def properties_from_row(
self, row: t.List[openpyxl.cell.Cell]
) -> t.Optional[RowObject]:
self.row_prop_separators = [
self.db.dataset["CognatesetTable", k].separator for k in self.row_header
]
data = [clean_cell_value(cell) for cell in row[: self.left - 1]]
properties: t.Dict[t.Optional[str], t.Any] = {
n: (v if sep is None else v.split(sep))
for n, sep, v in zip(self.row_header, self.row_prop_separators, data)
}
if not any(properties.values()):
return None
# delete all possible None entries coming from row_header
cogset: t.Dict[str, t.Any] = {
key: value for key, value in properties.items() if key is not None
}
while None in properties.keys():
del properties[None]
comments: t.List[str] = []
for cell in row[: self.left - 1]:
c = get_cell_comment(cell)
if c is not None:
comments.append(c)
comment = "\t".join(comments).strip()
cogset[self.db.dataset["CognatesetTable", "comment"].name] = comment
return CogSet(cogset)
def language_from_column(self, column: t.List[openpyxl.cell.Cell]) -> Language:
data = [clean_cell_value(cell) for cell in column[: self.top - 1]]
# Do we need to know language comments? – comment = get_cell_comment(column[0])
return Language(
{
self.db.dataset["LanguageTable", "name"].name: data[0],
}
)
def language_from_column(self,
column: t.List[openpyxl.cell.Cell]) -> Language:
data = [clean_cell_value(cell) for cell in column[:self.top - 1]]
comment = get_cell_comment(column[0])
id = string_to_id(data[0])
return Language(
# an id candidate must be provided, which is transformed into a unique id
ID=id,
Name=data[0],
Comment=comment,
)
def parse(
self, cell: openpyxl.cell.Cell, language_id: str, cell_identifier: str = ""
) -> t.Iterable[Form]:
"""Return form properties for every form in the cell"""
text = clean_cell_value(cell)
if not text:
return []
for element in self.separate(
text, context=cell_identifier and f"{cell_identifier}: "
):
try:
form = self.parse_form(element, language_id, cell_identifier)
except KeyError:
continue
if form:
yield form
def properties_from_row(
self, row: t.List[openpyxl.cell.Cell]) -> t.Optional[RowObject]:
row_object = self.row_object()
c_id = self.db.dataset[row_object.__table__, "id"].name
c_comment = self.db.dataset[row_object.__table__, "comment"].name
c_name = self.db.dataset[row_object.__table__, "name"].name
data = [clean_cell_value(cell) for cell in row[:self.left - 1]]
properties = dict(zip(self.row_header, data))
# delete all possible None entries coming from row_header
while None in properties.keys():
del properties[None]
# fetch cell comment
comment = get_cell_comment(row[0])
properties[c_comment] = comment
# cldf_name serves as cldf_id candidate
properties[c_id] = properties[c_name]
# create new row object
return self.row_object(properties)
def import_data_from_sheet(
sheet,
sheet_header,
implicit: t.Mapping[Literal["languageReference", "id", "value"], str] = {},
entries_to_concepts: t.Mapping[str, str] = KeyKeyDict(),
concept_column: t.Tuple[str, str] = ("Concept_ID", "Concept_ID"),
) -> t.Iterable[Form]:
row_iter = sheet.iter_rows()
# TODO?: compare header of this sheet to format of given data set process
# row. Maybe unnecessary. In any case, do not complain about the unused
# variable.
header = next(row_iter) # noqa: F841
assert (
concept_column[1] in sheet_header
), f"Could not find concept column {concept_column[0]} in your excel sheet {sheet.title}."
for row in row_iter:
data = Form({k: clean_cell_value(cell) for k, cell in zip(sheet_header, row)})
if "value" in implicit:
data[implicit["value"]] = "\t".join(map(str, data.values()))
try:
concept_entry = data.pop(concept_column[1])
data[concept_column[0]] = entries_to_concepts[concept_entry]
except KeyError:
logger.warning(
f"Concept {concept_entry} was not found. Please add it to the concepts table manually. The corresponding form was ignored and not added to the dataset."
)
data[concept_column[0]] = concept_entry
continue
if "id" in implicit:
data[implicit["id"]] = None
if "languageReference" in implicit:
data[implicit["languageReference"]] = sheet.title
yield data
def parse(
self, cell: openpyxl.cell.Cell, language_id: str, cell_identifier: str = ""
) -> t.Iterable[Judgement]:
try:
url = cell.hyperlink.target
text = clean_cell_value(cell)
comment = get_cell_comment(cell)
if "{" not in text:
slice, alignment = alignment_from_braces("{" + text + "}")
else:
slice, alignment = alignment_from_braces(text)
properties = {
self.c["c_id"]: url.split("/")[-1],
self.c.get("c_segments"): ",".join(
"{:}:{:}".format(i, j) for i, j in slice
),
self.c.get("c_alignment"): alignment,
self.c.get("c_comment"): comment,
}
properties.pop(None, None)
yield Judgement(properties)
except AttributeError:
pass
def cells_are_empty(cells: t.Iterable[openpyxl.cell.Cell]) -> bool:
return not any([clean_cell_value(cell) for cell in cells])