Open
Description
I wrote a quick and dirty script that checks to see which pandas-stubs functions are missing from the public API reference. Note that this only looks at top-level functions, not methods.
The list of undocumented functions is as follows:
add_ops
align_terms
all_indexes_same
all_none
all_not_none
any_none
any_not_none
apply_if_callable
array_strptime
asarray_tuplesafe
assert_almost_equal
assert_attr_equal
assert_categorical_equal
assert_class_equal
assert_contains_all
assert_copy
assert_datetime_array_equal
assert_dict_equal
assert_equal
assert_interval_array_equal
assert_is_sorted
assert_is_valid_plot_return_object
assert_numpy_array_equal
assert_period_array_equal
assert_sp_array_equal
assert_timedelta_array_equal
cast_scalar_indexer
check_bool_indexer
check_setitem_lengths
clean_backtick_quoted_toks
clean_column_name
clean_fill_method
clean_interp_method
clean_reindex_fill_method
coerce_to_array
consensus_name_attr
convert_from_missing_indexer_tuple
convert_missing_indexer
convert_to_index_sliceable
count_not_none
create_valid_python_identifier
deprecate_ndim_indexing
ensure_clean
ensure_index_from_sequences
ensure_scope
extract_array
factorize_from_iterable
factorize_from_iterables
fill_binop
get_array_op
get_callable_name
get_fill_func
get_objs_combined_axis
get_op_result_name
get_period_alias
get_rename_function
get_test_result
get_versions
get_writer
index_labels_to_array
interpolate_2d
ints_to_td64ns
invalid_comparison
is_array_like
is_bool_indexer
is_empty_data
is_empty_indexer
is_full_slice
is_label_like
is_list_like_indexer
is_nested_tuple
is_null_slice
is_scalar_indexer
is_term
is_terminal
is_true_slices
kleene_and
kleene_or
kleene_xor
localize_pydatetime
logical_op
make_invalid_op
mask_missing
maybe_convert_dtype
maybe_convert_indices
maybe_convert_ix
maybe_dispatch_ufunc_to_dunder_op
maybe_droplevels
maybe_expression
maybe_extract_name
maybe_infer_freq
maybe_iterable_to_list
maybe_make_list
na_logical_op
need_slice
not_none
objects_to_td64ns
parse_time_string
raise_for_nan
reconstruct_object
register_converter_cb
register_plotting_backend_cb
register_writer
result_type_many
rewrite_exception
sanitize_array
sequence_to_td64ns
set_numexpr_threads
set_test_mode
set_use_numexpr
should_extension_dispatch
standardize_mapping
table_schema_cb
tokenize_backtick_quoted_string
tokenize_string
tz_to_dtype
union_indexes
unpack_zerodim_and_defer
use_bottleneck_cb
use_inf_as_na_cb
use_numexpr_cb
validate_indices
validate_tz_from_dtype
Additionally, there are a few functions that are only mentioned in the "what's new" release notes, but aren't actually documented.
cartesian_product
ensure_index
length_of_indexer
objects_to_datetime64ns
Should they be removed from these stubs? Or should we reach out to get them documented?
Metadata
Metadata
Assignees
Labels
No labels