Forum Discussion
Cannot make relationship active
HotChilli That is a great point with the fact and dimension tables and I will definitely clean the data. But in fact, the model acts in that way even though there are some unnecessary columns.Jednotící tabulka (table) is a fact table, tables like Alza CZ or Datart CZ are dimension tables.
Unfortunately I cannot see "2 paths linking the same 2 tables" like you. Jednotící tabulka - Datart CZ (path A) is a fact table to dimension table type of connection and Pořadí RRP table - Datart CZ (path D) is a dimension table to another dimension table type of connection.
Maybe I am just missing something, but I cannot think of a way how to connect these tables differently. In Jednotící tabulka table there are unique Product IDs (column Product ID) with a recommended selling price. In other tables such as Datart CZ or Alza CZ there are prices that those eshops are using (scraped from web) and also an information (that is a function) if the eshop price is higher or lower than the recommended price. Pořadí RRP table just acts as a dimension table with a single purpose, namely to order those strings (price is higher, price is lower, price is the same) generated by a function in a logical way and this order should be used in a legend of graphs.
So to sum it up:
Jednotící tabulka - fact table, using mainly Product ID and Recommended price
Datart CZ - dimension table, connected via Product ID to fact table, important columns are eshop price (scraped from web) and a function determining whether the eshop price is higher or lower than the recommended one, also lot of redundant columns in that table, I will fix that
Pořadí RRP - dimension table used to order strings (otherwise they are ordered alphabeticaly in the legend) containing information about the price being higher or lower
Maybe you could argue that since Jednotící tabulka has the same amount of rows as the other tables (Alza CZ, Datart CZ and many others) and many identical columns, I can merge it and have this one table (Jednotící tabulka) with just different columns for different eshops. This solution however is not good because I want to be able to refresh the tables one by one because it takes almost an hour to scrape 400 products from one eshop.
PaulDBrown I cannot change the direction. This pops up.
- PaulDBrown4 years agoCommunity Champion
You must first change the cardinality to one-to-many ("one" being the table doing the filtering) and then you can change the direction to "Single"