Add a resize operation

[ThomasJClark]

No description provided.

[codecov-io]

Current coverage is 47.96%

Merging #372 into master will increase coverage by +0.14% as of 0e5d82a

@@            master    #372   diff @@
======================================
  Files          127     128     +1
  Stmts         3808    3840    +32
  Branches       417     417       
  Methods          0       0       
======================================
+ Hit           1821    1842    +21
- Partial        114     115     +1
- Missed        1873    1883    +10

Review entire Coverage Diff as of 0e5d82a

---

Uncovered Suggestions

1. +0.71% via ...loyerController.java#102...128
2. +0.58% via ...InstanceManager.java#182...203
3. +0.55% via .../ExceptionAlert.java#83...103
4. See 7 more...

Powered by Codecov. Updated on successful CI builds.

[AustinShalit]

What is the order of the operations?

[JLLeitschuh]

What do you mean?

[AustinShalit]

What is the sort order of the operations? It should have some set order so people can expect where the operations are in the list.

[JLLeitschuh]

We don't have a sort order the order is defined by the order that we list them in this file and the generated operations file.

[JLLeitschuh]

To rephrase @AustinShalit's question.
We should have some sort of policy for ordering and defining how we want the palette ordered.
Why we are putting specific operations in a specified order.

Do we have a policy like this? Can we come up with one?

[AustinShalit]

How is this operation different from the Open CV Resize operation? Do we want to document the differences?

[ThomasJClark]

The generated OpenCV one:

• Requires you to use a "new size" operation to specify the width and hight
• Has two different ways of resizing (using a size, or with an x and y scale). It's really unintuitive how it ignores one if the other is specified.
• Has unfriendly socket names
• Has unfriendly interpolation mode names ("INTER_LINEAR" instead of "Linear")
• Has a bunch of "interpolation modes" that don't work and actually just make the operation crash (like "FILL_OUTLIERS", which isn't actually an interpolation mode in OpenCV)

[AustinShalit]

@ThomasJClark Should the OpenCV operation be removed if this PR is merged?

[ThomasJClark]

Probably eventually, but removing it would break compatibility with existing projects that use it

[AustinShalit]

A solution to that would be removing it from the operation palate but leaving it as a valid operation.

[JLLeitschuh]

These should both be final variables.

[ThomasJClark]

k