U
|
¬ý°dúA ã @ sþ d Z ddlmZ ddlZddlZddlmZmZ ddlm Z m
|
Z
|
m Z m Z m Z ddlmZ ddlmZ ddlmZ dd lmZ dd
|
lmZ dd lmZ dd lmZm Z er´dd l m Z eed ed d dG dd dZ G dd de Z G dd de Z!dS )zF
|
:mod:`pandas.io.formats.xml` is a module for formatting data in XML.
|
é )Ú annotationsN)Ú TYPE_CHECKINGÚAny)ÚCompressionOptionsÚFilePathÚ
|
ReadBufferÚStorageOptionsÚ WriteBuffer©ÚAbstractMethodError)Údoc)Ú is_list_like)Úisna)Ú _shared_docs)Ú
|
get_handle)Úget_data_from_filepathÚpreprocess_data)Ú DataFrameÚstorage_optionsÚcompression_optionsÚpath_or_buffer)r r c @ sö e Zd ZdZd8dd d
|
d d d d d d d ddddddddddZddddZdddd Zddd d Zd dd d!Zddd"d#Z ddd$d%Z
|
d&dd'd(Z d)d*d*d+d,d-Z d.dd/d0d1Z d)d*dd+d2d3Zd)d*dd+d4d5Zd dd6d7ZdS )9ÚBaseXMLFormattera¼
|
Subclass for formatting data in XML.
|
|
Parameters
|
----------
|
path_or_buffer : str or file-like
|
This can be either a string of raw XML, a valid URL,
|
file or file-like object.
|
|
index : bool
|
Whether to include index in xml document.
|
|
row_name : str
|
Name for root of xml document. Default is 'data'.
|
|
root_name : str
|
Name for row elements of xml document. Default is 'row'.
|
|
na_rep : str
|
Missing data representation.
|
|
attrs_cols : list
|
List of columns to write as attributes in row element.
|
|
elem_cols : list
|
List of columns to write as children in row element.
|
|
namespaces : dict
|
The namespaces to define in XML document as dicts with key
|
being namespace and value the URI.
|
|
prefix : str
|
The prefix for each element in XML document including root.
|
|
encoding : str
|
Encoding of xml object or document.
|
|
xml_declaration : bool
|
Whether to include xml declaration at top line item in xml.
|
|
pretty_print : bool
|
Whether to write xml document with line breaks and indentation.
|
|
stylesheet : str or file-like
|
A URL, file, file-like object, or a raw string containing XSLT.
|
|
{compression_options}
|
|
.. versionchanged:: 1.4.0 Zstandard support.
|
|
{storage_options}
|
|
See also
|
--------
|
pandas.io.formats.xml.EtreeXMLFormatter
|
pandas.io.formats.xml.LxmlXMLFormatter
|
|
NTÚdataÚrowúutf-8Úinferr z7FilePath | WriteBuffer[bytes] | WriteBuffer[str] | NoneÚboolz
|
str | Nonezlist[str] | Nonez dict[str | None, str] | NoneÚstrz bool | Nonez5FilePath | ReadBuffer[str] | ReadBuffer[bytes] | Noner r ÚNone)Úframer ÚindexÚ root_nameÚrow_nameÚna_repÚ attr_colsÚ elem_colsÚ
|
namespacesÚprefixÚencodingÚxml_declarationÚ pretty_printÚ
|
stylesheetÚ compressionr Úreturnc C s || _ || _|| _|| _|| _|| _|| _|| _| | _|
|
| _ | | _
|
| | _ | | _ || _ || _|| _| j j ¡ | _| ¡ | _| ¡ | ¡ | ¡ | _| ¡ d S ©N)r r r r! r" r# r$ r% r& r' r( r) r* r+ r, r ÚcolumnsÚtolistÚ orig_colsÚprocess_dataframeÚ frame_dictsÚvalidate_columnsÚvalidate_encodingÚget_prefix_uriÚ
|
prefix_uriÚhandle_indexes)Úselfr r r r! r" r# r$ r% r& r' r( r) r* r+ r, r © r: úLd:\z\workplace\vscode\pyvenv\venv\Lib\site-packages\pandas/io/formats/xml.pyÚ__init__f s,
|
|
zBaseXMLFormatter.__init__Úbytes©r- c C s t | dS )ú
|
Build tree from data.
|
|
This method initializes the root and builds attributes and elements
|
with optional namespaces.
|
Nr
|
©r9 r: r: r; Ú
|
build_tree s zBaseXMLFormatter.build_treec C sP | j r&t| j s&tt| j j d| jrLt| jsLtt| jj ddS )zä
|
Validate elems_cols and attrs_cols.
|
|
This method will check if columns is list-like.
|
|
Raises
|
------
|
ValueError
|
* If value is not a list and less then length of nodes.
|
z" is not a valid type for attr_colsz" is not a valid type for elem_colsN)r$ r Ú TypeErrorÚtypeÚ__name__r% r@ r: r: r; r4 s ÿÿz!BaseXMLFormatter.validate_columnsc C s t | j¡ dS )z×
|
Validate encoding.
|
|
This method will check if encoding is among listed under codecs.
|
|
Raises
|
------
|
LookupError
|
* If encoding is not available in codecs.
|
N)ÚcodecsÚlookupr( r@ r: r: r; r5 ° s z"BaseXMLFormatter.validate_encodingz dict[int | str, dict[str, Any]]c C s6 | j }| jr| ¡ }| jdk r*| | j¡}|jddS )zÊ
|
Adjust Data Frame to fit xml output.
|
|
This method will adjust underlying data frame for xml output,
|
including optionally replacing missing values and including indexes.
|
Nr )Zorient)r r Z reset_indexr# ZfillnaÚto_dict)r9 Zdfr: r: r; r2 ¾ s
|
z"BaseXMLFormatter.process_dataframec s\ j s
|
dS tt j} fdd j| ¡ D } jrF| j _ jrX| j _dS )zd
|
Handle indexes.
|
|
This method will add indexes into attr_cols or elem_cols.
|
Nc s g | ]}| j kr|qS r: )r1 )Ú.0Úxr@ r: r; Ú
|
<listcomp>Û s
|
z3BaseXMLFormatter.handle_indexes.<locals>.<listcomp>)r ÚnextÚiterr3 Úkeysr$ r% )r9 Z first_keyZindexesr: r@ r; r8 Ð s
|
ÿ z BaseXMLFormatter.handle_indexesc C s t | dS )zâ
|
Get uri of namespace prefix.
|
|
This method retrieves corresponding URI to prefix in namespaces.
|
|
Raises
|
------
|
KeyError
|
*If prefix is not included in namespace dict.
|
Nr
|
r@ r: r: r; r6 å s z BaseXMLFormatter.get_prefix_uriÚdictc C sL i }| j r(| jdkr(dd | j ¡ D }| j rH| jrHdd | j ¡ D }|S )zÃ
|
Define other namespaces.
|
|
This method will build dictionary of namespaces attributes
|
for root element, conditionally with optional namespaces and
|
prefix.
|
Nc S s i | ]\}}|d krd|qS ©Ú Úxmlnsr: ©rH ÚpÚnr: r: r; Ú
|
<dictcomp>þ s z5BaseXMLFormatter.other_namespaces.<locals>.<dictcomp>c S s i | ]\}}|d krd|qS rO r: rR r: r: r; rU s )r& r' Úitems)r9 Z nmsp_dictr: r: r; Úother_namespacesó s z!BaseXMLFormatter.other_namespacesúdict[str, Any]r ©ÚdÚelem_rowr- c C sj | j s
|
|S | j D ]T}| |¡}z"t|| s>t|| |j|< W q tk
|
rb td| Y qX q|S )z´
|
Create attributes of row.
|
|
This method adds attributes using attr_cols to row element and
|
works with tuples for multindex or hierarchical columns.
|
úno valid column, )r$ Ú_get_flat_col_namer r ÚattribÚKeyError)r9 rZ r[ ÚcolÚ attr_namer: r: r; Ú build_attribs s
|
|
z BaseXMLFormatter.build_attribsz str | tuple)r` r- c C sT |}t |trFd|kr.d dd |D ¡ ¡ nd dd |D ¡ ¡ }| j | S )NrP c S s g | ] }t |qS r: ©r ©rH Úcr: r: r; rJ s z7BaseXMLFormatter._get_flat_col_name.<locals>.<listcomp>Ú_c S s g | ] }t |qS r: rc rd r: r: r; rJ s )Ú
|
isinstanceÚtupleÚjoinÚstripr7 )r9 r` Zflat_colr: r: r; r] s
|
ÿýz#BaseXMLFormatter._get_flat_col_namec C s t | dS )z¼
|
Create child elements of row.
|
|
This method adds child elements using elem_cols to row element and
|
works with tuples for multindex or hierarchical columns.
|
Nr
|
)r9 rZ r[ r: r: r; Ú build_elems# s z BaseXMLFormatter.build_elemsc C s | j s
|
d S | j D ]j}| |¡}z8t|| s8|| dkr<d n
|
t|| }||||_W q tk
|
rx td| Y qX qd S )NrP r\ )r% r] r r Útextr_ )r9 Zsub_element_clsrZ r[ r` Z elem_nameÚvalr: r: r; Ú _build_elems- s
|
|
(z BaseXMLFormatter._build_elemsc C sZ | ¡ }| jd k rFt| jd| j| jdd}|j |¡ W 5 Q R X d S | | j¡ ¡ S d S )NÚwbF)r, r Zis_text)
|
rA r r r, r ÚhandleÚwriteÚdecoder( Úrstrip)r9 Zxml_docZhandlesr: r: r; Ú write_output9 s
|
ûz BaseXMLFormatter.write_output)NTr r NNNNNr TTNr N)rD Ú
|
__module__Ú __qualname__Ú__doc__r< rA r4 r5 r2 r8 r6 rW rb r] rk rn rt r: r: r: r; r & s: > ï0,
|
|
r c @ sh e Zd ZdZddddZddddZd d
|
d d d dZddddZddddZddddZ dS )ÚEtreeXMLFormatterú
|
Class for formatting data in xml using Python standard library
|
modules: `xml.etree.ElementTree` and `xml.dom.minidom`.
|
r= r> c C sò ddl m}m}m} || j | j | ¡ d| _| j ¡ D ]\}|| j| j | j
|
}| j s| j st | ¡ | _ | ||¡ q<| ||¡}| ||¡ q<|| jd| jd| _| jr¾| ¡ | _| jrÐ| ¡ | _n
|
| ¡ | _| jd k rìtd| jS )Nr ©ÚElementÚ
|
SubElementÚtostring)r^ Úxml)Úmethodr( zBTo use stylesheet, you need lxml installed and selected as parser.)Úxml.etree.ElementTreer{ r| r} r7 r! rW Úrootr3 Úvaluesr" r$ r% ÚlistrM rk rb r( Úout_xmlr* Ú prettify_treer) Úadd_declarationÚremove_declarationr+ Ú
|
ValueError©r9 r{ r| r} rZ r[ r: r: r; rA Q s. ÿ
|
|
|
ÿz EtreeXMLFormatter.build_treer c C s¤ ddl m} d}| jr | j ¡ D ]&\}}t|tr t|tr ||| q | jrzd| j| j d}W q tk
|
r t| j dY q X nd| jd d}|S )Nr )Úregister_namespacerP Ú{Ú}ú is not included in namespaces)r r r& rV rg r r' r_ )r9 r ÚurirS rT r: r: r; r6 x s z EtreeXMLFormatter.get_prefix_urirX r r rY c C s ddl m} | |||¡ d S ©Nr )r| )r r| rn ©r9 rZ r[ r| r: r: r; rk s z EtreeXMLFormatter.build_elemsc C s& ddl m} || j}|jd| jdS )z
|
Output tree for pretty print format.
|
|
This method will pretty print xml with line breaks and indentation.
|
r )Ú parseStringz )Úindentr( )Zxml.dom.minidomr r Z toprettyxmlr( )r9 r Údomr: r: r; r
s
|
z EtreeXMLFormatter.prettify_treec C s2 d| j d}| j d¡r | jS | | j ¡| j S )z¶
|
Add xml declaration.
|
|
This method will add xml declaration of working tree. Currently,
|
xml_declaration is supported in etree starting in Python 3.8.
|
z <?xml version="1.0" encoding="z"?>
|
s <?xml)r( r Ú
|
startswithÚencode)r9 Údeclr: r: r; r s
|
|
ÿýz!EtreeXMLFormatter.add_declarationc C s | j d¡d ¡ S )z¦
|
Remove xml declaration.
|
|
This method will remove xml declaration of working tree. Currently,
|
pretty_print is not supported in etree.
|
s ?>éÿÿÿÿ)r Úsplitrj r@ r: r: r; r « s z$EtreeXMLFormatter.remove_declarationN)
|
rD ru rv rw rA r6 rk r
r r r: r: r: r; rx K s ' rx c sp e Zd ZdZdd fdd ZddddZddd d
|
Zd dd d ZddddddZddddZ Z
|
S )ÚLxmlXMLFormatterry r r> c s t j|| | ¡ d S r. )Úsuperr< Úconvert_empty_str_key)r9 ÚargsÚkwargs©Ú __class__r: r; r< ¼ s zLxmlXMLFormatter.__init__r= c C sÎ ddl m}m}m} || j | j | jd| _| j ¡ D ]\}|| j| j | j
|
}| j s~| j s~t | ¡ | _ | ||¡ q:| ||¡}| ||¡ q:|| j| jd| j| jd| _| jdk rÈ| ¡ | _| jS )r? r rz )Znsmapr~ )r* r r( r) N)Ú
|
lxml.etreer{ r| r} r7 r! r& r r3 r r" r$ r% r rM rk rb r* r( r) r r+ Ú transform_docr r: r: r; rA Á s& û
|
|
zLxmlXMLFormatter.build_treec C s, | j r(d| j ¡ kr(| j dd¡| j d< dS )z¿
|
Replace zero-length string in `namespaces`.
|
|
This method will replace '' with None to align to `lxml`
|
requirement that empty string prefixes are not allowed.
|
rP ÚdefaultN)r& rM Úpopr@ r: r: r; r è s z&LxmlXMLFormatter.convert_empty_str_keyr c C sf d}| j rb| jrPzd| j | j d}W qb tk
|
rL t| j dY qbX nd| j d d}|S )NrP r r r )r& r' r_ )r9 r r: r: r; r6 ó s z LxmlXMLFormatter.get_prefix_urirX r rY c C s ddl m} | |||¡ d S r )r r| rn r r: r: r; rk s z LxmlXMLFormatter.build_elemsc C s¬ ddl m}m}m}m} | j}|dk s*tt|| j| j | j
|
d}t |D}|| jd}t |t jr||| ¡ | j¡|d} n |||d} W 5 Q R X || }
|
|
|
| j} t| S )zõ
|
Parse stylesheet from file or buffer and run it.
|
|
This method will parse stylesheet object into tree for parsing
|
conditionally by its specific object type, then transforms
|
original tree with XSLT script.
|
r )ÚXSLTÚ XMLParserÚ
|
fromstringÚparseN)Zfilepath_or_bufferr( r, r )r( )Úparser)r r¤ r¥ r¦ r§ r+ ÚAssertionErrorr r( r, r r rg ÚioÚStringIOÚgetvaluer r r= ) r9 r¤ r¥ r¦ r§ Z style_docÚ handle_dataÚxml_dataZ curr_parserZxsl_docZ transformerZnew_docr: r: r; r¡ s( ü
|
ÿ
|
z LxmlXMLFormatter.transform_doc) rD ru rv rw r< rA r r6 rk r¡ Ú __classcell__r: r: r r; r ¶ s ' r )"rw Ú
|
__future__r rE rª Útypingr r Zpandas._typingr r r r r Z pandas.errorsr Zpandas.util._decoratorsr Zpandas.core.dtypes.commonr Zpandas.core.dtypes.missingr Zpandas.core.shared_docsr Zpandas.io.commonr Z pandas.io.xmlr r Zpandasr r rx r r: r: r: r; Ú<module> s.
|
þ #k
|
