x\c@sdZddlZddlmZmZddlZddlZddlZddlZddl j Z ddl m Z m Z m Z mZddlmZmZmZmZmZddlmZmZddlmZeZdd d d d d dddddddddddhZe jrddlmZmZeZ ddl!m"Z#ddl!m$Z$m%Z%m&Z&m'Z'm(Z(ddl)m*Z*ddl+m,Z,nddl-mZ ddl.m'Z'mZddl"m"Z#dd l"m$Z$m%Z%m&Z&m(Z(ddl-m*Z*ddl/m,Z,dd!lmZmZdd"l0m1Z1ed#Ze2e$e%e&Z3e3j4dd$e5fd%YZ6e jsd&e6_7nd'Z8d(Z9d)Z:d*Z;d+Z<d,Z=dddd-Z?d.Z@id/d06d1d26d3d46d5d66ZAd7ZBddeCeDd8ZEd9ejFe fd:YZGd;e6fd<YZHe jsmd=eH_7nd>e6fd?YZIe jrejJd@dAZKejJd@dBZLn,dCe6fdDYZKdEe5fdFYZLdS(GsCommon IO api utilitiesiN(tclosingtcontextmanager(tBytesIOtStringIOt string_typest text_type(tAbstractMethodErrort DtypeWarningtEmptyDataErrort ParserErrort ParserWarning(t is_file_liket is_number(t pprint_things-1.#INDs1.#QNANs1.#INDs-1.#QNANs#N/A N/As#N/AsN/Asn/atNAs#NAtNULLtnulltNaNs-NaNtnans-nant(turlopent pathname2url(turlparse(t uses_relativet uses_netloct uses_paramst urlencodeturljoin(tURLError(t HTTPException(R(RR(RRRR(RR(twrapscos'tt|| }|VWdQXdS(N(Rt_urlopen(targstkwargstf((s/lib/python2.7/site-packages/pandas/io/common.pyR2st BaseIteratorcBs eZdZdZdZRS(sSubclass this and provide a "__next__()" method to obtain an iterator. Useful only when the object being iterated is non-reusable (e.g. OK for a parser, not for an in-memory table, yes for its iterator).cCs|S(N((tself((s/lib/python2.7/site-packages/pandas/io/common.pyt__iter__AscCst|dS(N(R(R$((s/lib/python2.7/site-packages/pandas/io/common.pyt__next__Ds(t__name__t __module__t__doc__R%R&(((s/lib/python2.7/site-packages/pandas/io/common.pyR#<s cCs |jS(N(R&(R$((s/lib/python2.7/site-packages/pandas/io/common.pytIscCs0yt|jtkSWntk r+tSXdS(sCheck 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. N(t parse_urltschemet _VALID_URLSt ExceptiontFalse(turl((s/lib/python2.7/site-packages/pandas/io/common.pyt_is_urlLs  cCs#t|trtjj|S|S(s[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 (t isinstanceRtostpatht expanduser(tfilepath_or_buffer((s/lib/python2.7/site-packages/pandas/io/common.pyt _expand_user^s cCs"t|trtdndS(NsPassing 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(R2tboolt TypeError(theader((s/lib/python2.7/site-packages/pandas/io/common.pyt_validate_header_argpscCsyddl}t}Wntk r/t}nXyddlm}t}Wntk rct}nXt|dr}|jS|rt||j rt |S|rt||r|j St |S(s_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. For backwards compatibility with older pythons, pathlib.Path and py.path objects are specially coerced. Any other object is passed through unchanged, which includes bytes, strings, buffers, or anything else that's not even path-like. iN(tlocalt __fspath__( tpathlibtTruet ImportErrorR/tpy.pathR<thasattrR=R2tPathRtstrpathR7(R6R>t_PATHLIB_INSTALLEDt LocalPatht_PY_PATH_INSTALLED((s/lib/python2.7/site-packages/pandas/io/common.pyt_stringify_pathxs"         cCs0yt|jdkSWntk r+tSXdS(s Check for an s3, s3n, or s3a urlts3ts3nts3aN(RIRJRK(R+R,R.R/(R0((s/lib/python2.7/site-packages/pandas/io/common.pyt is_s3_urls cCs0yt|jdkSWntk r+tSXdS(sCheck for a gcs urltgcstgsN(RMRN(R+R,R.R/(R0((s/lib/python2.7/site-packages/pandas/io/common.pyt is_gcs_urls c Csjt|}t|rzt|}|jjdd }|dkrNd}nt|j}|j|||t fSt |rddl m }|j |d|d|d|St|rddl m}|j |d|d|d|St|tjtjtjfr't|d |tfSt|sZd } t| jd t|n|d |tfS( s 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 encoding : the encoding to use to decode py3 bytes, default is 'utf-8' mode : str, optional Returns ------- tuple of ({a filepath_ or buffer or S3File instance}, encoding, str, compression, str, should_close, bool) sContent-Encodingtgzipi(RItencodingt compressiontmode(RMs0Invalid file path or buffer object type: {_type}t_typeN(RHR1RtheaderstgettNoneRtreadtcloseR?RLt pandas.ioRItget_filepath_or_bufferRORMR2tcompatRt binary_typetmmapR7R/R t ValueErrortformatttype( R6RQRRRStreqtcontent_encodingtreaderRIRMtmsg((s/lib/python2.7/site-packages/pandas/io/common.pyR[s:            !cCstdt|S(s converts an absolute native path to a FILE URL. Parameters ---------- path : a path in native format Returns ------- a valid FILE URL sfile:(RR(R4((s/lib/python2.7/site-packages/pandas/io/common.pytfile_path_to_urls s.gzRPs.bz2tbz2s.ziptzips.xztxzcCs|dkrdS|dkrrt|}t|tjs>dSx-tjD]\}}|j|rK|SqKWdS|tkr|Sdj|}ddgt t}|dj|7}t |dS(s 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 : a path (str) or buffer compression : {'infer', 'gzip', 'bz2', 'zip', 'xz', None} If 'infer' and `filepath_or_buffer` is path-like, then detect compression from the following extensions: '.gz', '.bz2', '.zip', or '.xz' (otherwise no compression). Returns ------- string or None : compression method Raises ------ ValueError on invalid compression specified tinfers!Unrecognized compression type: {}s Valid compression types are {}N( RWRHR2R\Rt_compression_to_extensiontitemstendswithR`tsortedR_(R6RRt extensionRetvalid((s/lib/python2.7/site-packages/pandas/io/common.pyt_infer_compressions     cCsy ddlm}t|f}Wntk r<tf}nXt}|} t|}t|tj} | rt ||}n|rtj r| r|rd} t | n|dkrddl } | r| j ||} q| jd|} n|dkrqddl} | r.| j||} qtj r_t| j|j} |jq| j|} n|dkr;t||}|j||jd kr|} q|jd kr|j}t|d kr|j |j} q8t|d kr t d j|q8t dj|qnH|dkrhtj}|j||} ndj|} t | |j| n| r3tj r|d krdn|}t ||} nW|rt ||d|dd} n3|rt ||dddd} nt ||} |j| ntjr|r|sWt| |rddlm }|| d|dd} |j| n|rt!| dry t"| }| j|} Wqt#k rqXn| |fS(s Get file handle for given path/buffer and mode. Parameters ---------- path_or_buf : a path (str) or buffer mode : str mode to open path_or_buf with encoding : str or None compression : {'infer', 'gzip', 'bz2', 'zip', 'xz', None}, default None If 'infer' and `filepath_or_buffer` is path-like, then detect compression from the following extensions: '.gz', '.bz2', '.zip', or '.xz' (otherwise no compression). memory_map : boolean, default False See parsers._parser_params for more information. is_text : boolean, default True whether file/buffer is in text format (csv, json, etc.), or in binary mode (pickle, etc.) Returns ------- f : file-like A file-like object handles : list of file-like objects A list of file-like object that were opened in this function. i(tS3Files:compression with encoding is not yet supported in Python 2RPNtfileobjRgRhtwtriisZero files found in ZIP file {}s;Multiple files found in ZIP file. Only one file per ZIP: {}Ris!Unrecognized compression type: {}twbRQtnewlineRterrorstreplace(t TextIOWrappertfileno($ts3fsRrRR@tlistRHR2R\RRqtPY2R_RPtopentGzipFileRgtBZ2FileRt decompressRXRYt BytesZipFiletappendRStnamelisttlentpopR`t import_lzmatLZMAFiletPY3tioRzRBt MMapWrapperR.(t path_or_bufRSRQRRt memory_maptis_textRrtneed_text_wrappingthandlesR"tis_pathReRPRgtzft zip_namestlzmaRztg((s/lib/python2.7/site-packages/pandas/io/common.pyt _get_handle:s                        RcBs5eZdZejdZdZedZRS(s  Wrapper for standard library class ZipFile and allow the returned file-like handle to accept byte strings via `write` method. BytesIO provides attributes of file-like object and ZipFile.writestr writes bytes strings into a member of the archive. cKsD|dkr!|jdd}ntt|j||||dS(NRvtrbtbR(RvR(RytsuperRt__init__(R$tfileRSRRR!((s/lib/python2.7/site-packages/pandas/io/common.pyRs cCs tt|j|j|dS(N(RRtwritestrtfilename(R$tdata((s/lib/python2.7/site-packages/pandas/io/common.pytwritescCs |jdkS(N(tfpRW(R$((s/lib/python2.7/site-packages/pandas/io/common.pytcloseds( R'R(R)tzipfilet ZIP_DEFLATEDRRtpropertyR(((s/lib/python2.7/site-packages/pandas/io/common.pyRs RcBs2eZdZdZdZdZdZRS(s Wrapper for the Python's mmap class so that it can be properly read in by Python's csv.reader class. Parameters ---------- f : file object File object to be mapped onto memory. Must support the 'fileno' method or have an equivalent attribute cCs(tj|jddtj|_dS(Nitaccess(R^R{t ACCESS_READ(R$R"((s/lib/python2.7/site-packages/pandas/io/common.pyRscCst|j|S(N(tgetattrR^(R$tname((s/lib/python2.7/site-packages/pandas/io/common.pyt __getattr__scCs|S(N((R$((s/lib/python2.7/site-packages/pandas/io/common.pyR%scCsC|jj}tjr*tj|}n|dkr?tn|S(NR(R^treadlineR\Rt bytes_to_strt StopIteration(R$Rw((s/lib/python2.7/site-packages/pandas/io/common.pyR&s    (R'R(R)RRR%R&(((s/lib/python2.7/site-packages/pandas/io/common.pyRs     cCs |jS(N(R&(R$((s/lib/python2.7/site-packages/pandas/io/common.pyR*st UTF8RecodercBs5eZdZdZddZdZdZRS(sP Iterator that reads an encoded stream and reencodes the input to UTF-8 cCstj|||_dS(N(tcodecst getreaderRd(R$R"RQ((s/lib/python2.7/site-packages/pandas/io/common.pyRsicCs|jj|jdS(Nsutf-8(RdRXtencode(R$tbytes((s/lib/python2.7/site-packages/pandas/io/common.pyRXscCs|jjjdS(Nsutf-8(RdRR(R$((s/lib/python2.7/site-packages/pandas/io/common.pyRscCst|jjdS(Nsutf-8(tnextRdR(R$((s/lib/python2.7/site-packages/pandas/io/common.pyRs(R'R(R)RRXRR(((s/lib/python2.7/site-packages/pandas/io/common.pyRs    sutf-8cKstj|d||S(Ntdialect(tcsvRd(R"RRQtkwds((s/lib/python2.7/site-packages/pandas/io/common.pyt UnicodeReaderscKstj|d||S(NR(Rtwriter(R"RRQR((s/lib/python2.7/site-packages/pandas/io/common.pyt UnicodeWriter sRcBs)eZdZejddZdZRS(s A CSV reader which will iterate over lines in the CSV file "f", which is encoded in the given encoding. On Python 3, this is replaced (below) by csv.reader, which handles unicode. sutf-8cKs.t||}tj|d|||_dS(NR(RRRd(R$R"RRQR((s/lib/python2.7/site-packages/pandas/io/common.pyR-scCs2t|j}g|D]}tj|d^qS(Nsutf-8(RRdR\R(R$trowts((s/lib/python2.7/site-packages/pandas/io/common.pyR&1s(R'R(R)RtexcelRR&(((s/lib/python2.7/site-packages/pandas/io/common.pyR#sRcBs2eZdZejddZdZdZRS(su A CSV writer which will write rows to CSV file "f", which is encoded in the given encoding. sutf-8cKsat|_tj|jd|||_||_tj||_|jdd|_ dS(NRtquoting( RtqueueRRtstreamRtgetincrementalencodertencoderRVRWR(R$R"RRQR((s/lib/python2.7/site-packages/pandas/io/common.pyR<s   csfd}g|D]-}||r.|nt|jd^q}jjg|D] }|^qYjj}|jd}jj|}jj |jj ddS(Ncs+jtjkrt|p*t|tS(N(RRtQUOTE_NONNUMERICR R2tstr(tx(R$(s/lib/python2.7/site-packages/pandas/io/common.pyt _check_as_isEssutf-8i( R RRtwriterowRtgetvaluetdecodeRRRttruncate(R$RRRRR((R$s/lib/python2.7/site-packages/pandas/io/common.pyRDs7#csfd}xXt|D]J\}}g|D]-}||rG|nt|jd^q/||s~     "(  ("      +  9   6'