A robust, high-performance and user-friendly alternative to the traditional curl-based Stream Load.

• Parallel Loading: Split data files automatically and perform parallel loading
• Support for Multiple Files and Directories: Support multiple files and directories load with one shot
• Path Traversal Support: Support path traversal when the source files are in directories
• Resilience and Continuity: Resume loading from previous failures and cancellations
• Automatic Retry Mechanism: Retry automatically when failure
• Comprehensive and Concise Input Parameters

doris-streamloader --source_file={FILE_LIST} --url={FE_OR_BE_SERVER_URL}:{PORT} --header={STREAMLOAD_HEADER} --db={TARGET_DATABASE} --table={TARGET_TABLE}

• FILE_LIST: directory or file list, support * wildcard
• FE_OR_BE_SERVER_URL & PORT: Doris FE or BE hostname or IP and HTTP port
• STREAMLOAD_HEADER: supports all headers as curl Stream Load does，multiple headers are separated by '?'
• TARGET_DATABASE & TARGET_TABLE: indicate the target database and table where the data will be loaded

e.g.:

doris-streamloader --source_file="data.csv" --url="http://localhost:8330" --header="column_separator:|?columns:col1,col2" --db="testdb" --table="testtbl"

For additional details and options, refer to our comprehensive docs below.

User Guide

中文使用文档

To build Streamloader, ensure you have golang installed (version >= 1.19.9). For example, on CentOS:

yum install golang

Then, navigate to the doris-streamloader directory and execute:

cd doris-streamloader && sh build.sh

Apache License, Version 2.0