如何通过Java互操作在Clojure中从Google BigQuery查询数据？

首先确保你的项目中已添加以下依赖（以Leiningen为例）：

[com.google.cloud/google-cloud-bigquery "2.16.0"]

完整示例代码 #

(ns bigquery-demo.core
  (:import (com.google.cloud.bigquery BigQuery BigQueryException BigQueryOptions QueryJobConfiguration TableResult)))

(defn query-bigquery [query]
  (try
    ;; 初始化BigQuery客户端，建议全局复用而非每次查询创建
    (let [bigquery (.getService (BigQueryOptions/getDefaultInstance))
          query-config (.build (QueryJobConfiguration/newBuilder query))
          results (.query bigquery query-config)]
      ;; 遍历查询结果并打印
      (.forEach (.iterateAll results)
                (fn [row]
                  (.forEach row
                            (fn [val]
                              (printf "%s," (.toString val))))))
      (println "\n查询执行成功。"))
    (catch BigQueryException e
      (println "查询执行失败:\n" (.toString e)))
    (catch InterruptedException e
      (println "查询被中断:\n" (.toString e)))))

(defn -main []
  ;; 替换为你的项目、数据集和表名
  (let [project-id "MY_PROJECT_ID"
        dataset-name "MY_DATASET_NAME"
        table-name "MY_TABLE_NAME"
        query (str "SELECT name, SUM(number) as total_people\n"
                   " FROM `" project-id "." dataset-name "." table-name "`\n"
                   " WHERE state = 'TX'\n"
                   " GROUP BY name, state\n"
                   " ORDER BY total_people DESC\n"
                   " LIMIT 20")]
    (query-bigquery query)))

代码说明 #

• 客户端初始化：BigQueryOptions/getDefaultInstance会自动读取环境变量、本地密钥文件等配置创建客户端，建议全局复用该实例以提升性能。
• 查询配置：通过QueryJobConfiguration/newBuilder构建查询任务配置，还可按需设置超时、目标数据集等额外参数。
• 结果处理：TableResult的iterateAll方法返回所有查询结果行，通过嵌套的forEach遍历每行的字段值。
• 异常处理：捕获BigQueryException处理BigQuery服务端错误，捕获InterruptedException处理查询过程中的中断异常。

内容的提问来源于stack exchange，提问作者Edward J. Stembler