def read(self, reader): rows = iter(reader) header = rows.next() if self.ref_field_name not in header: raise MissingFieldError(self.ref_field_name) for field_name in self.transformer.output_field_names: if field_name not in header: raise MissingFieldError(self.ref_field_name) # first field is ID field_maps = FieldMaps() for input_field_name in self.field_names: field_maps.add(input_field_name, input_field_name) map_transformer = SimpleTransformer(field_maps) map_transformer.bind(header) count = 0 values = dict() for row in rows: transformed_row = map_transformer.transform(row) ref = int(transformed_row[0]) value = transformed_row[1:] values[value] = ref count += 1 if count != len(values): raise DuplicateValuesError() if count != len(set(values.values())): raise DuplicateRefsError() self.values = values self.next_ref = max(values.values()) + 1
class Map(object): changed = False def __init__(self, map_field_maps, ref_field_name): self.transformer = SimpleTransformer(map_field_maps) self.ref_field_name = ref_field_name self.values = dict() self.next_ref = 0 def read(self, reader): rows = iter(reader) header = rows.next() if self.ref_field_name not in header: raise MissingFieldError(self.ref_field_name) for field_name in self.transformer.output_field_names: if field_name not in header: raise MissingFieldError(self.ref_field_name) # first field is ID field_maps = FieldMaps() for input_field_name in self.field_names: field_maps.add(input_field_name, input_field_name) map_transformer = SimpleTransformer(field_maps) map_transformer.bind(header) count = 0 values = dict() for row in rows: transformed_row = map_transformer.transform(row) ref = int(transformed_row[0]) value = transformed_row[1:] values[value] = ref count += 1 if count != len(values): raise DuplicateValuesError() if count != len(set(values.values())): raise DuplicateRefsError() self.values = values self.next_ref = max(values.values()) + 1 def write(self, writer): writer.writerow(self.field_names) for (value, ref) in self.values.iteritems(): writer.writerow(tuple([ref]) + tuple(value)) def translate(self, input_row): key = self.transformer.transform(input_row) ref = self.values.setdefault(key, self.next_ref) if ref == self.next_ref: self.next_ref += 1 self.changed = True return ref def bind(self, header): self.transformer.bind(header) @property def field_names(self): return ( tuple([self.ref_field_name]) + self.transformer.output_field_names)
class Map(object): changed = False def __init__(self, map_field_maps, ref_field_name): self.transformer = SimpleTransformer(map_field_maps) self.ref_field_name = ref_field_name self.values = dict() self.next_ref = 0 def read(self, reader): rows = iter(reader) header = rows.next() if self.ref_field_name not in header: raise MissingFieldError(self.ref_field_name) for field_name in self.transformer.output_field_names: if field_name not in header: raise MissingFieldError(self.ref_field_name) # first field is ID field_maps = FieldMaps() for input_field_name in self.field_names: field_maps.add(input_field_name, input_field_name) map_transformer = SimpleTransformer(field_maps) map_transformer.bind(header) count = 0 values = dict() for row in rows: transformed_row = map_transformer.transform(row) ref = int(transformed_row[0]) value = transformed_row[1:] values[value] = ref count += 1 if count != len(values): raise DuplicateValuesError() if count != len(set(values.values())): raise DuplicateRefsError() self.values = values self.next_ref = max(values.values()) + 1 def write(self, writer): writer.writerow(self.field_names) for (value, ref) in self.values.iteritems(): writer.writerow(tuple([ref]) + tuple(value)) def translate(self, input_row): key = self.transformer.transform(input_row) ref = self.values.setdefault(key, self.next_ref) if ref == self.next_ref: self.next_ref += 1 self.changed = True return ref def bind(self, header): self.transformer.bind(header) @property def field_names(self): return (tuple([self.ref_field_name]) + self.transformer.output_field_names)