We now type this module with django version which allows us to.

This fixes the debian buster build [1]

[1] https://jenkins.softwareheritage.org/job/debian/job/packages/job/DDEP/job/gbp-buildpackage/94/consoleFull

Closes T3072

python3-django all 1:1.11.29-1~deb10u1 (debian stable) and `django.urls.reverse` are not
going well together in test apparently.

[1] https://jenkins.softwareheritage.org/job/debian/job/packages/job/DDEP/job/gbp-buildpackage/92/

The internal detail of zip format revealed a limitation when it comes to date anterior
to 1980. The tar format seems to overcome that limitation so we switch to internally use
such format.

Related to T3070

Needs djangorestframework-stubs>=1.4 for proper mypy checking.

This explicitely checks for the presence of swh:origin_to_add or swh:create_origin
within the provided metadata file.

If not present, this logs a warning about it.

Related to T2860

This no longer checks the error messages.

Fixes build [1]

[1] https://jenkins.softwareheritage.org/job/DDEP/job/tests/1273/console

The --slug behavior is marked as deprecated but still usable. The new --create-origin
flag allows to generate the proper xml stanza when the client does not provide any
metadata xml file.

Related to T2860

deposit.parent can be true in two cases:

1. because we got <swh:add_to_origin> -> then the deprecation warning is raised
   before this check is reached, because 'deposit.origin_url' is set
2. because we got a slug -> then we shouldn't raise this error (it breaks
   existing clients that didn't stop using <external_identifier> yet)

Related to T2928

Related to T2929

They were only handled in non-multipart ones.

It will be reused by _multipart_upload in the next commit, and makes _atom_entry smaller.

It makes it clearer which are the important arguments for the test.

They were not aligned properly, which causes Black to crash:

```
$ black --version
black, version 20.8b1

$ cat test.py
def f():
    # fmt: off
    if True:
        print()
        # fmt: on

def g():
    pass

$ black test.py
error: cannot format test.py: INTERNAL ERROR: Black produced code that is not equivalent to the source.  Please report a bug on https://github.com/psf/black/issues.  This diff might be helpful: /tmp/blk_phpp_27o.log
Oh no!   
1 file failed to reformat.
```

To group them more logically.

Otherwise, querying /1/private/<deposit_id>/meta/ will crash because it fails
to parse the date.

Resolves T2906.

This should allow to deprecate or simplify those when we start using a `swh
deposit --config-file` or `swh --config-file deposit` or even moving those
flags up to the main `swh deposit` command.

Related to T2886

Otherwise, a check task is scheduled for checks and current metadata-only
deposit are rejected with a "Deposit without archive is rejected" message.

Which in this case is not correct.

Related to T2886

Related to T2886

Related to T2886

So it can be reused in other non django part of the deposit.

Related to T2886

Prior to this commit, running this script would not work when actually
triggered on a db with data:

```
swhdeposit@deposit:~$ django-admin migrate --settings=swh.deposit.settings.production
Operations to perform:
  Apply all migrations: auth, contenttypes, deposit, sessions
Running migrations:
  Applying deposit.0021_deposit_origin_url_20201124_1438...Traceback (most recent call last):
...
TypeError: guess_deposit_origin_url() missing 1 required positional argument: 'deposit'
```

eg. 'atom:authorblahblah' should not be accepted when we expect 'atom:author'

Accept <codemeta:name> and <codemeta:author> as alternatives to <atom:name>/<atom:title> and <atom:author>.

This was broken by a8e86a92,
as the check_metadata() checks whether there is a tag that *contains*
the expected name, checking for 'author' (the old name of 'atom:author')
accidentally matched 'codemeta:author' as well.

This resulted in the right behavior in the majority of the cases
(accepted 'codemeta:author'), but for the wrong reason, and explicitly
renaming to atom:author broke this.

Ditto for name.

A future commit will remove the substring matching to remove
false positives (eg. 'atom:authorblahblah' should not be accepted
as 'atom:author')

Motivations:

1. it was never implemented because we don't currently need it
2. if we ever need it, we might go for SWORDv3's ByReference instead

Computing origin urls based on provider url + slug is brittle.

Additionally, the Slug is not designed to define relationships between deposits.