script_utils.getObjects() sorts by IDs

[will-moore]

See bug #10745. We now sort objects by IDs in script_utils.getObjects().

To test:

• Select a bunch of images and open split view figure dialog in Web.
• Re-order the images by dragging & dropping rows (handle to left of each row)
• Grab a screen-shot (makes comparing easier)
• Run script and check that the figure generated has the same order of rows as the preview.

[sbesson]

In terms of logic, sorting objects returned by the server by input ids is sensible and should fix bugs in other scripts than the one ticketed.
Testing scenario described above was successfully tested against howe.

👍 Ready to merge.

NB: I tried to test this PR using the Combine Images script (with non-defaults parameters) and realized this script is not working anymore.

[will-moore]

Ticket for Combine_Images script: https://trac.openmicroscopy.org.uk/ome/ticket/10960

[joshmoore]

NB: Assuming objects is a list objects.sort(lambda left, right: cmp(left.id, right.id)) would suffice.

[will-moore]

I'm not trying to sort the objects by their own IDs. I'm trying to sort them so that they are in the same order as the list of ids that I used to retrieve them. E.g. ids = [5, 3, 4] then I want to sort objects to be: [<ImageWrapper id:5>, <ImageWrapper id:3>, <ImageWrapper id:4>]
I thought there might be a 'one-liner' to do it, but I couldn't work it out.

[joshmoore]

Ugh. Ok. Agreed. As interesting as that sounds, I bet it's going to be way more confusing than your two liner! Sorry for the noise.

...objects.sort(lambda l,r: cmp(ids.index(l.id), ids.index(r.id)))
comes close but will throw a ValueError if the id isn't in ids
but so would yours.

[joshmoore]

Being quiet now.