How to use the isbnlib.dev.vias.parallel function in isbnlib

def _fake_provider_merge(isbn):
    """Fake provider 'merge' service."""
    try:  # pragma: no cover
        named_tasks = (('openl', _oed), ('thingl', _ted), ('wiki', _wiki))
        results = vias.parallel(named_tasks, isbn)
        odata = results.get('openl', set())
        tdata = results.get('thingl', set())
        wdata = results.get('wiki', set())
        return list(odata | tdata | wdata)
    except Exception:  # pragma: no cover
        LOGGER.error("Some error on editions 'merge' service for %s!", isbn)
        return [isbn]

def query(isbn, processor=None):
    """Query function for the 'merge provider' (waterfall model)."""
    if not processor:
        processor = config.options.get('VIAS_MERGE', processor).lower()
        if not processor:  # pragma: no cover
            processor = 'parallel'

    named_tasks = (('wcat', qwcat), ('goob', qgoob), ('openl', qopen))
    if processor == 'parallel':
        results = vias.parallel(named_tasks, isbn)
    elif processor == 'serial':
        results = vias.serial(named_tasks, isbn)
    elif processor == 'multi':
        results = vias.multi(named_tasks, isbn)

    rw = results.get('wcat')
    rg = results.get('goob') or results.get('openl')

    if not rw and not rg:  # pragma: no cover
        return None

    md = Metadata(rw) if rw else None

    if md and rg:
        # Overwrite with Authors, Publisher and Year from Google
        md.merge(rg, overwrite=('Authors', 'Publisher', 'Year'))

def query(isbn,
          qserv1,
          qserv2,
          overwrite=('Authors', 'Publisher', 'Language'),
          processor=None):
    """Query function for the 'merge provider' (waterfall model)."""
    if not processor:
        processor = config.options.get('VIAS_MERGE', processor).lower()
        if not processor:  # pragma: no cover
            processor = 'serial'

    named_tasks = (('serv1', qserv1), ('serv2', qserv2))
    if processor == 'parallel':
        results = vias.parallel(named_tasks, isbn)
    elif processor == 'serial':
        results = vias.serial(named_tasks, isbn)
    elif processor == 'multi':
        results = vias.multi(named_tasks, isbn)

    r1 = results.get('serv1')
    r2 = results.get('serv2')

    if not r1 and not r2:
        return {}

    md = Metadata(r1) if r1 else {}

    if md and r2:
        # Try to complete Authors, Publisher and Language from serv2
        md.merge(r2, overwrite=overwrite)