使用Visual Studio查询构建器过滤数据

Question

Currently I have two tables created in MS SQL Server( postings table and user table) outputting to visual studio 2012 to a grid view and details view. I am trying to figure a way to ensure that when the details view is paged into a certain user (there is only two users) then the only the postings of made by that user in the past will show in the gridview.

The postings table has an ID (primary key) that is in two parts with the first part being a unique identifier of that poster with the second part being a unique identifier of that POST. For Poster1 has ID of 45678987654_98765456789, 45678987654_234565432123, 45678987654_999999999 etc and Poster 2 has the id 98765678987_987654866, 98765678987_2828288282, 98765678987_1111111111 all up about 700 entries.

In the User table there is a unique ID as well that is the same for that Poster eg. Poster1 has id 45678987654_98765456789 and Poster2 has 98765678987_987654866. I tried to filter in query builder in visual studio by =, Like% and CONTAINS because in theory if the first digits of the ID for both POST and USER were analysed and compared then the specifications could be achieved. As the current ID in the details view could be compared to all the gridview IDs and only those that are alike and share the first 15 or so digits would then be accepted onto the details view.

However all of the ways I tried to do this in query builder using Post_Id = table_Id or like/contains formats didn't work. If there is another way to do it that or if there is a slight modification to the current method then that would be great to hear about.

Answer 1

The first thing you need to do is to change the postings table. Where you have concatenated two values to make the PK, split it out into two columns, the first containing the userId, and the second the unique id of the post that you've generated.

Having done that, set the PK to be the combination of the two, and tell the db that the userId in postings is an FK referencing the PK of User table.

Once this is done, the data will be in a format that VS can probably understand better, and the db design will certainly be improved.

EDIT

SQL Code to create the FK:

ALTER TABLE postings  WITH CHECK ADD  CONSTRAINT [FK_postings_user] FOREIGN KEY([userId])
REFERENCES user ([userId])
GO

ALTER TABLE postings CHECK CONSTRAINT [FK_postings_user]
GO

If there are values in postings.UserId which have no matching value in user.UserId then this will fail. To check, run something like:

select * from postings p left outer join user u on u.UserId = p.UserId where u.UserId is null

This will give a list of posting rows where there is no matching user row - ones where your data conversion maybe had a problem? Or there was 'orphaned' data there already.