3
���h?) � @ s d Z ddlmZmZmZmZmZ ddlZddlj j
Z
ddl m Z m
Z
mZ ddlmZmZmZmZmZmZmZmZmZ ddlmZ ddlm Z ddl j j Z er�ddl!m"Z" e
j#Z#e e$d �d
d �Z%d d
� Z&dd� Z'dd� Z(de
e)ee) e)ee$ef d�dd�Z*dd� Z+dS )zM
Table Schema builders
https://specs.frictionlessdata.io/json-table-schema/
� )�
TYPE_CHECKING�Any�Dict�Optional�castN)�DtypeObj�
FrameOrSeries�JSONSerializable) �
is_bool_dtype�is_categorical_dtype�is_datetime64_dtype�is_datetime64tz_dtype�is_integer_dtype�is_numeric_dtype�is_period_dtype�is_string_dtype�is_timedelta64_dtype)�CategoricalDtype)� DataFrame)�
MultiIndex)�x�returnc C sl t | �r dS t| �rdS t| �r$dS t| �s<t| �s<t| �r@dS t| �rLdS t| �rXdS t| �rddS dS dS ) a�
Convert a NumPy / pandas type to its corresponding json_table.
Parameters
----------
x : np.dtype or ExtensionDtype
Returns
-------
str
the Table Schema data types
Notes
-----
This table shows the relationship between NumPy / pandas dtypes,
and Table Schema dtypes.
============== =================
Pandas type Table Schema type
============== =================
int64 integer
float64 number
bool boolean
datetime64[ns] datetime
timedelta64[ns] duration
object str
categorical any
=============== =================
�integer�boolean�number�datetime�duration�any�stringN) r r
r r r
r r r r )r � r �>/tmp/pip-build-5_djhm0z/pandas/pandas/io/json/_table_schema.py�as_json_table_type" s r! c C s� t j| jj� rf| jj}t|�dkr:| jjdkr:tjd� n(t|�dkrbtdd� |D ��rbtjd� | S | j � } | jj
dkr�dd� t | jj�D �}|| j_n| jjp�d| j_| S ) z?Sets index names to 'index' for regular, or 'level_x' for Multi� �indexz,Index name of 'index' is not round-trippablec s s | ]}|j d �V qdS )�level_N)�
startswith)�.0r r r r � <genexpr>X s z$set_default_names.<locals>.<genexpr>z;Index names beginning with 'level_' are not round-trippablec S s& g | ] \}}|d k r|nd|� ��qS )Nr$ r )r& �i�namer r r �
<listcomp>_ s z%set_default_names.<locals>.<listcomp>) �comZ all_not_noner# �names�lenr) �warnings�warnr �copy�nlevels� enumerate)�dataZnmsr, r r r �set_default_namesR s
r4 c C s� | j }| jd krd}n| j}|t|�d�}t|�rX|j}|j}dt|�i|d< ||d< n*t|�rn|jj |d< nt
|�r�|j j |d< |S )N�values)r) �type�enum� constraints�ordered�freq�tz)
�dtyper) r! r �
categoriesr9 �listr r: Zfreqstrr
r; �zone)Zarrr< r) �fieldZcatsr9 r r r �!convert_pandas_type_to_json_fieldh s
rA c C s� | d }|dkrdS |dkr dS |dkr,dS |dkr8d S |d
krDd S |d krl| j d
�rfd| d
� d�S dS n4|dkr�d| kr�d| kr�t| d d | d d�S dS td|� ���dS )a�
Converts a JSON field descriptor into its corresponding NumPy / pandas type
Parameters
----------
field
A JSON field descriptor
Returns
-------
dtype
Raises
------
ValueError
If the type of the provided field is unknown or currently unsupported
Examples
--------
>>> convert_json_field_to_pandas_type({'name': 'an_int',
'type': 'integer'})
'int64'
>>> convert_json_field_to_pandas_type({'name': 'a_categorical',
'type': 'any',
'constraints': {'enum': [
'a', 'b', 'c']},
'ordered': True})
'CategoricalDtype(categories=['a', 'b', 'c'], ordered=True)'
>>> convert_json_field_to_pandas_type({'name': 'a_datetime',
'type': 'datetime'})
'datetime64[ns]'
>>> convert_json_field_to_pandas_type({'name': 'a_datetime_with_tz',
'type': 'datetime',
'tz': 'US/Central'})
'datetime64[ns, US/Central]'
r6 r �objectr Zint64r Zfloat64r �boolr � timedelta64r r; zdatetime64[ns, �]zdatetime64[ns]r r8 r9 r7 )r= r9 z#Unsupported or invalid field type: N)�getr �
ValueError)r@ �typr r r �!convert_json_field_to_pandas_type� s* %
rI T)r3 r# � primary_key�versionr c C s, |dkrt | �} i }g }|r�| jjdkrrtd| j�| _xJt| jj| jj�D ]"\}}t|�}||d< |j|� qJW n|jt| j�� | j dkr�x2| j
� D ]\} }
|jt|
�� q�W n|jt| �� ||d< |o�| jj o�|dk�r| jjdkr�| jj g|d< n | jj|d< n|dk �r||d< |�r(d|d < |S )
a
Create a Table schema from ``data``.
Parameters
----------
data : Series, DataFrame
index : bool, default True
Whether to include ``data.index`` in the schema.
primary_key : bool or None, default True
Column names to designate as the primary key.
The default `None` will set `'primaryKey'` to the index
level or levels if the index is unique.
version : bool, default True
Whether to include a field `pandas_version` with the version
of pandas that generated the schema.
Returns
-------
schema : dict
Notes
-----
See `Table Schema
<https://pandas.pydata.org/docs/user_guide/io.html#table-schema>`__ for
conversion types.
Timedeltas as converted to ISO8601 duration format with
9 decimal places after the seconds field for nanosecond precision.
Categoricals are converted to the `any` dtype, and use the `enum` field
constraint to list the allowed values. The `ordered` attribute is included
in an `ordered` field.
Examples
--------
>>> df = pd.DataFrame(
... {'A': [1, 2, 3],
... 'B': ['a', 'b', 'c'],
... 'C': pd.date_range('2016-01-01', freq='d', periods=3),
... }, index=pd.Index(range(3), name='idx'))
>>> build_table_schema(df)
{'fields': [{'name': 'idx', 'type': 'integer'},
{'name': 'A', 'type': 'integer'},
{'name': 'B', 'type': 'string'},
{'name': 'C', 'type': 'datetime'}],
'pandas_version': '0.20.0',
'primaryKey': ['idx']}
Tr" r r) �fieldsN�
primaryKeyz0.20.0Zpandas_version)
r4 r# r1 r �zipZlevelsr, rA �append�ndim�itemsZ is_uniquer) ) r3 r# rJ rK �schemarL �levelr) Z new_field�column�sr r r �build_table_schema� s4 5
rV c C s� t | |d�}dd� |d d D �}t|d |d�| }dd � |d d D �}td
d � |j� D ��rjtd ��d
|j� kr~td��|j|�}d|d kr�|j|d d �}t|jj �dkr�|jj
dkr�d|j_
ndd� |jj D �|j_ |S )a
Builds a DataFrame from a given schema
Parameters
----------
json :
A JSON table schema
precise_float : boolean
Flag controlling precision when decoding string to double values, as
dictated by ``read_json``
Returns
-------
df : DataFrame
Raises
------
NotImplementedError
If the JSON table schema contains either timezone or timedelta data
Notes
-----
Because :func:`DataFrame.to_json` uses the string 'index' to denote a
name-less :class:`Index`, this function sets the name of the returned
:class:`DataFrame` to ``None`` when said string is encountered with a
normal :class:`Index`. For a :class:`MultiIndex`, the same limitation
applies to any strings beginning with 'level_'. Therefore, an
:class:`Index` name of 'index' and :class:`MultiIndex` names starting
with 'level_' are not supported.
See Also
--------
build_table_schema : Inverse function.
pandas.read_json
)�
precise_floatc S s g | ] }|d �qS )r) r )r& r@ r r r r* > s z&parse_table_schema.<locals>.<listcomp>rR rL r3 )�columnsc S s i | ]}t |�|d �qS )r) )rI )r& r@ r r r �
<dictcomp>A s z&parse_table_schema.<locals>.<dictcomp>c s s | ]}t |�jd �V qdS )zdatetime64[ns, N)�strr% )r& r r r r r' G s z%parse_table_schema.<locals>.<genexpr>z-table="orient" can not yet read timezone datarD z<table="orient" can not yet read ISO-formatted Timedelta datarM r" r# Nc S s g | ]}|j d �rdn|�qS )r$ N)r% )r& r r r r r* Y s ) �loadsr r r5 �NotImplementedErrorZastypeZ set_indexr- r# r, r) )�jsonrW �tableZ col_orderZdfZdtypesr r r �parse_table_schema s$ $
r_ )TNT),�__doc__Ztypingr r r r r r. Zpandas._libs.jsonZ_libsr] Zpandas._typingr r r Zpandas.core.dtypes.commonr
r r r
r r r r r Zpandas.core.dtypes.dtypesr Zpandasr Zpandas.core.common�core�commonr+ Zpandas.core.indexes.multir r[ rZ r! r4 rA rI rC rV r_ r r r r �<module> s( , 0B T |