I am working on an Ignition Perspective Table component.
My table data is coming from a Dataset Tag:
[default]IGBT_ROBOT_CELL_PROD/SAP_RES_Dataset
The table contains a column:
"Prod. Order No"
I also have another tag:
[default]IGBT_ROBOT_CELL_PROD/OrderNumber
Requirement:
Whenever the value of OrderNumber tag matches any row's "Prod. Order No", I want the entire row to be highlighted dynamically.
I tried:
Script Transform on props.data
props.rows.style.classes expression binding
Row style bindings
But since the table data originates from a Dataset Tag and Perspective internally converts rows into objects containing value/style properties, I am facing issues with row styling and expression syntax.
What is the recommended Perspective approach to dynamically highlight a row based on matching tag value with a dataset column value?
The canonical approach would be an expression structure binding on your table's props.data referencing the raw dataset and the order number tag. That would have a script transform to generate the row objects, styled or not per the comparison.
I don't like script transforms (mainly for performance), so I would use my Integration Toolkit to perform the above in a single expression binding. Something like this:
transform(
{[default]IGBT_ROBOT_CELL_PROD/OrderNumber},
forEach(
{[default]IGBT_ROBOT_CELL_PROD/SAP_RES_Dataset},
if(
it()['Prod. Order No'] = value(),
asMap(
forEach(
it(), // key/value pairs of the row
it()[0], // column key
asMap(
'value', asMap(it()[1]), // column value nested
'style', asMap('backgroundColor', '#ff3333')
)
)
),
asMap(it()) // optimized unstyled conversion of a whole row
)
)
)
# 1. Safely unpack the values from our Expression Structure
dataset = value.get("myDataset")
target_order = value.get("currentOrder")
# Return empty if data is missing or bad quality
if dataset is None or target_order is None:
return []
# 2. Define the highlight style (you can use background colors or Style Classes)
# Highlight style for a match:
highlight_style = {
"backgroundColor": "#FFFFCC" # Soft yellow highlight
}
# Standard style for non-matching rows:
normal_style = {}
# 3. Convert the dataset to a PyDataSet so we can iterate through it easily
py_data = system.dataset.toPyDataSet(dataset)
# Get column names to structure our row objects properly
headers = dataset.getColumnNames()
# 4. Build the JSON structure Perspective expects
output_rows = []
for row in py_data:
# Build the cell data dict for this row
row_cells = {}
for header in headers:
row_cells[header] = row[header]
# Check if this row's order number matches our target tag
# Replace "Prod. Order No" with the exact spelling of your column header
if str(row["Prod. Order No"]) == str(target_order):
row_style = highlight_style
else:
row_style = normal_style
# Combine the cells and the dynamic row style into the required Perspective object
output_rows.append({
"value": row_cells,
"style": row_style
})
return output_rows
Yes, Perspective, in the designer, doesn't like keys for mapping objects that have spaces or special characters (basically, keys that are not valid variable names). Not a problem for datasets, but you can't style a dataset.
(This is true whether you generate the problem column names with a transform script, or my expressions, or just by having a query binding return JSON format.)
The simplest solution would be to arrange for the source dataset to have simple column names (this is what I would do).
A more complex solution would be to substitute simple names on the fly in the expression. Would need a custom property to hold the pairs of substitutions.
