上传文件

def load_and_split_text(
    local_file_path: str,
    document_set_name: str | None = None,
    metadata: dict | None = None,
    splitter_process: SplitterProcess | None = None,
    timeout: float | None = None
) -> DocumentSet

import tcvectordb
from tcvectordb.model.ai_database import AIDatabase
from tcvectordb.model.collection_view import CollectionView
from tcvectordb.model.document import Filter, Document
from tcvectordb.model.document_set import DocumentSet
from tcvectordb.model.enum import ReadConsistency
from tcvectordb.model.collection_view import SplitterProcess

#create a database client object
client = tcvectordb.RPCVectorDBClient(url='http://10.0.X.X:80', username='root', key='eC4bLRy2va******************************', read_consistency=ReadConsistency.EVENTUAL_CONSISTENCY, timeout=30)

# Specify the database name
db = client.database('db-test-ai')
coll_view = db.collection_view('coll-ai-files')
# 上传文件
# 1. local_file_path：指定当前文件在客户端的路径
# 2. document_set_name：配置文件存储于向量数据库的名称
# 3. metadata：定义文件 meta 信息字段，示例中指定了author 与 tags
res = coll_view.load_and_split_text(
            local_file_path="/data/home/../腾讯云向量数据库.md", 
            document_set_name="腾讯云向量数据库.md",
            metadata={
                'author': 'Tencent',
                'tags': ['向量', 'Embedding', 'AI']
            },
            splitter_process=SplitterProcess(
                append_keywords_to_chunk=True,
                append_title_to_chunk=False,
                # chunk_splitter 参数以正则表达式的方式配置文档拆分方式，如下：\\n{2,} 代表以两个及以上的换行进行拆分，常用在QA对文件拆分中。
                # chunk_splitter="\\n{2,}",
            ),
    )
print(vars(res))