def test_cell_comments():
dataset, _ = copy_to_temp(
Path(__file__).parent / "data/cldf/minimal/cldf-metadata.json")
excel_filename = Path(
__file__).parent / "data/excel/judgement_cell_with_note.xlsx"
ws = openpyxl.load_workbook(excel_filename).active
import_cognates_from_excel(ws, dataset)
cognates = {
cog["ID"]: {
k: v
for k, v in cog.items()
if k in {"Form_ID", "Cognateset", "Comment"}
}
for cog in dataset["CognateTable"]
}
assert cognates == {
"autaa_Woman-cogset": {
"Cognateset": "cogset",
"Comment": "Comment on judgement",
"Form_ID": "autaa_Woman",
}
}
cognatesets = {
cog["ID"]: {k: v
for k, v in cog.items()}
for cog in dataset["CognatesetTable"]
}
assert cognatesets == {
"cogset": {
"Name": "cogset",
"Comment": "Cognateset-comment",
"ID": "cogset"
}
}
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_cell_comments():
dataset, _ = copy_to_temp(
Path(__file__).parent / "data/cldf/minimal/cldf-metadata.json")
ws_test = Path(
__file__).parent / "data/excel/judgement_cell_with_note.xlsx"
import_cognates_from_excel(ws_test, dataset)
cognates = {
cog["ID"]: {k: v
for k, v in cog.items()}
for cog in dataset["CognateTable"]
}
assert cognates == {
"autaa_Woman-cogset": {
"Cognateset": "cogset",
"Comment": "Comment on judgement",
"Form_ID": "autaa_Woman",
"ID": "autaa_Woman-cogset",
"Segment_Slice": None,
"Alignment": None,
}
}
cognatesets = {
cog["ID"]: {k: v
for k, v in cog.items()}
for cog in dataset["CognatesetTable"]
}
assert cognatesets == {
"cogset": {
"Name": "cogset",
"Comment": "Cognateset-comment",
"ID": "cogset"
}
}
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_cell_comments_and_comment_column(caplog):
dataset, _ = copy_to_temp(
Path(__file__).parent / "data/cldf/minimal/cldf-metadata.json")
excel_filename = Path(
__file__).parent / "data/excel/judgement_cell_with_note.xlsx"
sheet = openpyxl.load_workbook(excel_filename).active
sheet.insert_cols(2)
sheet.cell(row=1, column=2, value="Comment")
sheet.cell(row=2, column=2, value="Comment")
with caplog.at_level(logging.INFO):
import_cognates_from_excel(sheet, dataset)
assert "from the cell comments" in caplog.text
cognates = {
cog["ID"]: {
k: v
for k, v in cog.items()
if k in {"Form_ID", "Cognateset", "Comment"}
}
for cog in dataset["CognateTable"]
}
assert cognates == {
"autaa_Woman-cogset": {
"Cognateset": "cogset",
"Comment": "Comment on judgement",
"Form_ID": "autaa_Woman",
}
}
cognatesets = {
cog["ID"]: {k: v
for k, v in cog.items()}
for cog in dataset["CognatesetTable"]
}
assert cognatesets == {
"cogset": {
"Name": "cogset",
"Comment": "Cognateset-comment",
"ID": "cogset"
}
}
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_excel_messy_row(caplog):
# Build a dataset with forms F1, F2, F3 in languages L1, L2 and
# CognateTable columns ID and Status
dataset = util.fs.new_wordlist(
FormTable=[
{
"ID": "F1",
"Language_ID": "L1",
"Form": "f1",
"Parameter_ID": "C"
},
{
"ID": "F2",
"Language_ID": "L2",
"Form": "f1",
"Parameter_ID": "C"
},
{
"ID": "F3",
"Language_ID": "L1",
"Form": "f1",
"Parameter_ID": "C"
},
],
LanguageTable=[{
"ID": "L1",
"Name": "L1"
}, {
"ID": "L2",
"Name": "L2"
}],
ParameterTable=[{
"ID": "C"
}],
CognateTable=[],
CognatesetTable=[],
)
# TODO: Ensure FormTable does not need a value
dataset.add_columns("FormTable", "value")
dataset["FormTable", "value"].required = False
dataset.add_columns("CognatesetTable", "Status")
dataset.add_columns("CognatesetTable", "comment")
# Construct a sheet with a messy cognateset header
messy_sheet = MockSingleExcelSheet([
[
"CogSet",
"Status",
"L1",
"L2",
],
[
"S1",
"valid",
"F1",
"F2",
],
[
"",
"invalid",
"F3",
],
])
for cell in [(2, 3), (3, 3), (2, 4)]:
messy_sheet.cell(*cell).hyperlink = "/{:}".format(
messy_sheet.cell(*cell).value)
# Cognate-import this dataset
with caplog.at_level(logging.INFO):
import_cognates_from_excel(
messy_sheet,
dataset,
)
# Check that cognateset S1 contains form F3
assert ("F3", "S1") in [(j["Form_ID"], j["Cognateset_ID"])
for j in dataset["CognateTable"]]
# Check for warning in caplog
assert re.search("[Rr]ow 3 .* no cognate ?set .*'Status': 'invalid'",
caplog.text)