How to go from a dataset to a dictionary?

Ignition
1

I have a dataset from a query binding and I want to turn it into a dictionary in a script. How can I do this? Ignition 7.9.

3

That’s only in 8.0 I believe, I don’t have that option.

4

There’s no obvious mapping from dataset to dictionary, so you’ll need to tell us what your dataset looks like and what “shape” you want the resulting dictionary to be.

5

It’s two columns, I would like the first column to be the key and the second column to be the value.

6

2 columns and 1 row or multiple rows?

7

It has multiple rows.

8

So something like this?

{
"column 1 name" : [values, in, each, row, for, column 1],
"column 2 name" : [values, in, each, row, for, column 2]
}

where the keys are the column name, and the value is an array of the values from the rows of that column?

9

The dataset is such

id     description
1      label1
2      label2
3      label3
...

and so on. I’d like the dictionary to be

{
   "1": "label1",
   "2": "label2",
   "3": "label3",
   ...
}
10

Psuedocode-ish:

dict = {}

for row in dataset:
	dict[row['id']] = row['description']

You could also just use the indices, i.e. row[0] and row[1].

and I think for the iteration to work it needs to be a PyDataSet, otherwise you’ll have to use some of the other more verbose methods to get the data out.

3 Likes
11

That was it, thanks.

12

Hope this helps, for one narrow kind of input as described:

def to_dataset(array_of_objects):
	"""Returns a Dataset made from an array of PyDictionaries that have only key-value pairs--no children"""
	if array_of_objects == []: raise ValueError("Needs at least one item with keys. Cannot derive keys from an empty array.")
	rows = [item.values() for item in array_of_objects]
	return system.dataset.toDataSet(array_of_objects[0].keys(), rows)
13

I've realized since this post (which is 6 years old ) I should just use PyDataSet each row is functionally a dictionary. You can simply do

#all these return a pydataset now
data = system.db.runNamedQuery/runQuery/runPrepQuery(query) 
for row in data:
    # Access like a dictionary
    someValue = row['someVal']

Manually converting to a dictionary is wasting time and computation.

Or if you know your defined columns, then unpacking in the for loop

for (column1, column2, ..., columnN) in data:
    # now we have access to the column data directly
14

If you're gonna put the append on the same line, you might as well use a comprehension:

rows = [item.values() for item in arrayOfObjects]
3 Likes
15

Thanks, love it! I've modified my code above.