Summary: 249 instances, 239 unique

Text	Count
# TODO: there are many places that rely on these private methods existing in	1
# TODO: consider re-implementing freq._should_cache for fastpath	1
# TODO: what if a level contains tuples??	1
# TODO: Fix join unit generation so we aren't passed this.	1
# TODO: cleanup this special case.	1
# TODO: Can we do this before the is_integer_dtype check?	1
# TODO: handle cell width and height: needs support in pandas.io.excel	1
# TODO: Can we make the allowed_types arg unnecessary?	1
# FIXME: use Index.delete as soon as it uses fastpath=True	1
# TODO: sort=False here for backwards compat. It may	1
# TODO: disallow unit-less timedelta64	1
# TODO: mixed type case	1
# TODO: Could skip verify_integrity here.	1
# TODO: could use -Xlinker here, if it's supported	2
s = s.to_sparse()  # TODO: specify kind?	1
# TODO: this works around .astype(bool) not working properly (gh-9847)	1
# FIXME: real_t is unused	1
/* TODO: How to do this warning message for other compilers? */	1
# TODO: Support 'ab'	1
# TODO: implement get_f90flags and use it in _compile similarly to get_f77flags	1
# TODO: this should be seriously cythonized	1
# TODO: should probably raise on `how` here, so we don't ignore it.	1
# TODO: GH 15362	1
# TODO: these two lines can rejplace the code below but	1
# TODO: may need to coerce array to fill value	1
# TODO: consider making the results of broadcast_arrays readonly to match	1
# FIXME: optimization potential	1
# TODO: delete sparse stuff in core/ops.py	1
# TODO: this can be combined with DataFrame.sort_index impl as	1
# TODO: simplify, figure out type of values	1
# TODO: watch out for overflows	1
# TODO: Remove after 0.23.x	1
# TODO: why not timedelta?	1
# TODO: I don't think the functions defined by bool_method are tested	1
# FIXME:	1
# TODO: bug?	1
# TODO: standardize `_offset` vs `offset` naming convention	1
#XXX TODO: --inplace support for sdist command	1
# FIXME: GH#5284, GH#5035, GH#19448	1
# TODO: this prob needs some better checking	1
# TODO: We have no tests for these	1
# FIXME: This seems like it will copy strings around	1
# TODO: use to_period_array	1
# TODO: support for multiIndex	1
# TODO: what about datetime64tz?  PeriodDtype?	1
# TODO: we shouldn't be setting attributes like this;	1
# TODO: transformations??	1
# 3. TODO: resolve other font-relative units	1
# TODO: sort=True here for backwards compat. It may	1
TODO:	2
TODO:	1
# TODO: When FrozenNDArray is removed, add	1
# TODO: rmod? rdivmod?	1
# TODO: If/when pandas supports more than datetime64[ns], this should be	1
# TODO: deprecate	1
# TODO: Add option for bins like value_counts()	1
# TODO: speed up on homogeneous DataFrame objects	1
# TODO: perhaps allow for special properties	1
# TODO: the actual managing of mgr_locs is a PITA	1
t = c[j]/(2*j + 1)  # FIXME: t never used	1
# TODO: https://github.com/pandas-dev/pandas/issues/23023	1
# TODO: See if we can avoid these copies	1
# TODO: Do this timedelta properly in objToJSON.c See GH #15137	1
# TODO: Refactor to remove code duplication with code	1
# TODO: Does this make sense for the general case?  It would help	1
# TODO: See if we can avoid these copies	1
# TODO: what about the existing index?	1
# TODO: This is no longer hit internally; does it need to be retained	1
# TODO: Document public vs. private API	1
# TODO: skipna is broken with max.	1
# TODO: handle record value which are lists, at least error	1
# TODO: Investigate an alternative implementation with	1
# TODO: Refactor to use _get_column_name_list from	1
# TODO: infer dtypes other than float64 from fill_value	1
# TODO: there may be a more performant way to do this	1
# TODO: lexsort depth needs to be 2!!	1
# TODO: only copy DataFrames when modification necessary	1
# TODO: Use the fact that axis can have units to simplify the process	1
# TODO: namespace all the pandas keys	1
# TODO:	1
# TODO: disentangle the fill_value dtype inference from	1
# TODO emit portability warning?	1
# TODO: this expands to a tuple index, see if we can	1
# TODO: Remove import when statsmodels updates #18264	1
# TODO: this path can never be reached	1
# TODO: Make other agg func handle axis=None properly	2
# TODO: be a bit more intelligent here	1
# TODO: Refactor to use _get_column_name_list from	1
# FIXME: optimization potential in case all mgrs contain slices and	1
# TODO: See if this can be shared	1
# TODO: docstring?	2
# TODO: fill value handling	1
# FIXME: cache results of indexer == -1 checks.	1
# TODO: we shouldn't be setting attributes like this;	1
# TODO: We do not have tests specific to string-dtypes,	1
# GH#23917 TODO: needs tests for case where lk is integer-dtype	1
# TODO: used only once in indexing; belongs elsewhere?	1
# TODO: Look into deprecating this in favor of `to_dense`.	1
#FIXME This is no longer needed? Assume it was for use in tests.	1
# FIXME: r not used	1
# TODO: I think we can avoid densifying when masking a	1
# TODO: cases where we need to do another pass through this func,	1
# TODO: if we are a MultiIndex, we can do better	1
# TODO: Most of this should be refactored and shared with Series	1
# TODO speed up Series case	1
# TODO: memoize?	1
# TODO: When this deprecation is enforced, PeriodIndex.freq can	1
idx += length  # TODO: wraparound	1
# TODO: remove except block OR modify pc.load to use compat	1
# TODO: Consider making Period an instance of PeriodDtype	1
# TODO: Needs a better name; too many modules are already called "concat"	1
# TODO: deprecate this behavior to instead treat symmetrically	1
# TODO: Consider deprecating these in favor of `union` (xref gh-15506)	1
# TODO: don't lowercase case sensitive parts of values (strings)	1
# TODO: gh-15585: consider making the checks stricter.	1
# TODO: text-indent, padding-left -> alignment.indent	1
# TODO: Refactor when convert_objects is removed since there will be 1 path	1
# TODO: going through __new__ raises on call to _validate_frequency;	3
# TODO: avoid double copy when dtype forces cast.	1
# TODO: figure out how to handle this case, all nan's?	1
# TODO: This is basically the same as BQuarterEnd	1
# FIXME: DTA/TDA/PA inplace methods should actually be inplace, GH#24115	1
# TODO: handle the optional third argument for __pow__?	1
# TODO: arithmetic operations	1
# TODO: implement at Cython level?	1
# TODO:  This should be more robust.  Handles case where path includes	1
# TODO: Do attrs get handled reliably?	1
# TODO: update this.	1
# TODO: Document difference from Series.__getitem__, deprecate,	1
# TODO: this duplicates the C append_metastr_to_string	1
# TODO: This should be moved to the array?	1
# TODO: remove this when we drop Python 2 support (functools.wraps	1
# TODO: same for tuples?	1
# TODO: support other fill patterns	1
# GH#23917 TODO: Needs tests for non-matching dtypes	1
# TODO: Doesn't handle compressed files!	1
# TODO: test .eq., .neq., etc replacements.	1
# TODO: Figure out the end of this sente	1
# TODO: Better unicode/repr for GroupBy object	1
# TODO: There is no support for opening a file for writing which	1
# FIXME: lapack_atlas is unused	1
# TODO: Use contextlib.ExitStack once we drop Python 2	1
# FIXME: mgr_groupby_blknos must return mgr_locs in ascending order,	1
# TODO: combine with hasattr(result, 'dtype') further down	1
# TODO: verify whether any path hits this except #18819 (invalid)	1
# TODO: https://github.com/asweigart/pyperclip/issues/43	1
# TODO: how to do this better? cleanly slice nonnull_labels given the	1
# TODO: _level is unused?	1
# TODO: can we make a no-copy implementation?	1
# TODO: This try/except will be repeated.	1
# TODO: expand this to handle a default datetime format?	1
# TODO: validate this in a more general way (thinking of future engines	1
# TODO: make kind=None, and use data.kind?	1
# TODO: integrate bottleneck	1
# TODO: min_count	1
# TODO: .zip support, .tar support?	1
- bool -> uint64 (TODO this should be uint8)	1
# TODO: remove?	1
# TODO: _shallow_copy(subset)?	1
# TODO: Combine this with BusinessMixin version by defining a whitelisted	1
# FIXME:RELATIVE_IMPORT	1
# TODO: for int-dtypes we make a copy, but for everything else this	1
# TODO: this logic is surely elsewhere	1
# TODO: should this be deepcopy?	1
# FIXME _sharedmask is never used.	1
# TODO:	1
# TODO: use Matplotlib public API when available	1
# TODO: Should Tick have its own apply_index?	1
# TODO: Why is DatetimeIndex being imported here?	1
# TODO: Add a ``subdir`` parameter for specifying the subdirectory	1
# TODO: handle uint8	1
# FIXME: In the case where there are more than two packages,	1
# TODO: expand to handle datetime to integer conversion	2
# TODO: remove	1
# TODO: make this more flexible than just ndarray...	1
# TODO: this can be combined with Series.sort_index impl as	1
# TODO: Filter the warnings we actually care about here.	1
# FIXME: ordered_funcs_api is unused	1
# TODO: is the next line needed above in the data(...) method?	1
# TODO: remove when the default Categorical.take behavior changes	1
# _shared_docs['pivot_table'] will not yet exist.  TODO: Fix this dependency	1
# TODO: is_dtype_union_equal is a hack around	1
# FIXME: pyperclip currently does not support Cygwin,	1
# TODO: Use .array	1
# TODO: infer freq?	2
# TODO: Refactor to remove code duplication with code block	1
# TODO: should we do this inside II?	1
# TODO: should this be ju.block._can_hold_na?	1
# TODO: consider privatizing (discussion in GH#23113)	1
# TODO: raising on floats is tested, but maybe not useful.	1
# TODO: a generic formatter wld b in DataFrameFormatter	1
# FIXME complex numbers may also have exponents	1
#FIXME when yield tests are gone.	1
# TODO: default is for backward compat	1
# TODO: handle index names!	1
operator.mod: None,  # TODO: Why None for mod but '%' for rmod?	1
# TODO: Can we simplify/generalize these cases at all?	1
# TODO: watch out for overflows when converting from lower-resolution	1
# TODO: flexible with index=None and/or items=None	1
# TODO: hash_array doesn't handle mixed types. It casts	1
# TODO: Why does this handle the 0 case the opposite of others?	1
# TODO: Refine integer size.	1
* FIXME: This should check for a flag on the data-type that	1
# TODO: make more efficient	1
# TODO: support %	1
# TODO: Check the following : is it really info['fmt'] ?	1
# TODO: Consider making Interval an instance of IntervalDtype	1
# TODO: NonConsolidatableMixin shape	1
# TODO: Why self.values and not values?	1
# TODO deprecate numeric_only argument for Categorical and use	1
# TODO: These if-block and else-block are almost same.	1
# TODO: Support other joins	1
# TODO: Decide if we care about having different examples for different	1
# TODO: remove once pd.concat sort default changes	1
# TODO: Refactor to combine type with format	1
# TODO: copy	1
# FIXME: document this. If pkgname is defined in the variables section, and	1
# TODO: look into _wrap_result	1
# TODO: If we knew what was going in to **d, we might be able to	1
# TODO: Remove Datetime & DatetimeTZ formatters.	1
# TODO: we're stuck with disabling math formatting until we handle	1
# TODO: consider re-implementing _cached_range; GH#17914	1
# TODO: Not sure if above is correct - need someone to confirm.	1
# TODO: We can do some of these with no-copy / coercion?	1
# TODO: what if they both have np.nan for their names?	1
# FIXME: refactor, clearly separate broadcasting & zip-like assignment	1
# TODO: unused?	1
# TODO: Consider combining QuarterOffset and YearOffset __init__ at some	1
# TODO: SparseSeries.isna is Sparse, while Series.isna is dense	1
# TODO: Remove this conditional when #23918 is fixed	1
# TODO: Figure out whether this can be reached.	1
# TODO: what about self.name?  tz? if so, use shallow_copy?	1
# TODO: this could be more efficient	1
# TODO: https://github.com/pandas-dev/pandas/issues/22314	1
# TODO: integrate with categorical and make generic	2
# TODO: Index.union returns other when `len(self)` is 0.	1
# TODO: implement this is a non-naive way!	1
# TODO: handle n here...	2
# TODO: https://github.com/pandas-dev/pandas/issues/22791	1