New logic for outputBag upload

If md5 location already exists in hatrac, re-use the URL... but URL is the key for Output_Bag table... so update old ermrest entry with that url

re-Hongsuda 1/21/21

added Doing data-hub pipeline labels

changed the description

changed milestone to %v2.0.0

added To Do label and removed Doing label

added Doing label and removed To Do label

added To Do label and removed Doing label

chaise uploader saves outputBag's in a folder based on date... uploadOutputBag process uploads based on study and rep RIDs...

that means that a check for outputBag exist means an iterative HEAD across all sub-namespaces of the hatrac location... that seems a waste as hatrac already does a head before put...

if the namespace was consistent then iterative HEAD wouldn't be necessary...

trying to schedule a meeting with Hongsuda and Laura on 20200126 about this...

re-Hongsuda 20210125 ignore chaise upload location... only check for uploadOutputBag process duplicates

hatrac api will automatically do a HEAD and if the md5sum is identical it won't version the file, so the new logic is pretty much there already

logic needs to be:

• upload to hatrac (auto checks for it already being there)
• check ermrest for that url and the update the executionRun etc

added Doing label and removed To Do label

mentioned in merge request !67 (merged)

added Test label and removed Doing label

mentioned in commit fca601e4

closed via merge request !67 (merged)