DS-3511 Fix HTTP 500 errors on REST API bitstream updates

[minurmin]

https://jira.duraspace.org/browse/DS-3511 and
https://groups.google.com/forum/#!topic/dspace-tech/ZEvYkjqbx8s

When trying to update the content of a bitstream or delete a bitstream policy HTTP 500 error occurs (Dspace 6.0-6.2). Fixed by adding context.complete() to the end of the respective methods (adapted the original patch submitted to DS-3511).

[tdonohue]

Please rebase this PR against the latest master branch. The master branch now requires adherence to our new Java Code Style documented at: https://wiki.duraspace.org/display/DSPACE/Code+Style+Guide

If you have any questions, feel free to ask in this PR or on our Slack: https://wiki.duraspace.org/display/DSPACE/Slack

[terrywbrady]

@minurmin , thank you for submitting this PR. I believe your intention was to apply this fix to DSpace 6.x. If that is correct, could you update the base branch to dspace-6_x? That will remove the error that is currently in place and will allow a review to take place.

[AndrewBennet]

I've cherry-picked this commit onto dspace-6_x and created another pull request: #1996

[minurmin]

Yes, the PR is supposed to be used for Dspace 6. Sorry to have missed the comments, originally I wasn't sure what branch to merge into since the code apparently works on master branch as well. For what it's worth, this PR should now be mergeable with current 6.x branch as well (fixed by force push).

[kshepherd]

+1
Thanks @minurmin for the fix, and thanks @AndrewBennet for the cherrypick to 6.x
@minurmin , i think we'll still want another rebased PR for master (as per @tdonohue 's comments)

[tdonohue]

The code changes look fine to me, however, Travis CI seems to having issues with the PR. But, the alternative/cherry-picked PR (#1996). Seems to be perfectly fine. Should we be concentrating on reviewing/merging #1996?

[minurmin]

By all means, please proceed with #1996. Perhaps the force push caused some problems, I have no idea what could cause the problem considering that this is just a 2-line change.

[tdonohue]

I'll give my general thumbs up here, 👍 , code looks fine.. I suspect the Travis issues in this PR are actually a Travis hiccup, and no fault of the PR itself. As noted, the same code in #1996 seems to be 100% fine in Travis. I'll leave it up to @kshepherd which one of these to merge...but one should get merged and the other closed.

[minurmin]

I suggest to close this one and use #1996 for 6.3. I have submitted another PR #2001 rebased to master.