Rep frame helper

[kelleyl]

Adds two functions to video_document_helper.py. Functions are used to extract the image for the representative timepoint for a given timeframe annotation.

[codecov]

Codecov Report

Attention: Patch coverage is 83.33333% with 3 lines in your changes are missing coverage. Please review.

Project coverage is 91.41%. Comparing base (262015b) to head (fa069fc).

Files	Patch %	Lines
mmif/utils/video_document_helper.py	83.33%	3 Missing ⚠️

Additional details and impacted files

@@             Coverage Diff             @@
##           develop     #278      +/-   ##
===========================================
- Coverage    91.54%   91.41%   -0.13%     
===========================================
  Files           10       10              
  Lines         1265     1282      +17     
===========================================
+ Hits          1158     1172      +14     
- Misses         107      110       +3     

Flag	Coverage Δ
unittests	91.41% <83.33%> (-0.13%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[keighrim]

looks good! I left a small suggestion to use long_id and Mmif.get() for retrieving annotation objects.

[keighrim]

this can be much shorter if using

mmif-python/mmif/serialize/mmif.py

Lines 564 to 575 in 2a5bffc

	def __getitem__(self, item: str) -> Union[Document, View, Annotation]:
	"""
	getitem implementation for Mmif. When nothing is found, this will raise an error
	rather than returning a None (although pytype doesn't think so...)
	:raises KeyError: if the item is not found or if the search results are ambiguous
	:param item: the search string, a document ID, a view ID, or a view-scoped annotation ID
	:return: the object searched for
	"""
	if item in self._named_attributes():
	return self.__dict__[item]
	split_attempt = item.split(self.id_delimiter)

and

mmif-python/mmif/serialize/annotation.py

Lines 130 to 135 in 693972a

	@property
	def long_id(self) -> str:
	if self.parent is not None and len(self.parent) > 0:
	return f"{self.parent}{self.id_delimiter}{self.id}"
	else:
	return self.id

[keighrim]

can we provide an argument to set the rule to pick the "top"? I'm thinking of

1. get all fnums from all elements in the rep list.
2. get the fnum of the first rep element (as a single list maybe, only to keep the return type simple)
3. get the fnum of the "middle" rep element

We might need to think more carefully how this argument can be propagated in the following extract_representative_frame wrapper fn.