def create_concepticon_for_concepts(
dataset: pycldf.Dataset,
language: t.Iterable,
concepticon_glosses: bool,
overwrite: bool,
status_update: t.Optional[str],
):
# add Status_Column if status update
if status_update:
add_status_column_to_table(dataset=dataset, table_name="ParameterTable")
# add Concepticon_ID column to ParameterTable
if dataset.column_names.parameters.concepticonReference is None:
# Create a concepticonReference column
dataset.add_columns("ParameterTable", "Concepticon_ID")
c = dataset["ParameterTable"].tableSchema.columns[-1]
c.valueUrl = "http://concepticon.clld.org/parameters/{Concepticon_ID}"
c.propertyUrl = URITemplate(
"http://cldf.clld.org/v1.0/terms.rdf#concepticonReference"
)
dataset.write_metadata()
if not language:
language = [(dataset.column_names.parameters.id, "en")]
gloss_languages: t.Dict[str, str] = dict(language)
add_concepticon_references(
dataset,
gloss_languages=gloss_languages,
status_update=status_update,
overwrite=overwrite,
)
if concepticon_glosses:
add_concepticon_names(dataset)
def add_table_with_columns(
table: str,
column_names: t.Set[str],
data: pycldf.Dataset) -> None:
"""Add a table with the given columns to the dataset.
If such a table already exists, only add the columns that do not exist
yet.
"""
delete = True
try:
data[table]
delete = False
except KeyError:
data.add_component(table)
columns = data[table].tableSchema.columns
for c in range(len(columns) - 1, -1, -1):
column = columns[c]
expected_name = "cldf_{}".format(
column.propertyUrl.uri.split("#")[-1].lower())
if expected_name not in column_names and delete:
del columns[c]
else:
column_names.remove(expected_name)
for column_name in column_names:
data.add_columns(
table,
column_name.replace(
"cldf_", "http://cldf.clld.org/v1.0/terms.rdf#"))
def add_status_column_to_table(dataset: pycldf.Dataset,
table_name: str) -> None:
if "Status_Column" not in dataset[table_name].tableSchema.columndict.keys(
):
dataset.add_columns(table_name, "Status_Column")
else:
cli.logger.info(
f"Table {table_name} already contains a Status_Column.")
def add_segments_to_dataset(dataset: pycldf.Dataset, transcription: str,
overwrite_existing: bool):
if dataset.column_names.forms.segments is None:
# Create a Segments column in FormTable
dataset.add_columns("FormTable", "Segments")
c = dataset["FormTable"].tableSchema.columns[-1]
c.separator = " "
c.propertyUrl = URITemplate(
"http://cldf.clld.org/v1.0/terms.rdf#segments")
dataset.write_metadata()
write_back = []
c_f_segments = dataset["FormTable", "Segments"].name
for row in dataset["FormTable"]:
if row[c_f_segments] and not overwrite_existing:
continue
else:
if row[transcription]:
form = row[transcription].strip()
row[dataset.column_names.forms.segments] = segment_form(form)
write_back.append(row)
dataset.write(FormTable=write_back)
def add_concepticon_definitions(
dataset: pycldf.Dataset,
column_name: str = "Concepticon_Definition",
logger: cli.logging.Logger = cli.logger,
) -> None:
concepticon_ids = dataset.column_names.parameters.concepticonReference
if concepticon_ids is None:
logger.error(
"Your concepts table has no #concepticonReference column, so I cannot add any definitions from Concepticon to it. Try running lexedata.edit.add_concepticon to have me guess those references."
)
return
# Create a concepticon_definition column
try:
dataset["ParameterTable", column_name]
logger.info("Overwriting existing {:} column in concepts table".format(
column_name))
except KeyError:
dataset.add_columns("ParameterTable", column_name)
dataset.write_metadata()
# Now if this throws an exception, it's an unexpected exception.
# write concepticon definitions
write_back = []
for row in cli.tq(
dataset["ParameterTable"],
task="Write concepts with concepticon definitions to dataset",
):
try:
row[column_name] = concepticon.api.conceptsets[
row[concepticon_ids]].definition
except KeyError:
pass
write_back.append(row)
dataset.write(ParameterTable=write_back)
def add_status_column_to_table(dataset: pycldf.Dataset,
table_name: str) -> None:
if "Status_Column" not in dataset[table_name].tableSchema.columndict.keys(
):
dataset.add_columns(table_name, "Status_Column")