def test_adding_central_concepts():
dataset = pycldf.Dataset.from_metadata(
Path(__file__).parent /
"data/cldf/smallmawetiguarani/cldf-metadata.json")
dirname = Path(tempfile.mkdtemp(prefix="lexedata-test"))
excel_writer = ExcelWriter(dataset=dataset, add_central_concepts=True)
output = dirname / "out.xlsx"
excel_writer.create_excel(out=output)
# load central concepts from output
ws = op.load_workbook(output).active
concept_index = 0
for row in ws.iter_rows(min_row=1, max_row=1):
for cell in row:
if cell.value == "Central_Concept":
concept_index = cell.column
# when accessing the row as a tuple the index is not 1-based as for excel sheets
central_concepts = [
row[concept_index - 1].value for row in ws.iter_rows(min_row=2)
]
assert central_concepts == [
"one",
"one",
"one",
"one",
"two",
"three",
"two",
"three",
"four_1",
"four",
"five",
]
def test_roundtrip(cldf_wordlist, working_and_nonworking_bibfile):
filled_cldf_wordlist = working_and_nonworking_bibfile(cldf_wordlist)
dataset, target = filled_cldf_wordlist
c_formReference = dataset["CognateTable", "formReference"].name
c_cogsetReference = dataset["CognateTable", "cognatesetReference"].name
old_judgements = {(row[c_formReference], row[c_cogsetReference])
for row in dataset["CognateTable"].iterdicts()}
writer = ExcelWriter(dataset,
database_url="https://example.org/lexicon/{:}")
forms = util.cache_table(filled_cldf_wordlist[0])
languages = util.cache_table(filled_cldf_wordlist[0],
"LanguageTable").values()
judgements = util.cache_table(filled_cldf_wordlist[0],
"CognateTable").values()
cogsets = util.cache_table(filled_cldf_wordlist[0],
"CognatesetTable").values()
writer.create_excel(rows=cogsets,
judgements=judgements,
forms=forms,
languages=languages)
# Reset the existing cognatesets and cognate judgements, to avoid
# interference with the the data in the Excel file
dataset["CognateTable"].write([])
dataset["CognatesetTable"].write([])
import_cognates_from_excel(writer.ws, dataset)
new_judgements = {(row[c_formReference], row[c_cogsetReference])
for row in dataset["CognateTable"].iterdicts()}
assert new_judgements == old_judgements
def test_adding_singleton_cognatesets():
dataset = pycldf.Dataset.from_metadata(
Path(__file__).parent /
"data/cldf/smallmawetiguarani/cldf-metadata.json")
dirname = Path(tempfile.mkdtemp(prefix="lexedata-test"))
excel_writer = ExcelWriter(dataset=dataset, singleton_cognate=True)
output = dirname / "out.xlsx"
excel_writer.create_excel(out=output)
# load central concepts from output
ws = op.load_workbook(output).active
cogset_index = 0
for row in ws.iter_rows(min_row=1, max_row=1):
for cell in row:
if cell.value == "CogSet":
cogset_index = cell.column
# when accessing the row as a tuple the index is not 1-based as for excel sheets
cogset_ids = [
row[cogset_index - 1].value for row in ws.iter_rows(min_row=2)
]
assert cogset_ids == [
"one1",
"one1",
"one2",
"one6",
"two1",
"three1",
"two8",
"three9",
"four1",
"four8",
"five5",
"X1_old_paraguayan_guarani",
"X2_ache",
"X3_paraguayan_guarani",
]
def test_roundtrip_separator_column(cldf_wordlist):
"""Test whether a CognatesetTable column with separator survives a roundtrip."""
dataset, target = copy_to_temp(cldf_wordlist)
dataset.add_columns("CognatesetTable", "CommaSeparatedTags")
dataset["CognatesetTable", "CommaSeparatedTags"].separator = ","
c_id = dataset["CognatesetTable", "id"].name
write_back = list(dataset["CognatesetTable"])
tags = []
for tag, row in zip(
itertools.cycle([["two", "tags"], ["single-tag"], [],
["tag;containing;other;separator"]]),
write_back,
):
tags.append((row[c_id], tag))
row["CommaSeparatedTags"] = tag
dataset.write(CognatesetTable=write_back)
writer = ExcelWriter(dataset)
_, out_filename = tempfile.mkstemp(".xlsx", "cognates")
writer.create_excel(out_filename)
import_cognates_from_excel(out_filename, dataset)
reread_tags = [(c[c_id], c["CommaSeparatedTags"])
for c in dataset["CognatesetTable"]]
reread_tags.sort(key=lambda x: x[0])
tags.sort(key=lambda x: x[0])
assert reread_tags == tags
def test_toexcel_runs(cldf_wordlist):
filled_cldf_wordlist = copy_to_temp(cldf_wordlist)
writer = ExcelWriter(
dataset=filled_cldf_wordlist[0],
database_url=str(filled_cldf_wordlist[1]),
)
_, out_filename = tempfile.mkstemp(".xlsx", "cognates")
writer.create_excel(out_filename)
def test_no_comment_column():
dataset, _ = copy_to_temp(
Path(__file__).parent /
"data/cldf/smallmawetiguarani/cldf-metadata.json")
dataset.remove_columns("FormTable", "comment")
writer = ExcelWriter(dataset=dataset, )
forms = util.cache_table(dataset).values()
for form in forms:
assert writer.form_to_cell_value(
form).strip() == "{ e t a k ɾ ã } ‘one, one’"
break
def test_adding_singleton_cognatesets_with_status(caplog):
dataset = get_dataset(
Path(__file__).parent /
"data/cldf/smallmawetiguarani/cldf-metadata.json")
dataset.add_columns("CognatesetTable", "Status_Column")
with caplog.at_level(logging.WARNING):
excel_writer = ExcelWriter(dataset=dataset)
cogsets, judgements = create_singletons(
dataset,
status="NEW",
by_segment=True,
)
properties_as_key(cogsets,
dataset["CognatesetTable"].tableSchema.columns)
properties_as_key(judgements,
dataset["CognateTable"].tableSchema.columns)
forms = util.cache_table(dataset)
languages = util.cache_table(dataset, "LanguageTable").values()
excel_writer.create_excel(rows=cogsets,
judgements=judgements,
forms=forms,
languages=languages)
assert re.search("no Status_Column to write", caplog.text) is None
cogset_index = 0
for row in excel_writer.ws.iter_rows(min_row=1, max_row=1):
for cell in row:
if cell.value == "Status_Column":
cogset_index = cell.column - 1
# when accessing the row as a tuple the index is not 1-based as for excel sheets
status = [
row[cogset_index].value for row in excel_writer.ws.iter_rows(min_row=2)
]
assert status == [
None,
None,
None,
None,
None,
None,
None,
None,
None,
None,
None,
"NEW",
"NEW",
"NEW",
"NEW",
]
def test_adding_singleton_cognatesets(caplog):
dataset = get_dataset(
Path(__file__).parent /
"data/cldf/smallmawetiguarani/cldf-metadata.json")
with caplog.at_level(logging.WARNING):
excel_writer = ExcelWriter(dataset=dataset, )
cogsets, judgements = create_singletons(
dataset,
status="NEW",
by_segment=False,
)
properties_as_key(cogsets,
dataset["CognatesetTable"].tableSchema.columns)
properties_as_key(judgements,
dataset["CognateTable"].tableSchema.columns)
forms = util.cache_table(dataset)
languages = util.cache_table(dataset, "LanguageTable").values()
excel_writer.create_excel(rows=cogsets,
judgements=judgements,
forms=forms,
languages=languages)
assert re.search("No Status_Column", caplog.text)
# load central concepts from output
cogset_index = 0
for row in excel_writer.ws.iter_rows(min_row=1, max_row=1):
for cell in row:
if cell.value == "CogSet":
cogset_index = cell.column - 1
# when accessing the row as a tuple the index is not 1-based as for excel sheets
cogset_ids = [
row[cogset_index].value for row in excel_writer.ws.iter_rows(min_row=2)
]
assert cogset_ids == [
"one1",
"one1",
"one2",
"one6",
"two1",
"three1",
"two8",
"three9",
"four1",
"four8",
"five5",
"X_old_paraguayan_guarani_two_1",
"X_paraguayan_guarani_five_1",
]
def test_missing_required_column():
dataset, _ = copy_to_temp(
Path(__file__).parent /
"data/cldf/smallmawetiguarani/cldf-metadata.json")
dataset.remove_columns("FormTable", "ID")
# TODO: switch to pycldf.dataset.SchemaError
with pytest.raises(KeyError):
excel_writer = ExcelWriter(dataset=dataset)
forms = util.cache_table(dataset)
languages = util.cache_table(dataset, "LanguageTable").values()
judgements = util.cache_table(dataset, "CognateTable")
cogsets = util.cache_table(dataset, "CognatesetTable")
excel_writer.create_excel(rows=cogsets,
judgements=judgements,
forms=forms,
languages=languages)
def test_included_segments(caplog):
ds = util.fs.new_wordlist(FormTable=[],
CognatesetTable=[],
CognateTable=[])
E = ExcelWriter(dataset=ds)
E.form_to_cell_value({"form": "f", "parameterReference": "c"})
with caplog.at_level(logging.WARNING):
cell = E.form_to_cell_value({
"id": "0",
"cognateReference": "j",
"form": "fo",
"parameterReference": "c",
"segments": ["f", "o"],
"segmentSlice": ["3:1"],
})
assert cell == "{ f o } ‘c’"
assert re.search("segment slice '3:1' is invalid", caplog.text) is None
def test_no_cognate_table(caplog):
dataset, _ = empty_copy_of_cldf_wordlist(
Path(__file__).parent /
"data/cldf/smallmawetiguarani/cldf-metadata.json")
dataset.remove_table("CognateTable")
with pytest.raises(SystemExit):
ExcelWriter(dataset=dataset, )
assert "presupposes a separate CognateTable" in caplog.text
assert "lexedata.edit.add_cognate_table" in caplog.text
def test_toexcel_runs(cldf_wordlist, working_and_nonworking_bibfile):
filled_cldf_wordlist = working_and_nonworking_bibfile(cldf_wordlist)
writer = ExcelWriter(
dataset=filled_cldf_wordlist[0],
database_url=str(filled_cldf_wordlist[1]),
)
forms = util.cache_table(filled_cldf_wordlist[0])
languages = util.cache_table(filled_cldf_wordlist[0],
"LanguageTable").values()
judgements = util.cache_table(filled_cldf_wordlist[0],
"CognateTable").values()
cogsets = util.cache_table(filled_cldf_wordlist[0],
"CognatesetTable").values()
writer.create_excel(rows=cogsets,
judgements=judgements,
forms=forms,
languages=languages)
_, out_filename = tempfile.mkstemp(".xlsx", "cognates")
writer.wb.save(filename=out_filename)
def test_no_cognateset_table(caplog):
dataset, _ = empty_copy_of_cldf_wordlist(
Path(__file__).parent /
"data/cldf/smallmawetiguarani/cldf-metadata.json")
dataset.remove_table("CognatesetTable")
# TODO: SystemExit or dataset error?
with pytest.raises((SystemExit, SchemaError)) as exc_info:
ExcelWriter(dataset=dataset, )
if exc_info.type == SystemExit:
assert "presupposes a separate CognatesetTable" in caplog.text
assert "lexedata.edit.add_table" in caplog.text
def test_roundtrip(cldf_wordlist):
filled_cldf_wordlist = copy_to_temp(cldf_wordlist)
dataset, target = filled_cldf_wordlist
c_formReference = dataset["CognateTable", "formReference"].name
c_cogsetReference = dataset["CognateTable", "cognatesetReference"].name
old_judgements = {(row[c_formReference], row[c_cogsetReference])
for row in dataset["CognateTable"].iterdicts()}
writer = ExcelWriter(dataset)
_, out_filename = tempfile.mkstemp(".xlsx", "cognates")
writer.create_excel(out_filename)
# Reset the existing cognatesets and cognate judgements, to avoid
# interference with the the data in the Excel file
dataset["CognateTable"].write([])
dataset["CognatesetTable"].write([])
import_cognates_from_excel(out_filename, dataset)
new_judgements = {(row[c_formReference], row[c_cogsetReference])
for row in dataset["CognateTable"].iterdicts()}
assert new_judgements == old_judgements
def test_roundtrip_separator_column(cldf_wordlist,
working_and_nonworking_bibfile):
"""Test whether a CognatesetTable column with separator survives a roundtrip."""
dataset, target = working_and_nonworking_bibfile(cldf_wordlist)
dataset.add_columns("CognatesetTable", "CommaSeparatedTags")
dataset["CognatesetTable", "CommaSeparatedTags"].separator = ","
c_id = dataset["CognatesetTable", "id"].name
write_back = list(dataset["CognatesetTable"])
tags = []
for tag, row in zip(
itertools.cycle([["two", "tags"], ["single-tag"], [],
["tag;containing;other;separator"]]),
write_back,
):
tags.append((row[c_id], tag))
row["CommaSeparatedTags"] = tag
dataset.write(CognatesetTable=write_back)
writer = ExcelWriter(dataset,
database_url="https://example.org/lexicon/{:}")
_, out_filename = tempfile.mkstemp(".xlsx", "cognates")
forms = util.cache_table(dataset)
languages = util.cache_table(dataset, "LanguageTable").values()
judgements = util.cache_table(dataset, "CognateTable").values()
cogsets = util.cache_table(dataset, "CognatesetTable").values()
writer.create_excel(rows=cogsets,
judgements=judgements,
forms=forms,
languages=languages)
import_cognates_from_excel(writer.ws, dataset)
reread_tags = [(c[c_id], c["CommaSeparatedTags"])
for c in dataset["CognatesetTable"]]
reread_tags.sort(key=lambda x: x[0])
tags.sort(key=lambda x: x[0])
assert reread_tags == tags
def test_cell_comments_export():
dataset, _ = copy_to_temp(
Path(__file__).parent / "data/cldf/minimal/cldf-metadata.json")
_, out_filename = tempfile.mkstemp(".xlsx", "cognates")
writer = ExcelWriter(dataset,
database_url="https://example.org/lexicon/{:}")
forms = util.cache_table(dataset)
languages = sorted(util.cache_table(dataset, "LanguageTable").values(),
key=lambda x: x["name"])
judgements = util.cache_table(dataset, "CognateTable").values()
cogsets = util.cache_table(dataset, "CognatesetTable").values()
writer.create_excel(rows=cogsets,
judgements=judgements,
forms=forms,
languages=languages)
for col in writer.ws.iter_cols():
pass
assert (
col[-1].comment and col[-1].comment.content
), "Last row of last column should contain a judgement, with a comment attached to it."
assert (col[-1].comment.content == "A judgement comment"
), "Comment should match the comment from the cognate table"
def test_cell_comments_export():
dataset, _ = copy_to_temp(
Path(__file__).parent / "data/cldf/minimal/cldf-metadata.json")
_, out_filename = tempfile.mkstemp(".xlsx", "cognates")
E = ExcelWriter(dataset)
E.set_header()
E.create_excel(out_filename, size_sort=False, language_order="Name")
ws_out = openpyxl.load_workbook(out_filename).active
for col in ws_out.iter_cols():
pass
assert col[
-1].comment.content, "Last row of last column should contain a judgement, with a comment attached to it."
assert (col[-1].comment.content == "A judgement comment"
), "Comment should match the comment from the cognate table"