def test_fromexcel_correct(excel_wordlist):
lexicon, cogsets, (empty_dataset, original_md) = excel_wordlist
original = pycldf.Wordlist.from_metadata(original_md)
# TODO: parameterize original, like the other parameters, over possible
# test datasets.
f.load_dataset(Path(empty_dataset.tablegroup._fname), str(lexicon),
str(cogsets))
form_ids_from_excel = {form["ID"] for form in empty_dataset["FormTable"]}
form_ids_original = {form["ID"] for form in original["FormTable"]}
cognate_ids_from_excel = {
cognate["ID"]
for cognate in empty_dataset["CognateTable"]
}
cognate_ids_original = {form["ID"] for form in original["CognateTable"]}
assert form_ids_original == form_ids_from_excel, "{:} and {:} don't match.".format(
empty_dataset["FormTable"]._parent._fname.parent /
str(empty_dataset["FormTable"].url),
original["FormTable"]._parent._fname.parent /
str(empty_dataset["FormTable"].url),
)
assert (cognate_ids_original == cognate_ids_from_excel
), "{:} and {:} don't match.".format(
empty_dataset["CognateTable"]._parent._fname.parent /
str(empty_dataset["CognateTable"].url),
original["CognateTable"]._parent._fname.parent /
str(empty_dataset["CognateTable"].url),
)
def test_matrix_import_skips_question_marks():
data = [ # noqa
["Concept", "L1", "L2"],
["C1", "?", "Form1"],
["C2", "Form2", "Form3"],
]
# create excel with data
wb = op.Workbook()
ws = wb.active
for row in data:
ws.append(row)
dirname = Path(tempfile.mkdtemp(prefix="lexedata-test"))
target = dirname / "test.xlsx"
wb.save(target)
# TODO: struggeling to make a matrix importer run on such a simple structure in a way whitout too much coding ... this might even be a bad sign?
# create simple metadata
dataset = pycldf.Wordlist.in_dir(dirname)
dataset.write(FormTable=[{
"Concept": "",
"L1": "",
"L2": "",
"value": ""
}]) # noqa
# metadata = dataset.tablegroup._fname
metadata = Path(__file__).parent / "data/cldf/minimal/cldf-metadata.json"
# load the dataset
load_dataset(
metadata=metadata,
lexicon=target,
)
dataset = pycldf.Dataset.from_metadata(metadata)
forms = {f for f in dataset["FormTable"]}
print(forms)
assert True
def test_no_first_row_in_excel(empty_excel):
original = Path(__file__).parent / "data/cldf/minimal/cldf-metadata.json"
copy = copy_metadata(original=original)
with pytest.raises(
AssertionError,
match="Your first data row didn't have a name. Please check your format specification or ensure the "
"first row has a name.",
):
f.load_dataset(metadata=copy, lexicon=empty_excel)
def test_no_dialect_excel_cognate_parser(tmp_path, caplog, empty_excel):
# ExcelCognateParser
path = tmp_path / "invented_path"
path.open("w").write("{}")
with pytest.raises(ValueError):
# mock empty json file
f.load_dataset(metadata=path, lexicon=None, cognate_lexicon=empty_excel)
assert re.search("User-defined format specification .* missing", caplog.text)
assert re.search("default parser", caplog.text)
def test_dialect_missing_key_excel_cognate_parser(tmp_path, caplog, empty_excel):
path = tmp_path / "invented_path"
path.open("w").write("""{"special:fromexcel": {}}""")
# CognateExcelParser
with pytest.raises(ValueError):
f.load_dataset(path, lexicon=None, cognate_lexicon=empty_excel)
assert re.search(
r"User-defined format specification in the json-file was missing the key .*falling back to default parser.*",
caplog.text,
)
def test_dialect_missing_key_excel_parser(tmp_path, caplog, empty_excel):
# ExcelParser
path = tmp_path / "invented_path"
path.open("w").write("""{"special:fromexcel": {}}""")
with pytest.raises(ValueError):
f.load_dataset(path, lexicon=empty_excel)
assert re.search(
"User-defined format specification in the json-file was missing the key lang_cell_regexes, "
"falling back to default parser",
caplog.text,
)
def test_no_wordlist_and_no_cogsets(tmp_path):
# mock empty json file
path = tmp_path / "invented_path"
path.open("w").write("{}")
with pytest.raises(
argparse.ArgumentError,
match="At least one of WORDLIST and COGNATESETS excel files must be specified.*",
):
f.load_dataset(
metadata=path,
lexicon=None,
cognate_lexicon=None,
)
def test_fromexcel_runs(excel_wordlist):
lexicon, cogsets, (empty_dataset, original) = excel_wordlist
f.load_dataset(Path(empty_dataset.tablegroup._fname), str(lexicon),
str(cogsets))