I started a discussion on this before, you can check it in the mail group https://lists.apache.org/thread/o770bc3k623kyfks2mzkt21qsc4g6328 In order to facilitate everyone to organize the documents, I created a new-docs directory under the incubator-doris directory. The new directory structure is below this. I just created a directory structure here, which needs to be rearranged. In the data import scenario, in order to take into account the viewing habits of previous users, the import is organized in two ways: 1. According to the usage scenario: This will give users clearer guidance. For example, the user is the data source of kafka, then the user can directly select the routine to load 2. According to the import method: it is the introduction of the various import methods we provided before In order to facilitate everyone to run and debug locally, I migrated the entire .vuepress under the original document. After completion, you only need to delete the original docs directory and rename the new new-docs directory to docs. At the same time, you can also run it locally, so that you can organize documents and know the content of each document directory. In the local debugging execution, switch to the new-docs directory and execute the following command: ```` npm install npm run dev ```` then through the browser http://ip:port/zh-CN http://ip:port/en
1.6 KiB
1.6 KiB
title, language
| title | language |
|---|---|
| bitmap_union | en |
bitmap_union function
description
Aggregate function, used to calculate the grouped bitmap union. Common usage scenarios such as: calculating PV, UV.
Syntax
BITMAP BITMAP_UNION(BITMAP value)
Enter a set of bitmap values, find the union of this set of bitmap values, and return.
example
mysql> select page_id, bitmap_union(user_id) from table group by page_id;
Combined with the bitmap_count function, the PV data of the web page can be obtained
mysql> select page_id, bitmap_count(bitmap_union(user_id)) from table group by page_id;
When the user_id field is int, the above query semantics is equivalent to
mysql> select page_id, count(distinct user_id) from table group by page_id;
keyword
BITMAP_UNION, BITMAP