Merge with origin

Use hashutil.hashdata for hashcode computation in ObjStorage

Rename previous_content to last_content.

See previous diff; I made the update from the wrong branch.

No changes, but current diff needs to be updated as the parent commit as changed due to a rebase in top of previously pushed diff.

Perform some minor corrections.
There is still some work to do about the prototypes.

Remove docstring in subclasses

Remove pointless docstrings when abc's one can be used

Make the BaseContentChecker class an Abtract Base Class

• Factorize hash computation into objstorage
• Rename an argument that didn't follow the API

Correct some docstring and use ABC for the base class.

Correct some mistakes in docstrings

Remove a forgotten TODO

I've made some investigations : S3 does have GET and PUT operations on blobs that are referenced by a name (could be sha1). This is sufficient for our ObjStorage needs.

Submit identical diff due to merge with master

Remove cli arguments & correct config override in tests

Change the way Archiver's config is overriden in the tests

Remove outdated manual test for the archiver

• archiver.worker: Remove additional config

• The Director and Worker configuration are now parsed using swh.core.config.SWHConfig mixin class
• A bug where ongoing contents could be used as sources have been corrected

Correct json format and move integrity check from copier to worker (previous position was a mistake)

Corrections of the specs and add the json schema description

Add the update in sql/archiver/swh-archiver-schema.sql

Correct an import

Make the worker correclty catch corrupted or missing contents that were referenced as 'present' into the db

Note that the current version does not yet change the status of a content when it is found to be corrupted.

Closed by D83 & D84