Crate fp_frame 

DataFrame / Series core for frankenpandas — the user-facing analytics surface, mirroring pandas’ DataFrame and Series APIs plus their groupby / window / resample / accessor families.

This is the largest workspace crate. It glues together:

• fp-types: scalar values, dtypes, nan* reductions
• fp-columnar: per-column storage (Column)
• fp-index: row labels and binary-op alignment
• fp-runtime: optional decision-recording policy

Most callers reach the contents of this crate via the frankenpandas umbrella prelude (use frankenpandas::prelude::*) or via direct imports (use fp_frame::DataFrame).

Core data structures

• DataFrame: columnar table with a row Index and an ordered column map. Constructors mirror pandas: from_dict, from_series, from_columns, from_records. Operations cover selection, mutation, reindexing, grouping, window-functions, joins, IO round-trips (via fp-io).
• Series: single-column labeled vector. Built-in arithmetic, reductions, and accessor families (string / datetime / categorical / sparse).
• FrameError: failure modes (column-not-found, length mismatch, dtype mismatch, alignment error, …).

GroupBy and window operations

• SeriesGroupBy (and the corresponding DataFrame-side wrappers): pandas .groupby() semantics including aggregation, transformation, filtering, and apply.
• Rolling / Expanding / Ewm: window-function builders for Series (rolling, expanding, exponentially-weighted moving stats).
• DataFrameRolling / DataFrameExpanding / DataFrameEwm: column-wise variants for DataFrame inputs.
• Resample / DataFrameResample: pandas .resample(freq) for time-indexed Series / DataFrames.

Accessors

Mirror pandas .str / .dt / .cat / .sparse namespaces:

• StringAccessor: vectorized string ops on Utf8 Series.
• DatetimeAccessor: per-element datetime field extraction. See also TzLocalizeOptions / TzAmbiguousPolicy / TzNonexistentPolicy for tz-aware localization.
• CategoricalAccessor + CategoricalMetadata: ordered / unordered categorical type machinery.
• SparseAccessor: paired with fp-columnar’s SparseColumn.

Type coercion (pandas pd.to_* analogs)

• to_numeric: best-effort string -> Int64/Float64 with errors='raise'|'coerce' semantics.
• to_datetime / to_datetime_with_format / to_datetime_with_unit / to_datetime_with_options: parse a Series of dates with explicit format / unit / origin (ToDatetimeOrigin) controls. Options bundle: ToDatetimeOptions.
• to_timedelta / to_timedelta_with_unit: parse ISO-8601 / pandas-shorthand durations. ToTimedeltaOptions / ToTimedeltaErrors for tunables.

Frame combination

concat_dataframes, concat_series, concat_dataframes_with_* variants mirror pd.concat(..., axis=, join=, ignore_index=, keys=). Merge / join helpers live in the sibling fp-join crate (merge_dataframes_on, merge_asof, merge_ordered, join_series).

SQL IO

Read / write IO for DataFrames lives in fp-io, which exposes the DataFrameIoExt extension trait so users can call df.to_parquet(path), df.to_sql(conn, ..), etc. directly on values from this crate.

Structs

BoxPlotSpec

Backend-neutral boxplot request produced by boxplot() hooks.

CategoricalAccessor

Accessor for categorical operations on a Series.

CategoricalMetadata

Metadata for categorical (factor) data.

DataFrame

DataFrameCsvReadOptions

DataFrameDictSplit

DataFrameDictTight

Pandas 2.2 to_dict(orient='tight') payload.

DataFrameEwm

Exponentially weighted moving window over a DataFrame’s numeric columns.

DataFrameExpanding

Expanding window aggregation over a DataFrame’s numeric columns.

DataFrameFlags

DataFrameGroupBy

Deferred GroupBy object for DataFrame.

DataFrameResample

Time-based resampling view over a DataFrame’s numeric columns.

DataFrameRolling

Rolling window aggregation over a DataFrame’s numeric columns.

DataFrameXArrayDataset

DataFrameXArrayVariable

DatetimeAccessor

Ewm

Exponentially weighted moving window aggregation over a Series.

Expanding

Expanding window aggregation over a Series.

GroupByEwm

Grouped exponentially weighted window over a DataFrameGroupBy.

GroupByExpanding

Grouped expanding window over a DataFrameGroupBy.

GroupByResample

Grouped resampling over a DataFrameGroupBy.

GroupByRolling

Grouped rolling window over a DataFrameGroupBy.

HistogramSpec

Backend-neutral histogram request produced by hist() hooks.

ListAccessor

Accessor for list operations on a Series.

OnlineEwm

Streaming EWM accumulator. Returned by Ewm::online.

PlotSeriesSpec

One logical series ready for a future plotting backend.

PlotSpec

Backend-neutral plot request produced by plot() hooks.

Resample

Time-based resampling view over a Series.

Rolling

Rolling window aggregation over a Series.

Series

SeriesGroupBy

GroupBy for Series, grouping by values of another Series.

SeriesGroupByEwm

Grouped exponentially weighted window over a SeriesGroupBy.

SeriesGroupByExpanding

Grouped expanding window over a SeriesGroupBy.

SeriesGroupByResample

Grouped resampling over a SeriesGroupBy.

SeriesGroupByRolling

Grouped rolling window over a SeriesGroupBy.

SeriesXArrayDataArray

xarray DataArray representation of a Series.

SparseAccessor

Accessor for sparse operations on a Series.

SparseDataFrameView

StringAccessor

Created by Series::str(). Provides string manipulation methods analogous to pandas Series.str namespace.

StructAccessor

Accessor for struct operations on a Series.

StyledDataFrame

ToDatetimeOptions

ToTimedeltaOptions

Options for to_timedelta conversion.

TzLocalizeOptions

Enums

ConcatJoin

CsvQuoting

DataFrameArithmeticOperand

DataFrameColumnInput

DataFrameComparisonOperand

DataFrameDictAxisLabels

DataFrameDictResult

DropNaHow

FrameError

PlotKind

Logical plot kind requested by a pandas-style plotting hook.

SeriesResetIndexResult

ToDatetimeOrigin

ToTimedeltaErrors

Error handling mode for to_timedelta conversions.

TzAmbiguousPolicy

Datetime accessor for Series, analogous to pd.Series.dt.

TzNonexistentPolicy

Traits

DataFrameApplyArg

Functions

concat_dataframes

Concatenate DataFrames along axis 0 (row-wise).

concat_dataframes_with_axis

Concatenate DataFrames along the requested axis.

concat_dataframes_with_axis_join

Concatenate DataFrames along axis with explicit join policy.

concat_dataframes_with_ignore_index

Concatenate DataFrames along axis 0 with optional index reset.

concat_dataframes_with_keys

Concatenate DataFrames with keys for hierarchical labeling.

concat_series

Concatenate multiple Series along axis 0 (row-wise).

concat_series_with_ignore_index

Concatenate Series with optional index reset.

cut

index_to_frame

Convert an Index to a single-column DataFrame.

index_to_series

Convert an Index to a Series.

qcut

Quantile-based binning.

timedelta_total_seconds

Convert a timedelta-like Series to total seconds as Float64.

to_datetime

Convert a Series of strings or integers to normalized ISO 8601 datetime strings.

to_datetime_with_format

Convert to datetime with an explicit format string.

to_datetime_with_options

Convert to datetime with explicit pandas-like options.

to_datetime_with_unit

Convert numeric epoch values using an explicit unit.

to_numeric

Convert a Series to a numeric dtype.

to_timedelta

Convert a Series of strings or numbers to Timedelta64 representation.

to_timedelta_with_options

Convert to timedelta with explicit options.

to_timedelta_with_unit

Convert numeric values to timedelta using explicit unit.