select case ( a when a is not null then a else b end) from my_table错在哪

在 SQL 语法中，CASE 表达式应该是这样的：

SELECT CASE WHEN a IS NOT NULL THEN a ELSE b END FROM my_table;

其中，CASE 表达式的语法是：

CASE
    WHEN condition_1 THEN result_1
    WHEN condition_2 THEN result_2
    ...
    ELSE default_result
END

在这个例子中，我们使用了 WHEN 和 THEN 子句来定义条件和结果，并在 CASE 表达式的结尾使用了 ELSE 子句来定义默认结果。注意，CASE 表达式必须以 END 结尾。

另外，根据 SQL 标准，NULL 值之间不能使用等于运算符来比较，应该使用 IS NULL 或 IS NOT NULL。因此，我们在这个例子中使用了 IS NOT NULL 来判断 a 是否为 NULL。

相关问题

你能把pg视图改为sqlserver的视图吗 SELECT c.relname AS table_name, a.attname AS column_name, d.description AS column_comment, CASE WHEN a.attnotnull AND con.conname IS NULL THEN 1 ELSE 0 END AS is_required, CASE WHEN con.conname IS NOT NULL THEN 1 ELSE 0 END AS is_pk, a.attnum AS sort, CASE WHEN "position"(pg_get_expr(ad.adbin, ad.adrelid), ((c.relname::text || '_'::text) || a.attname::text) || '_seq'::text) > 0 THEN 1 ELSE 0 END AS is_increment, btrim( CASE WHEN t.typelem <> 0::oid AND t.typlen = '-1'::integer THEN 'ARRAY'::text ELSE CASE WHEN t.typtype = 'd'::"char" THEN format_type(t.typbasetype, NULL::integer) ELSE format_type(a.atttypid, NULL::integer) END END, '"'::text) AS column_type FROM pg_attribute a JOIN (pg_class c JOIN pg_namespace n ON c.relnamespace = n.oid) ON a.attrelid = c.oid LEFT JOIN pg_description d ON d.objoid = c.oid AND a.attnum = d.objsubid LEFT JOIN pg_constraint con ON con.conrelid = c.oid AND (a.attnum = ANY (con.conkey)) LEFT JOIN pg_attrdef ad ON a.attrelid = ad.adrelid AND a.attnum = ad.adnum LEFT JOIN pg_type t ON a.atttypid = t.oid WHERE (c.relkind = ANY (ARRAY['r'::"char", 'p'::"char"])) AND a.attnum > 0 AND n.nspname = 'public'::name ORDER BY c.relname, a.attnum

抱歉，我无法将此查询转换为SQL Server的视图。这是因为该查询涉及到了一些PostgreSQL特定的函数和表达式，这些在SQL Server中并不存在。你可能需要手动重新编写查询以适应SQL Server的语法和函数。以下是一个可能的SQL Server等效查询的示例，供参考：

SELECT 
    t.name AS table_name,
    c.name AS column_name,
    CAST(ep.value AS NVARCHAR(MAX)) AS column_comment,
    CASE
        WHEN c.is_nullable = 0 AND pk.constraint_column_id IS NULL THEN 1
        ELSE 0
    END AS is_required,
    CASE
        WHEN pk.constraint_column_id IS NOT NULL THEN 1
        ELSE 0
    END AS is_pk,
    c.column_id AS sort,
    CASE
        WHEN COLUMNPROPERTY(c.object_id, c.name, 'IsIdentity') = 1 THEN 1
        ELSE 0
    END AS is_increment,
    CASE
        WHEN t.name = 'varchar' AND c.max_length <> -1 THEN t.name + '(' + CAST(c.max_length AS NVARCHAR(MAX)) + ')'
        WHEN t.name = 'nvarchar' AND c.max_length <> -1 THEN t.name + '(' + CAST(c.max_length / 2 AS NVARCHAR(MAX)) + ')'
        ELSE t.name
    END AS column_type
FROM sys.tables t
INNER JOIN sys.columns c ON t.object_id = c.object_id
LEFT JOIN sys.extended_properties ep ON ep.major_id = t.object_id AND ep.minor_id = c.column_id AND ep.name = 'MS_Description'
LEFT JOIN sys.indexes i ON t.object_id = i.object_id AND i.is_primary_key = 1
LEFT JOIN sys.index_columns ic ON i.object_id = ic.object_id AND i.index_id = ic.index_id AND c.column_id = ic.column_id
LEFT JOIN sys.key_constraints pk ON t.object_id = pk.parent_object_id AND ic.index_column_id = pk.unique_index_id
WHERE t.type = 'U'
AND SCHEMA_NAME(t.schema_id) = 'dbo'
ORDER BY t.name, c.column_id

请注意，这只是一个示例查询，具体取决于你的数据库结构和需求，你可能需要进行一些调整和修改。建议在转换之前备份你的数据库，以防出现意外情况。

mismatched input 'count' expecting <EOF>(line 1, pos 379) == SQL == insert overwrite table spark_odc_data.D_RE_BA_RAD_MAP_H partition(p_hour = '2023060600') select '202306060000' TIME_ID, a.IMSI, a.GJ, a.YYS, count(case when substr(b.IMSI,1,5)<>46000 and b.operate_code=2 then a.IMSI else null end) as MRWZGXQQCS, count(case when substr(b.IMSI,1,5)<>46000 and b.operate_code=2 and b.result<>1 then a.IMSI else null end) as MRWZGXCGCS, case when count(case when substr(b.IMSI,1,5)<>46000 and b.operate_code=2 then a.IMSI else null end)=0 then 0 else count(case when substr(b.IMSI,1,5)<>46000 and b.operate_code=2 and b.result<>1 then a.IMSI else null)/count(case when substr(b.IMSI,1,5)<>46000 and b.operate_code=2 then a.IMSI else null end) as MRWZGXCGL, sum(c.MRZJCS) as MRZJCS, sum(c.MRZJHJCGCS) as MRZJHJCGCS, sum(d.MRBJCS) as MRBJCS, sum(d.MRBJHJCGCS) as MRBJHJCGCS, case when sum(c.MRZJCS)=0 then 0 else sum(c.MRZJHJCGCS)/sum(c.MRZJCS) end as MRZJWLJTL, case when sum(c.MRZJCS)=0 then 0 else sum(d.MRBJHJCGCS)/sum(d.MRBJCS) end as MRBJWLJTL, count(case when substr(b.msc,1,2)<>86 and a.GJ is not null and b.operate_code=2 then a.IMSI else null end) as MCWZGXQQCS, count(case when substr(b.msc,1,2)<>86 and a.GJ is not null and b.operate_code=2 and b.result<>1 then a.IMSI else null end) as MCWZGXCGCS, case when count(case when substr(b.msc,1,2)<>86 and a.GJ is not null and b.operate_code=2 then a.IMSI else null end)=0 then 0 else count(case when substr(b.msc,1,2)<>86 and a.GJ is not null and b.operate_code=2 and b.result<>1 then a.IMSI else null end)/ count(case when substr(b.msc,1,2)<>86 and a.GJ is not null and b.operate_code=2 then a.IMSI else null end) as MCWZGXCGL, now() LOAD_TIME

这是一条 SQL 语句，用于将数据插入到名为 spark_odc_data.D_RE_BA_RAD_MAP_H 的表中，表的分区为 p_hour = '2023060600'。其中包含了多个字段和计算表达式，用于统计各种数据指标。具体的指标包括 MRWZGXQQCS、MRWZGXCGCS、MRWZGXCGL、MRZJCS、MRZJHJCGCS、MRZJWLJTL、MRBJCS、MRBJHJCGCS、MRBJWLJTL、MCWZGXQQCS、MCWZGXCGCS、MCWZGXCGL。最后还有一个 LOAD_TIME 字段，表示数据的加载时间。