def test_exit_with_ConfigError_wtf(tmp_path):
"""2022-05-13: It appears that this test, to succeed,
would need to change away from directory with the bad
config file created in the previous test (above).
"""
with pytest.raises(ConfigError):
get_config()
def test_exit_with_ConfigError_if_default_configfile_found_with_bad_yaml(
tmp_path):
"""Exit with ConfigError if default config file has bad YAML."""
os.chdir(tmp_path)
bad_config_yaml = "DELIBE\nRATELY BAD: -: ^^YAML CONTENT^^\n"
Path(DEFAULT_CONFIGFILE_NAME).write_text(bad_config_yaml)
with pytest.raises(ConfigError):
get_config()
def test_exit_with_ConfigError_if_specified_configfile_found_with_bad_yaml(
tmp_path):
"""Exit with ConfigError if config file specified as argument has bad YAML."""
os.chdir(tmp_path)
bad_config_yaml = "DELIBE\nRATELY BAD: -: ^^YAML CONTENT^^\n"
nondefault_configfile_name = "dctap_settings.yml"
Path(nondefault_configfile_name).write_text(bad_config_yaml)
with pytest.raises(ConfigError):
get_config(configfile_name=nondefault_configfile_name)
def test_get_rows_with_unknown_column(tmp_path):
"""Non-DCTAP elements kept by _get_rows (but dropped by _get_shapes)."""
os.chdir(tmp_path)
config_dict = get_config()
csvfile_path = Path(tmp_path).joinpath("some.csv")
csvfile_path.write_text(
("shapeID,propertyID,valueConstraint,value Gestalt\n"
":book,dc:creator,,:author\n"
",dc:type,so:Book,\n"
":author,foaf:name,,\n"))
csvfile_obj = open(csvfile_path)
expected_rows_list = [{
'shapeID': ':book',
'propertyID': 'dc:creator',
'valueConstraint': '',
'valuegestalt': ':author'
}, {
'shapeID': '',
'propertyID': 'dc:type',
'valueConstraint': 'so:Book',
'valuegestalt': ''
}, {
'shapeID': ':author',
'propertyID': 'foaf:name',
'valueConstraint': '',
'valuegestalt': ''
}]
actual_rows_list, actual_warnings = _get_rows(csvfile_obj, config_dict)
assert actual_rows_list == expected_rows_list
def test_get_rows_with_simple_csvfile(tmp_path):
"""Another simple CSV with three columns."""
os.chdir(tmp_path)
config_dict = get_config()
csvfile_path = Path(tmp_path).joinpath("some.csv")
csvfile_path.write_text(("shapeID,propertyID,valueNodeType\n"
":a,dct:creator,URI\n"
":a,dct:subject,URI\n"
":a,dct:date,String\n"))
csvfile_obj = open(csvfile_path)
expected_rows_list = [{
'shapeID': ':a',
'propertyID': 'dct:creator',
'valueNodeType': 'URI'
}, {
'shapeID': ':a',
'propertyID': 'dct:subject',
'valueNodeType': 'URI'
}, {
'shapeID': ':a',
'propertyID': 'dct:date',
'valueNodeType': 'String'
}]
actual_rows_list, actual_warnings = _get_rows(csvfile_obj, config_dict)
assert actual_rows_list == expected_rows_list
def test_list_item_separator_defaults_to_single_blank():
"""Setting list_item_separator of None defaults to single blank."""
sc = TAPStatementTemplate()
config_dict = get_config()
config_dict["list_elements"] = ["valueNodeType"]
config_dict["list_item_separator"] = None
sc.valueNodeType = "iri bnode"
sc._parse_elements_configured_as_list_elements(config_dict)
assert sc.valueNodeType == ["iri", "bnode"]
def test_normalize_element_name():
"""Element names not recognized as aliases are left unchanged."""
config_dict = get_config()
element_aliases_dict = config_dict.get("element_aliases")
assert _normalize_element_name("SHAPE ID",
element_aliases_dict) == "shapeID"
assert _normalize_element_name("SHAPE___ID",
element_aliases_dict) == "shapeID"
assert _normalize_element_name("rid", element_aliases_dict) == "rid"
def test_get_TAPShape_elements_plus_extras_when_config_dict_specified():
"""List TAPShape elements plus extra shape elements."""
expected_main_shems = ["shapeID", "shapeLabel"]
expected_xtra_shems = ["closed", "start"]
#config_dict = dict(extra_shape_elements=["closed", "start"])
config_dict = get_config()
config_dict["extra_shape_elements"] = ["closed", "start"]
assert get_shems(TAPShape, config_dict)[0] == expected_main_shems
assert get_shems(TAPShape, config_dict)[1] == expected_xtra_shems
def test_get_rows_raises_exception_if_first_line_has_no_propertyid(tmp_path):
"""Raises exception if first line of CSV has no propertyID."""
os.chdir(tmp_path)
csvfile_path = Path(tmp_path).joinpath("some.csv")
csvfile_path.write_text(("shapeID,propertyIdentifier,valueNodeType\n"
":a,dct:creator,URI\n"))
csvfile_obj = open(csvfile_path)
config_dict = get_config()
with pytest.raises(SystemExit):
_get_rows(csvfile_obj, config_dict)
def test_get_rows_minimal(tmp_path):
"""Get list of rows, as dicts, from one-row, one-column CSV."""
os.chdir(tmp_path)
config_dict = get_config()
csvfile_path = Path(tmp_path).joinpath("some.csv")
csvfile_path.write_text(("PropertyID\n"
"http://purl.org/dc/terms/creator\n"))
csvfile_obj = open(csvfile_path)
expected_rows_list = [{'propertyID': 'http://purl.org/dc/terms/creator'}]
actual_rows_list, actual_warnings = _get_rows(csvfile_obj, config_dict)
assert actual_rows_list == expected_rows_list
def test_get_rows_given_customized_element_alias(tmp_path):
"""Using customized element alias, normalized for case, dashes, underscores."""
os.chdir(tmp_path)
config_dict = get_config()
csvfile_path = Path(tmp_path).joinpath("some.csv")
csvfile_path.write_text(("Prop_ID\n" "http://purl.org/dc/terms/creator\n"))
csvfile_obj = open(csvfile_path)
config_dict["element_aliases"].update({"propid": "propertyID"})
expected_rows_list = [{'propertyID': 'http://purl.org/dc/terms/creator'}]
actual_rows_list, actual_warnings = _get_rows(csvfile_obj, config_dict)
assert actual_rows_list == expected_rows_list
def test_get_config_from_builtins():
"""Get config dict from built-in settings."""
config_dict = get_config()
config_keys = list(config_dict.keys())
assert "prefixes" in config_keys
assert config_dict.get("csv_elements") # computed from dataclasses
assert config_dict.get("shape_elements") # computed from dataclasses
assert config_dict.get(
"statement_template_elements") # computed from dataclasses
assert config_dict.get("element_aliases") # computed from dataclasses
assert "element_aliases" in config_keys # computed and configurable
def test_extra_shape_elements(tmp_path):
"""2022-05-13: os.chdir(tmp_path) is needed here because
a previous pytest (above) wrote a bad config file to tmp_path.
"""
os.chdir(tmp_path)
config_dict = get_config()
config_dict["extra_shape_elements"] = ["closed", "start"]
assert config_dict["shape_elements"] == ["shapeID", "shapeLabel"]
config_dict["shape_elements"].extend(config_dict["extra_shape_elements"])
assert config_dict["shape_elements"] == [
"shapeID", "shapeLabel", "closed", "start"
]
def test_get_config_from_default_config_file_if_present(tmp_path):
"""Get config dict from config file DEFAULT_CONFIGFILE_NAME if present."""
os.chdir(tmp_path)
Path(DEFAULT_CONFIGFILE_NAME).write_text(NONDEFAULT_CONFIG_YAMLDOC)
config_dict = get_config()
assert "prefixes" in list(config_dict.keys())
assert config_dict.get("default_shape_identifier")
assert config_dict.get("csv_elements") # computed
assert config_dict.get("shape_elements") # computed
assert config_dict.get("statement_template_elements") # computed
assert config_dict.get("element_aliases") # asserted/computed
assert config_dict.get("value_node_types") is None
def test_valueConstraintType_languagetag_item_separator_pipe(tmp_path):
"""@@@"""
config_dict = get_config()
config_dict["list_item_separator"] = "|"
config_dict["default_shape_identifier"] = "default"
os.chdir(tmp_path)
csvfile_path = Path(tmp_path).joinpath("some.csv")
csvfile_path.write_text(('PropertyID,valueConstraintType,valueConstraint\n'
'ex:foo,languagetag,"fr|it|de"\n'))
value_constraint = csvreader(
open(csvfile_path), config_dict
)[0]["shapes"][0]["statement_templates"][0]["valueConstraint"]
assert value_constraint == ["fr", "it", "de"]
def test_valueConstraintType_list_item_separator_pipe(tmp_path):
"""Picklist values are split on pipe character if so configured."""
config_dict = get_config()
config_dict["list_item_separator"] = "|"
config_dict["default_shape_identifier"] = "default"
os.chdir(tmp_path)
csvfile_path = Path(tmp_path).joinpath("some.csv")
csvfile_path.write_text(('PropertyID,valueConstraintType,valueConstraint\n'
'ex:foo,picklist,"one|two|three"\n'))
value_constraint = csvreader(
open(csvfile_path), config_dict
)[0]["shapes"][0]["statement_templates"][0]["valueConstraint"]
assert value_constraint == ["one", "two", "three"]
def test_get_rows_fills_in_short_headers_first_with_empty_header(tmp_path):
"""Where headers shorter than rows, adds one empty header."""
os.chdir(tmp_path)
config_dict = get_config()
csvfile_path = Path(tmp_path).joinpath("some.csv")
csvfile_path.write_text(("shapeID,propertyID,\n" ":a,dct:creator,URI\n"))
csvfile_obj = open(csvfile_path)
expected_rows_list = [{
'shapeID': ':a',
'propertyID': 'dct:creator',
'': 'URI'
}]
actual_rows_list, actual_warnings = _get_rows(csvfile_obj, config_dict)
assert actual_rows_list == expected_rows_list
def test_get_rows_including_header_not_in_DCTAP(tmp_path):
"""Get rows where one header is not part of the DCTAP model."""
os.chdir(tmp_path)
config_dict = get_config()
csvfile_path = Path(tmp_path).joinpath("some.csv")
csvfile_path.write_text(("PropertyID,Ricearoni\n"
"dc:creator,SFO treat\n"))
csvfile_obj = open(csvfile_path)
expected_rows_list = [{
'propertyID': 'dc:creator',
'ricearoni': 'SFO treat',
}]
actual_rows_list, actual_warnings = _get_rows(csvfile_obj, config_dict)
assert actual_rows_list == expected_rows_list
def test_get_rows_with_complete_csvfile(tmp_path):
"""Simple CSV with all columns."""
os.chdir(tmp_path)
config_dict = get_config()
csvfile_path = Path(tmp_path).joinpath("some.csv")
csvfile_path.write_text(
("shapeID,shapeLabel,propertyID,"
"propertyLabel,mandatory,repeatable,valueNodeType,"
"valueDataType,valueConstraint,valueConstraintType,valueShape,note\n"
":a,Book,dct:creator,Creator,1,0,URI,,,,:b,Typically the author.\n"
":b,Person,ex:name,Name,1,0,Literal,xsd:string,,,,\n"))
csvfile_obj = open(csvfile_path)
expected_rows_list = [
{
"shapeID": ":a",
"shapeLabel": "Book",
"propertyID": "dct:creator",
"propertyLabel": "Creator",
"mandatory": "1",
"repeatable": "0",
"valueNodeType": "URI",
"valueDataType": "",
"valueConstraint": "",
"valueConstraintType": "",
"valueShape": ":b",
"note": "Typically the author.",
},
{
"shapeID": ":b",
"shapeLabel": "Person",
"propertyID": "ex:name",
"propertyLabel": "Name",
"mandatory": "1",
"repeatable": "0",
"valueNodeType": "Literal",
"valueDataType": "xsd:string",
"valueConstraint": "",
"valueConstraintType": "",
"valueShape": "",
"note": "",
},
]
actual_rows_list, actual_warnings = _get_rows(csvfile_obj, config_dict)
assert isinstance(actual_rows_list, list)
assert isinstance(expected_rows_list, list)
assert actual_rows_list == expected_rows_list
assert actual_rows_list[0]["mandatory"]
assert len(actual_rows_list) == 2
assert len(expected_rows_list) == 2
def test_get_rows_fills_in_short_rows_with_None_values(tmp_path):
"""Fills in short rows with None values."""
os.chdir(tmp_path)
config_dict = get_config()
csvfile_path = Path(tmp_path).joinpath("some.csv")
csvfile_path.write_text(("shapeID,propertyID,valueNodeType\n"
":a,dct:creator\n"))
csvfile_obj = open(csvfile_path)
expected_rows_list = [{
'shapeID': ':a',
'propertyID': 'dct:creator',
'valueNodeType': None
}]
actual_rows_list, actual_warnings = _get_rows(csvfile_obj, config_dict)
assert actual_rows_list == expected_rows_list
def test_get_rows_fills_in_short_headers_subsequently_with_None(tmp_path):
"""Where headers shorter than rows, extra values collected under header None."""
os.chdir(tmp_path)
config_dict = get_config()
csvfile_path = Path(tmp_path).joinpath("some.csv")
csvfile_path.write_text("shapeID,propertyID,\n"
":a,dct:creator,URI,comment,comment two\n")
csvfile_obj = open(csvfile_path)
expected_rows_list = [{
'shapeID': ':a',
'propertyID': 'dct:creator',
'': 'URI',
None: ['comment', 'comment two']
}]
actual_rows_list, actual_warnings = _get_rows(csvfile_obj, config_dict)
assert actual_rows_list == expected_rows_list
def test_mkshapes_returns_tapshape_object_even_in_absence_of_propertyID(
tmp_path):
"""Populates TAPShape object even in the absence of a propertyID."""
os.chdir(tmp_path) # precaution to avoid interference among pytests
config_dict = get_config()
assert config_dict["shape_elements"] == ["shapeID", "shapeLabel"]
one_row = {
"shapeID": ":a",
"shapeLabel": "Book",
}
assert _mkshape(row_dict=one_row,
config_dict=config_dict) == TAPShape(shapeID=':a',
shapeLabel='Book',
state_list=[],
shape_warns={},
shape_extras={})
def test_get_rows_correct_a_real_mess(tmp_path):
"""Messiness in headers (extra spaces, punctuation, wrong case) is corrected."""
os.chdir(tmp_path)
config_dict = get_config()
csvfile_path = Path(tmp_path).joinpath("some.csv")
csvfile_path.write_text(
"S hape ID,pr-opertyID___,valueShape ,wildCard \n"
":book,dcterms:creator,:author,Yeah yeah yeah\n")
csvfile_obj = open(csvfile_path)
expected_rows_list = [{
'shapeID': ':book',
'propertyID': 'dcterms:creator',
'valueShape': ':author',
'wildcard': 'Yeah yeah yeah',
}]
actual_rows_list, actual_warnings = _get_rows(csvfile_obj, config_dict)
assert actual_rows_list == expected_rows_list
def test_warns_if_header_not_recognized(tmp_path):
"""@@@"""
os.chdir(tmp_path)
config_dict = get_config()
config_dict["default_shape_identifier"] = "default"
csvfile_path = Path(tmp_path).joinpath("some.csv")
csvfile_path.write_text(("propertyID,ricearoni\n" "dc:date,SFO treat\n"))
csvfile_obj = open(csvfile_path)
expected_rows_list = [
{
'propertyID': 'dc:date',
'ricearoni': 'SFO treat',
},
]
actual_rows_list, actual_warnings = _get_rows(csvfile_obj, config_dict)
assert actual_rows_list == expected_rows_list
assert len(actual_warnings) == 1
def test_mkshape_recognizes_only_shape_elements_so_configured(tmp_path):
"""Populates TAPShape object but ignores any statement template elements in row."""
os.chdir(tmp_path) # precaution to avoid interference among pytests
config_dict = get_config()
config_dict["extra_shape_elements"] = ["closed"]
one_row = {
"shapeID": ":a",
"shapeLabel": "Book",
"closed": False,
"start": True,
}
assert _mkshape(one_row, config_dict=config_dict) == TAPShape(
shapeID=':a',
shapeLabel='Book',
state_list=[],
shape_warns={},
shape_extras={"closed": False})
def test_does_not_warn_if_non_dctap_header_configured_as_extra(tmp_path):
"""@@@"""
os.chdir(tmp_path)
config_dict = get_config()
config_dict["default_shape_identifier"] = "default"
config_dict["extra_statement_template_elements"] = ["ricearoni"]
csvfile_path = Path(tmp_path).joinpath("some.csv")
csvfile_path.write_text(("propertyID,ricearoni\n" "dc:date,SFO treat\n"))
csvfile_obj = open(csvfile_path)
expected_rows_list = [
{
'propertyID': 'dc:date',
'ricearoni': 'SFO treat',
},
]
actual_rows_list, actual_warnings = _get_rows(csvfile_obj, config_dict)
assert actual_rows_list == expected_rows_list
assert len(actual_warnings) == 0
def test_mkshape_extra_shape_elements_that_are_empty_are_passed_through(
tmp_path):
"""Empty shape elements are passed through, but not unasserted elements."""
os.chdir(tmp_path) # precaution to avoid interference among pytests
config_dict = get_config()
assert config_dict["shape_elements"] == ["shapeID", "shapeLabel"]
config_dict["extra_shape_elements"] = ["closed", "start"]
one_row = {
"shapeID": ":a",
"shapeLabel": "",
"closed": "",
}
assert _mkshape(row_dict=one_row, config_dict=config_dict) == TAPShape(
shapeID=':a',
shapeLabel='',
state_list=[],
shape_warns={},
shape_extras={"closed": ""})
def test_liststatements_with_csv_column_outside_dctap_model_are_ignored(
tmp_path):
"""CSV columns not part of the DC TAP model are simply ignored."""
os.chdir(tmp_path)
config_dict = get_config()
csvfile_path = Path(tmp_path).joinpath("some.csv")
csvfile_path.write_text(("shapeID,propertyID,confidential\n"
":a,dct:subject,True\n"))
csvfile_obj = open(csvfile_path)
expected_rows_list = [
{
"shapeID": ":a",
"propertyID": "dct:subject",
"confidential": "True"
},
]
actual_rows_list, actual_warnings = _get_rows(csvfile_obj, config_dict)
assert actual_rows_list == expected_rows_list
def test_mkshape_sets_shape_elements_only(tmp_path):
"""Populates TAPShape object but ignores any statement template elements in row."""
os.chdir(tmp_path) # precaution to avoid interference among pytests
config_dict = get_config()
config_dict["extra_shape_elements"] = ["closed", "start"]
one_row = {
"shapeID": ":a",
"shapeLabel": "Book",
"closed": False,
"start": True,
"propertyID": "ex:name",
"valueNodeType": "literal",
}
shape = _mkshape(one_row, config_dict)
assert shape.shapeID == ":a"
assert shape.shapeLabel == "Book"
assert shape.shape_warns == {}
assert shape.shape_extras == {"closed": False, "start": True}
assert shape.state_list == [
] # _mkshape() sets shape fields only, not ST fields
def test_mkshape_reads_all_extra_shape_elements_so_configured(tmp_path):
"""Reads all elements configured as extra shape elements."""
os.chdir(tmp_path) # precaution to avoid interference among pytests
config_dict = get_config()
assert config_dict["shape_elements"] == ["shapeID", "shapeLabel"]
config_dict["extra_shape_elements"] = ["closed", "start"]
one_row = {
"shapeID": ":a",
"shapeLabel": "Book",
"closed": False,
"start": True,
}
assert _mkshape(row_dict=one_row,
config_dict=config_dict) == TAPShape(shapeID=':a',
shapeLabel='Book',
state_list=[],
shape_warns={},
shape_extras={
"closed": False,
"start": True
})
