system.dataset.toDataSet pyfloat cannot be cast

Ignition
1

I’m trying to make a dynamic table, based on the script below, this is just for one product, but depending on what location I pick, only want to run queries for available products. I keep getting this:
java.lang.ClassCastException: java.lang.ClassCastException: class org.python.core.PyFloat cannot be cast to class org.python.core.PySequence (org.python.core.PyFloat and org.python.core.PySequence are in unnamed module of loader java.net.URLClassLoader @37827677)

Here is the script I’m using. Right now it’s only for one location, but I’ll loop thru it as I add locations, or do I need to scrap this and try a different approach.

if (event.propertyName == 'yearly' or 'monthly' or 'daily') and (event.newValue == 1):
	start=event.source.parent.getComponent('LOADS').start
	end=event.source.parent.getComponent('LOADS').end
	loc=event.source.parent.Loc
	prods=system.db.runPrepQuery("select * from Location where location="+"'"+loc+"'")
	
	qry="SELECT  coalesce(sum(weight/2000.0),0.0) as "+'"'+"Total Tons" + '"' + " from truck_counts where location="+ "'" + loc + "'" +"  and t_stamp between " + "'" + start + "'" + "and " + "'" + end + "'"

	master=system.db.runPrepQuery(qry)
	total=master[0][0]
	toth='"'+"<html><br>&nbsp;<html>&nbsp;<br><Center>Total&nbsp"+'"'
	data=[]
	data.append(total)
	header=[]
	header.append(toth)
	if prods[0][4]==1:
			prod1='SELECT  coalesce(sum(weight/2000.0),0.0) as '+'"'+"Product I/II"+'"'+" from truck_counts where location="+"'"+loc+"'"+" and product_type='Product_I_II' and t_stamp between "+"'"+start+"'"+ "and "+"'"+end+"'"
			p1=system.db.runPrepQuery(prod1)
			p1d=p1[0][0]
			p1h='"'+"<html><br><Center>Product&nbsp;<br><Center>I/II&nbsp"+'"'
			data.extend([p1d])
			header.extend(p1h)
		
	d1=system.dataset.toDataSet(header ,data)
	
	event.source.data=d1
2

Haven't looked at anything else yet, but if (event.propertyName == 'yearly' or 'monthly' or 'daily') isn't doing what you think it is - it's checking:

does event.propertyName equal 'yearly'
or is "monthly" true
or is "daily" true

In Python, any non-zero length string is true, so your check isn't doing what it's supposed to. It should be
if event.propertyName in {"yearly", "monthly", "daily"} (the curly braces initialize a set; you could also use square brackets to initialize a list, or parentheses for a tuple)

That aside, is there a line number on the exception you're getting?

1 Like
3

The top part is working, just let’s me know when to run the script. I’m getting the error when I go to create the dataset, so line 27.
This is what each list returns:
[20.0, 0.0] # data
[’
  
Total&nbsp’, ‘
Product 
I/II&nbsp’] #header I have this header static in another script for something else and works great for increasing the column header height on the table so I can multi-line it.

4

Another aside, this is not the proper use of prepared SQL statements and leaves you open for an injection attack if loc, start, or end are human inputs

query="""
    SELECT COALESCE(SUM(weight/2000.0),0.0) as 'Product I/II'
    FROM truck_counts
    WHERE 
        location = ? 
        AND product_type = 'Product_I_II' 
        AND t_stamp BETWEEN ? AND ?
"""
args = [location, start_time, end_time]
p1=system.db.runPrepQuery(query, args)
1 Like
5

There’s no line 27 in what you posted, but I’m pretty sure the issue is this:
header.extend(p1h)
If you only have a single item, you should be using append - extend is used to add all the elements from one sequence to the end of another.

2 Likes
6

The top part is working, just let’s me know when to run the script

As long as we acknowledge that 'working' and 'doing what you expect' are two separate things :wink:

2 Likes
7

Figured out what the problem was…gotta thank Nick Mudge in another post for jogging a few brain cells. I forgot the data part of that fuction is a list of lists. Just created another list, and after appending everything in my original list, appended that list to the new one list. Works great now.

8

I had something very similar.


testsHeaders 	= [ 'Sample '+str(i)		for i,j in valuesIn] 
testsData 		= [[ float(j)				for i,j in valuesIn]]
print testsHeaders
print testsData
print len(testsHeaders),len(testsData)
datasetTests = system.dataset.toDataSet(testsHeaders, testsData)

When I added an additional set of brackets for testsData it worked.
Without them, I got the error about the floats.
There are 240 i, and 240 j.
i is an integer. j is a float.
Why did I need the second brackets to make this work?

Snip of results. I don't understand why it works with 240, 1 for lengths instead of 240, 240.
image

These are just he rules of the function?

9

Because, system.dataset.toDataSet() expects a List of Lists as the second argument.

Each list is a row, each item in each list is a value in that column.

You get the error stating that a float can not be converted to a PySequence because the function treats it like a list of lists, but then the first "list" is actually a float.

1 Like
10

headers need to be a list, which is what you have.
data needs to be a list of lists, where each inner list is a row.

What you have here is a dataset with one row.

1 Like