def box_categorical_array(typ, val, c):
dtype = typ.dtype
mod_name = c.context.insert_const_string(c.builder.module, "pandas")
pd_class_obj = c.pyapi.import_module_noblock(mod_name)
# categories list e.g. ['A', 'B', 'C']
item_objs = _get_cat_obj_items(dtype.categories, c)
n = len(item_objs)
list_obj = c.pyapi.list_new(c.context.get_constant(types.intp, n))
for i in range(n):
idx = c.context.get_constant(types.intp, i)
c.pyapi.incref(item_objs[i])
c.pyapi.list_setitem(list_obj, idx, item_objs[i])
int_dtype = get_categories_int_type(dtype)
arr = box_array(types.Array(int_dtype, 1, 'C'), val, c)
pdcat_cls_obj = c.pyapi.object_getattr_string(pd_class_obj, "Categorical")
cat_arr = c.pyapi.call_method(pdcat_cls_obj, "from_codes", (arr, list_obj))
c.pyapi.decref(pdcat_cls_obj)
c.pyapi.decref(arr)
c.pyapi.decref(list_obj)
for obj in item_objs:
c.pyapi.decref(obj)
c.pyapi.decref(pd_class_obj)
return cat_arr
def box_int64_index(typ, val, c):
mod_name = c.context.insert_const_string(c.builder.module, "pandas")
pd_class_obj = c.pyapi.import_module_noblock(mod_name)
int64_index = cgutils.create_struct_proxy(typ)(c.context, c.builder, val)
data = box_array(typ.data, int64_index.data, c)
# dtype and copy params are not stored so use default values
dtype = c.pyapi.make_none()
copy = c.pyapi.bool_from_bool(c.context.get_constant(types.bool_, False))
if typ.is_named:
name = c.pyapi.from_native_value(types.unicode_type, int64_index.name)
else:
name = c.pyapi.make_none()
res = c.pyapi.call_method(pd_class_obj, "Int64Index",
(data, dtype, copy, name))
c.pyapi.decref(data)
c.pyapi.decref(dtype)
c.pyapi.decref(copy)
c.pyapi.decref(name)
c.pyapi.decref(pd_class_obj)
return res
def box_dataframe(typ, val, c):
context = c.context
builder = c.builder
n_cols = len(typ.columns)
col_names = typ.columns
arr_typs = typ.data
dtypes = [a.dtype for a in arr_typs] # TODO: check Categorical
dataframe = cgutils.create_struct_proxy(typ)(context, builder, value=val)
col_arrs = [builder.extract_value(dataframe.data, i) for i in range(n_cols)]
# df unboxed from Python
has_parent = cgutils.is_not_null(builder, dataframe.parent)
pyapi = c.pyapi
# gil_state = pyapi.gil_ensure() # acquire GIL
mod_name = context.insert_const_string(c.builder.module, "pandas")
class_obj = pyapi.import_module_noblock(mod_name)
df_dict = pyapi.dict_new()
for i, cname, arr, arr_typ, dtype in zip(range(n_cols), col_names, col_arrs, arr_typs, dtypes):
# df['cname'] = boxed_arr
# TODO: datetime.date, DatetimeIndex?
name_str = context.insert_const_string(c.builder.module, cname)
cname_obj = pyapi.string_from_string(name_str)
if dtype == string_type:
arr_obj = box_str_arr(arr_typ, arr, c)
elif isinstance(arr_typ, Categorical):
arr_obj = box_Categorical(arr_typ, arr, c)
# context.nrt.incref(builder, arr_typ, arr)
elif dtype == types.List(string_type):
arr_obj = box_list(list_string_array_type, arr, c)
# context.nrt.incref(builder, arr_typ, arr) # TODO required?
# pyapi.print_object(arr_obj)
else:
arr_obj = box_array(arr_typ, arr, c)
# TODO: is incref required?
# context.nrt.incref(builder, arr_typ, arr)
pyapi.dict_setitem(df_dict, cname_obj, arr_obj)
pyapi.decref(arr_obj)
pyapi.decref(cname_obj)
df_obj = pyapi.call_method(class_obj, "DataFrame", (df_dict,))
pyapi.decref(df_dict)
# set df.index if necessary
if typ.index != types.none:
arr_obj = _box_series_data(typ.index.dtype, typ.index, dataframe.index, c)
pyapi.object_setattr_string(df_obj, 'index', arr_obj)
pyapi.decref(arr_obj)
pyapi.decref(class_obj)
# pyapi.gil_release(gil_state) # release GIL
return df_obj
def _box_index_data(index_typ, val, c):
""" Boxes native value used to represent Pandas index into appropriate Python object.
Params:
index_typ: Numba type of native value
val: native value
c: LLVM context object
Returns: Python object native value is boxed into
"""
assert isinstance(index_typ, (RangeIndexType, StringArrayType, types.Array, types.NoneType))
if isinstance(index_typ, RangeIndexType):
index = box_range_index(index_typ, val, c)
elif isinstance(index_typ, types.Array):
index = box_array(index_typ, val, c)
elif isinstance(index_typ, StringArrayType):
index = box_str_arr(string_array_type, val, c)
else: # index_typ is types.none
index = c.pyapi.make_none()
return index
def _box_series_data(dtype, data_typ, val, c):
if isinstance(dtype, types.BaseTuple):
np_dtype = np.dtype(','.join(str(t) for t in dtype.types), align=True)
dtype = numba.np.numpy_support.from_dtype(np_dtype)
if dtype == string_type:
arr = box_str_arr(string_array_type, val, c)
elif isinstance(dtype, CategoricalDtypeType):
arr = box_Categorical(data_typ, val, c)
elif dtype == types.List(string_type):
arr = box_list(list_string_array_type, val, c)
else:
arr = box_array(data_typ, val, c)
if isinstance(dtype, types.Record):
o_str = c.context.insert_const_string(c.builder.module, "O")
o_str = c.pyapi.string_from_string(o_str)
arr = c.pyapi.call_method(arr, "astype", (o_str, ))
return arr
def box_Categorical(typ, val, c):
pandas_module_name = c.context.insert_const_string(c.builder.module,
"pandas")
pandas_module = c.pyapi.import_module_noblock(pandas_module_name)
constructor = c.pyapi.object_getattr_string(pandas_module, "Categorical")
empty_list = c.pyapi.list_new(c.context.get_constant(types.intp, 0))
args = c.pyapi.tuple_pack([empty_list])
categorical = c.pyapi.call(constructor, args)
dtype = box_CategoricalDtype(typ.pd_dtype, val, c)
c.pyapi.object_setattr_string(categorical, "_dtype", dtype)
codes = boxing.box_array(typ.codes, val, c)
c.pyapi.object_setattr_string(categorical, "_codes", codes)
c.pyapi.decref(codes)
c.pyapi.decref(dtype)
c.pyapi.decref(args)
c.pyapi.decref(empty_list)
c.pyapi.decref(constructor)
c.pyapi.decref(pandas_module)
return categorical
def box_Categorical(typ, val, c):
pandas_module_name = c.context.insert_const_string(c.builder.module,
"pandas")
pandas_module = c.pyapi.import_module_noblock(pandas_module_name)
categorical_class = c.pyapi.object_getattr_string(pandas_module,
"Categorical")
method_from_codes = c.pyapi.object_getattr_string(categorical_class,
"from_codes")
dtype = box_CategoricalDtype(typ.pd_dtype, val, c)
codes = boxing.box_array(typ.codes, val, c)
py_none = c.pyapi.make_none()
args = c.pyapi.tuple_pack([codes, py_none, py_none, dtype])
categorical = c.pyapi.call(method_from_codes, args=args)
c.pyapi.decref(args)
c.pyapi.decref(py_none)
c.pyapi.decref(codes)
c.pyapi.decref(dtype)
c.pyapi.decref(method_from_codes)
c.pyapi.decref(categorical_class)
c.pyapi.decref(pandas_module)
return categorical
