将表行合并为具有原始行的列标题的列 [英] joined table rows into columns with column titles from original rows

问题描述

我有三个经过简化的MYSQL表，其以下各列:

I have three MYSQL tables, simplified, with following columns:

ModuleEntries (MDE_ID)
ModuleFields (MDF_ID, MDF_Label)
ModuleEntryFieldValues (MFV_ID, MFV_MDE_ID, MFV_MDF_ID, MFV_Value)

从简化的列列表中可以看到，ModuleEntryFieldValues是说明对于条目，此字段具有此值"的表.

现在，我想在一个固定"记录集中检索此信息.我设法使事情正常进行，但并不能完全满足我的要求.

在另一篇SO文章的帮助下，我设法通过使用游标使可变数量的动态列开始工作.

As you can see from the simplified column list, ModuleEntryFieldValues is the table which states that "for an entry, this field has this value".

Now, I would like to retrieve this information in one "flat" recordset. I have managed to get things working, but not entirely to what i want.

With the help of a different SO article, I managed to get the dynamic, variable amount of columns to work, by using a cursor.

  declare finish int default 0;
  declare fldid int;
  declare str varchar(10000) default 'SELECT *,';
  declare curs cursor for select MDF_ID from ModuleFields group by MDF_ID;
  declare continue handler for not found set finish = 1;
  open curs;
  my_loop:loop
    fetch curs into fldid;
    if finish = 1 then
      leave my_loop;
    end if;
    set str = concat(str, 'max(case when MFV_MDF_ID = ',fldid,' then MFV_Value else null end) as field_',fldid,',');
  end loop;
  close curs;
  set str = substr(str,1,char_length(str)-1);
  set @str = concat(str, ' from ModuleEntries LEFT join ModuleEntryFieldValues ON MDE_ID = MDF_MDE_ID GROUP BY MDE_ID');
  prepare stmt from @str;
  execute stmt;
  deallocate prepare stmt;

这段代码没有要做的是，允许我将MDF_Label的列值作为行的实际列标题.现在，上面的代码为我提供了"field_1，field_2，..."

是否可以联接这三个表，并以MDF_Label作为现在已联接表中的列的行的列标题?

我想要这个...

What this code doesn't do, is allow me to put the column values of MDF_Label as the actual column headers of the rows. Now, the above code gives me "field_1, field_2, ..."

Is it possible to join these three tables, and have the MDF_Label as column header for the rows that are now columns in the joined table ?

I want this...

ModuleEntries  |   ModuleFields         |   ModuleEntryFieldValues
-------------  |   ------------------   |   -----------------------------------
MDE_ID         |   MDF_ID - MDF_Label   |   MFV_MDE_ID - MFV_MDF_ID - MFV_Value
1              |   1      - Height      |   1          - 1          - 120cms
               |   2      - Width       |   1          - 2          - 30cms
               |   3      - Weight      |   

进入此

Recordset
---------
MDE_ID - Height - Width - Weight
1      - 120cms - 30cms - null

我希望我的问题足够清楚.如果没有，请发表评论，如果可以的话，我会在需要的地方提供更多信息.

I hope my question was clear enough. If not please comment and I will give more information where needed, if I can.

推荐答案

我只使用

I'd just use GROUP_CONCAT() instead of cursors to generate the prepared statement:

SELECT CONCAT(
 ' SELECT MDE_ID,'

,  GROUP_CONCAT(
     't', MDF_ID, '.MFV_Value AS `', REPLACE(MDF_Label, '`', '``'), '`'
   )

,' FROM ModuleEntries '

, GROUP_CONCAT(
    'LEFT JOIN ModuleEntryFieldValues AS t', MDF_ID, '
       ON t', MDF_ID, '.MFV_MDE_ID = ModuleEntries.MDE_ID
      AND t', MDF_ID, '.MFV_MDF_ID = ', MDF_ID
  SEPARATOR ' ')

) INTO @qry FROM ModuleFields;

保存以进行空白编辑以使其更具可读性，并且示例数据@qry将包含:

Save for whitespace editing to make it more readable, with your sample data @qry would then contain:

SELECT MDE_ID,
       t1.MFV_Value AS `Height`,
       t2.MFV_Value AS `Width`,
       t3.MFV_Value AS `Weight`
FROM   ModuleEntries
  LEFT JOIN ModuleEntryFieldValues AS t1
         ON t1.MFV_MDE_ID = ModuleEntries.MDE_ID AND t1.MFV_MDF_ID = 1
  LEFT JOIN ModuleEntryFieldValues AS t2
         ON t2.MFV_MDE_ID = ModuleEntries.MDE_ID AND t2.MFV_MDF_ID = 2
  LEFT JOIN ModuleEntryFieldValues AS t3
         ON t3.MFV_MDE_ID = ModuleEntries.MDE_ID AND t3.MFV_MDF_ID = 3

然后可以准备并执行该语句:

One can then prepare and execute that statement:

PREPARE stmt FROM @qry;
EXECUTE stmt;
DEALLOCATE PREPARE stmt;
SET @qry = NULL;

将产生以下结果:

MDE_ID    Height    Width    Weight
     1    120cms    30cms    (null)

在 sqlfiddle 上查看.

这篇关于将表行合并为具有原始行的列标题的列的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！