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 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 bind(self, header):
self.output_field_names = tuple(
field_name
for field_name in header
if field_name not in self.fields_to_remove)
field_maps = FieldMaps()
for field_name in self.output_field_names:
field_maps.add(field_name, field_name)
self.transformer = SimpleTransformer(field_maps)
self.transformer.bind(header)
def __init__(self, map_fields_spec, ref_field_spec, keep_fields=False):
field_maps = FieldMaps()
field_maps.parse_from(map_fields_spec)
self.fields_to_remove = (
set() if keep_fields else field_maps.input_field_names)
# TODO: this is ugly, beautify
ref_field_map = FieldMaps().parse_field_map_string(ref_field_spec)
self.map = Map(field_maps, ref_field_map.output_field_name)
self.ref_field_name = ref_field_map.input_field_name
self.transformer = None
def __init__(self, map_fields_spec, ref_field_spec, keep_fields=False):
field_maps = FieldMaps()
field_maps.parse_from(map_fields_spec)
self.fields_to_remove = (set() if keep_fields else
field_maps.input_field_names)
# TODO: this is ugly, beautify
ref_field_map = FieldMaps().parse_field_map_string(ref_field_spec)
self.map = Map(field_maps, ref_field_map.output_field_name)
self.ref_field_name = ref_field_map.input_field_name
self.transformer = None
def bind(self, header):
# TODO: DRY: copied from RemoveFields
# except for adding the ref field
# (extract common stuff into ProxyTransformer?)
input_fields_to_keep = tuple(
field_name
for field_name in header
if field_name not in self.fields_to_remove)
field_maps = FieldMaps()
for field_name in input_fields_to_keep:
field_maps.add(field_name, field_name)
field_maps.add(
input_field_name=None,
output_field_name=self.ref_field_name,
extractor_field=RefField(self.map))
self.transformer = SimpleTransformer(field_maps)
self.transformer.bind(header)
def bind(self, header):
# TODO: DRY: copied from RemoveFields
# except for adding the ref field
# (extract common stuff into ProxyTransformer?)
input_fields_to_keep = tuple(
field_name for field_name in header
if field_name not in self.fields_to_remove)
field_maps = FieldMaps()
for field_name in input_fields_to_keep:
field_maps.add(field_name, field_name)
field_maps.add(input_field_name=None,
output_field_name=self.ref_field_name,
extractor_field=RefField(self.map))
self.transformer = SimpleTransformer(field_maps)
self.transformer.bind(header)
def select(input_file, output_file, transform_spec):
reader = csv.reader(input_file)
writer = csv.writer(output_file)
field_maps = FieldMaps()
field_maps.parse_from(transform_spec)
SimpleTransformer(field_maps).process(reader, writer)
def simple_transformer(field_maps_string):
field_maps = FieldMaps()
field_maps.parse_from(field_maps_string)
return m.SimpleTransformer(field_maps)
def make_map(field_maps_spec, ref_field_name):
field_maps = FieldMaps()
field_maps.parse_from(field_maps_spec)
return m.Map(field_maps, ref_field_name)
