WHY IS THIS CHANGE NEEDED?:
- it complained that ImageDummy object has no .shape attribute when
  Tiler methods like ._run_tp() were called

HOW DOES THE CHANGE SOLVE THE PROBLEM?:
- first argument of the object to be returned by .dummy() is mistakenly
  an Image instance (i.e. ImageDummy); it should be a dask array like
  those returned by e.g. .from_image(), .from_h5()
- fixed by using the .data attribute, consistent with the other methods mentioned

WHAT SIDE EFFECTS DOES THIS CHANGE HAVE?:
- order of dimensions is inconsistent: it is tczxy in io.image.py but
  ctzxy in the .shape() method in tiler.py.  this may break things down
  the line

WHY IS THIS CHANGE NEEDED?:
- consistency

HOW DOES THE CHANGE SOLVE THE PROBLEM?:
- fill in unspecified channels with 'nil' -- normally this information
  is fetched from metadata, which we don't have for this dummy

WHY IS THIS CHANGE NEEDED?:
- currently has one time point, one channel, one z-stack and thus
  mimicking usual Image.data structure.  this isn't what we want to test.

HOW DOES THE CHANGE SOLVE THE PROBLEM?:
- use pad_array() to pad with zeros so that the shape is as desired

REFERENCES:
- MR !10

WHY IS THIS CHANGE NEEDED?:
- not obvious that there is an example

HOW DOES THE CHANGE SOLVE THE PROBLEM?:
- highlight example in ```
- remove unnecessary example detail that isn't needed to understand the function

REFERENCES:
- MR !10

HOW DOES THE CHANGE SOLVE THE PROBLEM?:
- remove FIXME

REFERENCES:
- MR !10

WHY IS THIS CHANGE NEEDED?:
- test dummy tiler, see if it crashes

REFERENCES:
- issue #53

REFERENCES:
- issue #53

WHY IS THIS CHANGE NEEDED?:
- need to test whether ImageDummy object makes Tiler happy
- existing implementation lacked a metadata argument, which Tiler expected

HOW DOES THE CHANGE SOLVE THE PROBLEM?:
- create a dummy_omero_metadata dict, based on the arguments of the
  metadata() property of the Image object from aliby.io.omero

WHAT SIDE EFFECTS DOES THIS CHANGE HAVE?:
- metadata dict is hard-coded, ideally take some info from ImageDummy
  object and parameters
- instantiating Tiler raises "Tiler: No z_perchannel data: 'zsections'",
  and several attributes are missing -- this may be expected

EVIDENCE THAT COMMIT WORKS:
- need to write tests, not sure what we should be looking for.  at the
  very least, an instantiation that doesn't break

REFERENCES:
- issue #53

WHY IS THIS CHANGE NEEDED?:
- consistent with standard indexing, and should make tiler happy

HOW DOES THE CHANGE SOLVE THE PROBLEM?:
- get the z-dimension in

REFERENCES:
- issue #53

WHY IS THIS CHANGE NEEDED?:
- https://git.ecdf.ed.ac.uk/swain-lab/aliby/aliby/-/jobs/56786 because
  AttributeError: 'function' object has no attribute 'shape'

REFERENCES:
- issue #53

WHY IS THIS CHANGE NEEDED?:
- https://git.ecdf.ed.ac.uk/swain-lab/aliby/aliby/-/jobs/56781 failed

REFERENCES:
- issue #53

WHY IS THIS CHANGE NEEDED?:
- https://git.ecdf.ed.ac.uk/swain-lab/aliby/aliby/-/jobs/56767 failed
  because ValueError: The truth value of a Array is ambiguous.

HOW DOES THE CHANGE SOLVE THE PROBLEM?:
- use numpy.array_equal() to compare arrays instead

REFERENCES:
- issue #53

WHY IS THIS CHANGE NEEDED?:
- https://git.ecdf.ed.ac.uk/swain-lab/aliby/aliby/-/jobs/56762 failed
  because TypeError: compute() takes 1 positional argument but 2 were given

HOW DOES THE CHANGE SOLVE THE PROBLEM?:
- corrected a stupid typo: compute() is a function that takes no
  arguments that gives out an array, and i want to index that array

REFERENCES:
- issue #53

WHY IS THIS CHANGE NEEDED?:
- https://git.ecdf.ed.ac.uk/swain-lab/aliby/aliby/-/jobs/56757 failed
  because TypeError raised: 'int' object is not iterable

HOW DOES THE CHANGE SOLVE THE PROBLEM?:
- @pytest.mark.parametrize() takes an iterable, e.g. a list -- I forgot
  about it.  fixed it by putting the values in a list.

REFERENCES:
- issue #53

EVIDENCE THAT COMMIT WORKS:
- issue #53

REFERENCES:

REFERENCES:
- issue #53

WHY IS THIS CHANGE NEEDED?:
- to help create a dummy image instance with dimensions tiler likes

HOW DOES THE CHANGE SOLVE THE PROBLEM?:
- decided to separate these two steps: (1) add more dimensions to loaded
  image, (2) pad the dask array.  otherwise, i'd have to tell the
  pad_array function the existing dimensions -- it's easier to just
  assume it has five before running
- pad_array: just concat zero arrays with same dimensions as image_array

REFERENCES:
- issue #53

WHY IS THIS CHANGE NEEDED?:
- an overview of what image instances do will help us understand and use
  such objects

HOW DOES THE CHANGE SOLVE THE PROBLEM?:
- add long docstring at beginning of module (similar to e.g. tiler)
- modify line in docstring for ImageDir.  it doesn't inherit from
  ImageLocal; it inherits from BaseLocalImage.

WHAT SIDE EFFECTS DOES THIS CHANGE HAVE?:
- need to check whether descriptions are accurate

WHY IS THIS CHANGE NEEDED?:
- organise notes from last commit & give context to writing code

REFERENCES:
- issue #53

WHY IS THIS CHANGE NEEDED?:
- organise notes from last commit & provide context to help with
  developing code

REFERENCES:
- issue #53

- clarification of what needs to be done as comments/docstrings

REFERENCES:
- issue #53

WHY IS THIS CHANGE NEEDED?:
- need a way to produce a dummy form of the next step after creating an
  image instance to test pipeline; this next step is the tiler

HOW DOES THE CHANGE SOLVE THE PROBLEM?:
- make template method

REFERENCES:
- issue #53

WHY IS THIS CHANGE NEEDED?:
- need function that creates the expected output from dummy image file,
  as an interface for other classes' get_data_lazy() methods

HOW DOES THE CHANGE SOLVE THE PROBLEM?:
- create abstract method that loads the dummy image file and returns
  2d dask array
- define function to get examples directory, following logfile_parser.legacy

WHAT SIDE EFFECTS DOES THIS CHANGE HAVE?:
- only able to handle one TIF file (i.e. 1 z-section)

EVIDENCE THAT COMMIT WORKS:
- created a test class that extends BaseLocalImage and has a
  get_data_lazy() function that extends the base class's function; it is
  able to load the dummy image into a dask array

REFERENCES:
- issue #53

WHY IS THIS CHANGE NEEDED?:
- need a TIFF image file to serve as a dummy image input for dummy image
  processing functions

HOW DOES THE CHANGE SOLVE THE PROBLEM?:
- add TIFF file from unit test (staffa:560)

REFERENCES:
- issue #53

refactor: Remove unused code and obsolete tests

See merge request aliby!9

Relates to #52

ci: implement and visualise test coverage

See merge request aliby!8

WHY IS THIS CHANGE NEEDED?:
- need a simple test to see if gitlab visualises coverage

HOW DOES THE CHANGE SOLVE THE PROBLEM?:
- write a test for interpolate post-process; this test checks that
  interpolated values are as expected by leveraging a dummy signal that
  is a gradient.

REFERENCES:
- issue #46
- merge request !8

WHY IS THIS CHANGE NEEDED?:
- branch coverage give additional information re test coverage -- more
  informative for writing tests

HOW DOES THE CHANGE SOLVE THE PROBLEM?:
- add --branch flag, according to https://coverage.readthedocs.io/en/7.0.4/branch.html#how-to-measure-branch-coverage

WHAT SIDE EFFECTS DOES THIS CHANGE HAVE?:
- will change the coverage report and the final percent coverage

REFERENCES:
- issue #46
- merge request !8