iddlmZddlZddlmZmZmZddlZddlZ ddl m Z ddl m cmZddlmZmZddlmZddlmZddlmZmZdd lmZmZmZmZdd lm Z dd l!m"Z"m#Z#ddl$m%cm&Z'dd l$m(Z(m)Z)dd l*m+Z+ddl,m-Z-ddl.m/Z/ddl0m1Z1ddl2m3Z3m4Z4m5Z5ddl6m7Z7ddl8m9Z9ddl:m;Z;mZ>m?Z?erddl@mAZAmBZBmCZCddlDmEZEddlFmGZGGddZH d' d(dZIed)d*dZJe d) d+dZJ d' d+dZJ d' d,dZK d-dZL d. d/d ZMd0d1d!ZNd2d"ZO d. d3d#ZP d4d$ZQd5d%ZR d6d&ZSy)7) annotationsN) TYPE_CHECKINGcastoverload) get_option)Pandas4WarningPerformanceWarning)cache_readonly)find_stack_level)find_common_type maybe_promote)ensure_platform_intis_1d_only_ea_dtype is_integerneeds_i8_conversion)ExtensionDtype)isnanotna) factorizeunique)NDArrayBackedExtensionArray)factorize_from_iterable)ensure_wrapped_if_datetimelike) DataFrame)Index MultiIndex default_index)concat)Series)compress_group_indexdecons_obs_group_idsget_compressed_idsget_group_indexget_group_index_sorter) ArrayLikeLevelnpt)ExtensionArray) FrozenListceZdZdZ d ddZe ddZeddZddZddZ eddZ eddZ dd Z dd Z dd Zedd ZeddZy ) _Unstackera Helper class to unstack data / pivot with multi-level index Parameters ---------- index : MultiIndex level : int or str, default last level Level to "unstack". Accepts a name for the level. fill_value : scalar, optional Default value to fill in missing values if subgroups do not have the same set of labels. By default, missing values will be replaced with the default fill value for that data type, NaN for float, NaT for datetimelike, etc. For integer types, by default data will converted to float and missing values will be set to NaN. constructor : object Pandas ``DataFrame`` or subclass used to create unstacked response. If None, DataFrame will be used. Examples -------- >>> index = pd.MultiIndex.from_tuples( ... [("one", "a"), ("one", "b"), ("two", "a"), ("two", "b")] ... ) >>> s = pd.Series(np.arange(1, 5, dtype=np.int64), index=index) >>> s one a 1 b 2 two a 3 b 4 dtype: int64 >>> s.unstack(level=-1) a b one 1 2 two 3 4 >>> s.unstack(level=0) one two a 1 3 b 2 4 Returns ------- unstacked : DataFrame cV||_||_|j|_|jj ||_d|jj |j v|_|jxr |j}|rdnd|_t|jj|_ t|jj|_ |jj|j |_|jj|j |_|j|j |_d|_|jst'|jj |j }|jr(|dk(}||}t)j*|d|_|j j-||_|j"j-||_t/drt1d|jD}|j j2} || z} | t)j4t(j6j0kDr(t9j:d| dt<t?|jAy) Nrperformance_warningsc34K|]}|jywN)size).0 index_levels X/app/cer_product_mecsu/.venv/lib/python3.12/site-packages/pandas/core/reshape/reshape.py z&_Unstacker.__init__..sU ;++Usz%The following operation may generate z& cells in the resulting pandas object. stacklevel)! constructorsortremove_unused_levelsindex_get_level_numberlevelcodeshas_nanliftlistlevelsnew_index_levelsnamesnew_index_namespop removed_name removed_levelremoved_level_fullunique_nan_indexrnp flatnonzerotakermaxr2iinfoint32warningswarnr r _make_selectors) selfr<r>r9r: should_lift unique_codesnan_masknum_rows num_columns num_cellss r5__init__z_Unstacker.__init__ys' //1 ZZ11%8 TZZ--djj99 ll0tyy $A! !%TZZ%6%6 7#DJJ$4$45 0044TZZ@!2266tzzB"',,tzz":%'yy'-djj.>.>tzz.J'KL||'2-+XI6 (*x(@(C%!%!3!3!8!8!FD &*&=&=&B&B<&PD # , - Ut?T?TUUH,,11K!;.I288BHH-111 ;I;G67&/1  c|j}t|jj}|js|Dcgc]}t |d}}t|jj }|d|||dzdz||gz}td|d|||dzdz||gzD}t||\}}t|} t|| } | |fScc}w)Nrr.c32K|]}t|ywr1lenr3xs r5r6z2_Unstacker._indexer_and_to_sort..sKc!fK) r>rBr<r?r:rrCtupler"rar$) rUvr?codelevsto_sortsizes comp_indexobs_idsngroupsindexers r5_indexer_and_to_sortz_Unstacker._indexer_and_to_sorts JJTZZ%%&yy49:DYt_Q':E:DJJ%%&)eAEGn,az9Kd2Aha!eg&>$q'&JKK0%@ Gg,(W=;sCc|j\}}|jr|Dcgc]}|j|c}S|Scc}wr1)ror:rN)rUrnrilines r5 sorted_labelsz_Unstacker.sorted_labelss=44 993:;4DIIg&; ; r]c@|j}|jdd}td|D}t||\}}t |}t |}|j j|j|jz}||f|_ |jd||zz|jz}tjtj|jt} | j!|d| j#t |j kr t%d||_| |_|j*r*|j-tj.||_ytj*tj2|dd|_y)Nr-c32K|]}t|ywr1r`rbs r5r6z-_Unstacker._make_selectors..s7qCF7rddtypeTz0Index contains duplicate entries, cannot reshape) return_indexr.)rDrrrer"rarr<levshaper>r@ full_shaperArLzerosprodboolputsum ValueError group_indexmaskr: searchsortedarange compressorr) rU new_levelsremaining_labels level_sizesrkrlrmstrideselectorrs r5rTz_Unstacker._make_selectorssD**  --cr27J77 01A;O Gg,(4 $$TZZ04<<?!6/%%b)FZ,??$))Kxx0= 4 88:DJJ 'OP P% 99(55bii6HIDO ggbii &Nq&QRDOr]cHt|jjSr1)rrall)rUs r5mask_allz_Unstacker.mask_allsDIIMMO$$r]ctjt|jtj}|j |d\}}||j dfS)Nr}r- fill_valuer)rLrrar<intpget_new_valuesany)rU dummy_arr new_valuesrs r5 arange_resultz_Unstacker.arange_resultsMIIc$**oRWW= ..yR.H D488A;&&r]c|j}|jdk(r|ddtjf}||jddk7r t d|j ||\}}|j|}|j}|j||||jd} t|tjr|j|j} } nBt|tr-|jj|jj} } nd\} } | | ur%| j j#|j | S)Nr.z-must pass column labels for multi-column dataF)r<columnsr~copy)r.)_valuesndimrLnewaxisshaperrget_new_columns new_indexr9r~ isinstancendarraybaser_ndarray_mgradd_references) rUobj value_columnsrrwrrxrr<resultrnew_bases r5 get_resultz_Unstacker.get_results ;;! ArzzM*F  V\\!_%9LM M++FJ? A&&}5!! eWJJ * 1 ! $1%$D!z% ~~U2 V*'7'9 "*- V*'7'9 ,e->?I# I>> 512YiRW  Ns)F-c(t|jt|jk7rJ|jj|j}|jrt j |dd}|St|j|jz}t j||jz }|jrL|js@|jdkDsJdd||j<||jdzdxxxdzccc|S)Nrr-z+`unique_nan_index` not properly initializedr.) rarJrI get_indexerr@rLrrrAr:rK)rUrrs r5rz_Unstacker._repeaters t&& '3t/A/A+B B..::4;M;MNH||99Xq"5++,t||;Fyy(4994H||DII,,r1A135../..245:5r]cn|jr|jdd}n9|j}t|jj }|d|||dzdz}|Dcgc]}|j |j}}t|jdk(ry|jd|d}}|dk(jr%|jt||j}|j |j|jdSt|j||jdScc}w)Nr-r.rFr)r:rrr>rBr<r?rNrrarDrrrrrFr)rUlabelsrfr?r result_codesr> level_codess r5rz_Unstacker.new_indexs 99'',F A))*E2AYq1uw/F=CDc1D D t$$ % *!%!6!6q!9<?;Er!&&( SZA::k*11$2F2Fq2IJ J((&&"   Es"D2)T)r<rr>r&r:rreturnNone)rz-tuple[npt.NDArray[np.intp], list[np.ndarray]])rzlist[np.ndarray])rw np.ndarrayrr)rr)rr)rz2tuple[npt.NDArray[np.intp], npt.NDArray[np.bool_]])rrr1)rz Index | None)rr)rMultiIndex | Index)__name__ __module__ __qualname____doc__r\r rorrrzrTrrrrrrrr]r5r+r+Js,^JN77(-7BF7 7r    ,  S8%%''4]$~! F.  r]r+ct|dk(r|S|j}tt|}||jvr|g}|Dcgc]}|j |}}t |jDcgc] }||vs| }}|Dcgc]}|j|}}|Dcgc]}|j|}}|Dcgc]}|j|} }|Dcgc]}|j|} }|Dcgc]}|j|} }|Dcgc]}|j|} }td|D} t|| dd}t|d\}}t||| |d}|st|dd}ntg| |g| |g| dd }t|t r5|j#d }||_|j%d|| }|}| }|}nt|j&trG|}|rA|j)d}|j%||| }|Dcgc]}||kr|n|d z }}|rA|S|j#d }||_|j%d|| }t|t r |jn |j&ttsJjdg|}|j&j*g| }jdg}|j-fd |Dt|||d }t|t r ||_|S||_|Scc}wcc}wcc}wcc}wcc}wcc}wcc}wcc}wcc}w)Nrc32K|]}t|ywr1r`rbs r5r6z$_unstack_multiple..s*Q#a&*rdF)r:xnullr:)r__placeholder__)rrrdeeprr:r.c3ZK|]"}|jjd$yw)r-N)rNr?)r3recunstcolss r5r6z$_unstack_multiple..s#N#(.."45Ns(+)rar<rrrEr=rangenlevelsrCr?rer#r r!rrrrrrrGrextend)dataclocsrr:r<irlocsclevelsccodescnamesrlevelsrcodesrnamesrrcomp_idsrl recons_codes dummy_indexdummy unstackedrrrrvalrfdummy_df new_columnsrs @r5_unstack_multiplers 5zQ  JJE U #E  16 7AU $ $Q ' 7E 7emm, ?1Q ?E ?(-.1u||A.G.&+ ,ekk!n ,F ,&+ ,ekk!n ,F ,(-.1u||A.G.&+ ,ekk!n ,F ,&+ ,ekk!n ,F , *'* *E!&%e5IK,[uEHg''5&PUVL G*;%H  &W&g&%F%H%.F.-."  $ u %! MM"3 QUMV    dllJ /Fiil JT(;@@Qa#g1q50@@M99%9($$$ *4%  i ( H ((H(J///ooa(373 \\&&00 ^^A&' NNN)eK)V$%  (  _ 8 ?. , ,. , ,JAs<L-4 L2>L2L7$L<MM8M M%Mcyr1rrr>rr:s r5rr,sPSr]cyr1rrs r5rr0sr]ct|ttfr"t|dk7rt ||||S|d}t |s |dk(s|j j|t|trCt|j trt||||S|jjSt|j ts"tdt|j dt|j rt#||||St%|j ||j&|}|j)|d| S) Nr.rrrz'index must be a MultiIndex to unstack, z was passedrr>r9r:rr)rrerBrarrr<r=rr_unstack_frameTstackrtyperr~_unstack_extension_seriesr+_constructor_expanddimr)rr>rr: unstackers r5rr6s%%' u:?%S%JTR R!HE e U.?%? ##E*#y! cii ,!#u$O O55;;=  : .5d399o5Fk R   syy ),S%$O O IIU0J0JQU ##Ct #SSr]cHt|jtsJt|j||j|}|j s:|j j||}|j||jS|j||j|S)Nrr)axesr) rr<rr+ _constructor_can_fast_transposerr_constructor_from_mgrrrr)rr>rr:rmgrs r5rrZs cii ,, , C,<,<4I  " "hhyZ@((388(<<## s{{z$  r]c|j}|j|||}|jjdg|_|S)an Unstack an ExtensionArray-backed Series. The ExtensionDtype is preserved. Parameters ---------- series : Series A Series with an ExtensionArray for values level : Any The level name or number. fill_value : Any The user-level (not physical storage) fill value to use for missing values introduced by the reshape. Passed to ``series.values.take``. sort : bool Whether to sort the resulting MuliIndex levels Returns ------- DataFrame Each column of the DataFrame will have the same dtype as the input Series. )r>rr:r)to_framerr_drop_level_numbers)seriesr>rr:dfrs r5rrksE8  B ZZe Z FF^^77.stack_factorizes; ??"))CJ// /3E:z5  r]) level_numdropnar:FrstrictTrr<)rrr=rr_stack_multi_columnsr<rBrCr?rrrLrravelrErzipmapr_is_homogeneous_typedtypesrrr_concat_same_typeitems"_reorder_for_extension_array_stackr_constructor_sliced)framer>r-r:r+NKr,rrrclevclabrrrCilabr?r5r~arrrxcolrrs r5rrs! ;;DAq //6I%--,# YvD   EKK ,%++,,- .3kk.?.?@sSZZ]@ @$U]]3 d$q)//12**+ ++,YiRW  # 5;; "> ? HL  t Aa 0 6 6 88;;##U]]%7%78"   ;;555ell**+q  e^ ,,,.C..+0;;=9C9J.s 7#3%--%% % 7s"%)r-r:c3<K|]}t|tywr1)rint)r3rs r5r6z!stack_multiple..s 3cZS ! 3srr.zTlevel should contain all level names or all level numbers, not a mixture of the two.)rrrr=rGr)r:r>r-r:rrrfs` r5stack_multiplerFs 7 77 BC63vDAF B: M/ 3U 3 3BGG#005GG))A,C63vDAF8==!!s(QA-=E= M  1  H>s "B7B<c .t|jdkr,|jdj|jdSdt |jdd|j dddD}t |d di}d t j|D}t |d di}tjt ||jdddDcgc]"\}}d|vrt||j n|$c}}|jdd Scc}}w) zBCreates a MultiIndex from the first N-1 levels of this MultiIndex.rrrc3`K|]!\}}|Dcgc]}|dk\r||ndc}#ycc}ww)rNr)r3rr?cs r5r6z,_stack_multi_column_index.. s9  C.3316Qt #3 3s .) .Nr-T)r.r.c3&K|] \}}| ywr1r)r3keyrxs r5r6z,_stack_multi_column_index..sAVS!SAsr}rE) rarCrrEr2r? itertoolsgroupbyr from_arraysrr~)rrhtuples unique_tuplesnew_levsnew_levrs r5_stack_multi_column_indexrTs 7>>a~~a ((gmmA.>(?? gnnSb17=="3EdS D $ $t $FAy'8'8'@AMM/$/H  ! !!$HgnnSb.A$ O 047/BE' + O mmCR   s'D c  dd}|jd}|j}t|tsJ||jdz k7rP|}t ||jdz D])}|||} ||dz|} |j | | }+|x|_}|js*|r(|d|} |j| d}|j}tt|}t|} i} |jd}t|jd}|rtj|}|j!t#|d}tj$||}t#|}g}| D]} |jj'|}t|t,s t#|}n|j.|j0z }||k7rk|j2dd|j|f}|j%|jjd|_|j5|j6}n|j8dd|f}t;|j<j?}t|t@r|jCjE|jGDcgc]"\}}|jHjK|d $c}}}|jL\}}tjN||zjQ||jRjQd}|j%|}n |j6}|jTdkDr|jW}|| |<t#|dkDr| jY|} t#|}t|jZtrqt]|jZj} t]|jZj^}!|jZjD"cgc]}"|"ja|}#}"nDtc|jZ\}$}%|%g} |$ja|g}#|jZjdg}!| j+||#j+tjf|||!j+|jj^|t| |#|!d }&|ji| |&| }'|jjdkDrJ|jjk|gj}(|'jjm|(s|'|(}'|r|'jodd }'|'S#t($r|j+|Y wxYwcc}}wcc}"w)Nc@||jvr|j|S|S)z Logic for converting the level number to something we can safely pass to swaplevel. If `level_num` matches a column name return the name from position `level_num`, otherwise return `level_num`. rL)r,rs r5_convert_level_numberz3_stack_multi_columns.._convert_level_number&s%  %==+ +r]Frr.r)r>rtr-)rrr)r<rr)rthow)r,rErr)8rrrrrr swaplevel _is_lexsorted sort_indexrrTrCrr?rLr:rrarNget_locKeyErrorrslicestopstartlocreindexrwilocr r5tolistrrr6r7rrrrrrrr1 differencer<rBrErrrrrr$equalsr-))r:r,r-r:rWthismi_cols roll_columnsrlev1lev2 level_to_sortr new_data level_valsrlevel_vals_nanlevel_vals_usedlevsize drop_colsrKra slice_lenchunk value_slicesubsetr~rxrcr;r<idxrrrr old_codes old_levelsrrdesired_columnss) r5r0r0#s  ::5: !DllG gz ** *GOOa'' y'//A"56 >A(L9D(Q =D'11$=L  > ".- w  "t.a9 ];,,:w'G+G4KH#Jr*+K ggk* &&s:=Nggnk:O+GI'$ ,,&&s+C#u%CI399,I  HHQ S 112E*// 0C0CB0GHEM---@GGKYYq#v&F$V]]%9%9%;>rB)..s3 $mm   a %++-K# O'$R 9~!,,Y7  D A$**j)$**++, ))* 48JJ4D4DESSZZ(E E 7 C : \ %%g./ ZZ__% j! RWW[!,- U]]((34)eI   ;  OF }}q --;;YKHOOQ~~$$_5O,FA51 MS    S !  0U,Fs1U#'U< VU98U9ctj||zj||jjd}|j |S)a Re-orders the values when stacking multiple extension-arrays. The indirect stacking method used for EAs requires a followup take to get the order correct. Parameters ---------- arr : ExtensionArray n_rows, n_columns : int The number of rows and columns in the original DataFrame. Returns ------- taken : ExtensionArray The original `arr` with elements re-ordered appropriately Examples -------- >>> arr = np.array(["a", "b", "c", "d", "e", "f"]) >>> _reorder_for_extension_array_stack(arr, 2, 3) array(['a', 'c', 'e', 'b', 'd', 'f'], dtype='>> _reorder_for_extension_array_stack(arr, 3, 2) array(['a', 'd', 'b', 'e', 'c', 'f'], dtype='t3|j&d \} } | g}t+t-j.| d|f} t|dkDr9t-j4|} t|tsJ|j7| } n|} | j9}t| tr'| j(}| j;j0}n|g}t3|d dg}|D cgc]!} t-j<| t|#}} t||z| |z|j&j>t+| j>zd |_t|}t|}t-j@|}t-j.||z|t-j<t-j@||z}|jC|}|jDd k(rd|jjt|k(rBt|jdk(rtG|j& }n|jHdddf}|jDdk(rd|_%|Scc}wcc} w) Nz8Columns with duplicate values are not supported in stackr.r-rF)orderr}F)use_na_sentinelrrr/)&rnuniquerarsetr$rrrrr4rblocksr~r9rr stack_reshaperr<rCrBrLrr?rargsort_reorder_ilevelsrdrop_duplicatesrrErrNrrrcr)r:r> set_levelsk stack_colsr~rratio index_levels index_codesr?uniquessorterordered_stack_colsordered_stack_cols_unique column_levels column_codeslen_df n_uniquesrnidxss r5stack_v3rsq }}#emm"44STT u: UJ22%--//!3R<Tq@STJ emmZ 0 u}}  !e&@&@JJ%%a(..EE** MM ! !"C ! 0+ ueZDA#f+U";E%++z*{{)) 2775;;#4#4q%jAB "5;;Fwy 27751e*56  5zA~E"*j111'88@' 2 9 9 ;$j1*11 )99;AA 23 !";USTUVW ?KKUBIIeSZ0KLKm+L(kk$'9'?'?"@@ FLZF-.Iii "G 776G#V ,ryy69JI/V VD [[ F{{aEMM11SZ? v~~ ! #&,,/F[[A&F {{a MQ U`Ls P,P,6&P1c t|jtsJt|d}g}|j D],}t |jdk(r|j d}nat |dk(r|f}t| t fdt|jjD}|jdd|f}t ||jjkr!|jj||_nI|jdk(r:|jdk(rd|_n#tt |j|_|j!|/t |dkDr|j"st%|d } npt ||jjkr*|jj|j } ndg} t'| |j(j* } t ||jjkrI|jj|j } | jj-| s| | } | S) a3Reshape the data of a frame for stack. This function takes care of most of the work that stack needs to do. Caller will sort the result once the appropriate index is set. Parameters ---------- frame: DataFrame DataFrame that is to be stacked. level: list of ints. Levels of the columns to stack. set_levels: set of ints. Same as level, but as a set. stack_cols: Index. Columns of the result when the DataFrame is stacked. Returns ------- The data of behind the stacked DataFrame. T)reverser.Frc3RK|]}|vr tn td ywr1)nextr^)r3rgenrs r5r6z stack_reshape..Ds,#*_S %+=#s$'Nr) ignore_index)rr~)rrrsortedrrariterrerrrar$rrrrrrrrr~rf) r:r>rr drop_levnumsbufrwrcolumn_indexerrr rzrs ` @r5rrs0 emmZ 00 0%.L C  " u}}  "::5:)D5zQfs)C"#u}}445#N99Q./D u: -- -<<;;LIDL   1 $yyA~ ,S->?  4+. 3x!|EKK$/ u: -- ---;;LIPPRK#K;emm6I6IJ 5zEMM)))--;;LIPPR~~$$_5O,F Mr])NT)rSeries | DataFramer:r)..)rrr:rrr)rrr:rrr)rrr:rrr)r%rr:rrr)r-TT)r:rr-rr:rrr)TT)r:rr-rr:r)rrrr) r:rr,rEr-rr:rrr)r@r(r|rEr}rErr()r:rr> list[int]rr) r:rr>rrzset[int]rrrr)T __future__rrMtypingrrrrRnumpyrLpandas._config.configrpandas._libs.reshape_libsrr pandas.errorsrr pandas.util._decoratorsr pandas.util._exceptionsr pandas.core.dtypes.castr r pandas.core.dtypes.commonrrrrpandas.core.dtypes.dtypesrpandas.core.dtypes.missingrrpandas.core.algorithmscore algorithmsrurrpandas.core.arrays._mixinsrpandas.core.arrays.categoricalrpandas.core.constructionrpandas.core.framerpandas.core.indexes.apirrrpandas.core.reshape.concatrpandas.core.seriesrpandas.core.sortingr r!r"r#r$pandas._typingr%r&r'pandas.core.arraysr(pandas.core.indexes.frozenr)r+rrrrrrFrTr0r8rrrr]r5rsM" ,))34 5 '&CBC' .% 25~ ~ D DH^ ^<@^B S S AD :>  CG!T !T;?!T!TJ:>  26  "$ $-1$$PCGNB NB(,NB;?NBNBb"J8NRA A!$A37AFJAAH" "!$"14""JOdF F&F4<FJOFFr]