||
response = self._client.retrieve(collection_name=self._collection_name, ids=[id])
|
||||
|
||||
return len(response) > 0
|
||||
|
||||
def search_by_vector(self, query_vector: list[float], **kwargs: Any) -> list[Document]:
|
||||
from qdrant_client.http import models
|
||||
|
||||
filter = models.Filter(
|
||||
must=[
|
||||
models.FieldCondition(
|
||||
key="group_id",
|
||||
match=models.MatchValue(value=self._group_id),
|
||||
),
|
||||
],
|
||||
)
|
||||
results = self._client.search(
|
||||
collection_name=self._collection_name,
|
||||
query_vector=query_vector,
|
||||
query_filter=filter,
|
||||
limit=kwargs.get("top_k", 4),
|
||||
with_payload=True,
|
||||
with_vectors=True,
|
||||
score_threshold=kwargs.get("score_threshold", 0.0),
|
||||
)
|
||||
docs = []
|
||||
for result in results:
|
||||
metadata = result.payload.get(Field.METADATA_KEY.value) or {}
|
||||
# duplicate check score threshold
|
||||
score_threshold = kwargs.get("score_threshold") or 0.0
|
||||
if result.score > score_threshold:
|
||||
metadata["score"] = result.score
|
||||
doc = Document(
|
||||
page_content=result.payload.get(Field.CONTENT_KEY.value),
|
||||
metadata=metadata,
|
||||
)
|
||||
docs.append(doc)
|
||||
# Sort the documents by score in descending order
|
||||
docs = sorted(docs, key=lambda x: x.metadata["score"], reverse=True)
|
||||
return docs
|
||||
|
||||
def search_by_full_text(self, query: str, **kwargs: Any) -> list[Document]:
|
||||
"""Return docs most similar by bm25.
|
||||
Returns:
|
||||
List of documents most similar to the query text and distance for each.
|
||||
"""
|
||||
from qdrant_client.http import models
|
||||
|
||||
scroll_filter = models.Filter(
|
||||
must=[
|
||||
models.FieldCondition(
|
||||
key="page_content",
|
||||
match=models.MatchText(text=query),
|
||||
)
|
||||
]
|
||||
)
|
||||
response = self._client.scroll(
|
||||
collection_name=self._collection_name,
|
||||
scroll_filter=scroll_filter,
|
||||
limit=kwargs.get("top_k", 2),
|
||||
with_payload=True,
|
||||
with_vectors=True,
|
||||
)
|
||||
results = response[0]
|
||||
documents = []
|
||||
for result in results:
|
||||
if result:
|
||||
document = self._document_from_scored_point(result, Field.CONTENT_KEY.value, Field.METADATA_KEY.value)
|
||||
document.metadata["vector"] = result.vector
|
||||
documents.append(document)
|
||||
|
||||
return documents
|
||||
|
||||
def _reload_if_needed(self):
|
||||
if isinstance(self._client, QdrantLocal):
|
||||
self._client = cast(QdrantLocal, self._client)
|
||||
self._client._load()
|
||||
|
||||
@classmethod
|
||||
def _document_from_scored_point(
|
||||
cls,
|
||||
scored_point: Any,
|
||||
content_payload_key: str,
|
||||
metadata_payload_key: str,
|
||||
) -> Document:
|
||||
return Document(
|
||||
page_content=scored_point.payload.get(content_payload_key),
|
||||
metadata=scored_point.payload.get(metadata_payload_key) or {},
|
||||
)
|
||||
|
||||
|
||||
class TidbOnQdrantVectorFactory(AbstractVectorFactory):
|
||||
def init_vector(self, dataset: Dataset, attributes: list, embeddings: Embeddings) -> TidbOnQdrantVector:
|
||||
tidb_auth_binding = (
|
||||
db.session.query(TidbAuthBinding).filter(TidbAuthBinding.tenant_id == dataset.tenant_id).one_or_none()
|
||||
)
|
||||
if not tidb_auth_binding:
|
||||
idle_tidb_auth_binding = (
|
||||
db.session.query(TidbAuthBinding)
|
||||
.filter(TidbAuthBinding.active == False, TidbAuthBinding.status == "ACTIVE")
|
||||
.limit(1)
|
||||
.one_or_none()
|
||||
)
|
||||
if idle_tidb_auth_binding:
|
||||
idle_tidb_auth_binding.active = True
|
||||
idle_tidb_auth_binding.tenant_id = dataset.tenant_id
|
||||
db.session.commit()
|
||||
TIDB_ON_QDRANT_API_KEY = f"{idle_tidb_auth_binding.account}:{idle_tidb_auth_binding.password}"
|
||||
else:
|
||||
with redis_client.lock("create_tidb_serverless_cluster_lock", timeout=900):
|
||||
tidb_auth_binding = (
|
||||
db.session.query(TidbAuthBinding)
|
||||
.filter(TidbAuthBinding.tenant_id == dataset.tenant_id)
|
||||
.one_or_none()
|
||||
)
|
||||
if tidb_auth_binding:
|
||||
TIDB_ON_QDRANT_API_KEY = f"{tidb_auth_binding.account}:{tidb_auth_binding.password}"
|
||||
|
||||
else:
|
||||
new_cluster = TidbService.create_tidb_serverless_cluster(
|
||||
dify_config.TIDB_PROJECT_ID,
|
||||
dify_config.TIDB_API_URL,
|
||||
dify_config.TIDB_IAM_API_URL,
|
||||
dify_config.TIDB_PUBLIC_KEY,
|
||||
dify_config.TIDB_PRIVATE_KEY,
|
||||
dify_config.TIDB_REGION,
|
||||
)
|
||||
new_tidb_auth_binding = TidbAuthBinding(
|
||||
cluster_id=new_cluster["cluster_id"],
|
||||
cluster_name=new_cluster["cluster_name"],
|
||||
account=new_cluster["account"],
|
||||
password=new_cluster["password"],
|
||||
tenant_id=dataset.tenant_id,
|
||||
active=True,
|
||||
status="ACTIVE",
|
||||
)
|
||||
db.session.add(new_tidb_auth_binding)
|
||||
db.session.commit()
|
||||
TIDB_ON_QDRANT_API_KEY = f"{new_tidb_auth_binding.account}:{new_tidb_auth_binding.password}"
|
||||
|
||||
else:
|
||||
TIDB_ON_QDRANT_API_KEY = f"{tidb_auth_binding.account}:{tidb_auth_binding.password}"
|
||||
|
||||
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)
|
||||
dataset.index_struct = json.dumps(self.gen_index_struct_dict(VectorType.TIDB_ON_QDRANT, collection_name))
|
||||
|
||||
config = current_app.config
|
||||
|
||||
return TidbOnQdrantVector(
|
||||
collection_name=collection_name,
|
||||
group_id=dataset.id,
|
||||
config=TidbOnQdrantConfig(
|
||||
endpoint=dify_config.TIDB_ON_QDRANT_URL,
|
||||
api_key=TIDB_ON_QDRANT_API_KEY,
|
||||
root_path=config.root_path,
|
||||
timeout=dify_config.TIDB_ON_QDRANT_CLIENT_TIMEOUT,
|
||||
grpc_port=dify_config.TIDB_ON_QDRANT_GRPC_PORT,
|
||||
prefer_grpc=dify_config.TIDB_ON_QDRANT_GRPC_ENABLED,
|
||||
),
|
||||
)
|
||||
|
||||
def create_tidb_serverless_cluster(self, tidb_config: TidbConfig, display_name: str, region: str):
|
||||
"""
|
||||
Creates a new TiDB Serverless cluster.
|
||||
:param tidb_config: The configuration for the TiDB Cloud API.
|
||||
:param display_name: The user-friendly display name of the cluster (required).
|
||||
:param region: The region where the cluster will be created (required).
|
||||
|
||||
:return: The response from the API.
|
||||
"""
|
||||
region_object = {
|
||||
"name": region,
|
||||
}
|
||||
|
||||
labels = {
|
||||
"tidb.cloud/project": "1372813089454548012",
|
||||
}
|
||||
cluster_data = {"displayName": display_name, "region": region_object, "labels": labels}
|
||||
|
||||
response = requests.post(
|
||||
f"{tidb_config.api_url}/clusters",
|
||||
json=cluster_data,
|
||||
auth=HTTPDigestAuth(tidb_config.public_key, tidb_config.private_key),
|
||||
)
|
||||
|
||||
if response.status_code == 200:
|
||||
return response.json()
|
||||
else:
|
||||
response.raise_for_status()
|
||||
|
||||
def change_tidb_serverless_root_password(self, tidb_config: TidbConfig, cluster_id: str, new_password: str):
|
||||
"""
|
||||
Changes the root password of a specific TiDB Serverless cluster.
|
||||
|
||||
:param tidb_config: The configuration for the TiDB Cloud API.
|
||||
:param cluster_id: The ID of the cluster for which the password is to be changed (required).
|
||||
:param new_password: The new password for the root user (required).
|
||||
:return: The response from the API.
|
||||
"""
|
||||
|
||||
body = {"password": new_password}
|
||||
|
||||
response = requests.put(
|
||||
f"{tidb_config.api_url}/clusters/{cluster_id}/password",
|
||||
json=body,
|
||||
auth=HTTPDigestAuth(tidb_config.public_key, tidb_config.private_key),
|
||||
)
|
||||
|
||||
if response.status_code == 200:
|
||||
return response.json()
|
||||
else:
|
||||
response.raise_for_status()
|
||||
250
api/core/rag/datasource/vdb/tidb_on_qdrant/tidb_service.py
Normal file
250
api/core/rag/datasource/vdb/tidb_on_qdrant/tidb_service.py
Normal file
|
|
@ -0,0 +1,250 @@
|
|||
import time
|
||||
import uuid
|
||||
|
||||
import requests
|
||||
from requests.auth import HTTPDigestAuth
|
||||
|
||||
from extensions.ext_database import db
|
||||
from extensions.ext_redis import redis_client
|
||||
from models.dataset import TidbAuthBinding
|
||||
|
||||
|
||||
class TidbService:
|
||||
@staticmethod
|
||||
def create_tidb_serverless_cluster(
|
||||
project_id: str, api_url: str, iam_url: str, public_key: str, private_key: str, region: str
|
||||
):
|
||||
"""
|
||||
Creates a new TiDB Serverless cluster.
|
||||
:param project_id: The project ID of the TiDB Cloud project (required).
|
||||
:param api_url: The URL of the TiDB Cloud API (required).
|
||||
:param iam_url: The URL of the TiDB Cloud IAM API (required).
|
||||
:param public_key: The public key for the API (required).
|
||||
:param private_key: The private key for the API (required).
|
||||
:param display_name: The user-friendly display name of the cluster (required).
|
||||
:param region: The region where the cluster will be created (required).
|
||||
|
||||
:return: The response from the API.
|
||||
"""
|
||||
|
||||
region_object = {
|
||||
"name": region,
|
||||
}
|
||||
|
||||
labels = {
|
||||
"tidb.cloud/project": project_id,
|
||||
}
|
||||
|
||||
spending_limit = {
|
||||
"monthly": 100,
|
||||
}
|
||||
password = str(uuid.uuid4()).replace("-", "")[:16]
|
||||
display_name = str(uuid.uuid4()).replace("-", "")[:16]
|
||||
cluster_data = {
|
||||
"displayName": display_name,
|
||||
"region": region_object,
|
||||
"labels": labels,
|
||||
"spendingLimit": spending_limit,
|
||||
"rootPassword": password,
|
||||
}
|
||||
|
||||
response = requests.post(f"{api_url}/clusters", json=cluster_data, auth=HTTPDigestAuth(public_key, private_key))
|
||||
|
||||
if response.status_code == 200:
|
||||
response_data = response.json()
|
||||
cluster_id = response_data["clusterId"]
|
||||
retry_count = 0
|
||||
max_retries = 30
|
||||
while retry_count < max_retries:
|
||||
cluster_response = TidbService.get_tidb_serverless_cluster(api_url, public_key, private_key, cluster_id)
|
||||
if cluster_response["state"] == "ACTIVE":
|
||||
user_prefix = cluster_response["userPrefix"]
|
||||
return {
|
||||
"cluster_id": cluster_id,
|
||||
"cluster_name": display_name,
|
||||
"account": f"{user_prefix}.root",
|
||||
"password": password,
|
||||
}
|
||||
time.sleep(30) # wait 30 seconds before retrying
|
||||
retry_count += 1
|
||||
else:
|
||||
response.raise_for_status()
|
||||
|
||||
@staticmethod
|
||||
def delete_tidb_serverless_cluster(api_url: str, public_key: str, private_key: str, cluster_id: str):
|
||||
"""
|
||||
Deletes a specific TiDB Serverless cluster.
|
||||
|
||||
:param api_url: The URL of the TiDB Cloud API (required).
|
||||
:param public_key: The public key for the API (required).
|
||||
:param private_key: The private key for the API (required).
|
||||
:param cluster_id: The ID of the cluster to be deleted (required).
|
||||
:return: The response from the API.
|
||||
"""
|
||||
|
||||
response = requests.delete(f"{api_url}/clusters/{cluster_id}", auth=HTTPDigestAuth(public_key, private_key))
|
||||
|
||||
if response.status_code == 200:
|
||||
return response.json()
|
||||
else:
|
||||
response.raise_for_status()
|
||||
|
||||
@staticmethod
|
||||
def get_tidb_serverless_cluster(api_url: str, public_key: str, private_key: str, cluster_id: str):
|
||||
"""
|
||||
Deletes a specific TiDB Serverless cluster.
|
||||
|
||||
:param api_url: The URL of the TiDB Cloud API (required).
|
||||
:param public_key: The public key for the API (required).
|
||||
:param private_key: The private key for the API (required).
|
||||
:param cluster_id: The ID of the cluster to be deleted (required).
|
||||
:return: The response from the API.
|
||||
"""
|
||||
|
||||
response = requests.get(f"{api_url}/clusters/{cluster_id}", auth=HTTPDigestAuth(public_key, private_key))
|
||||
|
||||
if response.status_code == 200:
|
||||
return response.json()
|
||||
else:
|
||||
response.raise_for_status()
|
||||
|
||||
@staticmethod
|
||||
def change_tidb_serverless_root_password(
|
||||
api_url: str, public_key: str, private_key: str, cluster_id: str, account: str, new_password: str
|
||||
):
|
||||
"""
|
||||
Changes the root password of a specific TiDB Serverless cluster.
|
||||
|
||||
:param api_url: The URL of the TiDB Cloud API (required).
|
||||
:param public_key: The public key for the API (required).
|
||||
:param private_key: The private key for the API (required).
|
||||
:param cluster_id: The ID of the cluster for which the password is to be changed (required).+
|
||||
:param account: The account for which the password is to be changed (required).
|
||||
:param new_password: The new password for the root user (required).
|
||||
:return: The response from the API.
|
||||
"""
|
||||
|
||||
body = {"password": new_password, "builtinRole": "role_admin", "customRoles": []}
|
||||
|
||||
response = requests.patch(
|
||||
f"{api_url}/clusters/{cluster_id}/sqlUsers/{account}",
|
||||
json=body,
|
||||
auth=HTTPDigestAuth(public_key, private_key),
|
||||
)
|
||||
|
||||
if response.status_code == 200:
|
||||
return response.json()
|
||||
else:
|
||||
response.raise_for_status()
|
||||
|
||||
@staticmethod
|
||||
def batch_update_tidb_serverless_cluster_status(
|
||||
tidb_serverless_list: list[TidbAuthBinding],
|
||||
project_id: str,
|
||||
api_url: str,
|
||||
iam_url: str,
|
||||
public_key: str,
|
||||
private_key: str,
|
||||
) -> list[dict]:
|
||||
"""
|
||||
Update the status of a new TiDB Serverless cluster.
|
||||
:param project_id: The project ID of the TiDB Cloud project (required).
|
||||
:param api_url: The URL of the TiDB Cloud API (required).
|
||||
:param iam_url: The URL of the TiDB Cloud IAM API (required).
|
||||
:param public_key: The public key for the API (required).
|
||||
:param private_key: The private key for the API (required).
|
||||
:param display_name: The user-friendly display name of the cluster (required).
|
||||
:param region: The region where the cluster will be created (required).
|
||||
|
||||
:return: The response from the API.
|
||||
"""
|
||||
clusters = []
|
||||
tidb_serverless_list_map = {item.cluster_id: item for item in tidb_serverless_list}
|
||||
cluster_ids = [item.cluster_id for item in tidb_serverless_list]
|
||||
params = {"clusterIds": cluster_ids, "view": "FULL"}
|
||||
response = requests.get(
|
||||
f"{api_url}/clusters:batchGet", params=params, auth=HTTPDigestAuth(public_key, private_key)
|
||||
)
|
||||
|
||||
if response.status_code == 200:
|
||||
response_data = response.json()
|
||||
cluster_infos = []
|
||||
for item in response_data["clusters"]:
|
||||
state = item["state"]
|
||||
userPrefix = item["userPrefix"]
|
||||
if state == "ACTIVE" and len(userPrefix) > 0:
|
||||
cluster_info = tidb_serverless_list_map[item["clusterId"]]
|
||||
cluster_info.status = "ACTIVE"
|
||||
cluster_info.account = f"{userPrefix}.root"
|
||||
db.session.add(cluster_info)
|
||||
db.session.commit()
|
||||
else:
|
||||
response.raise_for_status()
|
||||
|
||||
@staticmethod
|
||||
def batch_create_tidb_serverless_cluster(
|
||||
batch_size: int, project_id: str, api_url: str, iam_url: str, public_key: str, private_key: str, region: str
|
||||
) -> list[dict]:
|
||||
"""
|
||||
Creates a new TiDB Serverless cluster.
|
||||
:param project_id: The project ID of the TiDB Cloud project (required).
|
||||
:param api_url: The URL of the TiDB Cloud API (required).
|
||||
:param iam_url: The URL of the TiDB Cloud IAM API (required).
|
||||
:param public_key: The public key for the API (required).
|
||||
:param private_key: The private key for the API (required).
|
||||
:param display_name: The user-friendly display name of the cluster (required).
|
||||
:param region: The region where the cluster will be created (required).
|
||||
|
||||
:return: The response from the API.
|
||||
"""
|
||||
clusters = []
|
||||
for _ in range(batch_size):
|
||||
region_object = {
|
||||
"name": region,
|
||||
}
|
||||
|
||||
labels = {
|
||||
"tidb.cloud/project": project_id,
|
||||
}
|
||||
|
||||
spending_limit = {
|
||||
"monthly": 10,
|
||||
}
|
||||
password = str(uuid.uuid4()).replace("-", "")[:16]
|
||||
display_name = str(uuid.uuid4()).replace("-", "")
|
||||
cluster_data = {
|
||||
"cluster": {
|
||||
"displayName": display_name,
|
||||
"region": region_object,
|
||||
"labels": labels,
|
||||
"spendingLimit": spending_limit,
|
||||
"rootPassword": password,
|
||||
}
|
||||
}
|
||||
cache_key = f"tidb_serverless_cluster_password:{display_name}"
|
||||
redis_client.setex(cache_key, 3600, password)
|
||||
clusters.append(cluster_data)
|
||||
|
||||
request_body = {"requests": clusters}
|
||||
response = requests.post(
|
||||
f"{api_url}/clusters:batchCreate", json=request_body, auth=HTTPDigestAuth(public_key, private_key)
|
||||
)
|
||||
|
||||
if response.status_code == 200:
|
||||
response_data = response.json()
|
||||
cluster_infos = []
|
||||
for item in response_data["clusters"]:
|
||||
cache_key = f"tidb_serverless_cluster_password:{item['displayName']}"
|
||||
password = redis_client.get(cache_key)
|
||||
if not password:
|
||||
continue
|
||||
cluster_info = {
|
||||
"cluster_id": item["clusterId"],
|
||||
"cluster_name": item["displayName"],
|
||||
"account": "root",
|
||||
"password": password.decode("utf-8"),
|
||||
}
|
||||
cluster_infos.append(cluster_info)
|
||||
return cluster_infos
|
||||
else:
|
||||
response.raise_for_status()
|
||||
|
|
@ -9,8 +9,9 @@ from core.rag.datasource.vdb.vector_type import VectorType
|
|||
from core.rag.embedding.cached_embedding import CacheEmbedding
|
||||
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
|
||||
from models.dataset import Dataset, Whitelist
|
||||
|
||||
|
||||
class AbstractVectorFactory(ABC):
|
||||
|
|
@ -35,8 +36,18 @@ class Vector:
|
|||
|
||||
def _init_vector(self) -> BaseVector:
|
||||
vector_type = dify_config.VECTOR_STORE
|
||||
|
||||
if self._dataset.index_struct_dict:
|
||||
vector_type = self._dataset.index_struct_dict["type"]
|
||||
else:
|
||||
if dify_config.VECTOR_STORE_WHITELIST_ENABLE:
|
||||
whitelist = (
|
||||
db.session.query(Whitelist)
|
||||
.filter(Whitelist.tenant_id == self._dataset.tenant_id, Whitelist.category == "vector_db")
|
||||
.one_or_none()
|
||||
)
|
||||
if whitelist:
|
||||
vector_type = VectorType.TIDB_ON_QDRANT
|
||||
|
||||
if not vector_type:
|
||||
raise ValueError("Vector store must be specified.")
|
||||
|
|
@ -103,6 +114,10 @@ class Vector:
|
|||
from core.rag.datasource.vdb.analyticdb.analyticdb_vector import AnalyticdbVectorFactory
|
||||
|
||||
return AnalyticdbVectorFactory
|
||||
case VectorType.COUCHBASE:
|
||||
from core.rag.datasource.vdb.couchbase.couchbase_vector import CouchbaseVectorFactory
|
||||
|
||||
return CouchbaseVectorFactory
|
||||
case VectorType.BAIDU:
|
||||
from core.rag.datasource.vdb.baidu.baidu_vector import BaiduVectorFactory
|
||||
|
||||
|
|
@ -115,6 +130,10 @@ class Vector:
|
|||
from core.rag.datasource.vdb.upstash.upstash_vector import UpstashVectorFactory
|
||||
|
||||
return UpstashVectorFactory
|
||||
case VectorType.TIDB_ON_QDRANT:
|
||||
from core.rag.datasource.vdb.tidb_on_qdrant.tidb_on_qdrant_vector import TidbOnQdrantVectorFactory
|
||||
|
||||
return TidbOnQdrantVectorFactory
|
||||
case _:
|
||||
raise ValueError(f"Vector store {vector_type} is not supported.")
|
||||
|
||||
|
|
|
|||
|
|
@ -16,6 +16,8 @@ class VectorType(str, Enum):
|
|||
TENCENT = "tencent"
|
||||
ORACLE = "oracle"
|
||||
ELASTICSEARCH = "elasticsearch"
|
||||
COUCHBASE = "couchbase"
|
||||
BAIDU = "baidu"
|
||||
VIKINGDB = "vikingdb"
|
||||
UPSTASH = "upstash"
|
||||
TIDB_ON_QDRANT = "tidb_on_qdrant"
|
||||
|
|
|
|||
|
|
@ -21,7 +21,6 @@ from core.rag.extractor.unstructured.unstructured_eml_extractor import Unstructu
|
|||
from core.rag.extractor.unstructured.unstructured_epub_extractor import UnstructuredEpubExtractor
|
||||
from core.rag.extractor.unstructured.unstructured_markdown_extractor import UnstructuredMarkdownExtractor
|
||||
from core.rag.extractor.unstructured.unstructured_msg_extractor import UnstructuredMsgExtractor
|
||||
from core.rag.extractor.unstructured.unstructured_pdf_extractor import UnstructuredPDFExtractor
|
||||
from core.rag.extractor.unstructured.unstructured_ppt_extractor import UnstructuredPPTExtractor
|
||||
from core.rag.extractor.unstructured.unstructured_pptx_extractor import UnstructuredPPTXExtractor
|
||||
from core.rag.extractor.unstructured.unstructured_text_extractor import UnstructuredTextExtractor
|
||||
|
|
@ -103,7 +102,7 @@ class ExtractProcessor:
|
|||
if file_extension in {".xlsx", ".xls"}:
|
||||
extractor = ExcelExtractor(file_path)
|
||||
elif file_extension == ".pdf":
|
||||
extractor = UnstructuredPDFExtractor(file_path, unstructured_api_url, unstructured_api_key)
|
||||
extractor = PdfExtractor(file_path)
|
||||
elif file_extension in {".md", ".markdown"}:
|
||||
extractor = (
|
||||
UnstructuredMarkdownExtractor(file_path, unstructured_api_url, unstructured_api_key)
|
||||
|
|
@ -122,6 +121,8 @@ class ExtractProcessor:
|
|||
extractor = UnstructuredEmailExtractor(file_path, unstructured_api_url, unstructured_api_key)
|
||||
elif file_extension == ".ppt":
|
||||
extractor = UnstructuredPPTExtractor(file_path, unstructured_api_url, unstructured_api_key)
|
||||
# You must first specify the API key
|
||||
# because unstructured_api_key is necessary to parse .ppt documents
|
||||
elif file_extension == ".pptx":
|
||||
extractor = UnstructuredPPTXExtractor(file_path, unstructured_api_url, unstructured_api_key)
|
||||
elif file_extension == ".xml":
|
||||
|
|
|
|||
|
|
@ -234,7 +234,7 @@ class WordExtractor(BaseExtractor):
|
|||
def parse_paragraph(paragraph):
|
||||
paragraph_content = []
|
||||
for run in paragraph.runs:
|
||||
if hasattr(run.element, "tag") and isinstance(element.tag, str) and run.element.tag.endswith("r"):
|
||||
if hasattr(run.element, "tag") and isinstance(run.element.tag, str) and run.element.tag.endswith("r"):
|
||||
drawing_elements = run.element.findall(
|
||||
".//{http://schemas.openxmlformats.org/wordprocessingml/2006/main}drawing"
|
||||
)
|
||||
|
|
|
|||
|
|
@ -204,7 +204,7 @@ class ToolParameter(BaseModel):
|
|||
return str(value)
|
||||
|
||||
except Exception:
|
||||
raise ValueError(f"The tool parameter value {value} is not in correct type of {parameter_type}.")
|
||||
raise ValueError(f"The tool parameter value {value} is not in correct type.")
|
||||
|
||||
class ToolParameterForm(Enum):
|
||||
SCHEMA = "schema" # should be set while adding tool
|
||||
|
|
|
|||
|
|
@ -1,10 +1,3 @@
|
|||
"""
|
||||
语雀客户端
|
||||
"""
|
||||
|
||||
__author__ = "佐井"
|
||||
__created__ = "2024-06-01 09:45:20"
|
||||
|
||||
from typing import Any
|
||||
|
||||
import requests
|
||||
|
|
@ -29,14 +22,13 @@ class AliYuqueTool:
|
|||
session = requests.Session()
|
||||
session.headers.update({"accept": "application/json", "X-Auth-Token": token})
|
||||
new_params = {**tool_parameters}
|
||||
# 找出需要替换的变量
|
||||
|
||||
replacements = {k: v for k, v in new_params.items() if f"{{{k}}}" in path}
|
||||
|
||||
# 替换 path 中的变量
|
||||
for key, value in replacements.items():
|
||||
path = path.replace(f"{{{key}}}", str(value))
|
||||
del new_params[key] # 从 kwargs 中删除已经替换的变量
|
||||
# 请求接口
|
||||
del new_params[key]
|
||||
|
||||
if method.upper() in {"POST", "PUT"}:
|
||||
session.headers.update(
|
||||
{
|
||||
|
|
|
|||
|
|
@ -1,10 +1,3 @@
|
|||
"""
|
||||
创建文档
|
||||
"""
|
||||
|
||||
__author__ = "佐井"
|
||||
__created__ = "2024-06-01 10:45:20"
|
||||
|
||||
from typing import Any, Union
|
||||
|
||||
from core.tools.entities.tool_entities import ToolInvokeMessage
|
||||
|
|
|
|||
|
|
@ -13,7 +13,7 @@ description:
|
|||
|
||||
parameters:
|
||||
- name: book_id
|
||||
type: number
|
||||
type: string
|
||||
required: true
|
||||
form: llm
|
||||
label:
|
||||
|
|
|
|||
|
|
@ -1,11 +1,3 @@
|
|||
#!/usr/bin/env python3
|
||||
"""
|
||||
删除文档
|
||||
"""
|
||||
|
||||
__author__ = "佐井"
|
||||
__created__ = "2024-09-17 22:04"
|
||||
|
||||
from typing import Any, Union
|
||||
|
||||
from core.tools.entities.tool_entities import ToolInvokeMessage
|
||||
|
|
|
|||
|
|
@ -13,7 +13,7 @@ description:
|
|||
|
||||
parameters:
|
||||
- name: book_id
|
||||
type: number
|
||||
type: string
|
||||
required: true
|
||||
form: llm
|
||||
label:
|
||||
|
|
|
|||
|
|
@ -1,10 +1,3 @@
|
|||
"""
|
||||
获取知识库首页
|
||||
"""
|
||||
|
||||
__author__ = "佐井"
|
||||
__created__ = "2024-06-01 22:57:14"
|
||||
|
||||
from typing import Any, Union
|
||||
|
||||
from core.tools.entities.tool_entities import ToolInvokeMessage
|
||||
|
|
|
|||
|
|
@ -1,11 +1,3 @@
|
|||
#!/usr/bin/env python3
|
||||
"""
|
||||
获取知识库目录
|
||||
"""
|
||||
|
||||
__author__ = "佐井"
|
||||
__created__ = "2024-09-17 15:17:11"
|
||||
|
||||
from typing import Any, Union
|
||||
|
||||
from core.tools.entities.tool_entities import ToolInvokeMessage
|
||||
|
|
|
|||
|
|
@ -13,7 +13,7 @@ description:
|
|||
|
||||
parameters:
|
||||
- name: book_id
|
||||
type: number
|
||||
type: string
|
||||
required: true
|
||||
form: llm
|
||||
label:
|
||||
|
|
|
|||
|
|
@ -1,10 +1,3 @@
|
|||
"""
|
||||
获取文档
|
||||
"""
|
||||
|
||||
__author__ = "佐井"
|
||||
__created__ = "2024-06-02 07:11:45"
|
||||
|
||||
import json
|
||||
from typing import Any, Union
|
||||
from urllib.parse import urlparse
|
||||
|
|
@ -37,7 +30,6 @@ class AliYuqueDescribeDocumentContentTool(AliYuqueTool, BuiltinTool):
|
|||
book_slug = path_parts[-2]
|
||||
group_id = path_parts[-3]
|
||||
|
||||
# 1. 请求首页信息，获取book_id
|
||||
new_params["group_login"] = group_id
|
||||
new_params["book_slug"] = book_slug
|
||||
index_page = json.loads(
|
||||
|
|
@ -46,7 +38,7 @@ class AliYuqueDescribeDocumentContentTool(AliYuqueTool, BuiltinTool):
|
|||
book_id = index_page.get("data", {}).get("book", {}).get("id")
|
||||
if not book_id:
|
||||
raise Exception(f"can not parse book_id from {index_page}")
|
||||
# 2. 获取文档内容
|
||||
|
||||
new_params["book_id"] = book_id
|
||||
new_params["id"] = doc_id
|
||||
data = self.request("GET", token, new_params, "/api/v2/repos/{book_id}/docs/{id}")
|
||||
|
|
|
|||
|
|
@ -1,10 +1,3 @@
|
|||
"""
|
||||
获取文档
|
||||
"""
|
||||
|
||||
__author__ = "佐井"
|
||||
__created__ = "2024-06-01 10:45:20"
|
||||
|
||||
from typing import Any, Union
|
||||
|
||||
from core.tools.entities.tool_entities import ToolInvokeMessage
|
||||
|
|
|
|||
|
|
@ -14,7 +14,7 @@ description:
|
|||
|
||||
parameters:
|
||||
- name: book_id
|
||||
type: number
|
||||
type: string
|
||||
required: true
|
||||
form: llm
|
||||
label:
|
||||
|
|
|
|||
|
|
@ -1,11 +1,3 @@
|
|||
#!/usr/bin/env python3
|
||||
"""
|
||||
获取知识库目录
|
||||
"""
|
||||
|
||||
__author__ = "佐井"
|
||||
__created__ = "2024-09-17 15:17:11"
|
||||
|
||||
from typing import Any, Union
|
||||
|
||||
from core.tools.entities.tool_entities import ToolInvokeMessage
|
||||
|
|
|
|||
|
|
@ -13,7 +13,7 @@ description:
|
|||
|
||||
parameters:
|
||||
- name: book_id
|
||||
type: number
|
||||
type: string
|
||||
required: true
|
||||
form: llm
|
||||
label:
|
||||
|
|
|
|||
|
|
@ -1,10 +1,3 @@
|
|||
"""
|
||||
更新文档
|
||||
"""
|
||||
|
||||
__author__ = "佐井"
|
||||
__created__ = "2024-06-19 16:50:07"
|
||||
|
||||
from typing import Any, Union
|
||||
|
||||
from core.tools.entities.tool_entities import ToolInvokeMessage
|
||||
|
|
|
|||
|
|
@ -12,7 +12,7 @@ description:
|
|||
llm: Update doc in a knowledge base via ID/path.
|
||||
parameters:
|
||||
- name: book_id
|
||||
type: number
|
||||
type: string
|
||||
required: true
|
||||
form: llm
|
||||
label:
|
||||
|
|
|
|||
BIN
api/core/tools/provider/builtin/baidu_translate/_assets/icon.png
Normal file
BIN
api/core/tools/provider/builtin/baidu_translate/_assets/icon.png
Normal file
Binary file not shown.
|
After Width: | Height: | Size: 16 KiB |
|
|
@ -0,0 +1,11 @@
|
|||
from hashlib import md5
|
||||
|
||||
|
||||
class BaiduTranslateToolBase:
|
||||
def _get_sign(self, appid, secret, salt, query):
|
||||
"""
|
||||
get baidu translate sign
|
||||
"""
|
||||
# concatenate the string in the order of appid+q+salt+secret
|
||||
str = appid + query + salt + secret
|
||||
return md5(str.encode("utf-8")).hexdigest()
|
||||
|
|
@ -0,0 +1,17 @@
|
|||
from typing import Any
|
||||
|
||||
from core.tools.errors import ToolProviderCredentialValidationError
|
||||
from core.tools.provider.builtin.baidu_translate.tools.translate import BaiduTranslateTool
|
||||
from core.tools.provider.builtin_tool_provider import BuiltinToolProviderController
|
||||
|
||||
|
||||
class BaiduTranslateProvider(BuiltinToolProviderController):
|
||||
def _validate_credentials(self, credentials: dict[str, Any]) -> None:
|
||||
try:
|
||||
BaiduTranslateTool().fork_tool_runtime(
|
||||
runtime={
|
||||
"credentials": credentials,
|
||||
}
|
||||
).invoke(user_id="", tool_parameters={"q": "这是一段测试文本", "from": "auto", "to": "en"})
|
||||
except Exception as e:
|
||||
raise ToolProviderCredentialValidationError(str(e))
|
||||
|
|
@ -0,0 +1,39 @@
|
|||
identity:
|
||||
author: Xiao Ley
|
||||
name: baidu_translate
|
||||
label:
|
||||
en_US: Baidu Translate
|
||||
zh_Hans: 百度翻译
|
||||
description:
|
||||
en_US: Translate text using Baidu
|
||||
zh_Hans: 使用百度进行翻译
|
||||
icon: icon.png
|
||||
tags:
|
||||
- utilities
|
||||
credentials_for_provider:
|
||||
appid:
|
||||
type: secret-input
|
||||
required: true
|
||||
label:
|
||||
en_US: Baidu translate appid
|
||||
zh_Hans: Baidu translate appid
|
||||
placeholder:
|
||||
en_US: Please input your Baidu translate appid
|
||||
zh_Hans: 请输入你的百度翻译 appid
|
||||
help:
|
||||
en_US: Get your Baidu translate appid from Baidu translate
|
||||
zh_Hans: 从百度翻译开放平台获取你的 appid
|
||||
url: https://api.fanyi.baidu.com
|
||||
secret:
|
||||
type: secret-input
|
||||
required: true
|
||||
label:
|
||||
en_US: Baidu translate secret
|
||||
zh_Hans: Baidu translate secret
|
||||
placeholder:
|
||||
en_US: Please input your Baidu translate secret
|
||||
zh_Hans: 请输入你的百度翻译 secret
|
||||
help:
|
||||
en_US: Get your Baidu translate secret from Baidu translate
|
||||
zh_Hans: 从百度翻译开放平台获取你的 secret
|
||||
url: https://api.fanyi.baidu.com
|
||||
|
|
@ -0,0 +1,78 @@
|
|||
import random
|
||||
from hashlib import md5
|
||||
from typing import Any, Union
|
||||
|
||||
import requests
|
||||
|
||||
from core.tools.entities.tool_entities import ToolInvokeMessage
|
||||
from core.tools.provider.builtin.baidu_translate._baidu_translate_tool_base import BaiduTranslateToolBase
|
||||
from core.tools.tool.builtin_tool import BuiltinTool
|
||||
|
||||
|
||||
class BaiduFieldTranslateTool(BuiltinTool, BaiduTranslateToolBase):
|
||||
def _invoke(
|
||||
self,
|
||||
user_id: str,
|
||||
tool_parameters: dict[str, Any],
|
||||
) -> Union[ToolInvokeMessage, list[ToolInvokeMessage]]:
|
||||
"""
|
||||
invoke tools
|
||||
"""
|
||||
BAIDU_FIELD_TRANSLATE_URL = "https://fanyi-api.baidu.com/api/trans/vip/fieldtranslate"
|
||||
|
||||
appid = self.runtime.credentials.get("appid", "")
|
||||
if not appid:
|
||||
raise ValueError("invalid baidu translate appid")
|
||||
|
||||
secret = self.runtime.credentials.get("secret", "")
|
||||
if not secret:
|
||||
raise ValueError("invalid baidu translate secret")
|
||||
|
||||
q = tool_parameters.get("q", "")
|
||||
if not q:
|
||||
raise ValueError("Please input text to translate")
|
||||
|
||||
from_ = tool_parameters.get("from", "")
|
||||
if not from_:
|
||||
raise ValueError("Please select source language")
|
||||
|
||||
to = tool_parameters.get("to", "")
|
||||
if not to:
|
||||
raise ValueError("Please select destination language")
|
||||
|
||||
domain = tool_parameters.get("domain", "")
|
||||
if not domain:
|
||||
raise ValueError("Please select domain")
|
||||
|
||||
salt = str(random.randint(32768, 16777215))
|
||||
sign = self._get_sign(appid, secret, salt, q, domain)
|
||||
|
||||
headers = {"Content-Type": "application/x-www-form-urlencoded"}
|
||||
params = {
|
||||
"q": q,
|
||||
"from": from_,
|
||||
"to": to,
|
||||
"appid": appid,
|
||||
"salt": salt,
|
||||
"domain": domain,
|
||||
"sign": sign,
|
||||
"needIntervene": 1,
|
||||
}
|
||||
try:
|
||||
response = requests.post(BAIDU_FIELD_TRANSLATE_URL, headers=headers, data=params)
|
||||
result = response.json()
|
||||
|
||||
if "trans_result" in result:
|
||||
result_text = result["trans_result"][0]["dst"]
|
||||
else:
|
||||
result_text = f'{result["error_code"]}: {result["error_msg"]}'
|
||||
|
||||
return self.create_text_message(str(result_text))
|
||||
except requests.RequestException as e:
|
||||
raise ValueError(f"Translation service error: {e}")
|
||||
except Exception:
|
||||
raise ValueError("Translation service error, please check the network")
|
||||
|
||||
def _get_sign(self, appid, secret, salt, query, domain):
|
||||
str = appid + query + salt + domain + secret
|
||||
return md5(str.encode("utf-8")).hexdigest()
|
||||
|
|
@ -0,0 +1,123 @@
|
|||
identity:
|
||||
name: field_translate
|
||||
author: Xiao Ley
|
||||
label:
|
||||
en_US: Field translate
|
||||
zh_Hans: 百度领域翻译
|
||||
description:
|
||||
human:
|
||||
en_US: A tool for Baidu Field translate (Currently, the fields of "novel" and "wiki" only support Chinese to English translation. If the language direction is set to English to Chinese, the default output will be a universal translation result).
|
||||
zh_Hans: 百度领域翻译，提供多种领域的文本翻译（目前“网络文学领域”和“人文社科领域”仅支持中到英，如设置语言方向为英到中，则默认输出通用翻译结果）
|
||||
llm: A tool for Baidu Field translate
|
||||
parameters:
|
||||
- name: q
|
||||
type: string
|
||||
required: true
|
||||
label:
|
||||
en_US: Text content
|
||||
zh_Hans: 文本内容
|
||||
human_description:
|
||||
en_US: Text content to be translated
|
||||
zh_Hans: 需要翻译的文本内容
|
||||
llm_description: Text content to be translated
|
||||
form: llm
|
||||
- name: from
|
||||
type: select
|
||||
required: true
|
||||
label:
|
||||
en_US: source language
|
||||
zh_Hans: 源语言
|
||||
human_description:
|
||||
en_US: The source language of the input text
|
||||
zh_Hans: 输入的文本的源语言
|
||||
default: auto
|
||||
form: form
|
||||
options:
|
||||
- value: auto
|
||||
label:
|
||||
en_US: auto
|
||||
zh_Hans: 自动检测
|
||||
- value: zh
|
||||
label:
|
||||
en_US: Chinese
|
||||
zh_Hans: 中文
|
||||
- value: en
|
||||
label:
|
||||
en_US: English
|
||||
zh_Hans: 英语
|
||||
- name: to
|
||||
type: select
|
||||
required: true
|
||||
label:
|
||||
en_US: destination language
|
||||
zh_Hans: 目标语言
|
||||
human_description:
|
||||
en_US: The destination language of the input text
|
||||
zh_Hans: 输入文本的目标语言
|
||||
default: en
|
||||
form: form
|
||||
options:
|
||||
- value: zh
|
||||
label:
|
||||
en_US: Chinese
|
||||
zh_Hans: 中文
|
||||
- value: en
|
||||
label:
|
||||
en_US: English
|
||||
zh_Hans: 英语
|
||||
- name: domain
|
||||
type: select
|
||||
required: true
|
||||
label:
|
||||
en_US: domain
|
||||
zh_Hans: 领域
|
||||
human_description:
|
||||
en_US: The domain of the input text
|
||||
zh_Hans: 输入文本的领域
|
||||
default: novel
|
||||
form: form
|
||||
options:
|
||||
- value: it
|
||||
label:
|
||||
en_US: it
|
||||
zh_Hans: 信息技术领域
|
||||
- value: finance
|
||||
label:
|
||||
en_US: finance
|
||||
zh_Hans: 金融财经领域
|
||||
- value: machinery
|
||||
label:
|
||||
en_US: machinery
|
||||
zh_Hans: 机械制造领域
|
||||
- value: senimed
|
||||
label:
|
||||
en_US: senimed
|
||||
zh_Hans: 生物医药领域
|
||||
- value: novel
|
||||
label:
|
||||
en_US: novel (only support Chinese to English translation)
|
||||
zh_Hans: 网络文学领域（仅支持中到英）
|
||||
- value: academic
|
||||
label:
|
||||
en_US: academic
|
||||
zh_Hans: 学术论文领域
|
||||
- value: aerospace
|
||||
label:
|
||||
en_US: aerospace
|
||||
zh_Hans: 航空航天领域
|
||||
- value: wiki
|
||||
label:
|
||||
en_US: wiki (only support Chinese to English translation)
|
||||
zh_Hans: 人文社科领域（仅支持中到英）
|
||||
- value: news
|
||||
label:
|
||||
en_US: news
|
||||
zh_Hans: 新闻咨询领域
|
||||
- value: law
|
||||
label:
|
||||
en_US: law
|
||||
zh_Hans: 法律法规领域
|
||||
- value: contract
|
||||
label:
|
||||
en_US: contract
|
||||
zh_Hans: 合同领域
|
||||
|
|
@ -0,0 +1,95 @@
|
|||
import random
|
||||
from typing import Any, Union
|
||||
|
||||
import requests
|
||||
|
||||
from core.tools.entities.tool_entities import ToolInvokeMessage
|
||||
from core.tools.provider.builtin.baidu_translate._baidu_translate_tool_base import BaiduTranslateToolBase
|
||||
from core.tools.tool.builtin_tool import BuiltinTool
|
||||
|
||||
|
||||
class BaiduLanguageTool(BuiltinTool, BaiduTranslateToolBase):
|
||||
def _invoke(
|
||||
self,
|
||||
user_id: str,
|
||||
tool_parameters: dict[str, Any],
|
||||
) -> Union[ToolInvokeMessage, list[ToolInvokeMessage]]:
|
||||
"""
|
||||
invoke tools
|
||||
"""
|
||||
BAIDU_LANGUAGE_URL = "https://fanyi-api.baidu.com/api/trans/vip/language"
|
||||
|
||||
appid = self.runtime.credentials.get("appid", "")
|
||||
if not appid:
|
||||
raise ValueError("invalid baidu translate appid")
|
||||
|
||||
secret = self.runtime.credentials.get("secret", "")
|
||||
if not secret:
|
||||
raise ValueError("invalid baidu translate secret")
|
||||
|
||||
q = tool_parameters.get("q", "")
|
||||
if not q:
|
||||
raise ValueError("Please input text to translate")
|
||||
|
||||
description_language = tool_parameters.get("description_language", "English")
|
||||
|
||||
salt = str(random.randint(32768, 16777215))
|
||||
sign = self._get_sign(appid, secret, salt, q)
|
||||
|
||||
headers = {"Content-Type": "application/x-www-form-urlencoded"}
|
||||
params = {
|
||||
"q": q,
|
||||
"appid": appid,
|
||||
"salt": salt,
|
||||
"sign": sign,
|
||||
}
|
||||
|
||||
try:
|
||||
response = requests.post(BAIDU_LANGUAGE_URL, params=params, headers=headers)
|
||||
result = response.json()
|
||||
if "error_code" not in result:
|
||||
raise ValueError("Translation service error, please check the network")
|
||||
|
||||
result_text = ""
|
||||
if result["error_code"] != 0:
|
||||
result_text = f'{result["error_code"]}: {result["error_msg"]}'
|
||||
else:
|
||||
result_text = result["data"]["src"]
|
||||
result_text = self.mapping_result(description_language, result_text)
|
||||
|
||||
return self.create_text_message(result_text)
|
||||
except requests.RequestException as e:
|
||||
raise ValueError(f"Translation service error: {e}")
|
||||
except Exception:
|
||||
raise ValueError("Translation service error, please check the network")
|
||||
|
||||
def mapping_result(self, description_language: str, result: str) -> str:
|
||||
"""
|
||||
mapping result
|
||||
"""
|
||||
mapping = {
|
||||
"English": {
|
||||
"zh": "Chinese",
|
||||
"en": "English",
|
||||
"jp": "Japanese",
|
||||
"kor": "Korean",
|
||||
"th": "Thai",
|
||||
"vie": "Vietnamese",
|
||||
"ru": "Russian",
|
||||
},
|
||||
"Chinese": {
|
||||
"zh": "中文",
|
||||
"en": "英文",
|
||||
"jp": "日文",
|
||||
"kor": "韩文",
|
||||
"th": "泰语",
|
||||
"vie": "越南语",
|
||||
"ru": "俄语",
|
||||
},
|
||||
}
|
||||
|
||||
language_mapping = mapping.get(description_language)
|
||||
if not language_mapping:
|
||||
return result
|
||||
|
||||
return language_mapping.get(result, result)
|
||||
|
|
@ -0,0 +1,43 @@
|
|||
identity:
|
||||
name: language
|
||||
author: Xiao Ley
|
||||
label:
|
||||
en_US: Baidu Language
|
||||
zh_Hans: 百度语种识别
|
||||
description:
|
||||
human:
|
||||
en_US: A tool for Baidu Language, support Chinese, English, Japanese, Korean, Thai, Vietnamese and Russian
|
||||
zh_Hans: 使用百度进行语种识别，支持的语种：中文、英语、日语、韩语、泰语、越南语和俄语
|
||||
llm: A tool for Baidu Language
|
||||
parameters:
|
||||
- name: q
|
||||
type: string
|
||||
required: true
|
||||
label:
|
||||
en_US: Text content
|
||||
zh_Hans: 文本内容
|
||||
human_description:
|
||||
en_US: Text content to be recognized
|
||||
zh_Hans: 需要识别语言的文本内容
|
||||
llm_description: Text content to be recognized
|
||||
form: llm
|
||||
- name: description_language
|
||||
type: select
|
||||
required: true
|
||||
label:
|
||||
en_US: Description language
|
||||
zh_Hans: 描述语言
|
||||
human_description:
|
||||
en_US: Describe the language used to identify the results
|
||||
zh_Hans: 描述识别结果所用的语言
|
||||
default: Chinese
|
||||
form: form
|
||||
options:
|
||||
- value: Chinese
|
||||
label:
|
||||
en_US: Chinese
|
||||
zh_Hans: 中文
|
||||
- value: English
|
||||
label:
|
||||
en_US: English
|
||||
zh_Hans: 英语
|
||||
|
|
@ -0,0 +1,67 @@
|
|||
import random
|
||||
from typing import Any, Union
|
||||
|
||||
import requests
|
||||
|
||||
from core.tools.entities.tool_entities import ToolInvokeMessage
|
||||
from core.tools.provider.builtin.baidu_translate._baidu_translate_tool_base import BaiduTranslateToolBase
|
||||
from core.tools.tool.builtin_tool import BuiltinTool
|
||||
|
||||
|
||||
class BaiduTranslateTool(BuiltinTool, BaiduTranslateToolBase):
|
||||
def _invoke(
|
||||
self,
|
||||
user_id: str,
|
||||
tool_parameters: dict[str, Any],
|
||||
) -> Union[ToolInvokeMessage, list[ToolInvokeMessage]]:
|
||||
"""
|
||||
invoke tools
|
||||
"""
|
||||
BAIDU_TRANSLATE_URL = "https://fanyi-api.baidu.com/api/trans/vip/translate"
|
||||
|
||||
appid = self.runtime.credentials.get("appid", "")
|
||||
if not appid:
|
||||
raise ValueError("invalid baidu translate appid")
|
||||
|
||||
secret = self.runtime.credentials.get("secret", "")
|
||||
if not secret:
|
||||
raise ValueError("invalid baidu translate secret")
|
||||
|
||||
q = tool_parameters.get("q", "")
|
||||
if not q:
|
||||
raise ValueError("Please input text to translate")
|
||||
|
||||
from_ = tool_parameters.get("from", "")
|
||||
if not from_:
|
||||
raise ValueError("Please select source language")
|
||||
|
||||
to = tool_parameters.get("to", "")
|
||||
if not to:
|
||||
raise ValueError("Please select destination language")
|
||||
|
||||
salt = str(random.randint(32768, 16777215))
|
||||
sign = self._get_sign(appid, secret, salt, q)
|
||||
|
||||
headers = {"Content-Type": "application/x-www-form-urlencoded"}
|
||||
params = {
|
||||
"q": q,
|
||||
"from": from_,
|
||||
"to": to,
|
||||
"appid": appid,
|
||||
"salt": salt,
|
||||
"sign": sign,
|
||||
}
|
||||
try:
|
||||
response = requests.post(BAIDU_TRANSLATE_URL, params=params, headers=headers)
|
||||
result = response.json()
|
||||
|
||||
if "trans_result" in result:
|
||||
result_text = result["trans_result"][0]["dst"]
|
||||
else:
|
||||
result_text = f'{result["error_code"]}: {result["error_msg"]}'
|
||||
|
||||
return self.create_text_message(str(result_text))
|
||||
except requests.RequestException as e:
|
||||
raise ValueError(f"Translation service error: {e}")
|
||||
except Exception:
|
||||
raise ValueError("Translation service error, please check the network")
|
||||
|
|
@ -0,0 +1,275 @@
|
|||
identity:
|
||||
name: translate
|
||||
author: Xiao Ley
|
||||
label:
|
||||
en_US: Translate
|
||||
zh_Hans: 百度翻译
|
||||
description:
|
||||
human:
|
||||
en_US: A tool for Baidu Translate
|
||||
zh_Hans: 百度翻译
|
||||
llm: A tool for Baidu Translate
|
||||
parameters:
|
||||
- name: q
|
||||
type: string
|
||||
required: true
|
||||
label:
|
||||
en_US: Text content
|
||||
zh_Hans: 文本内容
|
||||
human_description:
|
||||
en_US: Text content to be translated
|
||||
zh_Hans: 需要翻译的文本内容
|
||||
llm_description: Text content to be translated
|
||||
form: llm
|
||||
- name: from
|
||||
type: select
|
||||
required: true
|
||||
label:
|
||||
en_US: source language
|
||||
zh_Hans: 源语言
|
||||
human_description:
|
||||
en_US: The source language of the input text
|
||||
zh_Hans: 输入的文本的源语言
|
||||
default: auto
|
||||
form: form
|
||||
options:
|
||||
- value: auto
|
||||
label:
|
||||
en_US: auto
|
||||
zh_Hans: 自动检测
|
||||
- value: zh
|
||||
label:
|
||||
en_US: Chinese
|
||||
zh_Hans: 中文
|
||||
- value: en
|
||||
label:
|
||||
en_US: English
|
||||
zh_Hans: 英语
|
||||
- value: cht
|
||||
label:
|
||||
en_US: Traditional Chinese
|
||||
zh_Hans: 繁体中文
|
||||
- value: yue
|
||||
label:
|
||||
en_US: Yue
|
||||
zh_Hans: 粤语
|
||||
- value: wyw
|
||||
label:
|
||||
en_US: Wyw
|
||||
zh_Hans: 文言文
|
||||
- value: jp
|
||||
label:
|
||||
en_US: Japanese
|
||||
zh_Hans: 日语
|
||||
- value: kor
|
||||
label:
|
||||
en_US: Korean
|
||||
zh_Hans: 韩语
|
||||
- value: fra
|
||||
label:
|
||||
en_US: French
|
||||
zh_Hans: 法语
|
||||
- value: spa
|
||||
label:
|
||||
en_US: Spanish
|
||||
zh_Hans: 西班牙语
|
||||
- value: th
|
||||
label:
|
||||
en_US: Thai
|
||||
zh_Hans: 泰语
|
||||
- value: ara
|
||||
label:
|
||||
en_US: Arabic
|
||||
zh_Hans: 阿拉伯语
|
||||
- value: ru
|
||||
label:
|
||||
en_US: Russian
|
||||
zh_Hans: 俄语
|
||||
- value: pt
|
||||
label:
|
||||
en_US: Portuguese
|
||||
zh_Hans: 葡萄牙语
|
||||
- value: de
|
||||
label:
|
||||
en_US: German
|
||||
zh_Hans: 德语
|
||||
- value: it
|
||||
label:
|
||||
en_US: Italian
|
||||
zh_Hans: 意大利语
|
||||
- value: el
|
||||
label:
|
||||
en_US: Greek
|
||||
zh_Hans: 希腊语
|
||||
- value: nl
|
||||
label:
|
||||
en_US: Dutch
|
||||
zh_Hans: 荷兰语
|
||||
- value: pl
|
||||
label:
|
||||
en_US: Polish
|
||||
zh_Hans: 波兰语
|
||||
- value: bul
|
||||
label:
|
||||
en_US: Bulgarian
|
||||
zh_Hans: 保加利亚语
|
||||
- value: est
|
||||
label:
|
||||
en_US: Estonian
|
||||
zh_Hans: 爱沙尼亚语
|
||||
- value: dan
|
||||
label:
|
||||
en_US: Danish
|
||||
zh_Hans: 丹麦语
|
||||
- value: fin
|
||||
label:
|
||||
en_US: Finnish
|
||||
zh_Hans: 芬兰语
|
||||
- value: cs
|
||||
label:
|
||||
en_US: Czech
|
||||
zh_Hans: 捷克语
|
||||
- value: rom
|
||||
label:
|
||||
en_US: Romanian
|
||||
zh_Hans: 罗马尼亚语
|
||||
- value: slo
|
||||
label:
|
||||
en_US: Slovak
|
||||
zh_Hans: 斯洛文尼亚语
|
||||
- value: swe
|
||||
label:
|
||||
en_US: Swedish
|
||||
zh_Hans: 瑞典语
|
||||
- value: hu
|
||||
label:
|
||||
en_US: Hungarian
|
||||
zh_Hans: 匈牙利语
|
||||
- value: vie
|
||||
label:
|
||||
en_US: Vietnamese
|
||||
zh_Hans: 越南语
|
||||
- name: to
|
||||
type: select
|
||||
required: true
|
||||
label:
|
||||
en_US: destination language
|
||||
zh_Hans: 目标语言
|
||||
human_description:
|
||||
en_US: The destination language of the input text
|
||||
zh_Hans: 输入文本的目标语言
|
||||
default: en
|
||||
form: form
|
||||
options:
|
||||
- value: zh
|
||||
label:
|
||||
en_US: Chinese
|
||||
zh_Hans: 中文
|
||||
- value: en
|
||||
label:
|
||||
en_US: English
|
||||
zh_Hans: 英语
|
||||
- value: cht
|
||||
label:
|
||||
en_US: Traditional Chinese
|
||||
zh_Hans: 繁体中文
|
||||
- value: yue
|
||||
label:
|
||||
en_US: Yue
|
||||
zh_Hans: 粤语
|
||||
- value: wyw
|
||||
label:
|
||||
en_US: Wyw
|
||||
zh_Hans: 文言文
|
||||
- value: jp
|
||||
label:
|
||||
en_US: Japanese
|
||||
zh_Hans: 日语
|
||||
- value: kor
|
||||
label:
|
||||
en_US: Korean
|
||||
zh_Hans: 韩语
|
||||
- value: fra
|
||||
label:
|
||||
en_US: French
|
||||
zh_Hans: 法语
|
||||
- value: spa
|
||||
label:
|
||||
en_US: Spanish
|
||||
zh_Hans: 西班牙语
|
||||
- value: th
|
||||
label:
|
||||
en_US: Thai
|
||||
zh_Hans: 泰语
|
||||
- value: ara
|
||||
label:
|
||||
en_US: Arabic
|
||||
zh_Hans: 阿拉伯语
|
||||
- value: ru
|
||||
label:
|
||||
en_US: Russian
|
||||
zh_Hans: 俄语
|
||||
- value: pt
|
||||
label:
|
||||
en_US: Portuguese
|
||||
zh_Hans: 葡萄牙语
|
||||
- value: de
|
||||
label:
|
||||
en_US: German
|
||||
zh_Hans: 德语
|
||||
- value: it
|
||||
label:
|
||||
en_US: Italian
|
||||
zh_Hans: 意大利语
|
||||
- value: el
|
||||
label:
|
||||
en_US: Greek
|
||||
zh_Hans: 希腊语
|
||||
- value: nl
|
||||
label:
|
||||
en_US: Dutch
|
||||
zh_Hans: 荷兰语
|
||||
- value: pl
|
||||
label:
|
||||
en_US: Polish
|
||||
zh_Hans: 波兰语
|
||||
- value: bul
|
||||
label:
|
||||
en_US: Bulgarian
|
||||
zh_Hans: 保加利亚语
|
||||
- value: est
|
||||
label:
|
||||
en_US: Estonian
|
||||
zh_Hans: 爱沙尼亚语
|
||||
- value: dan
|
||||
label:
|
||||
en_US: Danish
|
||||
zh_Hans: 丹麦语
|
||||
- value: fin
|
||||
label:
|
||||
en_US: Finnish
|
||||
zh_Hans: 芬兰语
|
||||
- value: cs
|
||||
label:
|
||||
en_US: Czech
|
||||
zh_Hans: 捷克语
|
||||
- value: rom
|
||||
label:
|
||||
en_US: Romanian
|
||||
zh_Hans: 罗马尼亚语
|
||||
- value: slo
|
||||
label:
|
||||
en_US: Slovak
|
||||
zh_Hans: 斯洛文尼亚语
|
||||
- value: swe
|
||||
label:
|
||||
en_US: Swedish
|
||||
zh_Hans: 瑞典语
|
||||
- value: hu
|
||||
label:
|
||||
en_US: Hungarian
|
||||
zh_Hans: 匈牙利语
|
||||
- value: vie
|
||||
label:
|
||||
en_US: Vietnamese
|
||||
zh_Hans: 越南语
|
||||
Some files were not shown because too many files have changed in this diff Show More
Loading…
Reference in New Issue
Block a user