Support for different data_types SQLiteResult.data_collections_by_output_name

[marctavenier]

Hi,

When parsing a list with outputs to SQLiteResult.data_collections_by_output_name like this;
["Zone Mean Air Temperature", "Zone Air Relative Humidity"]
it returns the data collections, but the RH data collections have Temperature as their data_type instead of Fraction. It seems the data_type part of the header is of the first item is duplicated to all other items.

Fyi: when setting the _output_names to list access on the HB Read Custom Result Grasshopper component the same issue appears.

Kind regards,
Marc

[chriswmackey]

Hey @marctavenier ,

This is kinda by design since the SQLite queries get a lot simpler and faster when I know that I'm always dealing with the same data type and units for all of the requested collections. Also, I designed the Grasshopper components such that this is always the case and people won't get cases of incorrect data types for the data collection.

Would you be alright with making two separate calls to the SQLiteResult.data_collections_by_output_name for your case here where you need two different data types? Or do you really need a method that checks the data type for each of the collections?

[marctavenier]

Hi @chriswmackey,

I indeed made two separate calls for the different data types to solve the issue for now.

If it's by design to make it faster then perhaps it's good to mention this somewhere or somehow throw a warning when 2 data types are input, if you can detect that at all. Because right now it's unexpected behavior at the least, given the output is wrong without mentioning so. But I'm curious to your opinion on this.

Perhaps the HB Read Custom Result Grasshopper component would get a significant speed improvement when accessing the entire list? I haven't tested this but I can imagine so given the overhead reduces on setting up a new GHPython cycle for each item in a list.