After you've designed a database, it is not usual to go back and make changes, either to add new elements or to delete things that are no longer needed. There are a number of features to help you do this.
...
When modifying an existing design, replace is a particularly useful feature. Simply drag a new raw data file on top of an existing raw data file to replace and your existing mappings will be retained. This saves considerable time and effort - to do this manually would require lots of un-mapping and re-mapping of individual dimensions, properties. etc.
- Identically named columns in the new file will remain mapped.
- New columns can appear anywhere in the new file. They'll need to be mapped manually.
- Any removed columns will be automatically unmapped.
- Columns do not have to be in the same order.
Widget Connector | ||||||
---|---|---|---|---|---|---|
|
Unmap vs. delete
When designing a database, the act of dragging and dropping a column will create both the structure and mapping at the same time. You probably won't even notice this has occurred. However, you are able to delete the structure and unmap separately. The following table illustrates this:
| The blue border means that 'Value' has been mapped from the raw data that appears at the bottom of your screen. Use the to unmap. There are three potential outcomes:
| |
2. Structure | In this situation, 'Value' exists as a structural element only. It isn't mapped, and you won't see it in Phocas after building the database. Often a structure exists as part of a template or a previous design. It often serves as a placeholder, awaiting a future mapping. Use the to permanently delete it. | |
3. Foreign Mapping | The blue dot means that 'Value' is mapped from a different raw data source - not the one that appears at the bottom of your screen. There is no ability to unmap or delete from here - first you'll need to locate the correct raw data file (it could be in another stream, or a different raw data file in the current stream). |
Working with missing raw data
Even if you have the perfect database design, it is possible for raw data files to get moved, deleted, or renamed independently, which could cause a build error.
From Administration > Databases, choose your database and then click Design. If there is a mismatch between the design and raw data, an error like the following will appear.
Clicking OK will take you into Designer, where the missing raw data items will turn red, as shown below. There are three options:
- Restore: Put the missing raw data item back. This could mean re-uploading it, re-Syncing it, and checking the file name.
- Replace: You can easily drag a new raw data file on top of the missing red file. The red will turn blue again, and any existing mappings will be retained. See Replace.
- Remove: If it is no longer needed, you can remove the red raw data file by clicking the X.
Filter
You have the ability to Filter data being loaded into a stream.
...
- If using ID instead of Name: The ID is an identifier displayed as the last part of the URL.
- Query is typically the name of the dimension you wish to filter on follow by '=%%'. And, if required, the property relating to the dimension. With a favorite, you might add the mode, etc.
- The syntax you enter must use the UK English spelling, which is favourite with a 'u', even if Phocas displays the word spelt in the usual way everywhere else in the product.
...