U sVc @sdZddlmZddlmZmZddlZddlZddlZddl Z ddl Z ddl m Z m Z mZmZmZmZddlZddlZddlmZddlZddlZddlmZmZmZmZmZmZmZm Z m!Z!m"Z"ddl#m$Z$m%Z&m'Z'm(Z(m)Z)ddl*Z*ddl+Z+dd l,m-Z-m.Z.m/Z/m0Z0m1Z1m2Z2m3Z3dd l4m5Z5dd l6m7Z7dd l8m9Z9dd l:m;Z;ddlZ>m?Z?m@Z@ddlAmBZBeCe)e'e(ZDeDEdeFdZGe de-dZHejIGdddZJejIGdddeeZKdddddZLe"ddddd ZMe"dddd!d ZMd"d"dd#d ZMdd$d%d&d'ZNe"d~d)ddd*d+d,ZOe"ddddd*d-d,ZOdd/dd"d*d0d,ZOd1d2ZPd3ddd4d5ZQe9eBd6eBd7d8d9dd3ddd?d@ZRdddAdBdCZSdDdDdDdDdEdFdGdHdIdJ ZTeCeTUZVdeZdZUdZded<ded<ded<ded<d Zd ed <d S) IOArgsz? Return value of io/common.py:_get_filepath_or_buffer. str | BaseBufferfilepath_or_bufferstrencodingmoder compressionFbool should_closeN)__name__ __module__ __qualname____doc____annotations__r6r<r<4/tmp/pip-unpacked-wheel-xj8nt62q/pandas/io/common.pyr.Ns r.c@sneZdZUdZded<ded<ejedZded<d Z d ed <d d ddZ dd ddZ dd dddZ dS) IOHandlesau Return value of io/common.py:get_handle Can be used as a context manager. This is used to easily close created buffers and to handle corner cases when TextIOWrapper is inserted. handle: The file handle to be used. created_handles: All file handles that are created by get_handle is_wrapped: Whether a TextIOWrapper needs to be detached. z IO[AnyStr]handlerr4)default_factoryzlist[IO[bytes] | IO[str]]created_handlesFr5 is_wrappedNonereturncCs\|jr8t|jtst|j|j|j|j|jD] }| q>g|_d|_dS)z Close all created buffers. Note: If a TextIOWrapper was inserted, it is flushed and detached to avoid closing the potentially user-created buffer. FN) rB isinstancer?r AssertionErrorflushdetachrAremoveclose)selfr?r<r<r=rKps    zIOHandles.closezIOHandles[AnyStr]cCs|SNr<rLr<r<r= __enter__szIOHandles.__enter__r )argsrEcGs |dSrM)rK)rLrPr<r<r=__exit__szIOHandles.__exit__N) r7r8r9r:r; dataclassesfieldlistrArBrKrOrQr<r<r<r=r>[s  r>objectr5)urlrEcCst|tsdSt|jtkS)z Check to see if a URL has a valid protocol. Parameters ---------- url : str or unicode Returns ------- isurl : bool If `url` has a valid protocol return True otherwise False. F)rFr1 parse_urlscheme _VALID_URLSrVr<r<r=is_urls r[r1)r0rEcCsdSrMr<r0r<r<r= _expand_usersr]cCsdSrMr<r\r<r<r=r]szstr | BaseBufferTcCst|trtj|S|S)a] Return the argument with an initial component of ~ or ~user replaced by that user's home directory. Parameters ---------- filepath_or_buffer : object to be converted if possible Returns ------- expanded_filepath_or_buffer : an expanded filepath or the input if not expandable )rFr1ospath expanduserr\r<r<r=r]s  rC)headerrEcCs|dkr dSt|r2tt|}|dkr.tddSt|ddr|tt|}ttt|s^tdtdd|DrxtddSt |rt d tddS) NrzUPassing negative integer to header is invalid. For no header, use header=None insteadF)Z allow_setsz*header must be integer or list of integerscss|]}|dkVqdS)rNr<).0ir<r<r= sz&validate_header_arg..z8cannot specify multi-index header with negative integerszPassing a bool to header is invalid. Use header=None for no header or header=int or list-like of ints to specify the row(s) making up the column names) r(rint ValueErrorr)rallmapanyr& TypeError)rar<r<r=validate_header_args*   rk.r)r0convert_file_likerEcCsdSrMr<r0rlr<r<r=stringify_pathsrncCsdSrMr<rmr<r<r=rnsFzFilePath | BaseBufferTcCs2|st|rtt|St|tjr*|}t|S)a Attempt to convert a path-like object to a string. Parameters ---------- filepath_or_buffer : object to be converted Returns ------- str_filepath_or_buffer : maybe a string version of the object Notes ----- Objects supporting the fspath protocol (python 3.6+) are coerced according to its __fspath__ method. Any other object is passed through unchanged, which includes bytes, strings, buffers, or anything else that's not even path-like. )r'rr,rFr^PathLike __fspath__r]rmr<r<r=rns    cOsddl}|jj||S)z` Lazy-import wrapper for stdlib urlopen, as that imports a big chunk of the stdlib. rN)urllib.requestrequesturlopen)rPkwargsurllibr<r<r=rssrszFilePath | BaseBuffercCs$t|to"tt|o"|d S)zR Returns true if the given URL looks like something fsspec can handle )zhttp://zhttps://)rFr1r5_RFC_3986_PATTERNmatch startswithrZr<r<r= is_fsspec_url s    rystorage_optionscompression_optionsr0)rzr{utf-8rrr )r0r2r4r3rzrEc Cst|}t|\}}t||}|rJt|drJd|krJtjdttdd}t||d}d|kr|dkr|d krtj|d |t td|}d |krd|kr|d7}t |t r.t |r.|pi}d dl }|jj||d }t|2} | jdd} | dkrddi}t| } W5QRXt| ||d|dSt|rXt |t sHt|dr`|dd}|drx|dd}td} g} z&tdd dlm}m}||tg} Wntk rYnXz$| j |fd|i|pi }Wn^t!| k rD|dkrddi}nt|}d|d<| j |fd|i|p6i }YnXt|||d|dS|rft"dt |t t#t$j$frtt%|||d|dSt|dst|dsdt&|}t"|t|||d|dS)a% If the filepath_or_buffer is a url, translate and return the buffer. Otherwise passthrough. Parameters ---------- filepath_or_buffer : a url, filepath (str, py.path.local or pathlib.Path), or buffer {compression_options} .. versionchanged:: 1.4.0 Zstandard support. encoding : the encoding to use to decode bytes, default is 'utf-8' mode : str, optional {storage_options} .. versionadded:: 1.2.0 ..versionchange:: 1.2.0 Returns the dataclass IOArgs. writebzDcompression has no effect when passing a non-binary object as input.) stacklevelN)methodw)bz2xz)zutf-16zutf-32z( will not write the byte order mark for tr)headerszContent-EncodinggziprT)r0r2r4r6r3zs3a://zs3://zs3n://fsspecZbotocore) ClientErrorNoCredentialsErrorr3Zanonz?storage_options passed with file object or non-fsspec file pathFreadz)Invalid file path or buffer object type: )'rnget_compression_methodinfer_compressionhasattrwarningswarnRuntimeWarningr%dictUnicodeWarningrFr1r[rqrrRequestrsrgetrrr.ryrGrxreplacer#Zbotocore.exceptionsrrPermissionError ImportErroropentuplerfbytesmmapr]type)r0r2r4r3rzcompression_methodZ fsspec_moderuZreq_inforeqcontent_encodingreaderrZerr_types_to_retry_with_anonrrZfile_objmsgr<r<r=_get_filepath_or_buffers"                r)r_rEcCsddlm}td||S)z converts an absolute native path to a FILE URL. Parameters ---------- path : a path in native format Returns ------- a valid FILE URL r) pathname2urlzfile:)rqrr)r_rr<r<r=file_path_to_urls rtarrrziprzstd) .tar.tar.gz.tar.bz2.tar.xz.gz.bz2.zip.xzz.zstz"tuple[str | None, CompressionDict])r4rEc Cs`t|trPt|}z|d}WqXtk rL}ztd|W5d}~XYqXXni}|}||fS)a Simplifies a compression argument to a compression method string and a mapping containing additional arguments. Parameters ---------- compression : str or mapping If string, specifies the compression method. If mapping, value at key 'method' specifies compression method. Returns ------- tuple of ({compression method}, Optional[str] {compression arguments}, Dict[str, Any]) Raises ------ ValueError on mapping missing 'method' key rz.If mapping, compression must have key 'method'N)rFrrpopKeyErrorrf)r4compression_argsrerrr<r<r=rs r)r{ str | None)r0r4rEcCs|dkr dS|dkrZt|dd}t|ts.dStD]\}}||r6|Sq6dS|tkrf|Sddgtt}d|d|}t |dS)a/ Get the compression method for filepath_or_buffer. If compression='infer', the inferred compression method is returned. Otherwise, the input compression method is returned unchanged, unless it's invalid, in which case an error is raised. Parameters ---------- filepath_or_buffer : str or file handle File path or object. {compression_options} .. versionchanged:: 1.4.0 Zstandard support. Returns ------- string or None Raises ------ ValueError on invalid compression specified. NZinferT)rlUnrecognized compression type: z Valid compression types are ) rnrFr1_extension_to_compressionitemslowerendswith_supported_compressionssortedrf)r0r4 extensionZvalidrr<r<r=rs    rz Path | strcCs&t|j}|s"td|ddS)z Check if parent directory of a file exists, raise OSError if it does not Parameters ---------- path: Path or str Path to check parent directory of z1Cannot save file into a non-existent directory: ''N)r parentis_dirOSError)r_rr<r<r=check_parent_directoryJs r)r2r4 memory_maperrorsrzzLiteral[False]zIOHandles[bytes]) path_or_bufr3r2r4ris_textrrzrEcCsdSrMr<rr3r2r4rrrrzr<r<r= get_handleXs r)r2r4rrrrzz Literal[True]zIOHandles[str]cCsdSrMr<rr<r<r=rgs z!IOHandles[str] | IOHandles[bytes]cCsdSrMr<rr<r<r=rvs rTcCs|pd}|pd}t||r*d|kr*|d7}t|t|trHt|t|||||d}|j} t| |\} }} t| t} t |j } | d}d|kr| rt t| |rR|dkr|j dd |_ n |dkrd|j kr|j d7_ |d kr2t| trtjf| |j d | } ntjf| |j d | } n|d krXtj| fd|j i| } n|dkrt| |j f| } | jj dkr6| | | j} t| dkr| j| } n$| std|ntd| nZ|dkr| d|j t| trtfd| i| } ntfd| i| } t| ts6td| jj kr6| | | j}t|dkr| j|d}|dk st|} n$|std|ntd|n|dkrt| |j } nl|dkr$t d}d|j krd|j!f| i}nd|j"f| i}|j| fd|j i|} nd|}t|t| trFt| | nNt| tr|j#rd|j krt| |j |j#|d d} n t| |j } | | d }|s|j d!krt| t$rt%| |j#d"} n~|rP|s|st| |j rPt&| d#rt&| d$rt&| d%st'| } t(| |j#|d d} | | t|jtpL|j) }d|j kr~t&| d&s~t*d't+|jd(| ,|j)rt|jtrt| |jt-| | ||j d)S)*a Get file handle for given path/buffer and mode. Parameters ---------- path_or_buf : str or file handle File path or object. mode : str Mode to open path_or_buf with. encoding : str or None Encoding to use. {compression_options} .. versionchanged:: 1.0.0 May now be a dict with key 'method' as compression mode and other keys as compression options if compression mode is 'zip'. .. versionchanged:: 1.1.0 Passing compression options as keys in dict is now supported for compression modes 'gzip', 'bz2', 'zstd' and 'zip'. .. versionchanged:: 1.4.0 Zstandard support. memory_map : bool, default False See parsers._parser_params for more information. Only used by read_csv. is_text : bool, default True Whether the type of the content passed to the file/buffer is string or bytes. This is not the same as `"b" not in mode`. If a string content is passed to a binary file/buffer, a wrapper is inserted. errors : str, default 'strict' Specifies how encoding and decoding errors are to be handled. See the errors argument for :func:`open` for a full list of options. storage_options: StorageOptions = None Passed to _get_filepath_or_buffer .. versionchanged:: 1.2.0 Returns the dataclass IOHandles r|strictr)r2r4r3rzrr}rrr+r)filenamer3)fileobjr3rr3rzZero files found in ZIP file z9Multiple files found in ZIP file. Only one file per ZIP: rnamerrNz Zero files found in TAR archive zDMultiple files found in TAR archive. Only one file per TAR archive: r zstandardZdctxZcctxr)r2rnewlineFrb)r2readablewritableseekablerz1Expected file path name or file-like object, got z type)r?rArBr4)._is_binary_modecodecslookuprFr1 lookup_errorrr0_maybe_memory_maprr4rrr3rrGzipFilerBZ2File _BytesZipFilebufferappendnamelistlenrrf setdefault _BytesTarFilerGgetnames extractfiler"r#ZstdDecompressorZZstdCompressorr2r _BytesIOWrapperr _IOWrapperr r6rjrreverser>)rr3r2r4rrrrzZioargsr?handlesZis_pathrZ zip_namesfilesfilerZ open_argsrrBr<r<r=rs>6                              cs8eZdZdZeddddZddfdd ZZS)_BufferedWriterz Some objects do not support multiple .write() calls (TarFile and ZipFile). This wrapper writes to the underlying buffer on close. rCrDcCsdSrMr<rNr<r<r=write_to_buffersz_BufferedWriter.write_to_bufferc sP|jr dS|r8|d|j|W5QRXn |jtdS)Nr)closedgetvalueseekrrrKsuperrN __class__r<r=rKs  z_BufferedWriter.close)r7r8r9r:rrrK __classcell__r<r<rr=rsrcsXeZdZdddddddfdd Zd d d d d ZddddZddddZZS)rNr}rzLiteral[('r', 'a', 'w', 'x')]z-ReadBuffer[bytes] | WriteBuffer[bytes] | NonerC)rr3r archive_namerEc s<t||_||_tjjf||||d||_dS)N)rr3r) r__init__rrtarfileTarFiler extend_moder)rLrr3rrrtrr<r=rs z_BytesTarFile.__init__r1)r3rEcCsP|dd}|dkr|S|jdk rLt|jj}|dkrL|d|dd}|S)Nrr+r)rrr:r)rrr suffix)rLr3rr<r<r=rs   z_BytesTarFile.extend_moderDcCsP|jdkrdSt|j}|jdkr.|djS|jdkrJ|ddjS|jS)z If an explicit archive_name is not given, we still want the file inside the zip file not to be named something.tar, because that causes confusion (GH39465). Nrr+)rrr)rr r with_suffixrLrr<r<r=infer_filenames     z_BytesTarFile.infer_filenamecCs>|jp|pd}tj|d}t||_|j||dS)Nrr) rrrTarInforrsizeraddfile)rLrtarinfor<r<r=rs z_BytesTarFile.write_to_buffer)Nr}NN)r7r8r9rrrrrr<r<rr=rs rcsFeZdZddddddfdd Zdd d d Zdd d d ZZS)rNz1FilePath | ReadBuffer[bytes] | WriteBuffer[bytes]r1rrC)rr3rrEc s@t|dd}||_|dtjtj||f||_dS)Nrr+r4) rrrrrzipfile ZIP_DEFLATEDZipFiler)rLrr3rrtrr<r=rs   z_BytesZipFile.__init__rDcCs@t|jjtjtfrr[r]rkrnrsryrrrvaluesrrrrrrrrrrrrr lru_cacherr<r<r<r=s   0 $         ,"   3 #8$$$$7$