# Mapping UUIDs to Display Names

In ENLYZE, every data entity—such as **Sites, Machines, Products, Downtimes**, etc.—has a **unique identifier (UUID)**. These UUIDs are essential for referencing elements across tables but can make the data model less readable.

When analyzing data in Power BI, it is often more convenient to use **display names** instead of UUIDs. This guide explains different methods to map UUIDs to their corresponding display names. Depending on where the mapping takes place, you may need to use **DAX formulas** or **Power Query transformations**.

There are two common scenarios where mapping UUIDs to display names is useful:

* **Adding a new column** within a table, such as including machine names in the `Production Runs` table.
* **Renaming columns** after querying time-series data for machine variables.

## Adding a New Column with Display Names

Let’s consider the `Production Runs` table, which stores information about:

* **Which machine** an order was running on.
* **Which product** was being produced.

Since the `Production Runs` table stores machine and product references as UUIDs, we want to replace them with readable names using the `Machines` and `Products` tables.

### **Method 1: Using DAX with LOOKUPVALUE**

In this method, we use **DAX (Data Analysis Expressions)** to look up and retrieve display names.

{% embed url="<https://www.loom.com/share/54160d6a3cf84f6da984a1722026cd71>" %}

To add the **machine name**, follow these steps:

1. Open **Power BI** and go to the **Table View**.
2. Select the **Production Runs** table.
3. Under **Table Tools** tab, click **New Column**.
4. Enter the following formula in the column definition bar:

   ```sql
   machine_name = LOOKUPVALUE(Machines[name], Machines[uuid], 'Production Runs'[machine])
   ```

**How It Works**

* `LOOKUPVALUE` searches for the **machine name** in the **Machines table**,
* It finds the row where `Machines[uuid]` matches `Production Runs[machine]`,
* It returns the corresponding machine **name**.

Now, the `Production Runs` table will have a `machine_name` column with readable names instead of UUIDs.

### **Method 2: Using a Merge Operation in Power Query Editor**

This method uses **Power Query** to merge data tables and replace UUIDs with display names.

{% embed url="<https://www.loom.com/share/d21c5c1dfc8d4150ae499bb237324cfa>" %}

To add the **product name**, follow these steps:

1. Open **Power Query Editor** (**Transform Data** in Power BI).
2. Select the `Production Runs` table.
3. Under **Combine**, click **Merge Queries** (**Merge Queries as New** if you want a new table).
4. In the pop-up window:
   * The `Production Runs` table will already be selected.
   * Select `Products` as the second table.
   * Choose `Product UUID` in both tables as the matching key.
   * Under **Join Kind**, select **Left Outer** to keep all records from `Production Runs`.
   * A message will appear showing how many rows found a match. Ideally, all rows should match.
5. Click **OK**.
6. Expand the newly created column by clicking the **expand button** on the right of the column name.
7. Select **Product Name** and click **OK**.

Now, the `Production Runs` table will include a `product_name` column with readable product names instead of UUIDs.

***

Both methods allow you to replace UUIDs with display names, improving readability and usability in Power BI:

* Use DAX **(`LOOKUPVALUE`)** if you prefer calculations inside your Power BI model.
* Use Power Query **(`Merge Queries`)** if you want to transform the data before loading it into Power BI.

## Renaming Columns in Time Series Data using Power Query

When using the `Timeseries` function from the connector, the column names don’t show variable names but variable UUIDs. Many times, It is convenient to rename those columns and use the machine and variable display names to improve readability of the data.

We have created a function that will do this name mapping automatically after querying the time series data.

### The Mapping Function

```powerquery

let
    // Step 1: Call the Timeseries function
    Source = Timeseries(FilteredVariables, #datetimezone(2025, 2, 25, 0, 0, 0, 1, 0), #datetimezone(2025, 2, 26, 0, 0, 0, 1, 0), "1m", "avg"),

    // Step 2: Rename Machine UUID columns to Machine Names
    MachineMapping = Table.SelectColumns(Machines, {"uuid", "name"}),
    MachineRenamePairs = Table.ToRows(MachineMapping),
    RenamedMachines = Table.RenameColumns(Source, MachineRenamePairs, MissingField.Ignore),

    // Step 3: Rename Variable UUIDs Inside Each Table
    RenamedTables = Table.TransformColumns(RenamedMachines, {}, (x) => try Table.RenameColumns(x, Table.ToRows(Table.SelectColumns(FilteredVariables, {"uuid", "display_name"})), MissingField.Ignore) otherwise x)
in
    RenamedTables
```

#### How it works

**1. Fetch Time-Series Data**

The function starts by calling the `Timeseries` function from the ENLYZE connector with:

* `FilteredVariables` – A filtered version of the `Variables` table with the variables that should be used in the query.
* A **start and end time** – Currently set from **February 25, 2025, to February 26, 2025**.
* A **sampling interval** of **1 minute**.
* An **aggregation method** of **average ("avg")**.

All these inputs can be modified as needed

**2. Rename Machine UUIDs to Machine Names**

* Machine UUIDs in the result are mapped to their corresponding **machine names**.
* The function retrieves the mapping from the `Machines` table (`uuid` → `name`).
* Machine column names are updated accordingly.

**3. Rename Variable UUIDs**

* Inside each machine column, **variable UUIDs** are replaced with their **display names**.
* The function extracts variable mappings from `FilteredVariables` (`uuid` → `display_name`).

#### How to use it

To use the function, you will need:

* The `Machines` table from the ENLYZE Connector
* The `Timeseries` function from the ENLYZE Connector
* A table called `FilteredVariables`, which should be a copy of the `Variables` tables from the ENLYZE Connector with a filter on the variables needed for the time series query

To use the function

1. Go to **New Source > Blank Query** to add a new empty query
2. Open the Advance Query editor under View > Advanced
3. Copy and paste the function code
4. Adapt the inputs of the `Timeseries` function to your necessities
5. Click on **Done**

After running the function, the **final table** will have machine and variable names instead of UUIDs in the column names