uj6-x UdZddlmZddlZddlmZmZddlmZddl Z ddl Z ddl Z ddl Z ddl Z ddlZddlZddlZddlmZmZmZmZmZmZmZmZmZmZddlZddlZddlm Z ddl!m"Z"dd l#m$Z$dd l%m&Z&dd l'm(Z(m)Z)m*Z*m+Z+m,Z,dd l-m.Z.dd l/m0Z0ddl1m2Z2m3Z3ddl4m5Z5m6Z6m7Z7m8Z8ddl9m:Z:ddl;mZ>ddl?m@Z@mAZAddlBmCZCddlDmEZEmFZFddlGmHZHddlImJZJddlKmLZLmMZMmNZNmOZOmPZPddlQmRZRddlSmTZTmUZUmVZVddlWmXZXmYZYddlZm[Z[m\Z\er>ddl]m^Z^m_Z_m`Z`maZambZbddlcmdZdddlemfZfmgZgmhZhmiZimjZjmkZkmlZlmmZmmnZnmoZoGd d!eeekd"#ZpneqZpGd$d%eZrd&d&d"dd'Zsd%etd(<Gd)d*eZud+d,dd-Zvd*etd.<d/hZwd0d1hZxeyhd2Zzdqd3Z{drd4Z|d5Z}d6Z~d7Zd8Zd9Zd:Zhd;Zedsdtd=Zedsdud>Zedsdvd?Z dw dxd@ZdydAZ dzdBZd{dCZd|dDZ d}dEZd~dFZ dw ddGZddHZ ddIZ ddJZed<d?ed@<yA)B _read_sharedstr | lib.NoDefault | Nonesep delimiter-int | Sequence[int] | Literal['infer'] | Noneheader)Sequence[Hashable] | lib.NoDefault | Nonenames"IndexLabel | Literal[False] | None index_colrHusecolsDtypeArg | NonedtypeCSVEngine | Noneengine#Mapping[HashableT, Callable] | None converters list | None true_values false_valuesboolskipinitialspace3list[int] | int | Callable[[Hashable], bool] | Noneskiprowsint skipfooter int | NonenrowsLHashable | Iterable[Hashable] | Mapping[Hashable, Iterable[Hashable]] | None na_valueskeep_default_na na_filterskip_blank_lines bool | Sequence[Hashable] | None parse_dates str | dict[Hashable, str] | None date_formatdayfirst cache_datesr? compression str | None thousandsstrdecimallineterminator quotecharquoting doublequote escapecharcommentencodingencoding_errors,str | csv.Dialect | type[csv.Dialect] | Nonedialect on_bad_lines low_memory memory_map.Literal['high', 'legacy', 'round_trip'] | Nonefloat_precisionStorageOptions | Nonestorage_optionsDtypeBackend | lib.NoDefault dtype_backendN__name__ __module__ __qualname____annotations__d/var/www/html/python-recuperauto-other/env/lib/python3.12/site-packages/pandas/io/parsers/readers.pyrJrJs('--==8855  77  !!EE X 5555'' "" ##==GG..33rrJF)totalc6eZdZUded<ded<ded<ded<y) _C_Parser_Defaults Literal[True]rirLiteral[False]rNonerNrrrrrrsrrT)rirrr_c_parser_defaultsc,eZdZUded<ded<ded<y) _Fwf_DefaultszLiteral['infer']colspecsz Literal[100] infer_nrowsrwidthsNrrrrrrs Lrrinferd)rrr _fwf_defaultsrcrr> roiteratorrirrrpryrhrjr_cRtj|xst|xr|dvS)N)r)ris_boolr)values r _is_bool_likers& ;;u  H*U"3"GHrcz|jD](\}}|tvst|rt||d*y)NF) none_allowed)items _bool_kwargsrr)kwdskwdrs r_validate_bool_kwargsrs:jjl@ U , }U'; s ?@riPii@ig?>rer{rrxror chunksizerirsrZrrrcrvrr_.cyNrnamevalmin_vals rvalidate_integerrHKrcyrrrs rrrrrcyrrrs rrr sTWrc||Sd|dd|d}t|r/t||k7r t|t|}t|St|r||k\s t|t|S)a Checks whether the 'name' parameter for parsing is either an integer OR float that can SAFELY be cast to an integer without losing accuracy. Raises a ValueError if that is not the case. Parameters ---------- name : str Parameter name (used for error reporting) val : int or float The value to check min_val : int Minimum allowed value (val < min_val will result in a ValueError) 'sz' must be an integer >=d)rrb ValueErrorr)rrrmsgs rrr$sw$ { d1X,WQK 8C} s8s?S/ !#h s8Oo#.o s8Orc|_t|tt|k7r tdt|ds&t |t j s tdyyy)aZ Raise ValueError if the `names` parameter contains duplicates or has an invalid data type. Parameters ---------- names : array-like or None An array containing a list of the names used for the output DataFrame. Raises ------ ValueError If names are not unique or are not ordered (e.g. set). Nz Duplicate names are not allowed.F) allow_setsz&Names should be an ordered collection.)lensetrr isinstancerKeysView)rQs r_validate_namesrDs[  u:SZ (?@ @ 5 1Zs||5TEF F6U 1 rct||jdd|jddd|d<nd|d<|jdd}|jdd}|jdd }t|ts!t d t |j |jd d k(r|r t d |t dtd|d}|jdd}t|jdd|jd$tjdtt|sN|L|Jt}|dkDr;t||r/t|}t|tsJ t!|||}||St)|fi|} |s|r| S| 5| j+|cdddS#t"t$t&f$rd}YKwxYw#1swYyxYw)zGeneric reader of line files.rlNrnFTrrr}strictz&encoding_errors must be a string, got rXpyarrowz@The 'iterator' option is not supported with the 'pyarrow' enginezAThe 'chunksize' option is not supported with the 'pyarrow' enginerrerQrzVThe 'float_precision' argument is deprecated. Use the default float precision instead. stacklevel)rgetrrtrtyperrrwarningswarnrr_default_n_workers_can_parallelize_csvr/_read_csv_parallelrUnicodeDecodeError OverflowErrorTextFileReaderread) filepath_or_bufferrrrerrorsre _n_workers _filepathresultparsers r_readr\s  $ xx t$, 88M4 ( 0"'D "&D xx E*Hd+IXX' 2F fc "4T&\5J5J4K L   xxY& R   S %[)Q? HHWd #EDHHWd+, xx!".  7 ')    )em') >23EtL&'9:Ii- -- +ItZH! . 7$ 7FH "{{5!""# !3]C  """s> F>#G>GGG#ctd}tjdk(ry||Stt j xsdt }t}| t||}|S)a Default worker count for a parallel ``read_csv``. ``mode.max_threads`` wins whenever it is set (except on Emscripten, which cannot spawn threads at all). Otherwise it is the smallest of the machine's logical CPU count, ``_MAX_DEFAULT_WORKERS``, and the CPUs actually available to the process (CPU affinity / cgroup limits) -- so that an embedded or containerised pandas does not oversubscribe its allocation. zmode.max_threads emscriptenr)rsysplatformminos cpu_count_MAX_DEFAULT_WORKERSr) max_threads n_workers availables rrrsd/0K |||#BLLN'a)=>I$%I 9- rct|ryt|}t|tsyd|vry|j dy|j dd}t|t r|j d}| t ||y |j dds|j d y|j d d vry|j d y|j d }||dk7ry|j d}|% tj|j}|dvry|j dd}|t|dkDr|dk7ryt|dk(rt|dkDry|j d}t|trt|dk(rt|dkDry|j dy|j dy|j dy|j dry|j ddsy|j dtjj k(ry|j d} | !t| t"t$j&fsyt|j ddt(t*fry|j d d!d!kDry|j d"d} | | dury|j d#y t,j.j1|} | t2kry |j dd} | dk(r|j d$d!nd} t#|j dxsd!} t7|| | }| |z t2kryt9|d%5}|j;|}dddt=d&j?Dryy#t$rYywxYw#t$rYywxYw#t4$rYywxYw#1swYXxYw)'a Return True when a ``read_csv`` call is eligible for parallel execution. Parallel reading works by splitting the file's data section into byte-range chunks at newline boundaries and processing them concurrently with threads. The following conditions must all hold: * *filepath_or_buffer* is a local file path (not a URL or file object). * The file is not compressed. * The C engine is selected (``engine='c'``, the default). * The call is not in iterator / chunked mode. * The entire file is being read (no ``nrows`` limit). * ``skiprows`` is a plain integer (or absent) - list/callable forms require tracking absolute line numbers across chunks. * ``header`` is a single integer or ``None`` - multi-level headers complicate the preamble boundary. * ``index_col`` is ``None`` or ``False`` - a column-based index would need its name propagated to non-first chunks. * ``usecols`` is ``None`` - column selection changes the mapping between raw column positions and names in non-first chunks. * The separator is a single ASCII character or ``r"\s+"``, and ``quotechar`` is a single ASCII character - anything else forces the python engine inside ``TextFileReader``. * The encoding is ``utf-8`` / ``utf-8-sig`` - chunk workers feed raw file bytes to the C tokenizer, which decodes words as UTF-8. ``ascii`` is excluded so a non-ASCII byte still raises rather than being masked. * ``comment`` is ``None`` and the preamble has no blank lines - full-line comments and blank lines shift where pandas locates the header relative to the physical line count used by :func:`_find_data_start_offset`. * ``escapechar`` is ``None`` and ``dialect`` is not used - an escaped literal newline would be split mid-field. * ``parse_dates`` is unset - datetime format inference is per-chunk and may disagree with the serial whole-column inference. * ``low_memory`` is truthy - a falsy ``low_memory`` guarantees whole-file type inference, which per-chunk parallel reading cannot honour (``low_memory=True`` already documents per-chunk inference divergence). * ``storage_options`` is ``None`` - it raises for local paths in the serial path, and that error must not be masked. * ``on_bad_lines`` is not ``"warn"`` - chunk workers would report chunk-relative (i.e. wrong) line numbers. * Both the file and its data section (i.e. excluding the header preamble) are at least ``_PARALLEL_READ_MIN_BYTES`` bytes large. Note that a chunk boundary landing on a newline embedded inside a quoted field leaves that chunk's parser inside an open quote at EOF, so the chunk fails and the read falls back to serial - it does not corrupt data silently. Fz://rNrqrmethodrrrX)cNrerv r|)utf-8z utf-8-sigrM,r\s+rwr{rzrrlrTrrarOrcrrSrTrQrbc3@K|]}|jd yw) N)rstrip).0lines r z'_can_parallelize_csv..s Ddt{{5! ! Ds) r#r/rrtrdictr.rcodecslookupr LookupErrorrordr2BadLineHandleMethod BLHM_WARNrbnpintegerlisttuplerpathgetsize_PARALLEL_READ_MIN_BYTESOSError_find_data_start_offsetopenrany splitlines)rrfilepathrqrvr| codec_namerMrwrarS file_size_header _skiprows data_startfdpreambles rrrsd&'01H h $  xx!".((='2K+t$!ooh/   ;7CD xx E"dhh{&;&G xx, xx$XX./N!n&<xx #H x055J 3 3 c*IS^a/I4G 9~s9~3%I)S!c)n&9c)ns>R xx & xx )  xx & xx  88L$ ' xx:#A#A#K#KKxx #HJx#rzz9J$K$((8W-e}= xx a 1$d+I%!7 xx &GGOOH- / / 0hhx)G'xx(0!dDHHZ(-A.I(7IFJ: 88 h '77:&' Dh.A.A.C DD W  D  h  ''sB6 N;N,(N;!O  N)(N), N87N8; OO Oc|d}nt|dz}t||z}t|d5}t|D]}|jrn|j cdddS#1swYyxYw)a9 Return the byte offset of the first data row in *filepath*. Reads ``int(skiprows) + (header + 1 if header is not None else 0)`` physical lines from the start of the file and returns the position after the last preamble byte. This is the byte from which each non-first chunk must begin. Nrrr)rbrrangereadlinetell)r rOra header_linespreamble_linesr_s rrrsu~ 6{Q ]\1N h ~& A;;= wwy sA*A**A3cdg|z}t||}t|D]"}|dz|z }ddtz |zz |||z |z<$|S)a Relative sizes for *n_chunks* chunks whose last *n_tapered* shrink. The chunks before the tail are full size; across the tail the size falls linearly to ``_PARALLEL_TAPER_RATIO`` of one, so the final chunk is the smallest piece of work in the file. g?r)rr_PARALLEL_TAPER_RATIO)n_chunks n_taperedweightspositionshares r_chunk_size_weightsr slehGIx(I)$ A* 3..%7 7 9$x/0 NrcNtjj|}||z }|g}|dks|dkr|j||S||kDrd|znd}t ||}t |} d} t |d5} td|D]z} | || dz z } |t|| z| z z} | |k\rnT| j| | j| j}||k\rn||dk(rj|j||ddd|j||S#1swYxYw)a Compute byte offsets that partition the data portion of *filepath* into *n_chunks* pieces aligned to newline boundaries. The pieces are equal-sized, except that when *n_workers* is given and the chunks take more than one round the last ``2 * n_workers`` taper down: the read finishes when the last chunk a worker picked up does, so ending on small chunks costs less than ending on a full one. Within a single round every chunk runs at once, so the makespan is the largest of them and any taper would only inflate it. Returns a list of ``n + 1`` offsets where the byte range ``[offsets[i], offsets[i+1])`` defines chunk *i*. rrgrN) rrrappendr sumrrrbseekrr)r rrrr  data_sizeoffsetsrr total_weightrunning_weightritargetposs r_find_chunk_byte_offsetsr.s8()IJ&IlG1} Qy!!)I!5I 1I!(I6GwN- h :J:LMNrcg} t||||}| t ||S#tttf$rt$r t |wxYw)a Read a large CSV file in parallel using *n_workers* threads. Every read this makes - the one-line name inference and each chunk - raises its own copy of the same ``ParserWarning``, and a worker raising one lands the stacklevel walk in threading internals. So the reads collect their warnings into a sink (see :attr:`TextReader.warning_sink`) and each distinct one is raised once here, from the caller's frame - except when the caller goes on to read the file serially, since that read raises them itself. GH#66259 The file's data section (everything after the header / skiprows preamble) is split into byte-range chunks aligned to newline boundaries, sized from *n_workers* and from the file's own rows, bytes and column count. Each chunk is parsed by an independent :class:`TextFileReader` / C-engine instance. Because the hot paths in ``pandas/_libs/parsers.pyx`` (tokenisation, int/float/bool conversion) are wrapped in ``with nogil:`` blocks, threads achieve real CPU-level parallelism. The caller must ensure the file is eligible via :func:`_can_parallelize_csv` before calling this function. Returns ``None`` when parallel reading turns out not to be applicable after all (the data section cannot be split, the engine falls back to python, the one-line sample used to infer column names has no columns, per-chunk dtype inference disagrees in a way that would not match the serial result, or any chunk fails to parse); the caller then reads serially. )_read_csv_chunksrrr Exceptionr4)r rrr1rs rrrseF57L !(D)\J& M +] ;  & s !*A cRSTUVWXYZ|jdd}|dk(r|jddnd}t|jdxsd}t|||}tjj ||z }g}t |d5} tdD][} | j||| zdzz| jd | jd } | sB|jt| ] ddd|j|r|t|d zn|} || z} td z| tz}|d kr|d tzkryd }i|dd dd }t |d5}|j!|}|j}ddddzj#ddvryt%j&||z}i|t(dt*j,d} t/|fi|}t3|j4t6s|j9y|j4_|j4j<_|j4j<j@r|j9y|j4jBJtE|j4jB}t|j!}|j9|dk7ry|t3|jdtFr|jIdd}t/t%j&|fii|dddd t(ddd d} tE|j!jJd}|j9ttM|t|k7ryt|tNztPtSt|dz| tz}|kDr | z z}tS||}tU|||XtXdz }|d kryt|i|d|dd dStWjXTt|D]}Tj[|dg|zYt]j^RgZd-RSTXYZfd }d.d}d/d} |jdVt3VtFxs1Vduxr+taVtbjdtbjffv}!g}"g}#gUg}$g}%d}&d }'t |d5} tijh| jkdthjl}(ddd tojp5})ts5}*tD+cgc]}+|)jutw(},}+|,D-cgc]}-|*jy||-c}-D]}.|.j{Rj}r^ dddddd|'sZD]}/ |/j9tojt5(j9dddytdZDr^ dddddd|'sZD]}/ |/j9tojt5(j9dddyZD/cgc]}/|*jy|/j8}0}/d}'tdYD1cgc]:}1|1dr3ttt|1djdkDr|1<}2}1|2s^ dddddd|'sZD]}/ |/j9tojt5(j9dddytE|2dd}%|2D+3cgc]\}+}3|3 }$}+}3tE|$dUUrtUfd |$D}&UD]}4|$D5chc]}5|5|4j}6}5t|6dkDs.td!|6DrAdddddd|'sZD]}/ |/j9tojt5(j9dddy|!si}7i}8tUD]\}9}4|$D5cgc] }5| |5|4}:}5|:dWWOtWfd"|:Dr;|7jWjgj|9td#|:|8|9<t|#j|9g};|7jD]\}<}=tcjt|=|&f|<$}>t|=D]6\}?}9d}@|8|9D]'}A|;j|>|?@|Af|@t|Az }@)8|"j|>tcj|=tbj$ft|; d%zz xsd}Btdt|;|BDCcgc]}C|*jy||;|C|CBz}D}C|DD]}.|.j{|0D]}.|.j{ dddddd|'sZD]}/ |/j9tojt5(j9dddt|&}E|!rt|$|%d &}Ft3VtFrtd'}G|GjVntVfd(}GFjDHIcic]H\}H}I|Ht|IEtaG|HtbjdtbjffvrGHndd )J}J}H}It|J|%Ed *S|#r|#D9cgc]}9U|9 }K}9|$D54cgc]}5KD4cic]}4|4|5|4 c}4}L}5}4t|LKd &}Mt|#|Kd+D]\}9}4M|4}Nt3|Ntbjr+Njtk(rtNEd ,j}Nt3NtbjrNjdd}N|"jNtcj|9gtbj$f|"DNOcgc]\}N}Ot|N|O}P}N}Ot|Pt|%Eg}Qtj|Q|QjS#1swY xYw#1swY 1xYw#t0$rYywxYw#|j9wxYw#1swYxYwcc}+wcc}-w#t~$rY/wxYw#1swYyxYw#t~$rYwxYw#1swYyxYwcc}/wcc}1w#t~$rY(wxYw#1swYyxYwcc}3}+wcc}5w#t~$rYewxYw#1swYyxYwcc}5wcc}Cw#1swYixYw#1swYnxYw#t~$rYywxYw#1swY?xYw#|'s'ZD]"}/ |/j9#t~$rY wxYwtojt5(j9dddw#1swYwxYwxYwcc}I}Hwcc}9wcc}4wcc}4}5wcc}O}Nw)0z Body of :func:`_read_csv_parallel`; see there for what the arguments mean and when the result is ``None``. Appends the ``ParserWarning``\s its parsers would have raised to *warning_sink* instead of raising them. rOrrQNrrar ir"F)rqrrrs r)rVrZrrrVT)keependsr#)rOrQrTrSrVrZrari)rOrQrarc@ttjdfi}t|jt sJd|jj _d|j_ |jj _ |j_ |jj _ j|js j}| | |dz}|jj j#| |dk(~d|j_ |jj'\}}}||f |<jsyy#tj $rYywxYw#t($rj+YywxYw)NrFrr) strip_bomT)rioBytesIOr_enginer5_readertrim_after_read wrap_deferred block_workers _warning_sinkr1r$is_set get_nowaitqueueEmpty load_buffer _first_chunkrr7r)mvreader chunk_idxdatar chunk_columnscol_dict chunk_failed chunk_kwds chunk_queuerr(resultsr1workers_readerss r_workerz!_read_csv_chunks.._workers| 3>:>&...99927. (-$09,'3$.:+v&%%' '224 gi(79q=+ABD NN " " . .twy?QUV?V . W+/FNN ' .4^^-@-@-B*=(#0!:GI +%%';;     " s$E(0F(E>=E>FFc<|D]\}}}||||t|zyrr)jobsdstr-pieces r _copy_piecesz&_read_csv_chunks.._copy_piecess. $ 0OCe*/CcCJ& ' 0rct|tjr|jtk(ry|St|t t fr0t|jtjr |jSyr)rrndarrayrVobjectr%r&_ndarray)r\s r_piece_ndarrayz(_read_csv_chunks.._piece_ndarray!s[ eRZZ ({{f$L em^< =* KKC >> !r)access) max_workersc3\K|]$}|jjj&ywr)r@rAna_left_literal)rrMs rrz#_read_csv_chunks..^s%;A&&66s*,z list[tuple]c3@K|]}t|dyw)rNrY)r chunk_dictcol_lists rrz#_read_csv_chunks..wsWZC 8A; 78Wsc3^K|]%}|jdvxrt|t 'yw)ifN)kindrr")rdts rrz#_read_csv_chunks..s25GGtOFJr:,F(FF5s+-c3\K|]#}|duxr|jjk(%ywrrV)rr\firsts rrz#_read_csv_chunks..s31!T)HekkU[[.HH1s),zlist[np.ndarray]ro) warn_mixedcyrrrrrz"_read_csv_chunks..rcSrr dtype_argsrrtz"_read_csv_chunks..s rindexrVcopycolumnsrzr{)r)rzr{)rL memoryviewreturnr)rZz(list[tuple[np.ndarray, int, np.ndarray]]rr)rznp.ndarray | None)`rrbrrrrrrr&rr$rsortr_PARALLEL_MIN_CHUNK_ROWSrrreplacer>r?rtr no_defaultrrrr@r5closerErAr1 leading_cols orig_namesrrr ilocr_PARALLEL_CHUNK_BYTES_PARALLEL_MAX_COLUMN_PIECESmaxr.rH SimpleQueueput threadingEventr!rstr_object_mmapfileno ACCESS_READ contextlib ExitStackr enter_contextr~submitrrFr7suppress BufferErrorrrnextitervaluesr%rVall enumerate setdefaultremptyarrayintpr(r6rupdater$r'zipr_r`_valuesreshaper*r+r) _from_mgraxes)[r rrr1rOrarr' line_lensfhprober bytes_per_rowest_rowsn_target base_kwdsrr first_linename_buf name_kwds name_reader col_names n_first_rows header_line raw_reader raw_names size_targetrrNrWr]rbneeds_series_wrap block_specs leftover_pos chunk_dictsr}rreaders_closingmmviewspoolr worker_viewsrLfutrM close_futuresres chunk_resultsrQrrh chunk_dtypesgroups col_nd_piecesr- nd_pieces copy_jobs group_dtype positionsvalues2drowoffsetr\batchstart copy_futuresrzrOrVkv new_col_dictleftover_namesleftover_dictsgatheredr placement block_objsmgrrRrSrTrirxrpr(rUrVs[ `` @@@@@@@@@rr6r6;sXXh (F hhw'/T488J',1-H)68DJ )J6II h ,1X ,E GGJU!2a!77 8 KK ;;w'D  T+  ,,NN6?Ic)n12YMM)H9q=(.F"FGH!| q33 3   I h #77:&[[] #  J&//== zz(Z/0H  I $X;;  k))> : (4K%/;K,""//     ) ) 55 5;..99:I {'')*Lqj'):DA))4)8< # JJ{ #    "" "    Z__.33A67I     s9~ #i. 0**#s3y>1'==,,KY$ 12Y> 8[)H&x:yQG7|aH!|Ix(I J&+%6%6%8K8_#  "#FX%G??$LO&;&;P0 !I#9d3R,y"9bggrzz=R"R "K LH K G EO h @ YYryy{Ad.>.> ?@z  "l &+ 9 5l 9= >C9=M89##JrN3LBM !Yl l \) LLN   -  HHJ  W=+A./G7DE 88EKEKN+HW;WW! ITU: 4 0 6 6U U|$q(5*52 Al l \) LLN   -  HHJ  } %57=? !*8!4 1ICKV!=Gz$'78!I!&aLE(S1%.1.))%++r:AA#F-12Di-P c*$++C0 1GI .4llnW*K!xxY(?{SH$-i$81S!"%23%71E%,,hsmVU-KL"c%j0F11  &&"((9BGG2T'UVWy>/i!m<=B"'q#i.%!@ KK i .NO   (!CJJL!$   Wl l \) LLN   -  HHJ  u E";EJ i &!,\!:E LL # 12E   1 v$E!H-"''2::1FF!H     we%PP3?@C(3-@@* 1? ?T:d# # ?  '~~RWX\>$G IICd^F&"**-&,,&2H e%@HH&"**-2.   #bgg(FG H IKVV5FVY+fi0VJV zL$95#A BC   sCHH --o ,,8##*   l    L@@IR!    !   sV \!   UF Vn!   _!8 Kl l l l d!   ) LLN    -  HHJ     A ? WsOA r4!r40"s' s&s<5s2:w v%v)!s? vt))vv%w*t t/vv% wt%t5v#"uv?uvv%#w3u u8v u' 0v u- v1vvv% wu2v"v?v Ev!v0v1v%9w v24w7A x; y0 y 9 y y $y4r>s  sss/2s<? v ttt"% t21t25t> v uuu$' v2 u?>u?v  vv" v%%v/*w2 v?>v?w x8w)(x8) w5 2x84w5 5x8x,# x8,x5 1x8y )rrc yrrrrrrs rread_csvrr)rc yrrrs rrrrr)rrc yrrrs rrrrc yrrrs rrr"%rpandas."rerror)+rLrMrOrQrSrTrVrXrZr\r]r_rarcrergrhrirjrlrnrorprrrqrsrurvrwrxryrzr{r|r}rrrrrrrc+ tj},|,d=|,d=t|%||||&|ddi|+| }-|,j|-t ||,S)a)^ Read a comma-separated values (csv) file into DataFrame. Also supports optionally iterating or breaking of the file into chunks. Additional help can be found in the online docs for `IO Tools `_. Parameters ---------- filepath_or_buffer : str, path object or file-like object Any valid string path is acceptable. The string could be a URL. Valid URL schemes include http, ftp, s3, gs, and file. For file URLs, a host is expected. A local file could be: file://localhost/path/to/table.csv. Certain URL schemes may require additional packages. For example, S3 URLs require the ``s3fs`` library. See :ref:`install.optional_dependencies` for a full list. If you want to pass in a path object, pandas accepts any ``os.PathLike``. By file-like object, we refer to objects with a ``read()`` method that accepts an optional size argument, such as a file handle (e.g. via builtin ``open`` function) or ``StringIO``. sep : str, default ',' Character or regex pattern to treat as the delimiter. ``sep=None`` detects the separator from the first valid row of the file with Python's builtin sniffer tool, ``csv.Sniffer``; it is supported only by the Python parsing engine, which will be used automatically. In addition, separators longer than 1 character and different from ``'\s+'`` will be interpreted as regular expressions and will force the use of the Python parsing engine. Note that regex delimiters are prone to ignoring quoted data. Regex example: ``'\r\t'``. delimiter : str, optional Alias for ``sep``. header : int, Sequence of int, 'infer' or None, default 'infer' Row index or indices in the file to use as DataFrame column labels. Indexing starts at 0 and counts only non-blank, non-commented lines when ``skip_blank_lines=True``, so ``header=0`` denotes the first line of data rather than the first line of the file. Valid arguments are: * ``int``: line index at which column labels are read. * Sequence of ``int``: line indices at which column labels are read and combined into a :class:`~pandas.MultiIndex` on the columns, e.g. ``[0, 1, 3]``. Intervening rows that are not specified will be skipped (e.g. row 2 in this example). * ``None``: no row in the file is interpreted as column labels. Columns are labelled by integer position, or by the values passed to ``names`` when provided. Use this for files with no header row. If the file has a header row that should be overridden by ``names``, pass ``header=0`` instead. * ``'infer'`` (default): equivalent to ``header=0`` when no ``names`` are passed, otherwise equivalent to ``header=None``. When inferred from the file contents, headers are kept distinct from each other by renaming duplicate names with a numeric suffix of the form ``".{count}"`` starting from 1, e.g. ``"foo"`` and ``"foo.1"``. Empty headers are named ``"Unnamed: {i}"`` or `` "Unnamed: {i}_level_{level}"`` in the case of MultiIndex columns. names : Sequence of Hashable, optional Sequence of column labels to apply. If the file contains a header row, then you should explicitly pass ``header=0`` to override the column names. Duplicates in this list are not allowed. index_col : Hashable, Sequence of Hashable or False, optional Column(s) to use as row label(s), denoted either by column labels or column indices. If a sequence of labels or indices is given, :class:`~pandas.MultiIndex` will be formed for the row labels. Note: ``index_col=False`` can be used to force pandas to *not* use the first column as the index, e.g., when you have a malformed file with delimiters at the end of each line. usecols : Sequence of Hashable or Callable, optional Subset of columns to select, denoted either by column labels or column indices. If list-like, all elements must either be positional (i.e. integer indices into the document columns) or strings that correspond to column names provided either by the user in ``names`` or inferred from the document header row(s). If ``names`` are given, the document header row(s) are not taken into account. For example, a valid list-like ``usecols`` parameter would be ``[0, 1, 2]`` or ``['foo', 'bar', 'baz']``. Element order is ignored, so ``usecols=[0, 1]`` is the same as ``[1, 0]``. To instantiate a :class:`~pandas.DataFrame` from ``data`` with element order preserved use ``pd.read_csv(data, usecols=['foo', 'bar'])[['foo', 'bar']]`` for columns in ``['foo', 'bar']`` order or ``pd.read_csv(data, usecols=['foo', 'bar'])[['bar', 'foo']]`` for ``['bar', 'foo']`` order. If callable, the callable function will be evaluated against the column names, returning names where the callable function evaluates to ``True``. An example of a valid callable argument would be ``lambda x: x.upper() in ['AAA', 'BBB', 'DDD']``. Using this parameter results in much faster parsing time and lower memory usage. dtype : dtype or dict of {Hashable : dtype}, optional Data type(s) to apply to either the whole dataset or individual columns. E.g., ``{'a': np.float64, 'b': np.int32, 'c': 'Int64'}`` Use ``str`` or ``object`` together with suitable ``na_values`` settings to preserve and not interpret ``dtype``. Specifying a ``dtype`` does not change how missing values are parsed; values recognized as missing (see ``na_values`` and ``keep_default_na``) are still read as ``NaN`` (or the dtype's NA value) rather than cast to the requested ``dtype``. If ``converters`` are specified, they will be applied INSTEAD of ``dtype`` conversion. Specify a ``defaultdict`` as input where the default determines the ``dtype`` of the columns which are not explicitly listed. engine : {'c', 'python', 'pyarrow'}, optional Parser engine to use. The C and pyarrow engines are faster, while the python engine is currently more feature-complete. The pyarrow engine is multithreaded, and the C engine reads sufficiently large files in parallel. Some features of the "pyarrow" engine are unsupported or may not work correctly. converters : dict of {Hashable : Callable}, optional Functions for converting values in specified columns. Keys can either be column labels or column indices. The function is applied to the raw text read from the file, so an empty field is passed as an empty string ``''``; ``na_values`` and ``keep_default_na`` are then applied to the value the function returns. true_values : list, optional Values to consider as ``True`` in addition to case-insensitive variants of 'True'. false_values : list, optional Values to consider as ``False`` in addition to case-insensitive variants of 'False'. skipinitialspace : bool, default False Skip spaces after delimiter. skiprows : int, list of int or Callable, optional Line numbers to skip (0-indexed) or number of lines to skip (``int``) at the start of the file. If callable, the callable function will be evaluated against the row indices, returning ``True`` if the row should be skipped and ``False`` otherwise. An example of a valid callable argument would be ``lambda x: x in [0, 2]``. skipfooter : int, default 0 Number of lines at bottom of file to skip (Unsupported with ``engine='c'``). nrows : int, optional Number of rows of file to read. Useful for reading pieces of large files. Refers to the number of data rows in the returned DataFrame, excluding: * The header row containing column names. * Rows before the header row, if ``header=1`` or larger. Example usage: * To read the first 999,999 (non-header) rows: ``read_csv(..., nrows=999999)`` * To read rows 1,000,000 through 1,999,999: ``read_csv(..., skiprows=1000000, nrows=999999)`` na_values : Hashable, Iterable of Hashable or dict of {Hashable : Iterable}, optional Additional strings to recognize as ``NA``/``NaN``. If ``dict`` passed, specific per-column ``NA`` values. By default the following values are interpreted as ``NaN``: empty string, "NaN", "N/A", "NULL", and other common representations of missing data. keep_default_na : bool, default True Whether or not to include the default ``NaN`` values when parsing the data. Depending on whether ``na_values`` is passed in, the behavior is as follows: * If ``keep_default_na`` is ``True``, and ``na_values`` are specified, ``na_values`` is appended to the default ``NaN`` values used for parsing. * If ``keep_default_na`` is ``True``, and ``na_values`` are not specified, only the default ``NaN`` values are used for parsing. * If ``keep_default_na`` is ``False``, and ``na_values`` are specified, only the ``NaN`` values specified ``na_values`` are used for parsing. * If ``keep_default_na`` is ``False``, and ``na_values`` are not specified, no strings will be parsed as ``NaN``. Note that if ``na_filter`` is passed in as ``False``, the ``keep_default_na`` and ``na_values`` parameters will be ignored. na_filter : bool, default True Detect missing value markers (empty strings and the value of ``na_values``). In data without any ``NA`` values, passing ``na_filter=False`` can improve the performance of reading a large file. skip_blank_lines : bool, default True If ``True``, skip over blank lines rather than interpreting as ``NaN`` values. parse_dates : bool, None, list of Hashable, default None The behavior is as follows: * ``bool``. If ``True`` -> try parsing the index. * ``None``. Behaves like ``True`` if ``date_format`` is specified. * ``list`` of ``int`` or names. e.g. If ``[1, 2, 3]`` -> try parsing columns 1, 2, 3 each as a separate date column. If a column or index cannot be represented as an array of ``datetime``, say because of an unparsable value or a mixture of timezones, the column or index will be returned unaltered as an ``object`` data type. For non-standard ``datetime`` parsing, use :func:`~pandas.to_datetime` after :func:`~pandas.read_csv`. Note: A fast-path exists for iso8601-formatted dates. date_format : str or dict of column -> format, optional Format to use for parsing dates and/or times when used in conjunction with ``parse_dates``. The strftime to parse time, e.g. :const:`"%d/%m/%Y"`. See `strftime documentation `_ for more information on choices, though note that :const:`"%f"`` will parse all the way up to nanoseconds. You can also pass: - "ISO8601", to parse any `ISO8601 `_ time string (not necessarily in exactly the same format); - "mixed", to infer the format for each element individually. This is risky, and you should probably use it along with `dayfirst`. .. versionadded:: 2.0.0 dayfirst : bool, default False DD/MM format dates, international and European format. cache_dates : bool, default True If ``True``, use a cache of unique, converted dates to apply the ``datetime`` conversion. May produce significant speed-up when parsing duplicate date strings, especially ones with timezone offsets. iterator : bool, default False Return ``TextFileReader`` object for iteration or getting chunks with ``get_chunk()``. chunksize : int, optional Number of lines to read from the file per chunk. Passing a value will cause the function to return a ``TextFileReader`` object for iteration. See the `IO Tools docs `_ for more information on ``iterator`` and ``chunksize``. compression : str or dict, default 'infer' For on-the-fly decompression of on-disk data. If 'infer' and 'filepath_or_buffer' is path-like, then detect compression from the following extensions: '.gz', '.bz2', '.zip', '.xz', '.zst', '.tar', '.tar.gz', '.tar.xz' or '.tar.bz2' (otherwise no compression). If using 'zip' or 'tar', the ZIP file must contain only one data file to be read in. Set to ``None`` for no decompression. Can also be a dict with key ``'method'`` set to one of {``'zip'``, ``'gzip'``, ``'bz2'``, ``'zstd'``, ``'xz'``, ``'tar'``} and other key-value pairs are forwarded to ``zipfile.ZipFile``, ``gzip.GzipFile``, ``bz2.BZ2File``, ``zstandard.ZstdDecompressor``, ``lzma.LZMAFile`` or ``tarfile.TarFile``, respectively. As an example, the following could be passed for Zstandard decompression using a custom compression dictionary: ``compression={'method': 'zstd', 'dict_data': my_compression_dict}``. thousands : str (length 1), optional Character acting as the thousands separator in numerical values. decimal : str (length 1), default '.' Character to recognize as decimal point (e.g., use ',' for European data). lineterminator : str (length 1), optional Character used to denote a line break. Only valid with C parser. quotechar : str (length 1), optional Character used to denote the start and end of a quoted item. Quoted items can include the ``delimiter`` and it will be ignored. quoting : {0, 1, 2, 3}, default csv.QUOTE_MINIMAL Control field quoting behavior per ``csv.QUOTE_*`` constants. Use one of ``0`` or ``csv.QUOTE_MINIMAL``, ``1`` or ``csv.QUOTE_ALL``, ``2`` or ``csv.QUOTE_NONNUMERIC``, or ``3`` or ``csv.QUOTE_NONE``. Default is ``csv.QUOTE_MINIMAL`` (i.e., 0) which implies that only fields containing special characters are quoted (e.g., characters defined in ``quotechar``, ``delimiter``, or ``lineterminator``. doublequote : bool, default True When ``quotechar`` is specified and ``quoting`` is not ``QUOTE_NONE``, indicate whether or not to interpret two consecutive ``quotechar`` elements INSIDE a field as a single ``quotechar`` element. escapechar : str (length 1), optional Character used to escape other characters. comment : str (length 1), optional Character indicating that the remainder of line should not be parsed. If found at the beginning of a line, the line will be ignored altogether. This parameter must be a single character. Like empty lines (as long as ``skip_blank_lines=True``), fully commented lines are ignored by the parameter ``header`` but not by ``skiprows``. For example, if ``comment='#'``, parsing ``#empty\na,b,c\n1,2,3`` with ``header=0`` will result in ``'a,b,c'`` being treated as the header. encoding : str, optional, default 'utf-8' Encoding to use for UTF when reading/writing (ex. ``'utf-8'``). `List of Python standard encodings `_ . encoding_errors : str, optional, default 'strict' How encoding errors are treated. `List of possible values `_ . dialect : str or csv.Dialect, optional If provided, this parameter will override values (default or not) for the following parameters: ``delimiter``, ``doublequote``, ``escapechar``, ``skipinitialspace``, ``quotechar``, and ``quoting``. If it is necessary to override values, a ``ParserWarning`` will be issued. See ``csv.Dialect`` documentation for more details. on_bad_lines : {'error', 'warn', 'skip'} or Callable, default 'error' Specifies what to do upon encountering a bad line (a line with too many fields). Allowed values are: - ``'error'``, raise an Exception when a bad line is encountered. - ``'warn'``, raise a warning when a bad line is encountered and skip that line. - ``'skip'``, skip bad lines without raising or warning when they are encountered. - Callable, function that will process a single bad line. - With ``engine='python'``, function with signature ``(bad_line: list[str]) -> list[str] | None``. ``bad_line`` is a list of strings split by the ``sep``. If the function returns ``None``, the bad line will be ignored. If the function returns a new ``list`` of strings with more elements than expected, a ``ParserWarning`` will be emitted while dropping extra elements. - With ``engine='pyarrow'``, function with signature as described in pyarrow documentation: `invalid_row_handler `_. .. versionchanged:: 2.2.0 Callable for ``engine='pyarrow'`` low_memory : bool, default True Internally process the file in chunks, resulting in lower memory use while parsing, but possibly mixed type inference. Because each chunk is type-inferred independently, the same literal can be parsed as an ``int`` in one chunk and a ``str`` in another; after concatenation the column has ``object`` dtype and contains both representations, so comparisons such as ``df["col"] == 12345`` or ``df["col"] == "12345"`` will match only a subset of the rows holding that value. A :class:`~pandas.errors.DtypeWarning` is emitted when this occurs. To ensure no mixed types either set ``False``, or specify the type with the ``dtype`` parameter. Note that the entire file is read into a single :class:`~pandas.DataFrame` regardless, use the ``chunksize`` or ``iterator`` parameter to return the data in chunks. (Only valid with C parser). memory_map : bool, default False If a filepath is provided for ``filepath_or_buffer``, map the file object directly onto memory and access the data directly from there. Using this option can improve performance because there is no longer any I/O overhead. float_precision : {'high', 'legacy', 'round_trip'}, optional Specifies which converter the C engine should use for floating-point values. The options are ``None`` or ``'high'`` for the ordinary converter, ``'legacy'`` for the original lower precision pandas converter, and ``'round_trip'`` for the round-trip converter. .. deprecated:: 3.1.0 All float precision modes now use the same converter. The ``float_precision`` argument will be removed in a future version. storage_options : dict, optional Extra options that make sense for a particular storage connection, e.g. host, port, username, password, etc. For HTTP(S) URLs the key-value pairs are forwarded to ``urllib.request.Request`` as header options. For other URLs (e.g. starting with "s3://", and "gcs://") the key-value pairs are forwarded to ``fsspec.open``. Please see ``fsspec`` and ``urllib`` for more details, and for more examples on storage options refer `here `_. dtype_backend : {'numpy_nullable', 'pyarrow'} Back-end data type applied to the resultant :class:`DataFrame` (still experimental). If not specified, the default behavior is to not use nullable data types. If specified, the behavior is as follows: * ``"numpy_nullable"``: returns nullable-dtype-backed :class:`DataFrame` * ``"pyarrow"``: returns pyarrow-backed nullable :class:`ArrowDtype` :class:`DataFrame` .. versionadded:: 2.0 Returns ------- DataFrame or TextFileReader A comma-separated values (csv) file is returned as two-dimensional data structure with labeled axes. See Also -------- DataFrame.to_csv : Write DataFrame to a comma-separated values (csv) file. read_table : Read general delimited file into DataFrame. read_fwf : Read a table of fixed-width formatted lines into DataFrame. Notes ----- Column labels read from a header row are always strings. To use column labels of another type, pass them explicitly with ``names`` and set ``header=0`` when the file contains a header row. Sufficiently large local uncompressed files read with the C engine may be parsed by multiple threads in parallel. Use the ``mode.max_threads`` option to cap or disable this; see :ref:`io.csv.parallel` for details. Examples -------- >>> pd.read_csv("data.csv") # doctest: +SKIP Name Value 0 foo 1 1 bar 2 2 #baz 3 Index and header can be specified via the `index_col` and `header` arguments. >>> pd.read_csv("data.csv", header=None) # doctest: +SKIP 0 1 0 Name Value 1 foo 1 2 bar 2 3 #baz 3 >>> pd.read_csv("data.csv", index_col="Value") # doctest: +SKIP Name Value 1 foo 2 bar 3 #baz Column types are inferred but can be explicitly specified using the dtype argument. >>> pd.read_csv("data.csv", dtype={"Value": float}) # doctest: +SKIP Name Value 0 foo 1.0 1 bar 2.0 2 #baz 3.0 True, False, and NA values, and thousands separators have defaults, but can be explicitly specified, too. Supply the values you would like as strings or lists of strings! >>> pd.read_csv("data.csv", na_values=["foo", "bar"]) # doctest: +SKIP Name Value 0 NaN 1 1 NaN 2 2 #baz 3 Comment lines in the input file can be skipped using the `comment` argument. >>> pd.read_csv("data.csv", comment="#") # doctest: +SKIP Name Value 0 foo 1 1 bar 2 By default, columns with dates will be read as ``object`` rather than ``datetime``. >>> df = pd.read_csv("tmp.csv") # doctest: +SKIP >>> df # doctest: +SKIP col 1 col 2 col 3 0 10 10/04/2018 Sun 15 Jan 2023 1 20 15/04/2018 Fri 12 May 2023 >>> df.dtypes # doctest: +SKIP col 1 int64 col 2 object col 3 object dtype: object Specific columns can be parsed as dates by using the `parse_dates` and `date_format` arguments. >>> df = pd.read_csv( ... "tmp.csv", ... parse_dates=[1, 2], ... date_format={"col 2": "%d/%m/%Y", "col 3": "%a %d %b %Y"}, ... ) # doctest: +SKIP >>> df.dtypes # doctest: +SKIP col 1 int64 col 2 datetime64[ns] col 3 datetime64[ns] dtype: object rrLrMrdefaultsrrvlocalsr{_refine_defaults_readrr.rrLrMrOrQrSrTrVrXrZr\r]r_rarcrergrhrirjrlrnrorprrrqrsrurvrwrxryrzr{r|r}rrrrrrrr kwds_defaultss. rrrshB 8==?D !" U )  s##% M KK  #T **rc yrrrs r read_tablerMrrc yrrrs rrrWrrc yrrrs rrrarrc yrrrs rrrkrrc+ tj},|,d=|,d=t|%||||&|ddi|+| }-|,j|-t ||,S)ah] Read general delimited file into DataFrame. Also supports optionally iterating or breaking of the file into chunks. Additional help can be found in the online docs for `IO Tools `_. Parameters ---------- filepath_or_buffer : str, path object or file-like object Any valid string path is acceptable. The string could be a URL. Valid URL schemes include http, ftp, s3, gs, and file. For file URLs, a host is expected. A local file could be: file://localhost/path/to/table.csv. Certain URL schemes may require additional packages. For example, S3 URLs require the ``s3fs`` library. See :ref:`install.optional_dependencies` for a full list. If you want to pass in a path object, pandas accepts any ``os.PathLike``. By file-like object, we refer to objects with a ``read()`` method that accepts an optional size argument, such as a file handle (e.g. via builtin ``open`` function) or ``StringIO``. sep : str, default '\t' (tab-stop) Character or regex pattern to treat as the delimiter. ``sep=None`` detects the separator from the first valid row of the file with Python's builtin sniffer tool, ``csv.Sniffer``; it is supported only by the Python parsing engine, which will be used automatically. In addition, separators longer than 1 character and different from ``'\s+'`` will be interpreted as regular expressions and will force the use of the Python parsing engine. Note that regex delimiters are prone to ignoring quoted data. Regex example: ``'\r\t'``. delimiter : str, optional Alias for ``sep``. header : int, Sequence of int, 'infer' or None, default 'infer' Row index or indices in the file to use as DataFrame column labels. Indexing starts at 0 and counts only non-blank, non-commented lines when ``skip_blank_lines=True``, so ``header=0`` denotes the first line of data rather than the first line of the file. Valid arguments are: * ``int``: line index at which column labels are read. * Sequence of ``int``: line indices at which column labels are read and combined into a :class:`~pandas.MultiIndex` on the columns, e.g. ``[0, 1, 3]``. Intervening rows that are not specified will be skipped (e.g. row 2 in this example). * ``None``: no row in the file is interpreted as column labels. Columns are labelled by integer position, or by the values passed to ``names`` when provided. Use this for files with no header row. If the file has a header row that should be overridden by ``names``, pass ``header=0`` instead. * ``'infer'`` (default): equivalent to ``header=0`` when no ``names`` are passed, otherwise equivalent to ``header=None``. When inferred from the file contents, headers are kept distinct from each other by renaming duplicate names with a numeric suffix of the form ``".{count}"`` starting from 1, e.g. ``"foo"`` and ``"foo.1"``. Empty headers are named ``"Unnamed: {i}"`` or ``"Unnamed: {i}_level_{level}"`` in the case of MultiIndex columns. names : Sequence of Hashable, optional Sequence of column labels to apply. If the file contains a header row, then you should explicitly pass ``header=0`` to override the column names. Duplicates in this list are not allowed. index_col : Hashable, Sequence of Hashable or False, optional Column(s) to use as row label(s), denoted either by column labels or column indices. If a sequence of labels or indices is given, :class:`~pandas.MultiIndex` will be formed for the row labels. Note: ``index_col=False`` can be used to force pandas to *not* use the first column as the index, e.g., when you have a malformed file with delimiters at the end of each line. usecols : Sequence of Hashable or Callable, optional Subset of columns to select, denoted either by column labels or column indices. If list-like, all elements must either be positional (i.e. integer indices into the document columns) or strings that correspond to column names provided either by the user in ``names`` or inferred from the document header row(s). If ``names`` are given, the document header row(s) are not taken into account. For example, a valid list-like ``usecols`` parameter would be ``[0, 1, 2]`` or ``['foo', 'bar', 'baz']``. Element order is ignored, so ``usecols=[0, 1]`` is the same as ``[1, 0]``. To instantiate a :class:`~pandas.DataFrame` from ``data`` with element order preserved use ``pd.read_csv(data, usecols=['foo', 'bar'])[['foo', 'bar']]`` for columns in ``['foo', 'bar']`` order or ``pd.read_csv(data, usecols=['foo', 'bar'])[['bar', 'foo']]`` for ``['bar', 'foo']`` order. If callable, the callable function will be evaluated against the column names, returning names where the callable function evaluates to ``True``. An example of a valid callable argument would be ``lambda x: x.upper() in ['AAA', 'BBB', 'DDD']``. Using this parameter results in much faster parsing time and lower memory usage. dtype : dtype or dict of {Hashable : dtype}, optional Data type(s) to apply to either the whole dataset or individual columns. E.g., ``{'a': np.float64, 'b': np.int32, 'c': 'Int64'}`` Use ``str`` or ``object`` together with suitable ``na_values`` settings to preserve and not interpret ``dtype``. Specifying a ``dtype`` does not change how missing values are parsed; values recognized as missing (see ``na_values`` and ``keep_default_na``) are still read as ``NaN`` (or the dtype's NA value) rather than cast to the requested ``dtype``. If ``converters`` are specified, they will be applied INSTEAD of ``dtype`` conversion. Specify a ``defaultdict`` as input where the default determines the ``dtype`` of the columns which are not explicitly listed. engine : {'c', 'python', 'pyarrow'}, optional Parser engine to use. The C and pyarrow engines are faster, while the python engine is currently more feature-complete. The pyarrow engine is multithreaded, and the C engine reads sufficiently large files in parallel. The 'pyarrow' engine is an *experimental* engine, and some features are unsupported, or may not work correctly, with this engine. converters : dict of {Hashable : Callable}, optional Functions for converting values in specified columns. Keys can either be column labels or column indices. The function is applied to the raw text read from the file, so an empty field is passed as an empty string ``''``; ``na_values`` and ``keep_default_na`` are then applied to the value the function returns. true_values : list, optional Values to consider as ``True`` in addition to case-insensitive variants of 'True'. false_values : list, optional Values to consider as ``False`` in addition to case-insensitive variants of 'False'. skipinitialspace : bool, default False Skip spaces after delimiter. skiprows : int, list of int or Callable, optional Line numbers to skip (0-indexed) or number of lines to skip (``int``) at the start of the file. If callable, the callable function will be evaluated against the row indices, returning ``True`` if the row should be skipped and ``False`` otherwise. An example of a valid callable argument would be ``lambda x: x in [0, 2]``. skipfooter : int, default 0 Number of lines at bottom of file to skip (Unsupported with ``engine='c'``). nrows : int, optional Number of rows of file to read. Useful for reading pieces of large files. Refers to the number of data rows in the returned DataFrame, excluding: * The header row containing column names. * Rows before the header row, if ``header=1`` or larger. Example usage: * To read the first 999,999 (non-header) rows: ``read_csv(..., nrows=999999)`` * To read rows 1,000,000 through 1,999,999: ``read_csv(..., skiprows=1000000, nrows=999999)`` na_values : Hashable, Iterable of Hashable or dict of {Hashable : Iterable}, optional Additional strings to recognize as ``NA``/``NaN``. If ``dict`` passed, specific per-column ``NA`` values. By default the following values are interpreted as ``NaN``: empty string, "NaN", "N/A", "NULL", and other common representations of missing data. keep_default_na : bool, default True Whether or not to include the default ``NaN`` values when parsing the data. Depending on whether ``na_values`` is passed in, the behavior is as follows: * If ``keep_default_na`` is ``True``, and ``na_values`` are specified, ``na_values`` is appended to the default ``NaN`` values used for parsing. * If ``keep_default_na`` is ``True``, and ``na_values`` are not specified, only the default ``NaN`` values are used for parsing. * If ``keep_default_na`` is ``False``, and ``na_values`` are specified, only the ``NaN`` values specified ``na_values`` are used for parsing. * If ``keep_default_na`` is ``False``, and ``na_values`` are not specified, no strings will be parsed as ``NaN``. Note that if ``na_filter`` is passed in as ``False``, the ``keep_default_na`` and ``na_values`` parameters will be ignored. na_filter : bool, default True Detect missing value markers (empty strings and the value of ``na_values``). In data without any ``NA`` values, passing ``na_filter=False`` can improve the performance of reading a large file. skip_blank_lines : bool, default True If ``True``, skip over blank lines rather than interpreting as ``NaN`` values. parse_dates : bool, None, list of Hashable, default None The behavior is as follows: * ``bool``. If ``True`` -> try parsing the index. * ``None``. Behaves like ``True`` if ``date_format`` is specified. * ``list`` of ``int`` or names. e.g. If ``[1, 2, 3]`` -> try parsing columns 1, 2, 3 each as a separate date column. If a column or index cannot be represented as an array of ``datetime``, say because of an unparsable value or a mixture of timezones, the column or index will be returned unaltered as an ``object`` data type. For non-standard ``datetime`` parsing, use :func:`~pandas.to_datetime` after :func:`~pandas.read_csv`. Note: A fast-path exists for iso8601-formatted dates. date_format : str or dict of column -> format, optional Format to use for parsing dates and/or times when used in conjunction with ``parse_dates``. The strftime to parse time, e.g. :const:`"%d/%m/%Y"`. See `strftime documentation `_ for more information on choices, though note that :const:`"%f"`` will parse all the way up to nanoseconds. You can also pass: - "ISO8601", to parse any `ISO8601 `_ time string (not necessarily in exactly the same format); - "mixed", to infer the format for each element individually. This is risky, and you should probably use it along with `dayfirst`. .. versionadded:: 2.0.0 dayfirst : bool, default False DD/MM format dates, international and European format. cache_dates : bool, default True If ``True``, use a cache of unique, converted dates to apply the ``datetime`` conversion. May produce significant speed-up when parsing duplicate date strings, especially ones with timezone offsets. iterator : bool, default False Return ``TextFileReader`` object for iteration or getting chunks with ``get_chunk()``. chunksize : int, optional Number of lines to read from the file per chunk. Passing a value will cause the function to return a ``TextFileReader`` object for iteration. See the `IO Tools docs `_ for more information on ``iterator`` and ``chunksize``. compression : str or dict, default 'infer' For on-the-fly decompression of on-disk data. If 'infer' and 'filepath_or_buffer' is path-like, then detect compression from the following extensions: '.gz', '.bz2', '.zip', '.xz', '.zst', '.tar', '.tar.gz', '.tar.xz' or '.tar.bz2' (otherwise no compression). If using 'zip' or 'tar', the ZIP file must contain only one data file to be read in. Set to ``None`` for no decompression. Can also be a dict with key ``'method'`` set to one of {``'zip'``, ``'gzip'``, ``'bz2'``, ``'zstd'``, ``'xz'``, ``'tar'``} and other key-value pairs are forwarded to ``zipfile.ZipFile``, ``gzip.GzipFile``, ``bz2.BZ2File``, ``zstandard.ZstdDecompressor``, ``lzma.LZMAFile`` or ``tarfile.TarFile``, respectively. As an example, the following could be passed for Zstandard decompression using a custom compression dictionary: ``compression={'method': 'zstd', 'dict_data': my_compression_dict}``. thousands : str (length 1), optional Character acting as the thousands separator in numerical values. decimal : str (length 1), default '.' Character to recognize as decimal point (e.g., use ',' for European data). lineterminator : str (length 1), optional Character used to denote a line break. Only valid with C parser. quotechar : str (length 1), optional Character used to denote the start and end of a quoted item. Quoted items can include the ``delimiter`` and it will be ignored. quoting : {0, 1, 2, 3}, default csv.QUOTE_MINIMAL Control field quoting behavior per ``csv.QUOTE_*`` constants. Use one of ``0`` or ``csv.QUOTE_MINIMAL``, ``1`` or ``csv.QUOTE_ALL``, ``2`` or ``csv.QUOTE_NONNUMERIC``, or ``3`` or ``csv.QUOTE_NONE``. Default is ``csv.QUOTE_MINIMAL`` (i.e., 0) which implies that only fields containing special characters are quoted (e.g., characters defined in ``quotechar``, ``delimiter``, or ``lineterminator``. doublequote : bool, default True When ``quotechar`` is specified and ``quoting`` is not ``QUOTE_NONE``, indicate whether or not to interpret two consecutive ``quotechar`` elements INSIDE a field as a single ``quotechar`` element. escapechar : str (length 1), optional Character used to escape other characters. comment : str (length 1), optional Character indicating that the remainder of line should not be parsed. If found at the beginning of a line, the line will be ignored altogether. This parameter must be a single character. Like empty lines (as long as ``skip_blank_lines=True``), fully commented lines are ignored by the parameter ``header`` but not by ``skiprows``. For example, if ``comment='#'``, parsing ``#empty\na,b,c\n1,2,3`` with ``header=0`` will result in ``'a,b,c'`` being treated as the header. encoding : str, optional, default 'utf-8' Encoding to use for UTF when reading/writing (ex. ``'utf-8'``). `List of Python standard encodings `_ . encoding_errors : str, optional, default 'strict' How encoding errors are treated. `List of possible values `_ . dialect : str or csv.Dialect, optional If provided, this parameter will override values (default or not) for the following parameters: ``delimiter``, ``doublequote``, ``escapechar``, ``skipinitialspace``, ``quotechar``, and ``quoting``. If it is necessary to override values, a ``ParserWarning`` will be issued. See ``csv.Dialect`` documentation for more details. on_bad_lines : {'error', 'warn', 'skip'} or Callable, default 'error' Specifies what to do upon encountering a bad line (a line with too many fields). Allowed values are: - ``'error'``, raise an Exception when a bad line is encountered. - ``'warn'``, raise a warning when a bad line is encountered and skip that line. - ``'skip'``, skip bad lines without raising or warning when they are encountered. - Callable, function that will process a single bad line. - With ``engine='python'``, function with signature ``(bad_line: list[str]) -> list[str] | None``. ``bad_line`` is a list of strings split by the ``sep``. If the function returns ``None``, the bad line will be ignored. If the function returns a new ``list`` of strings with more elements than expected, a ``ParserWarning`` will be emitted while dropping extra elements. - With ``engine='pyarrow'``, function with signature as described in pyarrow documentation: `invalid_row_handler `_. .. versionadded:: 2.2.0 Callable for ``engine='pyarrow'`` low_memory : bool, default True Internally process the file in chunks, resulting in lower memory use while parsing, but possibly mixed type inference. Because each chunk is type-inferred independently, the same literal can be parsed as an ``int`` in one chunk and a ``str`` in another; after concatenation the column has ``object`` dtype and contains both representations, so comparisons such as ``df["col"] == 12345`` or ``df["col"] == "12345"`` will match only a subset of the rows holding that value. A :class:`~pandas.errors.DtypeWarning` is emitted when this occurs. To ensure no mixed types either set ``False``, or specify the type with the ``dtype`` parameter. Note that the entire file is read into a single :class:`~pandas.DataFrame` regardless, use the ``chunksize`` or ``iterator`` parameter to return the data in chunks. (Only valid with C parser). memory_map : bool, default False If a filepath is provided for ``filepath_or_buffer``, map the file object directly onto memory and access the data directly from there. Using this option can improve performance because there is no longer any I/O overhead. float_precision : {'high', 'legacy', 'round_trip'}, optional Specifies which converter the C engine should use for floating-point values. The options are ``None`` or ``'high'`` for the ordinary converter, ``'legacy'`` for the original lower precision pandas converter, and ``'round_trip'`` for the round-trip converter. .. deprecated:: 3.1.0 All float precision modes now use the same converter. The ``float_precision`` argument will be removed in a future version. storage_options : dict, optional Extra options that make sense for a particular storage connection, e.g. host, port, username, password, etc. For HTTP(S) URLs the key-value pairs are forwarded to ``urllib.request.Request`` as header options. For other URLs (e.g. starting with "s3://", and "gcs://") the key-value pairs are forwarded to ``fsspec.open``. Please see ``fsspec`` and ``urllib`` for more details, and for more examples on storage options refer `here `_. dtype_backend : {'numpy_nullable', 'pyarrow'} Back-end data type applied to the resultant :class:`DataFrame` (still experimental). If not specified, the default behavior is to not use nullable data types. If specified, the behavior is as follows: * ``"numpy_nullable"``: returns nullable-dtype-backed :class:`DataFrame` * ``"pyarrow"``: returns pyarrow-backed nullable :class:`ArrowDtype` :class:`DataFrame` .. versionadded:: 2.0 Returns ------- DataFrame or TextFileReader A comma-separated values (csv) file is returned as two-dimensional data structure with labeled axes. See Also -------- DataFrame.to_csv : Write DataFrame to a comma-separated values (csv) file. read_csv : Read a comma-separated values (csv) file into DataFrame. read_fwf : Read a table of fixed-width formatted lines into DataFrame. Notes ----- Column labels read from a header row are always strings. To use column labels of another type, pass them explicitly with ``names`` and set ``header=0`` when the file contains a header row. Examples -------- >>> pd.read_table("data.csv") # doctest: +SKIP Name Value 0 foo 1 1 bar 2 2 #baz 3 Index and header can be specified via the `index_col` and `header` arguments. >>> pd.read_table("data.csv", header=None) # doctest: +SKIP 0 1 0 Name Value 1 foo 1 2 bar 2 3 #baz 3 >>> pd.read_table("data.csv", index_col="Value") # doctest: +SKIP Name Value 1 foo 2 bar 3 #baz Column types are inferred but can be explicitly specified using the dtype argument. >>> pd.read_table("data.csv", dtype={"Value": float}) # doctest: +SKIP Name Value 0 foo 1.0 1 bar 2.0 2 #baz 3.0 True, False, and NA values, and thousands separators have defaults, but can be explicitly specified, too. Supply the values you would like as strings or lists of strings! >>> pd.read_table("data.csv", na_values=["foo", "bar"]) # doctest: +SKIP Name Value 0 NaN 1 1 NaN 2 2 #baz 3 Comment lines in the input file can be skipped using the `comment` argument. >>> pd.read_table("data.csv", comment="#") # doctest: +SKIP Name Value 0 foo 1 1 bar 2 By default, columns with dates will be read as ``object`` rather than ``datetime``. >>> df = pd.read_table("tmp.csv") # doctest: +SKIP >>> df # doctest: +SKIP col 1 col 2 col 3 0 10 10/04/2018 Sun 15 Jan 2023 1 20 15/04/2018 Fri 12 May 2023 >>> df.dtypes # doctest: +SKIP col 1 int64 col 2 object col 3 object dtype: object Specific columns can be parsed as dates by using the `parse_dates` and `date_format` arguments. >>> df = pd.read_table( ... "tmp.csv", ... parse_dates=[1, 2], ... date_format={"col 2": "%d/%m/%Y", "col 3": "%a %d %b %Y"}, ... ) # doctest: +SKIP >>> df.dtypes # doctest: +SKIP col 1 int64 col 2 datetime64[ns] col 3 datetime64[ns] dtype: object rrLrM rrrs. rrrushr 8==?D !" U )  t$#% M KK  #T **r)rrrrc yrrrrrrrrrs rread_fwfr r)rrrrc yrrrs rrr rr)rrrrrc yrrrs rrr src | | td|dvr | td|&gd}}|D]}|j|||zf||z }|J|jd} | | tjurt | t |k7r|dk7rd} |jdB|jd} | dur-t | sd } n| tjusJt | } |jd %t | | zt |k7r td t|jd tjt||||d ||dzS)a Read a table of fixed-width formatted lines into DataFrame. Also supports optionally iterating or breaking of the file into chunks. Additional help can be found in the `online docs for IO Tools `_. Parameters ---------- filepath_or_buffer : str, path object, or file-like object Any valid string path is acceptable. The string could be a URL. Valid URL schemes include http, ftp, s3, and file. For file URLs, a host is expected. A local file could be: ``file://localhost/path/to/table.csv``. Certain URL schemes may require additional packages. For example, S3 URLs require the ``s3fs`` library. See :ref:`install.optional_dependencies` for a full list. If you want to pass in a path object, pandas accepts any ``os.PathLike``. By file-like object, we refer to objects with a ``read()`` method that accepts an optional size argument, such as a file handle (e.g. via builtin ``open`` function) or ``StringIO``. colspecs : list of tuple (int, int) or 'infer'. optional A list of tuples giving the extents of the fixed-width fields of each line as half-open intervals (i.e., [from, to] ). String value 'infer' can be used to instruct the parser to try detecting the column specifications from the first 100 rows of the data which are not being skipped via skiprows (default='infer'). widths : list of int, optional A list of field widths which can be used instead of 'colspecs' if the intervals are contiguous. infer_nrows : int, default 100 The number of rows to consider when letting the parser determine the ``colspecs``. Pass ``np.inf`` to infer column specifications from the entire file, which is useful when columns vary in width and a value wider than anything in the first ``infer_nrows`` rows would otherwise be truncated. iterator : bool, default False Return ``TextFileReader`` object for iteration or getting chunks with ``get_chunk()``. chunksize : int, optional Number of lines to read from the file per chunk. **kwds : optional Optional keyword arguments can be passed to ``TextFileReader``. Returns ------- DataFrame or TextFileReader A comma-separated values (csv) file is returned as two-dimensional data structure with labeled axes. See Also -------- DataFrame.to_csv : Write DataFrame to a comma-separated values (csv) file. read_csv : Read a comma-separated values (csv) file into DataFrame. Examples -------- >>> pd.read_fwf("data.csv") # doctest: +SKIP z&Must specify either colspecs or widths)Nrz4You must specify only one of 'widths' and 'colspecs'rrQrrSFrrTz-Length of colspecs must match length of namesr python-fwf)rrrXrr) rr$rrrrr rrr) rrrrrrrcolwrQ len_indexrSs rrr sVFNABB&6+=OPPA# A OOS#'N + 1HC     HHW E U#..8 u:X &8w+>Ixx $0!%+!6 E)' 2$%  )>>>$' N xx "*s5zI/EX/V !PQQHI   &" "     rceZdZdZ d ddZddZddZddZ ddZddZ d dd Z dd Z ddd Z ddd Z dd Z ddZy)ra Iterator over chunks of a delimited text file. Returned by :func:`read_csv`, :func:`read_table`, and :func:`read_fwf` when ``iterator=True`` or ``chunksize`` is given. Not intended to be constructed directly by users. A ``TextFileReader`` can be iterated over to yield successive chunks as :class:`DataFrame` objects, used as a context manager to ensure the underlying file handle is closed, or queried explicitly via :meth:`~TextFileReader.read` and :meth:`~TextFileReader.get_chunk`. Passed ``dialect`` overrides any of the related parser options. Parameters ---------- f : str, path object, or file-like object Source to read from. Accepts the same inputs as :func:`read_csv`. engine : {'c', 'python', 'pyarrow', 'python-fwf'}, optional Parser engine to use. If not specified, defaults to ``'python'``. **kwds Any keyword argument accepted by :func:`read_csv`, :func:`read_table`, or :func:`read_fwf`. See Also -------- read_csv : Read a comma-separated values (csv) file into a DataFrame. read_table : Read general delimited file into a DataFrame. read_fwf : Read a table of fixed-width formatted lines into a DataFrame. Examples -------- >>> with pd.read_csv("data.csv", chunksize=1000) as reader: # doctest: +SKIP ... for chunk in reader: ... process(chunk) >>> with pd.read_csv("data.csv", iterator=True) as reader: # doctest: +SKIP ... first = reader.get_chunk(5) Nc |d}nd}d}||_|jd||_t|t |t |}|'|dk(r t dt||}t||jdddk(r|jd d nd|d<||_d |_ |j|}|jd d|d <|jd d|_ |jd d|_ |j|||j||\|_|_d|vr|d|j d<d|_|j%||j|_y)NTpythonFengine_specifiedrz?The 'dialect' option is not supported with the 'pyarrow' enginerOrrQrrrrehas_index_names)rXr_engine_specifiedr_validate_skipfooter_extract_dialectr_merge_with_dialect_properties orig_options_currow_get_options_with_defaultspoprre_check_file_or_buffer_clean_optionsoptionshandles _make_enginer@)selffrXrrrrs r__init__zTextFileReader.__init__p sv  # F$  !%*<>N!O d#T""4(  " U2'4@D !$ ' 88Hg &' 1"&((7"3";QDN  11&9%)XX.?%F!" [$7[[$/  ""1f-$($7$7$H! dk  $.23D.EDLL* +)- ((DKK8 rc|j|jj|jjy)a Close the underlying file handle and parser engine. Called automatically when the ``TextFileReader`` is used as a context manager. See Also -------- TextFileReader.read : Read rows from the file into a DataFrame. TextFileReader.get_chunk : Read the next chunk of rows. Examples -------- >>> reader = pd.read_csv("data.csv", chunksize=5) # doctest: +SKIP >>> reader.close() # doctest: +SKIP N)rrr@rs rrzTextFileReader.close s." << # LL    rc|j}i}tjD]M\}}|j||}|dk(r,|tvr$||k7r|t |d|k7rt d|d|||<OtjD]K\}}||vr;||}|dk7r3||k7r.d|vr |tvrn!d|vr |tvrnt d|d|d|}|||<M|d k(r1tjD]\}}|j||||<|S) NrrzThe z2 option is not supported with the 'pyarrow' enginerrz" option is not supported with the z enginer) r r4rr_pyarrow_unsupportedgetattrrr_python_unsupportedr)rrXrrargnamedefaultrs rrz)TextFileReader._get_options_with_defaults sm  !0 5 5 7 % GWHHWg.E)#33W$WUGW== 7+%WX %GG  %!3 8 8 : % GW$W S=Ug%56)g=P.P"f,@T1T("7+-O%j1  $GG # %& \ !$1$7$7$9 > #'88GW#=  >rc t|dr]|j}|jjdd}|duxs|du}||k7r%|s"t |dd}t d|d|d|dyyy)Nr|rzThe specified reader encoding z is different from the encoding z of file r)hasattrr|r rrr)rrrX file_encodingorig_reader_encany_none file_paths rrz$TextFileReader._check_file_or_buffer s 1j !JJM"//33JEO$,G40GH/#Avt4  4_4EF))6y 1N9A/ "rc|j}d}|dk(r |ddkDrd}d}|d}| |dvrgd|d }d}n^t|d kDr |dk(r|d k(r d |d <|d=n=|dvr9d|d}d}n0d } t|jdd kDrd}|s |dvrd|d}d}|d}|>t |t t fr(t|d k(rt|dkDr |dvrd|d}d}|r|jr t||dk(rtD]}||=d|vr=tD]4}|r-||tj|k7rtd|d|d||=6|r(tjd|dt t#|d} |d} |d} |d } |d!} t%|d"| d ur td#t'| r(t | t(t*t,j.fs| g} | |d<| t)| n| } | 1t | t0s#t3d$t5| j6i} |d%}|d&k7}t9| ||'\} }|d&k(rt;| sF| Dtd(t;| r t=| } | t?} ntA| s t?| } | |d<| |d<| |d <||d)<| |d!<||fS#t$rd}Y8wxYw)*Nrrcrz*the 'c' engine does not support skipfooterrrM)rrzthe 'z"' engine does not support sep=NonerrTdelim_whitespacerrzI' engine does not support separators > 1 char, including regex separatorsrFz:the separator encoded in utf-8 is > 1 char long, and the 'z)' engine does not support such separatorsrwrzNord(quotechar) > 127, meaning the quotechar is larger than one byte, and the 'z)' engine does not support such quotecharsz,Falling back to the 'python' engine because z, but this causes z= to be ignored as it is not supported by the 'python' engine.z;; you can avoid this warning by specifying engine='python'.rrSrQrZrgrarOz)The value of index_col couldn't be 'True'z8Type converters must be a dict or subclass, input was a rhr)floatifyz@skiprows argument must be an integer when using engine='pyarrow' na_fvalues)!r{rencodeUnicodeEncodeErrorrrtbytesrr r_c_unsupportedrrrrrrrr0r3rrrr_r TypeErrorrr_clean_na_valuesrrrcallable)rrrXrfallback_reasonrL encodeablerwargrSrQrZrgrarhr*r+s rrzTextFileReader._clean_options s S=|$q("N!k" ;))$)&1S"T! X\}-1)*;'77F8$FF " J #szz'*+a/!&J&0H"H228:77 "K(  Z C<%HI!# NS(":: &x'PR " t55_- - S=% 3K  v * "vc{6H6L6LS6Q'Q$F*++=cWEQQ 3K   MMB&'(''+- K(  \* K( :&GH-.  HI I  "i$rzz)BC&K '{$0U e  !j$/##' #3#<#<"=? J""34Y& 0 !  : Y h'H,@ V(# ?5h'x= w)|'{)|%zv~]& #"  #s)J22 KKc` |jS#t$r|jwxYwr) get_chunk StopIterationrrs r__next__zTextFileReader.__next__ s/ >># #  JJL  s-c Jttttd}||vr t d|d|j dt |ts|dvxs%|dk(xr|jjddd v}|rd nd }t|||jjdd|jjd d|jjd d||jjdd|jjdd|_ |jJ|jj}n|dk7rdt|}t | |||fi|jS#t$r(|j|jjwxYw)N)rrrrzUnknown engine: z (valid options are )r)rr|r)NrrrrqrFr}rr)r|rqris_textrrrz)Invalid file path or buffer object type: )r5r8r1r7rkeysrrrrr-rhandlerr7r)rrrXmappingr=moders rrzTextFileReader._make_engine s  ")/ 0   "6(*>w||~>NaP !T"  88# QLL$$Z9P "3tD%))*d; LL,,]DA<<++L%@||''(98D $ 0 01BD I DL<<+ ++ ##A x =d1gYGCS/ ! "76?15 5 5 ||' ""$  s E111F"ct|r)rrs r_failover_to_pythonz"TextFileReader._failover_to_python s !$''rc4 |jdk(r |jj}|St d|} |jj|\}}}|T|rOt tt|j}t|j|j|z}nd}n t |}t|dr|jjdd nd t t rt#d}|j% n> :t' t(j*t(j,fvrt# fd}nd}|bi}|j/D]L\} } t'|| t(j*t(j,fvr|| nd} t1| || d || <Nn|}t3|||d }|xj|z c_ |S#t$r|j wxYw#t$r|j wxYw) a Read rows from the file into a :class:`DataFrame`. Advances the internal cursor by the number of rows returned, so successive calls yield later rows of the file. Used with readers created via :func:`read_csv` or :func:`read_table` with ``iterator=True`` or ``chunksize`` set. Parameters ---------- nrows : int, optional Number of rows to read. If ``None``, read until end of file. Ignored when the underlying engine is ``"pyarrow"``, which always reads the full file. Returns ------- DataFrame The parsed rows. See Also -------- TextFileReader.get_chunk : Read the next chunk of rows. read_csv : Read a comma-separated values (csv) file into a DataFrame. Examples -------- >>> with pd.read_csv("data.csv", iterator=True) as reader: # doctest: +SKIP ... df = reader.read() rreNrr rVcyrrrrrrtz%TextFileReader.read.. rurcSrrrwsrrtz%TextFileReader.read.. sIrFryr|)rXr@rr7rrrrrrr(rr"r rrrrrr!rrrrr$r') rredfrzr}rQnew_rowsrVrrrrrxs @rrzTextFileReader.read s> ;;) # \\&&(B y%We4E  LL%% }"4X__->(?#@AH&t||T\\H5LME Hu:t^, --11'4@  )T*#L1 Y'&< +B G,$$56 ! $NN,RDAq(a1bggrzz5JJa! '-Qe15&QLO R (  B LLH $L A      sGG<G9<Hc| |j}|j^|j|jk\rt||j|jz }n#t ||j|jz }|j |S)aK Read the next chunk of rows from the file. Convenience wrapper around :meth:`TextFileReader.read` that defaults ``size`` to the ``chunksize`` passed to :func:`read_csv` or :func:`read_table`, and respects a configured ``nrows`` limit by raising ``StopIteration`` once it has been reached. Parameters ---------- size : int, optional Number of rows to read in this chunk. Defaults to the ``chunksize`` passed when the reader was created. Returns ------- DataFrame The next chunk of parsed rows. Raises ------ StopIteration If ``nrows`` was set and has already been reached. See Also -------- TextFileReader.read : Read rows from the file into a DataFrame. read_csv : Read a comma-separated values (csv) file into a DataFrame. Examples -------- >>> with pd.read_csv("data.csv", iterator=True) as reader: # doctest: +SKIP ... first_five = reader.get_chunk(5) )re)rrerr8rr)rsizes rr7zTextFileReader.get_chunk, stF <>>D :: !||tzz)##|zzDLL04dll!:;yyty$$rc|Srrrs r __enter__zTextFileReader.__enter__Z s rc$|jyr)r)rexc_type exc_value tracebacks r__exit__zTextFileReader.__exit__] s rr)rz;FilePath | ReadCsvBuffer[bytes] | ReadCsvBuffer[str] | listrXrWrr)rr)rXr@rdict[str, Any])rXr@rr)rrRrXr@rz tuple[dict[str, Any], CSVEngine])rr')r)rz@FilePath | ReadCsvBuffer[bytes] | ReadCsvBuffer[str] | list | IOrXr@rr2)rerdrr')rJrdrr')rr )rNztype[BaseException] | NonerOzBaseException | NonerPzTracebackType | Nonerr)rrr__doc__rrrrrr9rrCrr7rLrQrrrrrG s&V$(19 F19!19  19f*,\ W%W/8W )Wr 0 K00  0d(cJ,%\,((   rc"d|d<t|i|S)a Converts lists of lists/tuples into DataFrames with proper type inference and optional (e.g. string to datetime) conversion. Also enables iterating lazily over chunks of large files Parameters ---------- data : file-like object or list delimiter : separator character to use dialect : str or csv.Dialect class / instance, optional Ignored if delimiter is longer than 1 character names : sequence, default header : int, default 0 Row to use to parse column labels. Defaults to the first row. Prior rows will be discarded index_col : int or list, optional Column or columns to use as the (possibly hierarchical) index has_index_names: bool, default False True if the cols defined in index_col have an index name and are not in the header. na_values : scalar, str, list-like, or dict, optional Additional strings to recognize as NA/NaN. keep_default_na : bool, default True thousands : str, optional Thousands separator comment : str, optional Comment out remainder of line parse_dates : bool, default False date_format : str or dict of column -> format, default ``None`` .. versionadded:: 2.0.0 skiprows : list of integers Row numbers to skip skipfooter : int Number of line at bottom of file to skip converters : dict, optional Dict of functions for converting values in certain columns. Keys can either be integers or column labels, values are functions that take one input argument, the cell (not column) content, and return the transformed content. encoding : str, optional Encoding to use for UTF when reading/writing (ex. 'utf-8') float_precision : str, optional Specifies which converter the C engine should use for floating-point values. The options are `None` or `high` for the ordinary converter, `legacy` for the original lower precision pandas converter, and `round_trip` for the round-trip converter. .. deprecated:: 3.1.0 All float precision modes now use the same converter. The ``float_precision`` argument will be removed in a future version. rrX)r)argsrs r TextParserrVf sjDN 4 (4 ((rc|!|rt}n t}t}||fSt|tr|j }i}|j D]6\}}t |s|g}|rt|tz}t||||<8|j Dcic]\}}|t|}}}||fSt |s|g}t||}|r |tz}t|}||fScc}}wr) rrrrr{rr _stringify_na_values_floatify_na_values)rgrhr*r+ old_na_valuesrrs rr1r1 s %IIU 6 j  5 It $!(  "'') =DAq?CF]*/8> ! 9SM%9 ^&3cnn+DUVV!S^^3DDMCNN" .5)"45'..y9 L ^i7H%H &BBKO@@I}M! !  CNN")[%[ #'  X#(  w)==HH^  )==GG^  )==GG^ ,  . .2  ,^9\N2NOPP &)D Krc|jdy|d}t|trB|tjvr,t dtj |}t||St|t d|}|S)za Extract concrete csv dialect instance. Returns ------- csv.Dialect or None rNz csv.Dialect)rrrtcsv list_dialectsr get_dialect_validate_dialect)rrtdialects rr r j sx xx "9oG'3Gs/@/@/B$B sw'?@(# O '" w/ Or)rMryrzr_rwrxcPtD]}t||rtd|dy)zx Validate csv dialect instance. Raises ------ ValueError If incorrect dialect is provided. zInvalid dialect z providedN)MANDATORY_DIALECT_ATTRSr"r)rparams rrnrn s4)Dw&/y BC CDrcv|j}tD]}t||}t|}|j ||}g}|||fvr4d|d|d|d}|dk(r|j dds|j ||r3tjdj|tt |||<|S) a Merge default kwargs in TextFileReader with dialect parameters. Parameters ---------- dialect : csv.Dialect Concrete csv dialect. See csv.Dialect documentation for more details. defaults : dict Keyword arguments passed to TextFileReader. Returns ------- kwds : dict Updated keyword arguments, merged with dialect parameters. zConflicting values for 'z': 'z+' was provided, but the dialect specifies 'z%'. Using the dialect-specified value.rMrbFz r) r{rqrr4rrr$rrjoinrr) rrrrr dialect_valparser_defaultprovided conflict_msgsrs rr r  s& ==?D("gu- (/88E>2  NK8 8*5'hZ@88C}E55 [(TXXne-L$$S)  MM M*MFVFX "U ;"< Krc|jdrJ|jds|jdr td|jdr tdyy)a Check whether skipfooter is compatible with other kwargs in TextFileReader. Parameters ---------- kwds : dict Keyword arguments passed to TextFileReader. Raises ------ ValueError If skipfooter is not compatible with other parameters. rcrrz('skipfooter' not supported for iterationrez''skipfooter' not supported with 'nrows'N)rr)rs rr r  sU xx  88J 488K#8GH H 88G FG G r)rr`rr^)rzMapping[str, Any]rr).)rrtrrrrbrr)rrtrr\rrbrrb)rrtrrdrrbrrd)r)rrtrzint | float | Nonerrbrrd)rQzSequence[Hashable] | Nonerr)r4FilePath | ReadCsvBuffer[bytes] | ReadCsvBuffer[str]rDataFrame | TextFileReader)rrb)rrrr^)r rtrOrdrarbrrb)rrbrrbrz list[float]) r rtrrbrrbrrbrz list[int])r1list[tuple[str, type[Warning]]]rr)r rtrrrrbrDataFrame | None) r rtrrrrbr1r|rr}) rrzrrrrdrUnpack[_read_shared[HashableT]]rr) rrzrr^rrbrr~rr) rrzrrrrrr~rr') rrzrr^rrdrr~rr{)ZrrzrLrKrMrKrOrNrQrPrSrRrTrHrVrUrXrWrZrYr\r[r]r[r_r^rar`rcrbrerdrgrfrhr^rir^rjr^rlrkrnrmror^rpr^rr^rrdrqr?rsrrrurtrvrrrwrtrxrbryr^rzrrr{rrr|rrr}rrrr~rrtrr^rr^rrrrrrrr{)rrzr&Sequence[tuple[int, int]] | str | NonerSequence[int] | Nonerrbrrrrdrr~rr)rrzrrrrrrbrr^rrbrr~rr)rrzrrrrrrbrrrrrr~rr')rrzrrrrrrbrr^rrdrr~rr{)rr)TT)rhr^r*r^)rz set[float])r*r^rzset[str | float])rr~rMrKrXrWrLrKrzstr | CallablerQrPrrRrrrvzstr | bytes | None)rz0dict[str, str | csv.Dialect | type[csv.Dialect]]rz&csv.Dialect | type[csv.Dialect] | None)rz%csv.Dialect | type[csv.Dialect] | strrr)rzcsv.Dialect | type[csv.Dialect]rrRrrR)rrRrr)rS __future__rr collectionsrrconcurrent.futuresrrrkr>rrrHrrtypingrrr r r r r rrrrnumpyrpandas._configr pandas._libsrpandas._libs.parsersrpandas.compat._cpur pandas.errorsrrrrrpandas.util._decoratorsrpandas.util._exceptionsrpandas.util._validatorsrrpandas.core.dtypes.commonrrr r!pandas.core.dtypes.dtypesr"pandas.core.dtypes.inferencer#rr$pandas.core.arraysr%r&pandas.core.framer'pandas.core.indexes.apir(r)pandas.core.internals.apir*pandas.core.internals.managersr+pandas.io.commonr,r-r.r/r0&pandas.io.parsers.arrow_parser_wrapperr1pandas.io.parsers.base_parserr2r3r4"pandas.io.parsers.c_parser_wrapperr5r6pandas.io.parsers.python_parserr7r8collections.abcr9r:r;r<r=typesr>pandas._typingr?r@rArBrCrDrErFrGrHrJrrrrrrr/r frozensetrrrrrrrrrrrrrrrrr r.r4rr6rr QUOTE_MINIMALrrIteratorrrVr1rYrXrr rqrnr r rrrrs  # 2    %.25  15(27F  $   -4y')"4E-4`L *&I -4CSWX }X#%67   I @+(#  ( K K K K W W89 &14@G0Q"LQ"Q"h4Rj  4, 3333 3  3lN 11 11 1hr.r. r.r.2 r.  r.j   L  ,     L  ,      # L  ,      %L%% % , %  % % H'*nn,/NN?q+BCq+DEq+H$Iq+JKq+LMq+NOq+PQq+RSq+TUq+VWq+XYq+Z[q+\ ]q+^:_q+bcq+fgq+hiq+jDkq+l+mq+n0oq+p qq+q+h  L  ,     L  ,      # L  ,      %L%% % , %  % % H'*nn,/NN?i+BCi+DEi+H$Ii+JKi+LMi+NOi+PQi+RSi+TUi+VWi+XYi+Z[i+\ ]i+^:_i+bci+fgi+hii+jDki+l+mi+n0oi+p qi+i+X 8;#& L 5  !       ,     8;#& L 5  !       ,     8;#&" L 5  !       ,     H8?#' zLz5z ! z  z  zz ,z zzz\S\\\~6)r"!J 6{ 9{){ { $ { ! { 5 {{0{'{| :+4 D3 ,333lHr