New endpoint that returns count, mb size, and nodes for each file type

[hkethi002]

Notes

• New endpoint POST /download/summary
• request body

[{
    "level": {"projects"|"sessions"|"acquisitions"|"analyses"},
     "_id": {ContainerId}
}]

• response

{
    "dicom": {
        "count" : 1,
        "mb_total": 23124,
        "_id": "dicom"
    }
    .
    .
    .
}

Review Checklist

• Tests were added to cover all code changes
• Documentation was added / updated
• Code and tests follow standards in CONTRIBUTING.md

[codecov-io]

Codecov Report

Merging #934 into master will increase coverage by 0.02%.
The diff coverage is 92.72%.

@@            Coverage Diff             @@
##           master     #934      +/-   ##
==========================================
+ Coverage   90.14%   90.16%   +0.02%     
==========================================
  Files          48       48              
  Lines        6462     6516      +54     
==========================================
+ Hits         5825     5875      +50     
- Misses        637      641       +4

Flag	Coverage Δ
#python	90.16% <92.72%> (+0.02%)	⬆️

Impacted Files	Coverage Δ
api/api.py	100% <ø> (ø)	⬆️
api/download.py	95.06% <92.72%> (-0.54%)	⬇️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 3a9ade2...d6116b7. Read the comment docs.

[nagem]

Looks good, but we don't actually need the node list returned with the summary results. Are you providing the node list to make the eventual download request more efficient?

[nagem]

The string version of the failure would get returned to the API user. In situations like this it's better to internally log the error (log.warning(e)) and then self.abort(500, 'Failure to load summary') or something like that so the user doesn't get sent a mongo error.

[nagem]

We should make sure we have tests if we're going to support file summaries for analysis downloads.

[nagem]

We won't want want analyses included in project/session/acquisition download summary (because they aren't included in the actual download), so it's okay to remove this.

[hkethi002]

That was the idea, it seemed like an easy place to generate it because we are already aggregating through the files anyways.

[nagem]

Unfortunately that list might be 1000s of acquisitions long for large projects, and the user might choose multiple filters, meaning a frontend join of multiple lists of acquisitions - which would lose the gain in efficiency.

[nagem]

Thanks for making the requested changes, LGTM!