models: HostBib: Properly handle 774s without $w

[awilfox]

Using .empty? still requires the value to not be nil. There is at least one record that has no w and therefore has a nil value; using .present? checks for nil and empty values.

Fixes: LIBIT-10644
Ref: AP-778

[anarchivist]

r+; looks like a reasonable change. do we need another spec/fixture to handle testing 774s without subfield $w?

the rubocop failure appears unrelated.

[davezuckerman]

Other than the rubocop issue it looks good to go.

[awilfox]

v2:

• Fix rubocop issue (prefer present over blank).

[awilfox]

r+; looks like a reasonable change. do we need another spec/fixture to handle testing 774s without subfield $w?

Yeah, I think that's a good idea. I'll add one.

[awilfox]

Since this should be processed like the other records, I'm not sure where to add it in spec/lib/bibliographic/report_spec.rb; would it be mmsid_succeeded_without_774w and no line in the log content?

[awilfox]

v3:

• Fix inverted logic caused by trying to fix the Rubocop issue, which thankfully triggered another Rubocop issue.

[awilfox]

v4:

• Add test.
• Test revealed that sometimes, we bubble up a nil into the subfields list, which isn't appropriate. Fix that as well.

[awilfox]

v5:

• Fix Rubocop warning.

[anarchivist]

r+. looks great; thanks !