mrrc 0.7.6

"""
Advanced unit tests for mrrc Python wrapper.
Tests for edge cases, format conversions, and comprehensive API coverage.
"""

import pytest
import mrrc
from mrrc import MARCReader, MARCWriter, Record, Field, Leader
import io


def create_field(tag, ind1='0', ind2='0', **subfields):
    """Helper to create a field with subfields."""
    field = Field(tag, ind1, ind2)
    for code, value in subfields.items():
        field.add_subfield(code, value)
    return field


class TestRecordEdgeCases:
    """Test edge cases for records."""

    def test_empty_record_serialization(self):
        """Test serializing an empty record."""
        record = Record()
        
        # Should serialize without error
        json_str = record.to_json()
        assert json_str is not None
        
        xml_str = record.to_xml()
        assert xml_str is not None
        
        marc_bytes = record.to_marc21()
        assert isinstance(marc_bytes, bytes)
        assert len(marc_bytes) >= 24  # At least leader


class TestRecordMultipleFields:
    """Test records with multiple fields."""

    def test_add_multiple_fields_same_tag(self):
        """Test adding multiple fields with the same tag."""
        record = Record()
        
        for i in range(5):
            field = Field('650', ' ', '0')
            field.add_subfield('a', f'Subject {i}')
            record.add_field(field)
        
        subjects = record.subjects
        assert len(subjects) >= 5


class TestFieldOperations:
    """Test field-level operations."""

    def test_field_subfields_iteration(self):
        """Test iterating over field subfields."""
        field = Field('245', '1', '0')
        field.add_subfield('a', 'Main title')
        field.add_subfield('b', 'Subtitle')
        field.add_subfield('c', 'Responsibility')
        
        subfields = field.subfields()
        assert len(subfields) == 3
        
        codes = [sf.code for sf in subfields]
        assert 'a' in codes
        assert 'b' in codes
        assert 'c' in codes


    def test_field_subfields_by_code(self):
        """Test getting subfields by code."""
        field = Field('300', ' ', ' ')
        field.add_subfield('a', 'Pages')
        field.add_subfield('c', 'Height')
        
        a_values = field.subfields_by_code('a')
        assert len(a_values) == 1
        assert a_values[0] == 'Pages'
        
        c_values = field.subfields_by_code('c')
        assert len(c_values) == 1
        assert c_values[0] == 'Height'


    def test_field_dict_like_access(self):
        """Test dictionary-like access to subfield values."""
        field = Field('245', '1', '0')
        field.add_subfield('a', 'Title Value')
        field.add_subfield('b', 'Subtitle Value')
        
        # Test __getitem__
        assert field['a'] == 'Title Value'
        assert field['b'] == 'Subtitle Value'
        
        # Test __contains__
        assert 'a' in field
        assert 'b' in field
        assert 'z' not in field
        
        # Test get with default
        assert field.get('a') == 'Title Value'
        assert field.get('missing', 'default') == 'default'


    def test_field_get_subfields(self):
        """Test getting multiple subfields at once."""
        field = Field('260', ' ', ' ')
        field.add_subfield('a', 'New York')
        field.add_subfield('b', 'Publisher')
        field.add_subfield('c', '2023')
        field.add_subfield('d', 'Distributor')
        
        # Get multiple codes at once
        values = field.get_subfields('a', 'b')
        assert 'New York' in values
        assert 'Publisher' in values


    def test_field_indicators_mutation(self):
        """Test modifying field indicators."""
        field = Field('245', '0', '0')
        assert field.indicator1 == '0'
        assert field.indicator2 == '0'
        
        # Modify indicators
        field.indicator1 = '1'
        field.indicator2 = '4'
        
        assert field.indicator1 == '1'
        assert field.indicator2 == '4'


class TestRecordRoundTrip:
    """Test round-trip serialization and deserialization."""

    def test_roundtrip_with_control_fields(self):
        """Test round-trip with control fields."""
        original = Record()
        original.add_control_field('001', 'control-123')
        original.add_control_field('003', 'ABC')
        original.add_control_field('005', '20231231120000.0')
        
        # Serialize
        marc_bytes = original.to_marc21()
        
        # Deserialize
        reader = MARCReader(io.BytesIO(marc_bytes))
        restored = reader.read_record()
        
        assert restored is not None
        assert restored.control_field('001') == 'control-123'
        assert restored.control_field('003') == 'ABC'


    def test_roundtrip_with_multiple_fields(self):
        """Test round-trip with multiple data fields."""
        original = Record()
        original.add_control_field('001', 'id-456')
        
        # Add various fields
        original.add_field(create_field('245', '1', '0', 
                                        a='Title', b='Subtitle'))
        original.add_field(create_field('100', '1', ' ', a='Author'))
        original.add_field(create_field('020', ' ', ' ', a='ISBN123'))
        original.add_field(create_field('650', ' ', '0', a='Subject 1'))
        original.add_field(create_field('650', ' ', '0', a='Subject 2'))
        
        # Serialize and deserialize
        marc_bytes = original.to_marc21()
        reader = MARCReader(io.BytesIO(marc_bytes))
        restored = reader.read_record()
        
        assert restored is not None
        assert restored.title is not None
        assert restored.author is not None
        assert restored.isbn is not None
        assert len(restored.subjects) >= 2


    def test_roundtrip_preserves_indicators(self):
        """Test that round-trip preserves field indicators."""
        original = Record()
        
        # Add field with specific indicators
        field = Field('245', '1', '4')
        field.add_subfield('a', 'The title')
        original.add_field(field)
        
        # Serialize and deserialize
        marc_bytes = original.to_marc21()
        reader = MARCReader(io.BytesIO(marc_bytes))
        restored = reader.read_record()
        
        restored_field = restored['245']
        assert restored_field is not None
        assert restored_field.indicator1 == '1'
        assert restored_field.indicator2 == '4'


class TestFormatConversions:
    """Test conversions to various formats."""

    def test_to_json_format(self):
        """Test JSON serialization produces valid output."""
        record = Record()
        record.add_field(create_field('245', '1', '0', a='Test Title'))
        
        json_str = record.to_json()
        assert isinstance(json_str, str)
        assert len(json_str) > 0
        
        # Should contain field data
        assert '245' in json_str or 'Test Title' in json_str


    def test_to_xml_format(self):
        """Test XML serialization produces valid output."""
        record = Record()
        record.add_field(create_field('245', '1', '0', a='Test Title'))
        
        xml_str = record.to_xml()
        assert isinstance(xml_str, str)
        assert len(xml_str) > 0
        assert '<' in xml_str


    def test_to_marcjson_format(self):
        """Test MARCJSON serialization."""
        record = Record()
        record.add_control_field('001', 'test-id')
        record.add_field(create_field('245', '1', '0', a='Title'))
        
        marcjson_str = record.to_marcjson()
        assert isinstance(marcjson_str, str)
        assert len(marcjson_str) > 0


    def test_dublin_core_conversion(self):
        """Test Dublin Core metadata conversion."""
        record = Record()
        record.add_field(create_field('245', '1', '0', a='Test Title'))
        record.add_field(create_field('100', '1', ' ', a='Test Author'))
        record.add_field(create_field('260', ' ', ' ', 
                                      b='Test Publisher', c='2023'))
        
        dc = record.to_dublin_core()
        
        assert isinstance(dc, dict)
        assert 'title' in dc
        assert 'creator' in dc
        assert 'publisher' in dc


class TestFormatConversionWrapping:
    """Test that format conversion functions return properly wrapped Python objects."""

    def _make_marcjson(self):
        """Create a MARCJSON string for testing."""
        import json
        return json.dumps([
            {"leader": "01826cam a2200421 a 4500"},
            {"001": "12345"},
            {"245": {"ind1": "1", "ind2": "0", "subfields": [{"a": "Test title /"}]}},
            {"650": {"ind1": " ", "ind2": "0", "subfields": [{"a": "Testing."}]}},
        ])

    def test_marcjson_to_record_returns_wrapped_record(self):
        """Test that marcjson_to_record returns a Python Record, not a raw Rust object."""
        from mrrc import marcjson_to_record
        record = marcjson_to_record(self._make_marcjson())
        assert type(record).__name__ == 'Record'
        assert type(record).__module__ == 'mrrc'

    def test_marcjson_to_record_fields_are_wrapped(self):
        """Test that fields from marcjson_to_record records support subscript access."""
        from mrrc import marcjson_to_record
        record = marcjson_to_record(self._make_marcjson())
        fields = record.get_fields('245')
        assert len(fields) == 1
        f = fields[0]
        assert type(f).__name__ == 'Field'
        assert type(f).__module__ == 'mrrc'
        # Subscript access should work
        assert f['a'] == 'Test title /'

    def test_marcjson_to_record_leader_is_wrapped(self):
        """Test that leader from marcjson_to_record records supports indexing."""
        from mrrc import marcjson_to_record
        record = marcjson_to_record(self._make_marcjson())
        ldr = record.leader()
        assert type(ldr).__name__ == 'Leader'
        assert type(ldr).__module__ == 'mrrc'
        assert ldr[9] is not None

    def test_marcjson_to_record_helper_methods_work(self):
        """Test that helper methods work on marcjson_to_record records."""
        from mrrc import marcjson_to_record
        record = marcjson_to_record(self._make_marcjson())
        assert record.title == 'Test title /'
        assert record.subjects == ['Testing.']

    def test_json_to_record_returns_wrapped_record(self):
        """Test that json_to_record returns a wrapped Python Record."""
        from mrrc import json_to_record
        record = Record()
        record.add_field(create_field('245', '1', '0', a='Test Title'))
        json_str = record.to_json()
        restored = json_to_record(json_str)
        assert type(restored).__name__ == 'Record'
        assert type(restored).__module__ == 'mrrc'
        assert restored.title == 'Test Title'
        fields = restored.get_fields('245')
        assert fields[0]['a'] == 'Test Title'

    def test_xml_to_record_returns_wrapped_record(self):
        """Test that xml_to_record returns a wrapped Python Record."""
        from mrrc import xml_to_record
        record = Record()
        record.add_field(create_field('245', '1', '0', a='Test Title'))
        xml_str = record.to_xml()
        restored = xml_to_record(xml_str)
        assert type(restored).__name__ == 'Record'
        assert type(restored).__module__ == 'mrrc'
        assert restored.title == 'Test Title'
        fields = restored.get_fields('245')
        assert fields[0]['a'] == 'Test Title'


class TestMarcxmlConformance:
    """MARCXML conformance tests against standard LOC format."""

    def test_output_has_xml_declaration(self):
        """Verify output starts with XML declaration."""
        record = Record()
        record.add_control_field('001', 'test')
        xml_str = record.to_xml()
        assert xml_str.startswith('<?xml version="1.0" encoding="UTF-8"?>')

    def test_output_has_xmlns(self):
        """Verify output contains xmlns namespace declaration."""
        record = Record()
        record.add_control_field('001', 'test')
        xml_str = record.to_xml()
        assert 'xmlns="http://www.loc.gov/MARC21/slim"' in xml_str

    def test_tag_ind_code_are_attributes(self):
        """Verify tag, ind1, ind2, code are XML attributes (not child elements)."""
        record = Record()
        record.add_control_field('001', '12345')
        record.add_field(create_field('245', '1', '0', a='Title'))
        xml_str = record.to_xml()
        assert '<controlfield tag="001">' in xml_str
        assert '<datafield tag="245" ind1="1" ind2="0">' in xml_str
        assert '<subfield code="a">' in xml_str

    def test_parse_standard_marcxml_no_namespace(self):
        """Parse MARCXML without namespace."""
        from mrrc import xml_to_record
        xml = '''<record>
            <leader>01234nam a2200289 a 4500</leader>
            <controlfield tag="001">12345</controlfield>
            <datafield tag="245" ind1="1" ind2="0">
                <subfield code="a">Test title</subfield>
            </datafield>
        </record>'''
        record = xml_to_record(xml)
        assert record.control_field('001') == '12345'
        assert record.title == 'Test title'

    def test_parse_marcxml_with_default_namespace(self):
        """Parse MARCXML with default xmlns."""
        from mrrc import xml_to_record
        xml = '''<record xmlns="http://www.loc.gov/MARC21/slim">
            <leader>01234nam a2200289 a 4500</leader>
            <controlfield tag="001">99999</controlfield>
            <datafield tag="245" ind1="0" ind2="0">
                <subfield code="a">Namespaced title</subfield>
            </datafield>
        </record>'''
        record = xml_to_record(xml)
        assert record.control_field('001') == '99999'
        assert record.title == 'Namespaced title'

    def test_parse_marcxml_with_prefix_namespace(self):
        """Parse MARCXML with marc: prefix namespace."""
        from mrrc import xml_to_record
        xml = '''<marc:record xmlns:marc="http://www.loc.gov/MARC21/slim">
            <marc:leader>01234nam a2200289 a 4500</marc:leader>
            <marc:controlfield tag="001">88888</marc:controlfield>
            <marc:datafield tag="245" ind1="1" ind2="0">
                <marc:subfield code="a">Prefixed title</marc:subfield>
            </marc:datafield>
        </marc:record>'''
        record = xml_to_record(xml)
        assert record.control_field('001') == '88888'
        assert record.title == 'Prefixed title'

    def test_parse_collection_with_default_namespace(self):
        """Parse <collection> wrapper with default namespace."""
        from mrrc import xml_to_records
        xml = '''<collection xmlns="http://www.loc.gov/MARC21/slim">
            <record>
                <leader>01234nam a2200289 a 4500</leader>
                <controlfield tag="001">rec1</controlfield>
            </record>
            <record>
                <leader>01234nam a2200289 a 4500</leader>
                <controlfield tag="001">rec2</controlfield>
            </record>
        </collection>'''
        records = xml_to_records(xml)
        assert len(records) == 2
        assert records[0].control_field('001') == 'rec1'
        assert records[1].control_field('001') == 'rec2'

    def test_parse_collection_with_prefix_namespace(self):
        """Parse <marc:collection> wrapper with prefix namespace."""
        from mrrc import xml_to_records
        xml = '''<marc:collection xmlns:marc="http://www.loc.gov/MARC21/slim">
            <marc:record>
                <marc:leader>01234nam a2200289 a 4500</marc:leader>
                <marc:controlfield tag="001">pfx1</marc:controlfield>
            </marc:record>
            <marc:record>
                <marc:leader>01234nam a2200289 a 4500</marc:leader>
                <marc:controlfield tag="001">pfx2</marc:controlfield>
            </marc:record>
        </marc:collection>'''
        records = xml_to_records(xml)
        assert len(records) == 2
        assert records[0].control_field('001') == 'pfx1'
        assert records[1].control_field('001') == 'pfx2'

    def test_parse_loc_collection_fixture(self):
        """Parse the LOC collection.xml fixture (marc: prefix, 2 records)."""
        import os
        from mrrc import xml_to_records
        fixture = os.path.join(os.path.dirname(__file__), '..', 'data', 'loc_collection.marcxml')
        with open(fixture, 'r', encoding='utf-8') as f:
            xml = f.read()
        records = xml_to_records(xml)
        assert len(records) == 2
        # First record: "The Great Ray Charles"
        assert records[0].control_field('001') == '5637241'
        assert 'The Great Ray Charles' in (records[0].title or '')
        # Second record: "The White House"
        assert records[1].control_field('001') == '12149120'
        assert 'The White House' in (records[1].title or '')

    def test_marcxml_roundtrip_preserves_data(self):
        """Roundtrip: record → MARCXML → record preserves all data."""
        from mrrc import xml_to_record
        record = Record()
        record.add_control_field('001', 'roundtrip-001')
        record.add_control_field('008', '230601s2023    xxu||||||||||||eng d')
        record.add_field(create_field('245', '1', '0', a='Roundtrip Title /', c='Author.'))
        record.add_field(create_field('650', ' ', '0', a='Testing.'))
        record.add_field(create_field('650', ' ', '0', a='Software.'))
        xml_str = record.to_xml()
        restored = xml_to_record(xml_str)
        assert restored.control_field('001') == 'roundtrip-001'
        assert restored.title == 'Roundtrip Title /'
        fields = restored.get_fields('245')
        assert fields[0]['c'] == 'Author.'
        subjects = restored.get_fields('650')
        assert len(subjects) == 2

    def test_parse_issue_15_example(self):
        """Parse the exact MARCXML from issue #15 (Introduction to Algorithms)."""
        from mrrc import xml_to_record
        xml = '''<?xml version="1.0" encoding="UTF-8"?>
        <record xmlns="http://www.loc.gov/MARC21/slim">
            <leader>01142cam  2200301 a 4500</leader>
            <controlfield tag="001">92005291</controlfield>
            <controlfield tag="008">920219s1990    mau           001 0 eng  </controlfield>
            <datafield tag="020" ind1=" " ind2=" ">
                <subfield code="a">0262031418</subfield>
            </datafield>
            <datafield tag="245" ind1="1" ind2="0">
                <subfield code="a">Introduction to algorithms /</subfield>
                <subfield code="c">Thomas H. Cormen ... [et al.].</subfield>
            </datafield>
            <datafield tag="650" ind1=" " ind2="0">
                <subfield code="a">Computer programming.</subfield>
            </datafield>
            <datafield tag="650" ind1=" " ind2="0">
                <subfield code="a">Computer algorithms.</subfield>
            </datafield>
        </record>'''
        record = xml_to_record(xml)
        assert record.control_field('001') == '92005291'
        assert record.title == 'Introduction to algorithms /'
        fields = record.get_fields('245')
        assert fields[0]['c'] == 'Thomas H. Cormen ... [et al.].'
        isbn = record.isbn
        assert isbn == '0262031418'
        subjects = record.get_fields('650')
        assert len(subjects) == 2

    def test_xml_to_records_returns_wrapped_records(self):
        """Verify xml_to_records returns properly wrapped Python Records."""
        from mrrc import xml_to_records
        xml = '''<collection>
            <record>
                <leader>01234nam a2200289 a 4500</leader>
                <controlfield tag="001">wrap1</controlfield>
                <datafield tag="245" ind1="1" ind2="0">
                    <subfield code="a">Title One</subfield>
                </datafield>
            </record>
        </collection>'''
        records = xml_to_records(xml)
        assert len(records) == 1
        assert type(records[0]).__name__ == 'Record'
        assert type(records[0]).__module__ == 'mrrc'
        assert records[0].title == 'Title One'


class TestControlFields:
    """Test control field operations."""

    def test_control_field_roundtrip(self):
        """Test control field preservation in round-trip."""
        record = Record()
        
        test_fields = {
            '001': '12345',
            '003': 'DLC',
            '005': '20231201',
            '006': 'fixed006value',
            '007': 'fixed007value',
            '008': '230601s2023    xxu||||||||||||eng d'
        }
        
        for tag, value in test_fields.items():
            record.add_control_field(tag, value)
        
        # Verify they're there
        for tag, expected_value in test_fields.items():
            actual = record.control_field(tag)
            assert actual == expected_value


    def test_multiple_control_fields(self):
        """Test getting all control fields."""
        record = Record()
        record.add_control_field('001', 'id1')
        record.add_control_field('003', 'source1')
        
        cfs = record.control_fields()
        assert len(cfs) >= 2


class TestRecordTypeDetection:
    """Test record type detection methods."""

    def test_is_book_detection(self):
        """Test book detection."""
        leader = Leader()
        leader.record_type = 'a'
        leader.bibliographic_level = 'm'
        record = Record(leader)
        
        assert record.is_book() is True


    def test_is_serial_detection(self):
        """Test serial detection."""
        leader = Leader()
        leader.record_type = 'a'
        leader.bibliographic_level = 's'
        record = Record(leader)
        
        assert record.is_serial() is True


    def test_is_music_detection(self):
        """Test music detection."""
        leader = Leader()
        leader.record_type = 'c'
        record = Record(leader)
        
        assert record.is_music() is True


    def test_is_audiovisual_detection(self):
        """Test audiovisual detection."""
        leader = Leader()
        leader.record_type = 'g'
        record = Record(leader)
        
        assert record.is_audiovisual() is True


class TestRecordRemoval:
    """Test field removal operations."""

    def test_remove_field_by_tag(self):
        """Test removing a field by tag."""
        record = Record()
        field = Field('245', '1', '0')
        field.add_subfield('a', 'Test')
        record.add_field(field)
        
        assert record['245'] is not None
        
        # Remove the field
        record.remove_field('245')
        
        # Verify it's gone (pymarc raises KeyError for missing tags)
        with pytest.raises(KeyError):
            record['245']


class TestFieldSerialization:
    """Test field-level serialization."""

    def test_field_with_repeated_subfields(self):
        """Test field with repeated subfield codes."""
        field = Field('300', ' ', ' ')
        field.add_subfield('a', 'Part 1')
        field.add_subfield('a', 'Part 2')
        field.add_subfield('c', 'Size')
        
        # Get all 'a' values
        a_values = field.subfields_by_code('a')
        assert len(a_values) == 2
        assert 'Part 1' in a_values
        assert 'Part 2' in a_values


class TestLeaderProperties:
    """Test Leader property access and modification."""

    def test_leader_defaults(self):
        """Test default leader values."""
        leader = Leader()
        
        assert leader.record_type == 'a'
        assert leader.bibliographic_level == 'm'
        assert leader.record_status == 'n'
        assert leader.character_coding == ' '


    def test_leader_modification(self):
        """Test modifying leader properties."""
        leader = Leader()
        
        leader.record_type = 'c'
        assert leader.record_type == 'c'
        
        leader.bibliographic_level = 'd'
        assert leader.bibliographic_level == 'd'
        
        leader.record_status = 'a'
        assert leader.record_status == 'a'


    def test_leader_encoding_level(self):
        """Test leader encoding level setting."""
        leader = Leader()
        leader.encoding_level = '4'
        assert leader.encoding_level == '4'


    def test_leader_cataloging_form(self):
        """Test leader cataloging form (descriptor_cataloging_form)."""
        leader = Leader()
        
        # Test via descriptor_cataloging_form property
        leader.descriptive_cataloging_form = 'a'
        assert leader.descriptive_cataloging_form == 'a'


class TestUnicodeAndEncoding:
    """Test unicode and special character handling."""

    def test_field_with_unicode_subfields(self):
        """Test fields with unicode characters."""
        field = Field('245', '1', '0')
        field.add_subfield('a', 'Tïtlé wíth üñíçödé')
        
        subfields = field.subfields_by_code('a')
        assert 'üñíçödé' in subfields[0]


    def test_record_with_unicode_fields(self):
        """Test record with unicode content."""
        record = Record()
        
        field = Field('245', '1', '0')
        field.add_subfield('a', '日本語タイトル')
        record.add_field(field)
        
        title = record.title
        assert title is not None
        assert '日本語' in title


    def test_roundtrip_preserves_unicode(self):
        """Test that round-trip preserves unicode characters."""
        original = Record()
        
        field = Field('245', '1', '0')
        field.add_subfield('a', 'Titel in Français')
        original.add_field(field)
        
        # Serialize and deserialize
        marc_bytes = original.to_marc21()
        reader = MARCReader(io.BytesIO(marc_bytes))
        restored = reader.read_record()
        
        restored_field = restored['245']
        assert restored_field is not None
        a_values = restored_field.subfields_by_code('a')
        assert 'Français' in a_values[0]


class TestMARCWriterIntegration:
    """Test MARCWriter integration."""

    def test_write_multiple_records(self):
        """Test writing multiple records to a stream."""
        buffer = io.BytesIO()
        writer = MARCWriter(buffer)
        
        # Write 3 records
        for i in range(3):
            record = Record()
            record.add_control_field('001', f'id-{i}')
            field = Field('245', '1', '0')
            field.add_subfield('a', f'Title {i}')
            record.add_field(field)
            writer.write(record)
        
        # Read them back
        buffer.seek(0)
        reader = MARCReader(buffer)
        
        count = 0
        for record in reader:
            assert record is not None
            count += 1
        
        assert count == 3


class TestFieldConveniences:
    """Test convenience methods for common fields."""

    def test_get_multiple_fields_by_tags(self):
        """Test getting multiple fields by multiple tags."""
        record = Record()
        
        record.add_field(create_field('245', '1', '0', a='Title'))
        record.add_field(create_field('100', '1', ' ', a='Author'))
        record.add_field(create_field('260', ' ', ' ', b='Publisher'))
        record.add_field(create_field('300', ' ', ' ', a='Pages'))
        
        # Get multiple tags at once
        fields = record.get_fields('245', '100', '260')
        assert len(fields) >= 3


    def test_all_fields_access(self):
        """Test getting all fields at once."""
        record = Record()
        
        record.add_field(create_field('245', '1', '0', a='Title'))
        record.add_field(create_field('100', '1', ' ', a='Author'))
        record.add_field(create_field('650', ' ', '0', a='Subject'))
        
        all_fields = record.get_fields()
        assert len(all_fields) >= 3


class TestLinkedFields:
    """Test 880 alternate graphic representation field linkage."""

    def _build_record_with_880(self, tag, ind1, ind2, occurrence,
                                romanized_subfields, script_subfields,
                                script_code=None):
        """Helper: build a record with one original field and one linked 880.

        Args:
            tag: Original field tag (e.g., '245')
            ind1, ind2: Indicators for both original and 880
            occurrence: Two-digit occurrence string (e.g., '01')
            romanized_subfields: dict of code->value for the original field
            script_subfields: dict of code->value for the 880 field
            script_code: Optional MARC script code (e.g., '(2/r' for Hebrew RTL)
        """
        record = Record()
        record.add_control_field('001', 'test-linked')

        # Build $6 values
        orig_6 = f'880-{occurrence}'
        if script_code:
            linked_6 = f'{tag}-{occurrence}/{script_code}'
        else:
            linked_6 = f'{tag}-{occurrence}'

        # Original field with $6 linkage
        orig = Field(tag, ind1, ind2)
        orig.add_subfield('6', orig_6)
        for code, value in romanized_subfields.items():
            orig.add_subfield(code, value)
        record.add_field(orig)

        # Linked 880 field
        linked = Field('880', ind1, ind2)
        linked.add_subfield('6', linked_6)
        for code, value in script_subfields.items():
            linked.add_subfield(code, value)
        record.add_field(linked)

        return record

    # ------------------------------------------------------------------
    # Hebrew (RTL) – Soncino Mishneh Torah example from the issue
    # ------------------------------------------------------------------

    def test_hebrew_title_linkage(self):
        """Test 880 linkage for Hebrew title (RTL script)."""
        record = self._build_record_with_880(
            '245', '1', '0', '01',
            romanized_subfields={'a': 'Mishneh Torah.'},
            script_subfields={'a': 'משנה תורה.'},
            script_code='(2/r',
        )
        f245 = record.get_fields('245')[0]
        linked = record.get_linked_fields(f245)
        assert len(linked) == 1
        assert linked[0].tag == '880'
        assert linked[0]['a'] == 'משנה תורה.'

    def test_hebrew_publisher_linkage(self):
        """Test 880 linkage for Hebrew publisher (RTL script)."""
        record = self._build_record_with_880(
            '260', ' ', ' ', '03',
            romanized_subfields={
                'a': 'Śontsino :',
                'b': 'Gershom ben Mosheh ish Śontsino,',
            },
            script_subfields={
                'a': 'שונצינו :',
                'b': 'גרשם בן משה איש שונצינו,',
            },
            script_code='(2/r',
        )
        f260 = record.get_fields('260')[0]
        linked = record.get_linked_fields(f260)
        assert len(linked) == 1
        assert 'שונצינו' in linked[0]['a']

    # ------------------------------------------------------------------
    # Arabic (RTL) – An Arabic novel
    # ------------------------------------------------------------------

    def test_arabic_title_linkage(self):
        """Test 880 linkage for Arabic title (RTL script)."""
        record = self._build_record_with_880(
            '245', '1', '0', '01',
            romanized_subfields={'a': 'Awlād ḥāratinā /'},
            script_subfields={'a': 'أولاد حارتنا /'},
            script_code='(3/r',
        )
        f245 = record.get_fields('245')[0]
        linked = record.get_linked_fields(f245)
        assert len(linked) == 1
        assert linked[0]['a'] == 'أولاد حارتنا /'

    def test_arabic_author_linkage(self):
        """Test 880 linkage for Arabic author (RTL script)."""
        record = self._build_record_with_880(
            '100', '1', ' ', '02',
            romanized_subfields={'a': 'Maḥfūẓ, Najīb,'},
            script_subfields={'a': 'محفوظ، نجيب،'},
            script_code='(3/r',
        )
        f100 = record.get_fields('100')[0]
        linked = record.get_linked_fields(f100)
        assert len(linked) == 1
        assert 'محفوظ' in linked[0]['a']

    # ------------------------------------------------------------------
    # CJK – Chinese book
    # ------------------------------------------------------------------

    def test_cjk_title_linkage(self):
        """Test 880 linkage for CJK (Chinese) title."""
        record = self._build_record_with_880(
            '245', '1', '0', '01',
            romanized_subfields={'a': 'Hong lou meng /'},
            script_subfields={'a': '紅樓夢 /'},
            script_code='$1',
        )
        f245 = record.get_fields('245')[0]
        linked = record.get_linked_fields(f245)
        assert len(linked) == 1
        assert linked[0]['a'] == '紅樓夢 /'

    def test_cjk_author_linkage(self):
        """Test 880 linkage for CJK (Chinese) author."""
        record = self._build_record_with_880(
            '100', '1', ' ', '02',
            romanized_subfields={'a': 'Cao, Xueqin,'},
            script_subfields={'a': '曹雪芹,'},
            script_code='$1',
        )
        f100 = record.get_fields('100')[0]
        linked = record.get_linked_fields(f100)
        assert len(linked) == 1
        assert '曹雪芹' in linked[0]['a']

    # ------------------------------------------------------------------
    # Cyrillic – Russian novel
    # ------------------------------------------------------------------

    def test_cyrillic_title_linkage(self):
        """Test 880 linkage for Cyrillic (Russian) title."""
        record = self._build_record_with_880(
            '245', '1', '0', '01',
            romanized_subfields={'a': 'Voĭna i mir /'},
            script_subfields={'a': 'Война и мир /'},
            script_code='(N',
        )
        f245 = record.get_fields('245')[0]
        linked = record.get_linked_fields(f245)
        assert len(linked) == 1
        assert linked[0]['a'] == 'Война и мир /'

    def test_cyrillic_author_linkage(self):
        """Test 880 linkage for Cyrillic (Russian) author."""
        record = self._build_record_with_880(
            '100', '1', ' ', '02',
            romanized_subfields={'a': 'Tolstoĭ, Lev Nikolaevich,'},
            script_subfields={'a': 'Толстой, Лев Николаевич,'},
            script_code='(N',
        )
        f100 = record.get_fields('100')[0]
        linked = record.get_linked_fields(f100)
        assert len(linked) == 1
        assert 'Толстой' in linked[0]['a']

    # ------------------------------------------------------------------
    # Linkage WITHOUT script identification code
    # (valid MARC — just occurrence number, no script/direction)
    # ------------------------------------------------------------------

    def test_linkage_without_script_code(self):
        """Test 880 linkage using bare occurrence numbers (no script code)."""
        record = self._build_record_with_880(
            '245', '1', '0', '01',
            romanized_subfields={'a': 'Romanized title'},
            script_subfields={'a': 'Vernacular title'},
            script_code=None,  # No script identification
        )
        f245 = record.get_fields('245')[0]
        linked = record.get_linked_fields(f245)
        assert len(linked) == 1
        assert linked[0]['a'] == 'Vernacular title'

    # ------------------------------------------------------------------
    # Subject field linkage (650)
    # ------------------------------------------------------------------

    def test_subject_field_linkage(self):
        """Test 880 linkage for subject headings (650)."""
        record = self._build_record_with_880(
            '650', ' ', '0', '04',
            romanized_subfields={'a': 'Filosofiyah Yehudit'},
            script_subfields={'a': 'פילוסופיה יהודית'},
            script_code='(2/r',
        )
        f650 = record.get_fields('650')[0]
        linked = record.get_linked_fields(f650)
        assert len(linked) == 1
        assert 'פילוסופיה' in linked[0]['a']

    # ------------------------------------------------------------------
    # Series field linkage (490)
    # ------------------------------------------------------------------

    def test_series_field_linkage(self):
        """Test 880 linkage for series statement (490)."""
        record = self._build_record_with_880(
            '490', '1', ' ', '05',
            romanized_subfields={'a': 'Mif ha-sifrut ha-ʻIvrit'},
            script_subfields={'a': 'מיף הספרות העברית'},
            script_code='(2/r',
        )
        f490 = record.get_fields('490')[0]
        linked = record.get_linked_fields(f490)
        assert len(linked) == 1

    # ------------------------------------------------------------------
    # Notes field linkage (500)
    # ------------------------------------------------------------------

    def test_notes_field_linkage(self):
        """Test 880 linkage for general note (500)."""
        record = self._build_record_with_880(
            '500', ' ', ' ', '06',
            romanized_subfields={'a': 'Includes index.'},
            script_subfields={'a': 'כולל מפתח.'},
            script_code='(2/r',
        )
        f500 = record.get_fields('500')[0]
        linked = record.get_linked_fields(f500)
        assert len(linked) == 1
        assert linked[0]['a'] == 'כולל מפתח.'

    # ------------------------------------------------------------------
    # Multiple linked pairs in one record
    # ------------------------------------------------------------------

    def test_multiple_linked_pairs(self):
        """Test record with multiple 880-linked field pairs."""
        record = Record()
        record.add_control_field('001', 'multi-link')

        # Pair 1: Title (245 <-> 880, occurrence 01)
        f245 = Field('245', '1', '0')
        f245.add_subfield('6', '880-01')
        f245.add_subfield('a', 'Mishneh Torah.')
        record.add_field(f245)

        f880_title = Field('880', '1', '0')
        f880_title.add_subfield('6', '245-01/(2/r')
        f880_title.add_subfield('a', 'משנה תורה.')
        record.add_field(f880_title)

        # Pair 2: Author (100 <-> 880, occurrence 02)
        f100 = Field('100', '1', ' ')
        f100.add_subfield('6', '880-02')
        f100.add_subfield('a', 'Maimonides,')
        record.add_field(f100)

        f880_author = Field('880', '1', ' ')
        f880_author.add_subfield('6', '100-02/(2/r')
        f880_author.add_subfield('a', 'רמב״ם,')
        record.add_field(f880_author)

        # Pair 3: Publisher (260 <-> 880, occurrence 03)
        f260 = Field('260', ' ', ' ')
        f260.add_subfield('6', '880-03')
        f260.add_subfield('a', 'Śontsino :')
        record.add_field(f260)

        f880_pub = Field('880', ' ', ' ')
        f880_pub.add_subfield('6', '260-03/(2/r')
        f880_pub.add_subfield('a', 'שונצינו :')
        record.add_field(f880_pub)

        # Verify each pair resolves correctly
        title_fields = record.get_fields('245')
        linked_title = record.get_linked_fields(title_fields[0])
        assert len(linked_title) == 1
        assert 'משנה' in linked_title[0]['a']

        author_fields = record.get_fields('100')
        linked_author = record.get_linked_fields(author_fields[0])
        assert len(linked_author) == 1
        assert 'רמב' in linked_author[0]['a']

        pub_fields = record.get_fields('260')
        linked_pub = record.get_linked_fields(pub_fields[0])
        assert len(linked_pub) == 1
        assert 'שונצינו' in linked_pub[0]['a']

    # ------------------------------------------------------------------
    # Edge cases
    # ------------------------------------------------------------------

    def test_field_without_subfield_6_returns_empty(self):
        """Field with no $6 linkage should return empty list."""
        record = Record()
        f245 = Field('245', '1', '0')
        f245.add_subfield('a', 'A plain title.')
        record.add_field(f245)

        result = record.get_linked_fields(record.get_fields('245')[0])
        assert result == []

    def test_subfield_6_with_no_matching_880_returns_empty(self):
        """Field with $6 but no matching 880 should return empty list."""
        record = Record()
        f245 = Field('245', '1', '0')
        f245.add_subfield('6', '880-01')
        f245.add_subfield('a', 'Orphan title.')
        record.add_field(f245)
        # No 880 field added

        result = record.get_linked_fields(record.get_fields('245')[0])
        assert result == []

    def test_linked_field_is_wrapped_python_field(self):
        """Returned linked fields should be wrapped Python Field objects."""
        record = self._build_record_with_880(
            '245', '1', '0', '01',
            romanized_subfields={'a': 'Romanized'},
            script_subfields={'a': 'Vernacular'},
        )
        f245 = record.get_fields('245')[0]
        linked = record.get_linked_fields(f245)

        assert len(linked) == 1
        # Must be a wrapped Field (supports subscript access)
        assert linked[0]['a'] == 'Vernacular'
        assert linked[0].tag == '880'
        assert hasattr(linked[0], 'subfields')

    def test_get_linked_fields_returns_list(self):
        """get_linked_fields always returns a list, even for single match."""
        record = self._build_record_with_880(
            '245', '1', '0', '01',
            romanized_subfields={'a': 'Title'},
            script_subfields={'a': 'כותרת'},
            script_code='(2/r',
        )
        f245 = record.get_fields('245')[0]
        result = record.get_linked_fields(f245)
        assert isinstance(result, list)

    # ------------------------------------------------------------------
    # Added-entry name linkage (700) — e.g., translator in original script
    # ------------------------------------------------------------------

    def test_added_entry_linkage(self):
        """Test 880 linkage for 700 added entry (translator)."""
        record = self._build_record_with_880(
            '700', '1', ' ', '07',
            romanized_subfields={'a': 'Ibn Tibbon, Shemuel,', 'e': 'translator.'},
            script_subfields={'a': 'אבן תבון, שמואל,', 'e': 'מתרגם.'},
            script_code='(2/r',
        )
        f700 = record.get_fields('700')[0]
        linked = record.get_linked_fields(f700)
        assert len(linked) == 1
        assert 'אבן תבון' in linked[0]['a']

    # ------------------------------------------------------------------
    # Greek script
    # ------------------------------------------------------------------

    def test_greek_title_linkage(self):
        """Test 880 linkage for Greek script."""
        record = self._build_record_with_880(
            '245', '1', '0', '01',
            romanized_subfields={'a': 'Politeia /'},
            script_subfields={'a': 'Πολιτεία /'},
            script_code='(S',
        )
        f245 = record.get_fields('245')[0]
        linked = record.get_linked_fields(f245)
        assert len(linked) == 1
        assert linked[0]['a'] == 'Πολιτεία /'


class TestLinkedFieldMARCJSON:
    """Test get_linked_fields with records loaded via marcjson_to_record."""

    def test_soncino_mishneh_torah(self):
        """Full Soncino Mishneh Torah example from issue #19."""
        import json
        from mrrc import marcjson_to_record

        marcjson = json.dumps([
            {"leader": "05723cam a22006251a 4500"},
            {"001": "2018751272"},
            {"245": {"ind1": "1", "ind2": "0", "subfields": [
                {"6": "880-01"}, {"a": "Mishneh Torah."}
            ]}},
            {"260": {"ind1": " ", "ind2": " ", "subfields": [
                {"6": "880-03"},
                {"a": "Śontsino :"},
                {"b": "Gershom ben Mosheh ish Śontsino,"},
                {"c": "r.ḥ. Nisan shenat 250 [March 23, 1490]"}
            ]}},
            {"880": {"ind1": "1", "ind2": "0", "subfields": [
                {"6": "245-01/(2/r"}, {"a": "משנה תורה."}
            ]}},
            {"880": {"ind1": " ", "ind2": " ", "subfields": [
                {"6": "260-03/(2/r"},
                {"a": "שונצינו :"},
                {"b": "גרשם בן משה איש שונצינו,"},
                {"c": "ר\"ח ניסן שנת נ\"ר"}
            ]}}
        ])
        record = marcjson_to_record(marcjson)

        # Look up linked 880 for title
        f245 = record.get_fields('245')[0]
        linked_title = record.get_linked_fields(f245)
        assert len(linked_title) == 1
        assert linked_title[0]['a'] == 'משנה תורה.'

        # Look up linked 880 for publisher
        f260 = record.get_fields('260')[0]
        linked_pub = record.get_linked_fields(f260)
        assert len(linked_pub) == 1
        assert 'שונצינו' in linked_pub[0]['a']


class TestSerializationRoundTrip:
    """Test that records from deserialization functions can be re-serialized.

    Regression tests for GitHub issue #57: record_to_json/xml fail on records
    returned by xml_to_record() with 'Record' object is not an instance of 'Record'.
    """

    def test_xml_to_record_then_record_to_json(self):
        """xml_to_record() result can be passed to record_to_json()."""
        xml = '''<record xmlns="http://www.loc.gov/MARC21/slim">
          <leader>00000cam a2200000 a 4500</leader>
          <controlfield tag="001">test123</controlfield>
          <datafield tag="245" ind1="1" ind2="0">
            <subfield code="a">Test title</subfield>
          </datafield>
        </record>'''
        rec = mrrc.xml_to_record(xml)
        json_str = mrrc.record_to_json(rec)
        assert '"245"' in json_str
        assert 'Test title' in json_str

    def test_xml_to_record_then_record_to_xml(self):
        """xml_to_record() result can be passed to record_to_xml()."""
        xml = '''<record xmlns="http://www.loc.gov/MARC21/slim">
          <leader>00000cam a2200000 a 4500</leader>
          <controlfield tag="001">test123</controlfield>
          <datafield tag="245" ind1="1" ind2="0">
            <subfield code="a">Test title</subfield>
          </datafield>
        </record>'''
        rec = mrrc.xml_to_record(xml)
        xml_out = mrrc.record_to_xml(rec)
        assert 'Test title' in xml_out

    def test_xml_to_record_then_record_to_marcjson(self):
        """xml_to_record() result can be passed to record_to_marcjson()."""
        xml = '''<record xmlns="http://www.loc.gov/MARC21/slim">
          <leader>00000cam a2200000 a 4500</leader>
          <controlfield tag="001">test123</controlfield>
          <datafield tag="245" ind1="1" ind2="0">
            <subfield code="a">Test title</subfield>
          </datafield>
        </record>'''
        rec = mrrc.xml_to_record(xml)
        mj = mrrc.record_to_marcjson(rec)
        assert 'Test title' in mj

    def test_xml_to_record_then_record_to_mods(self):
        """xml_to_record() result can be passed to record_to_mods()."""
        xml = '''<record xmlns="http://www.loc.gov/MARC21/slim">
          <leader>00000cam a2200000 a 4500</leader>
          <controlfield tag="001">test123</controlfield>
          <datafield tag="245" ind1="1" ind2="0">
            <subfield code="a">Test title</subfield>
          </datafield>
        </record>'''
        rec = mrrc.xml_to_record(xml)
        mods = mrrc.record_to_mods(rec)
        assert 'Test title' in mods

    def test_xml_to_record_then_record_to_dublin_core(self):
        """xml_to_record() result can be passed to record_to_dublin_core()."""
        xml = '''<record xmlns="http://www.loc.gov/MARC21/slim">
          <leader>00000cam a2200000 a 4500</leader>
          <controlfield tag="001">test123</controlfield>
          <datafield tag="245" ind1="1" ind2="0">
            <subfield code="a">Test title</subfield>
          </datafield>
        </record>'''
        rec = mrrc.xml_to_record(xml)
        dc = mrrc.record_to_dublin_core(rec)
        assert 'Test title' in dc['title'][0]

    def test_xml_to_record_then_record_to_dublin_core_xml(self):
        """xml_to_record() result can be passed to _mrrc.record_to_dublin_core_xml()."""
        from mrrc._mrrc import record_to_dublin_core_xml
        xml = '''<record xmlns="http://www.loc.gov/MARC21/slim">
          <leader>00000cam a2200000 a 4500</leader>
          <controlfield tag="001">test123</controlfield>
          <datafield tag="245" ind1="1" ind2="0">
            <subfield code="a">Test title</subfield>
          </datafield>
        </record>'''
        rec = mrrc.xml_to_record(xml)
        dc_xml = record_to_dublin_core_xml(rec)
        assert 'Test title' in dc_xml

    def test_json_to_record_then_record_to_xml(self):
        """json_to_record() result can be passed to record_to_xml()."""
        record = Record()
        record.add_field(create_field('245', '1', '0', a='JSON Test'))
        json_str = mrrc.record_to_json(record)
        restored = mrrc.json_to_record(json_str)
        xml_out = mrrc.record_to_xml(restored)
        assert 'JSON Test' in xml_out

    def test_raw_record_still_works(self):
        """Direct _mrrc.Record (unwrapped) still works with serialization."""
        from mrrc._mrrc import Record as _Record, Leader as _Leader
        raw = _Record(_Leader())
        # Should not raise
        mrrc.record_to_json(raw)


if __name__ == '__main__':
    pytest.main([__file__, '-v'])