E geddlmZddlZddlmZmZddlZddlZddl m cm Z ddl mZddlmZddlmZddlmZmZddlmZmZmZmZdd lmZdd lmZddlm cm!Z"dd lm#Z#m$Z$dd l%m&Z&dd l'm(Z(ddl)m*Z*ddl+m,Z,m-Z-m.Z.ddl/m0Z0ddl1m2Z2ddl3m4Z4m5Z5m6Z6m7Z7m8Z8erddl9m:Z:m;Z;mZ>ddl?m@Z@GddZA d9d:dZBd9d;dZC d9dd?d)ZFd@d?d*ZGdAd-ZH d>dBd0ZIdCd5ZJdDd8ZKdS)E) annotationsN) TYPE_CHECKINGcast)PerformanceWarning)cache_readonly)find_stack_level)find_common_type maybe_promote)ensure_platform_intis_1d_only_ea_dtype is_integerneeds_i8_conversion)ExtensionDtype)notna) factorizeunique)factorize_from_iterable)ensure_wrapped_if_datetimelike) DataFrame)Index MultiIndex RangeIndex)concat)Series)compress_group_indexdecons_obs_group_idsget_compressed_idsget_group_indexget_group_index_sorter) ArrayLikeLevelnpt)ExtensionArray) FrozenListceZdZdZ d d!d Zed"d Zed#dZd$dZdZ ed%dZ ed&dZ d'dZ d(dZ d)dZed*dZed+dZdS), _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 Tindexrlevelr!sortboolreturnNonec||_||_||_|j||_d|jj|jvrdnd|_t|jj |_ t|jj |_ |j |j|_|j |j|_|j |j|_|jsbt#|jj|j}|j||_|j||_t'jd|j D}|jj}||z}|t'jt&jjkr,t1jd|dt4t7|dS)Nrcg|] }|j S)size).0 index_levels P/var/www/sysmax/venv/lib/python3.11/site-packages/pandas/core/reshape/reshape.py z'_Unstacker.__init__..sUUU ;+UUUz%The following operation may generate z& cells in the resulting pandas object.) stacklevel) constructorr)remove_unused_levelsr'_get_level_numberr(codesliftlistlevelsnew_index_levelsnamesnew_index_namespop removed_name removed_levelremoved_level_fullrtakenpmaxr2iinfoint32warningswarnrr_make_selectors) selfr'r(r9r) unique_codesnum_rows num_columns num_cellss r5__init__z_Unstacker.__init__ns' //11 Z11%88 tz/ ;;;AA !%TZ%6 7 7#DJ$455 044TZ@@!266tzBB"',tz":y Q!$*"24:">??L!%!3!8!8!F!FD &*&=&B&B<&P&PD # 6UUt?TUUUVV(- {*  rx))- - - M3 333"+--      r7-tuple[npt.NDArray[np.intp], list[np.ndarray]]c|j}t|jj}t|jj}|d|||dzdz||gz}t d|d|||dzdz||gzD}t ||\}}t|}t||} | |fS)Nr/c34K|]}t|VdSNlenr3xs r5 z2_Unstacker._indexer_and_to_sort..s(KKc!ffKKKKKKr7) r(r>r'r<r?tuplerrZr) rOvr<levsto_sortsizes comp_indexobs_idsngroupsindexers r5_indexer_and_to_sortz_Unstacker._indexer_and_to_sorts JTZ%&&DJ%&&)eAEGGn,az9KKd2A2ha!egg&>$q'&JKKKKK0%@@ Gg,,(W==r7list[np.ndarray]cF|j\}|jrfd|DS|S)Nc:g|]}|Sr1rG)r3linerfs r5r6z,_Unstacker.sorted_labels..s%;;;4DIIg&&;;;r7)rgr))rOrarfs @r5 sorted_labelsz_Unstacker.sorted_labelss94 9 <;;;;7;;; ;r7values np.ndarraycZ|jr#|j\}}tj||d}|S|S)Nr)axis)r)rgalgostake_nd)rOrnrf_ sorted_valuess r5_make_sorted_valuesz_Unstacker._make_sorted_valuess9 9 !2JGQ!M&'BBBM  r7c:|j}|jdd}td|D}t||\}}t |}t |}|jj|j|j z}||f|_ |jd||zz|j z}tj tj |j t} | |d| t |jkrt#d||_| |_|jr.|tj||_dStjtj|dd|_dS)Nr.c34K|]}t|VdSrXrYr[s r5r]z-_Unstacker._make_selectors..s(77qCFF777777r7dtypeTz0Index contains duplicate entries, cannot reshape) return_indexr/)r@rmr^rrZr r'levshaper(r= full_shaperHzerosprodr*putsum ValueError group_indexmaskr) searchsortedarange compressorr) rO new_levelsremaining_labels level_sizesrcrdrestrideselectorrs r5rNz_Unstacker._make_selectorssl*  -crc277J77777 01A;OO Gg,,(44 $TZ049<!6/%b)FZ,??$)Kx00=== 4   88::DJ ' 'OPP P% 9 S(55bi6H6HIIDOOO gbi &N&N&Nq&QRRDOOOr7cNt|jSrX)r*rall)rOs r5mask_allz_Unstacker.mask_allsDIMMOO$$$r72tuple[npt.NDArray[np.intp], npt.NDArray[np.bool_]]ctjt|jtj}||d\}}||dfS)Nryr. fill_valuer)rHrrZr'intpget_new_valuesany)rO dummy_arr new_valuesrs r5 arange_resultz_Unstacker.arange_resultsTIc$*ooRW=== ..yR.HH D488A;;&&r7rc.|jdkr|ddtjf}| |jddkrt d|||\}}||}|j}|||||j S)Nr/z-must pass column labels for multi-column data)r'columnsrz) ndimrHnewaxisshaperrget_new_columns new_indexr9rz)rOrn value_columnsrrtrr's r5 get_resultz_Unstacker.get_results ;!  AAArzM*F  V\!_%9%9LMM M'' ;; &&}55 %     r7Nc |jdkr|ddtjf}||}|j\}}|jd}||z}||f}|j} |j} | rlt|r]| ||| dd |} tj |t} | | fS|j } | r|j } tj|| } nt| t r3| }||| } || dd<n>t'| |\} }tj|| } | || j}tj|t} t/|j r+|d}| d} n||d}t5j|| d|||| | dt/|j r>| d} t9| } | |j } | | fS) Nr/ryi8Fcopyu1zM8[ns])rrHrrvr}rrrrZreshapeswapaxesonesr*rzempty isinstancerconstruct_array_type_emptyr fillnamer~rviewastype libreshapeunstackr)rOrnrrulengthwidthr result_width result_shaperrrnew_maskrzclsrs r5rz_Unstacker.get_new_valuess ;!  AAArzM*F0088  av~  - y=  (F ( %%feV<<!Q&&  w|4888Hx' '   ,LE,e<<.Zs%MMM#*--MMMr7Fr?r<rAverify_integrity)r=rE_renamerDinsert _na_valuerenamerZrHrepeatrrrr?rFrAr<r _repeaterappendtile) rOrlevrrr new_names new_codesrepeaterrs @r5rz_Unstacker.get_new_columnsEsz  yA~~)11t7H1III$++AD4F4P+QQC::d/00 0T'((494M""Yry//88  mZ 0 0 %&-'1J&+t/@.BBIMMMM9LMMMII'J'+T->?I# I> 511222YiRW    r7cDt|jt|jkr=|j|j}|jrt j|dd}n8t|j|jz}t j||jz }|S)Nrr.)rZrFrE get_indexerr=rHrr)rOrrs r5rz_Unstacker._repeaterks t& ' '3t/A+B+B B B.::4;MNNHy 69Xq"55+,,ty8Fy((494Hr7cfdjddD}tjdkrjd|d}}|dkr(|t||j}||jdStj|jdS)NcDg|]}|jSr1)rGr)r3rrOs r5r6z(_Unstacker.new_index..}s'UUUc11UUUr7r.r/rFr) rmrZr@rrrrGrrBr)rO result_codesr( level_codess` r5rz_Unstacker.new_indexzsVUUUT=OPSQSPS=TUUU  t$ % % * *!%!6q!9<?;Er!&&(( B SZZAA::k**11$2Fq2IJJ J(&"     r7)T)r'rr(r!r)r*r+r,)r+rU)r+rh)rnror+ro)r+r*)r+r)r+rrX)rr)r+ro)r+r)__name__ __module__ __qualname____doc__rTrrgrmrvrNrrrrrrrr1r7r5r&r&@sh++\JN,,,,,\   ^ &^ SSS8%%%^%'''^'     M$M$M$M$^$ $ $ $ L   ^    ^   r7r&TdataSeries | DataFramer)r*ctdkr|S|jttjvrgfdDfdt jD}fdD}fdD}fdD}fd|D}fd|D} fd |D} td |D} t|| d d } t| d \} }t| || |d }|st|d}nt||gz| | gz| dgzd }t|tr;|}||_|d||}|}|}|}nt|jtrC|}r=d|||}fdD=|S|d }||_|d||}t|tr|jn|jttsJjdg|z}|jjg|z}jdg}|fd|Dt|||d }t|tr||_n||_|S)Nrc:g|]}|Sr1)r;r3ir's r5r6z%_unstack_multiple..s' 7 7 7AU $ $Q ' ' 7 7 7r7cg|]}|v| Sr1r1)r3rclocss r5r6z%_unstack_multiple..s ? ? ?1Qr7c*g|]}j|Sr1r?rs r5r6z%_unstack_multiple.....1u|A...r7c*g|]}j|Sr1r<rs r5r6z%_unstack_multiple.. , , ,ek!n , , ,r7c*g|]}j|Sr1rArs r5r6z%_unstack_multiple..rr7c*g|]}j|Sr1rrs r5r6z%_unstack_multiple..rr7c*g|]}j|Sr1rrs r5r6z%_unstack_multiple..rr7c*g|]}j|Sr1rrs r5r6z%_unstack_multiple..rr7c34K|]}t|VdSrXrYr[s r5r]z$_unstack_multiple..s(**Q#a&&******r7F)r)xnullr))r__placeholder__rrrr)c*g|]}|kr|n|dz Sr/r1)r3r_vals r5r6z%_unstack_multiple..s)@@@Qa#gg1q5@@@r7deepc3XK|]$}|jdV%dS)r.N)rGr<)r3recunstcolss r5r]z$_unstack_multiple..s6NN#(."455NNNNNNr7)rZr'rrrArangenlevelsr^rrrrrrrrrrCr?rr<extend)rrrr)rlocsclevelsccodescnamesrlevelsrcodesrnamesrrcomp_idsrd recons_codes dummy_indexdummy unstackedrrrresultdummy_df new_columnsr'rrs ` @@@r5_unstack_multipler s 5zzQ  JE U # #E   7 7 7 7 7 7 7E ? ? ? ?em,, ? ? ?E.......G , , , ,e , , ,F , , , ,e , , ,F.......G , , , ,e , , ,F , , , ,e , , ,F **'*** * *E!&%e5IIIK,[uEEEHg''5&PUVVVL   G*;<<<  gY&H:%-.."    $"O ! MM"3 QUMVV    dlJ / / F Aiill NN@@@@%@@@ A M99%9(($$$ *4%   i ( ( ) HH (H(J/////oa()G3 \&'&0 ^A&' NNNNNNNNNN)eK)V$$(% '  r7objct|ttfr.t|dkrt ||||S|d}t |s |dks|j|t|trHt|jtrt||||S|j dSt|jts%tdt|jdt|jrt#|||| St%|j||j| }||jd| S) Nr/rrrT) future_stackz'index must be a MultiIndex to unstack, z was passedrr(r9r)rr)rr^r>rZr r r'r;rr_unstack_frameTstackrtyper rz_unstack_extension_seriesr&_constructor_expanddimr_values)rr(rr) unstackers r5rrs%%'' u::??%S%JTRRR R!HE e  +U.?%?%? ##E***#y!! ci , , 2!#u$OOO O5;;D;11 1  : . .  Rd39oo R R R    sy ) ) P,S%$OOO O IU0JQU   ## Kt $   r7rr+c6t|jtsJt|j||j|}|js8|j||}|||j S| |j |j |S)Nrr)axesr) rr'rr& _constructor_can_fast_transpose_mgrr_constructor_from_mgrrrrr)rr(rr)rmgrs r5rr s ci , ,,,, C,<4I  " hyZ@@((38(<<<## Ks{z$   r7seriesrc|}||||}|jdg|_|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)r"r(rr)dfr s r5rrsJ8   B ZZe Z F FF^77<.stack_factorizeMsD ? 0")CJJ/// /3E::z5  r7) level_numr(r)c:g|]}|Sr1r)r3rKs r5r6zstack..^s#@@@sSZZ]]@@@r7Frrc"g|] \}}|j Sr1)r)r3rtcols r5r6zstack..}s999C999r7r')rrr;rr_stack_multi_columnsr'r>r?r<rrHrravelrArzipmaprr_is_homogeneous_typedtypesrrr_concat_same_typeitems"_reorder_for_extension_array_stackr_constructor_sliced)r'r(r(r)r-Nr.rrclevclabrrr?ilabr<r:rzarrrrr1s @r5rrCs!!! ;DAq //66I%-,, # YvD     EK , , %+,-- @@@@ek.?@@@ $_U]33 d$q))//11222*++ +,,,YiRW    #C%+u}9U$V$VW t Aa 0 0 6 6 8 88;#U]%78"     ;+55+el*++q  e^ , , /,,..C..995;;==999J.s- 7 7#3%-% % 7 7 7 7 7 7r7)r(r)c3@K|]}t|tVdSrX)rint)r3rs r5r]z!stack_multiple..s, 3 3cZS ! ! 3 3 3 3 3 3r7cDg|]}j|Sr1)rr;rFs r5r6z"stack_multiple..s)GGG#0055GGGr7rc*g|]}|kr|n|dz Srr1)r3r_rs r5r6z"stack_multiple..s)===!!s((QQA===r7zTlevel should contain all level names or all level numbers, not a mixture of the two.)rrrCr)r'r(r(r)r rs` @r5stack_multiplerKs 7 7 7 7 7 7 777  B BC63vDAAAFF B 3 3U 3 3 3 3 3 HGGGGGG >))A,,C63vDAAAF>===u===E  > 1   Mr7rrct|jdkr,|jd|jdSdt |jdd|jddD}t |}dt j|D}t |}tj dt ||jD|jdd S) zBCreates a MultiIndex from the first N-1 levels of this MultiIndex.rrrc2g|]\}fd|DS)c0g|]}|dkr|ndS)rNr1)r3crs r5r6z8_stack_multi_column_index...s)333166Qt333r7r1)r3r<rs @r5r6z-_stack_multi_column_index..sA    C 4333U333   r7Nr.c3 K|] \}}|V dSrXr1)r3keyrts r5r]z,_stack_multi_column_index..s&AAVS!SAAAAAAr7cLg|]!\}}d|vrt||jn|"S)Nry)rrz)r3new_levrs r5r6z-_stack_multi_column_index..sL   047/B/BE' + + + +   r7r) rZr?rrAr7r< itertoolsgroupbyr from_arrays)rr`tuples unique_tuplesnew_levss r5_stack_multi_column_indexrZs 7>a~a ((gmA.>(???  gnSbS17="3EFF   D $ZFAAy'8'@'@AAAMM"H  !  !$Hgn = =   mCRC    r7r.rHc $%dd}|d}|j}t|tsJ||jdz krW|}t ||jdz D]3}|||} ||dz|} || | }4|x|_}|s,|r*|d |} || d }|j}tt|}t|} i} |j d }t|j d }|rtj|}|t#|d}tj||}t#|%g}| D]} |j|}n%#t($r||Y@wxYwt|t,st#|}n|j|jz }|%krb|jdd|j|f}||jj d |_|| j}n|jdd|f}t;|j$t$t@r$!"$fd |#D}|j$\}}tj%||z&||j'(}||}n|j}|j)dkr|(}|| |<t#|d kr| *|} t#|}t|j+trKtY|j+j }tY|j+j-}%fd|j+j D}n=t]|j+\}} | g}|/%g}|j+j0g}|||tj1||||jj-|t|||d}!|2| |!| }"|jjdkrO|j3|g }#|"j4|#s|"|#}"|r|"5d d}"|"S)Nr.rHrrc2||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`. r)r.rs r5_convert_level_numberz3_stack_multi_columns.._convert_level_numbers$  % %=+ +r7Frr/r)r(rqr.)rcNg|]!\}}|jd"S)Fr)rr)r3rtr\rzs r5r6z(_stack_multi_columns..s2TTTTQQY%%e%%88TTTr7c:g|]}|Sr1r0)r3rlevsizes r5r6z(_stack_multi_columns..2s%EEESSZZ((EEEr7r)r'rr)rqhow)r.rHrr)6rrrrrr swaplevel _is_lexsorted sort_indexrrZr?rr<rHr)rrZrGget_locKeyErrorrslicestopstartlocreindexrnilocr r:tolistrrr;r<rrrrr6r differencer'r>rArrrrrr%equalsr()&r'r.r(r)r]thismi_cols roll_columnsrlev1lev2 level_to_sortr new_data level_valsrlevel_vals_nanlevel_vals_used drop_colsrQrj slice_lenchunk value_slicesubsetr?r1idxrrr old_codes old_levelsrr desired_columnsrzr`s& @@r5r5r5sm     ::5: ! !DlG gz * ****GOa''' y'/A"566 > >A((L99D((Q ==D'11$==LL!-- w  " "t.-a99 ];;,:w''G+G44KH#Jr*++K +gk** &&s:==Ngnk::O+GI'$'$ ,&&s++CC      S ! ! ! H #u%% -CII39,I   HQQQ S 112E*// 0CB0GHHEM---@@GKKYqqq#v&F$V]%9%9%;%;<>)..s33 $m  a  %++--K#  9~~!,,Y77  D A$*j))&$*+,, )** EEEEDJ4DEEE 7 C C : \ %%g../ Z_% j!!! RW[!,,--- U](3444)eI   ;  O OF }q  -;;YKHHOOQQ~$$_55 -O,F2A511 MsF--GGrCr#n_rows n_columnsctj||z||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='.|sCCCqAUNNNNNr7r.r/c3\K|]&}|vrtntdV'dSrX)nextrg)r3rgenr(s r5r]zstack_v3..sO##%ZZS U4[[######r7r)rrzF)use_na_sentinelcTg|]$}tj|t%Sr1)rHrrZ)r3r<r's r5r6zstack_v3..s+KKKUBIeSZZ00KKKr7rrr4))rnuniquerZrsortedr%rrrHargsort_reorder_ilevelsrriterr^rjrrrrrrrrrzrorr'rr?r>rr<rdrop_duplicatesrArrrGrrl)r'r( drop_levnums stack_colssorterordered_stack_colsstack_cols_uniqueordered_stack_cols_uniquebufrrcolumn_indexerr ratior r index_levels index_codesr<uniques column_levels column_codeslen_df n_uniquesrfidxsrs`` @r5stack_v3rus? }#em"4"444STTT%...L22CCCCE%-/00CCCDDbDIJ 5zzA~~E""'88@@'"))++ 2 9 9 ; ; C  u}   " "::<rs""""""))))))))),,,,,,222222444444 544444,,,,,,&&&&&&&&&CBBBBBCCCCCC'''''' .-----%%%%%%6 211111555555K K K K K K K K ^ DHXXXXXv! ! ! ! ! J:>     """""JJBJBJBJBJBZB8NRAAAAAH""""Jhhhhhhr7