Data table design problem
						
							
 now there are data tables as follows 
 data classification table 
 cate_id 
 cate_name 
 pid 
 Commodity list 
 goods_id 
 goods_name 
 cate_id 
 property sheet 
 attr_id 
 attr_name 
 goods_id 
 the requirements are as follows. Originally, attributes can be found indirectly through commodity table classification, but now the classification table needs to directly associate attributes and classify a pair of multiple attributes. How to design the data table data will not be redundant 
						 
												
					 
					
						
 how about this design? 
 disassociate the commodity table from the property sheet 
 create a new intermediate table 
 attr_id 
 attr_type [1 is category 2, commodity] 
 connect_id decides whether to associate goods_id or cate_id according to attr_type 
  this feeling still needs to design the table structure according to the actual requirements. 
 what is the relationship between property sheets and classifications. Model relationships, such as one-to-one, one-to-many, many-to-many relationships. 
 then you can design a reasonable data table. 
 
  @ chinasky  has a good solution. 
 tell me what I understand: 
 how many attributes does a commodity have? 
 if there is only one attribute, delete the property sheet directly and add the property field to the item table. 
 if there are multiple attributes. Then there is redundancy. Because classification is also associated with attributes, goods are also associated with attributes. Then the goods_id of the property sheet is useless. It has to be deleted. Then add an intermediate table, which I think is easy to understand and data separation. Build 2 intermediate tables. Classify and attribute, create an intermediate table. Goods and attributes, create an intermediate table. 
Is there an one-to-one relationship between the 
  property sheet and the commodity table? A commodity corresponds to an attribute? 
 if so, the direct commodity table and the property table are merged.