带包装的事实表

Question

I's like to ask about etl, and fact table.

Likely, my fact table's structure is like this :

Id   |    Name   |    QuestionAanswer    | QuestionBanswer   |  QuestionCanswer

I have dimension like this :

1. UserDimension
2. QuestionDimension
3. QuestionStaging

QuestionStaging data came from csv files, which the name is structurely randomed like A257_IDr, H1_ques1, H2_ques2, H3_ques3. ex :

 A001 | 1 | 2 | 1  |

where A001 -> UserID and 1,2,1 -> User's answer to following question.

QuestionDimension's structure is like:

SK_Question | ID_Question | IDAnswer

ex. I have the following question like =

How are yours company profit this year? Bad:Good:Best

then the data's structure in table is like :

Sk_Question | IDQuestion | IdAnswer | AnswerDetail
1                   Q1        1             Bad
2                   Q1        1             Good
3                   Q2        1             Best 

My Expectation FactTable is likely to :

SK_USer |Id   |    Name   |    QuestionAanswer    | QuestionBanswer   |  QuestionCanswer
1        A001     Richard             1                    1                     2

OR

1        A001     Richard             Bad                   Bad                 Good

What data flow component or control flow should I used for these situation?

Answer 1

您将需要一个平面文件连接管理器，数据流控制组件，平面文件数据源任务以及一个数据目标任务（可能是OLE DB），具体取决于您的数据库。