如何在 SQL Redshift 中对表进行分区后比较两列的值
Posted
技术标签:
【中文标题】如何在 SQL Redshift 中对表进行分区后比较两列的值【英文标题】：How to compare two columns values after partitioning the table in SQL Redshift 【发布时间】：2020-05-27 17:09:00 【问题描述】：我在一个表中有 4 列（Person_ID、Account_Name、Account_ID、Account_IDs）。对于每个人 id，我想查找该 person_ID 的 Account_IDs 列中不存在的所有 Account_ID。下面是一个示例表：
Person_id Account_Name Account_ID Account_IDs
--------------------------------------------------
123 Name3 ,000000ihi4MAQ, ,000000TF5MAHZ,000000TF5MAQQ,000000grVA6AM,000000ihi4MAQ,000016ILMhAO,
123 Name2 ,000016ILMhAO, ,000000TF5MAHZ,000000TF5MAQQ,000000grVA6AM,000000ihi4MAQ,000016ILMhAO,
123 Name1 ,000000grVA6AM, ,000000TF5MAHZ,000000TF5MAQQ,000000grVA6AM,000000ihi4MAQ,000016ILMhAO,
123 Name4 ,000000TF5MAQQ, ,000000TF5MAHZ,000000TF5MAQQ,000000grVA6AM,000000ihi4MAQ,000016ILMhAO,
123 Name5 ,000000TF5MAHZ, ,000000TF5MAHZ,000000TF5MAQQ,000000grVA6AM,000000ihi4MAQ,000016ILMhAO,
124 Name2 ,000016ILMhAO, ,000000frVA6AM,000016ILMhAO,
124 Name7 ,000024ILMhAO, ,000000frVA6AM,000016ILMhAO,
124 Name8 ,000000frVA7XZ, ,000000frVA6AM,000016ILMhAO,
124 Name5 ,000000TF5MAHZ, ,000000frVA6AM,000016ILMhAO,
124 Name6 ,000000frVA6AM, ,000000frVA6AM,000016ILMhAO,
125 Name11 ,000000frXC6A, ,000000frVA6BC,000024ILMhJZ,000000frXC6A,000024YTMhA,
125 Name9 ,000000frVA6BC, ,000000frVA6BC,000024ILMhJZ,000000frXC6A,000024YTMhA,
125 Name10 ,000024ILMhJZ, ,000000frVA6BC,000024ILMhJZ,000000frXC6A,000024YTMhA,
125 Name12 ,000024YTMhA, ,000000frVA6BC,000024ILMhJZ,000000frXC6A,000024YTMhA,
125 Name13 ,000024IXThJY, ,000000frVA6BC,000024ILMhJZ,000000frXC6A,000024YTMhA,
所以从这个示例中，答案应该是：
Person_ID Account_ID Account_Name
-----------------------------------------
124 000000TF5MAHZ Name5
124 000024ILMhAO Name7
124 000000frVA7XZ Name8
125 000024IXThJY Name13
我不明白如何在对表进行分区后比较两列中的值。
提前感谢您的帮助。
【问题讨论】：
您的列的数据类型是什么？您真的将列表存储为字符串吗？如果是这样，请修复您的数据模型！ 【参考方案1】：假设Account_IDs 是一个包含逗号分隔值的字符串，使用：
WHERE Account_IDs NOT LIKE '%' || Account_ID || '%'
【讨论】：
以上是关于如何在 SQL Redshift 中对表进行分区后比较两列的值的主要内容，如果未能解决你的问题，请参考以下文章