# Unable to convert row 1, column 2 to type class java.lang.integer

**URL:** <https://forum.inductiveautomation.com/t/unable-to-convert-row-1-column-2-to-type-class-java-lang-integer/112990>\
**Category:** Ignition\
**Tags:** perspective, scripting, reporting\
**Created:** [December 24, 2025, 10:58pm UTC](https://forum.inductiveautomation.com/t/unable-to-convert-row-1-column-2-to-type-class-java-lang-integer/112990 "2025-12-24T22:58:12Z")\
**Posts on this page:** 6\
**Page:** 1

<div class="post-metadata">

**Author:** ![bpomerhn](https://avatars.discourse-cdn.com/v4/letter/b/41988e/32.png) [@bpomerhn](https://forum.inductiveautomation.com/u/bpomerhn)\
**Post date:** [December 24, 2025, 10:58pm UTC](https://forum.inductiveautomation.com/t/unable-to-convert-row-1-column-2-to-type-class-java-lang-integer/112990/1 "2025-12-24T22:58:12Z")

</div>

Perspective 8.1.47

Working on doing some subtraction of daily meter readings for a report using a script data source. Raw data source is from a tag historian query (‘METER’).

I am getting an error message (subject of this post) and the data type I’m performing the operation on should be a floating point value. I believe the code I referenced might be generically adding an integer column, not sure? Do I need to use system.dataset.addColumn in lieu of headers = list(pyDataset.getColumnNames()) + [‘Difference’]? If so, how would I use system.dataset.addColumn here?

Another question is, I actually have a query and datakey where I have used a tag historian query to get the previous value. How would I reference that instead of initializing previousValue to 0? It is a query with a fixed sample size of 1.

Thank you for any help!

```python
def updateData(data, sample):
    #data['myKey']='Example'
    
    # Access the raw data from an existing data key
    rawDataset = data['METER'].getCoreResults()
    
    #Convert the dataset to a pydataset for easier raw iteration
    pyDataset = system.dataset.toPyDataSet(rawDataset)
    
    # Initialize a list to hold the new data, including a new difference column
    newData = []
    headers = list(pyDataset.getColumnNames()) + ['Difference']
    
    # Subtract the value in the current row from the value in the previous row for a 'Value' column
    previousValue = 0 # Initialize the 'previous' value
    
    for i, row in enumerate(pyDataset):
        currentValue = row['ty_total']
        if i == 0:
            difference = 0 # No previous row for the first entry
        else:
            difference = currentValue - previousValue
        
        # Append all existing values and the new difference value to the list
        rowList = list(row) + [difference]
        newData.append(rowList)
        
        previousValue = currentValue # Update the previous value for the next iteration
        
    #Convert the list of new data back into a standard dataset
    resultDataset = system.dataset.toDataSet(headers, newData)
    
    # Assign the new dataset to a new data key
    data['METERDIFFERENCE'] = resultDataset

```

---

<div class="post-metadata">

**Author:** ![pturmel](https://sea1.discourse-cdn.com/inductiveautomation/user_avatar/forum.inductiveautomation.com/pturmel/32/3295_2.png) [@pturmel](https://forum.inductiveautomation.com/u/pturmel)\
**Post date:** [December 25, 2025, 1:11am UTC](https://forum.inductiveautomation.com/t/unable-to-convert-row-1-column-2-to-type-class-java-lang-integer/112990/2 "2025-12-25T01:11:38Z")

</div>

The system functions that create new datasets, or add columns, use the values in the first row to **infer** the data types to use for the columns.

If that doesn't yield correct behavior (due to mixed types), you need to instead use the `DatasetBuilder` utility, where you **explicitly** set the column types.

---

<div class="post-metadata">

**Author:** ![bpomerhn](https://avatars.discourse-cdn.com/v4/letter/b/41988e/32.png) [@bpomerhn](https://forum.inductiveautomation.com/u/bpomerhn)\
**Post date:** [December 25, 2025, 8:50pm UTC](https://forum.inductiveautomation.com/t/unable-to-convert-row-1-column-2-to-type-class-java-lang-integer/112990/3 "2025-12-25T20:50:29Z")

</div>

Thank you! I will look into this solution.

---

<div class="post-metadata">

**Author:** ![tgarrison](https://sea1.discourse-cdn.com/inductiveautomation/user_avatar/forum.inductiveautomation.com/tgarrison/32/80965_2.png) [@tgarrison](https://forum.inductiveautomation.com/u/tgarrison)\
**Post date:** [December 26, 2025, 12:50pm UTC](https://forum.inductiveautomation.com/t/unable-to-convert-row-1-column-2-to-type-class-java-lang-integer/112990/4 "2025-12-26T12:50:17Z")

</div>

if it's just your difference column that you want to be a float, just set previousValue to 0.0 instead of just 0 so that it gets used as a float.

---

<div class="post-metadata">

**Author:** ![bpomerhn](https://avatars.discourse-cdn.com/v4/letter/b/41988e/32.png) [@bpomerhn](https://forum.inductiveautomation.com/u/bpomerhn)\
**Post date:** [December 26, 2025, 1:32pm UTC](https://forum.inductiveautomation.com/t/unable-to-convert-row-1-column-2-to-type-class-java-lang-integer/112990/5 "2025-12-26T13:32:57Z")

</div>

Thank you @tgarrison. Once I changed both previousValue to 0.0 and difference = 0.0, it cleared up the data type mismatch.

What would the syntax be to reference another tag historian query with sample size 1 that is the previous value? The dataset is called ‘PREVIOUS’ and the datakey is ‘ty\_total’. Instead of initializing to 0.0, what would I use? Thank you for any help.

---

<div class="post-metadata">

**Author:** ![Chris\_Bingham](https://avatars.discourse-cdn.com/v4/letter/c/e5b9ba/32.png) [@Chris\_Bingham](https://forum.inductiveautomation.com/u/Chris_Bingham)\
**Post date:** [December 26, 2025, 4:23pm UTC](https://forum.inductiveautomation.com/t/unable-to-convert-row-1-column-2-to-type-class-java-lang-integer/112990/6 "2025-12-26T16:23:59Z")

</div>

If it is an option, I recommend updating your initial query to have the database calculate & report the difference of two values at time-of-query, and NOT post-process the data. That aside...

> [@bpomerhn](#):
>
> If so, how would I use system.dataset.addColumn here?

The [system.dataset.addColumn](https://docs.inductiveautomation.com/docs/8.1/appendix/scripting-functions/system-dataset/system-dataset-addColumn) function allows you to specify the column type. You can check for the type from an existing dataset by using the [ds.getColumnType(colIndex)](https://docs.inductiveautomation.com/docs/8.1/platform/scripting/python-scripting/variables-datatypes-and-objects/datasets#accessing-data-in-a-dataset) method on an existing dataset:

```python
colName = 'ty_total'
colIndex = dataset.getColumnIndex(colName)
colType = dataset.getColumnType(colIndex)

```

Then, calculate your column data and add the new column:

```python
# Generate data for new column:
newColData = [None] # First row initial value - if your schema allows null values
# Get values from existing dataset, calculate new values & append:
for i in range(1, dataset.getRowCount()): # Start from second row
	previousValue = dataset.getValueAt(i-1, colName) # Previous row value
	currentValue = dataset.getValueAt(i, colName) # Current row value
	newColData .append(currentValue - previousValue)
newColName = 'Difference'
# Generate new dataset with added column (omit colIndex to add to end):
newDataset = system.dataset.addColumn(dataset, newColData, newColName, colType)

```

While I prefer to see `None` in a row where data cannot be calculated, you may utilize a default value of your choosing. The `addColumn` method will attempt to coerce all of your column data to the specified type, where possible.  
Disclaimer, untested... let me know if you find any mishaps, I'm happy to edit & update as necessary.