Una formula di array dinamico - Suggerimenti per Excel

Sommario

Nota

Questo è uno di una serie di articoli che descrivono in dettaglio le soluzioni inviate per la sfida Podcast 2316.

Lo YouTuber Rico S ha adottato un approccio completamente diverso. La sua tabella è generata da una singola formula di matrice dinamica.

Un array dinamico

Il totale in O non si riempie. Se avessi Charles Williams Speed Tools V4, potresti racchiudere la formula di Rico in TOTALI come mostrato di seguito.

SpeedTools V4

Postata qui, senza ulteriori commenti, è la formula di Rico in I13:

=CHOOSE((1,2,3,4,5,6),LEFT(INDEX($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee"),INT(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1)/4+1),MOD(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1),4)+1),FIND(":",INDEX($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee"),INT(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1)/4+1),MOD(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1),4)+1),1)-1),RIGHT(INDEX($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee"),INT(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1)/4+1),MOD(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1),4)+1),LEN(INDEX($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee"),INT(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1)/4+1),MOD(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1),4)+1))-FIND(":",INDEX($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee"),INT(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1)/4+1),MOD(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1),4)+1),1)),INDEX(UglyData,MATCH(LEFT(INDEX($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee"),INT(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1)/4+1),MOD(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1),4)+1),FIND(":",INDEX($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee"),INT(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1)/4+1),MOD(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1),4)+1),1)-1),$A$3:$A$9,0),MATCH(RIGHT(INDEX($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee"),INT(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1)/4+1),MOD(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1),4)+1),LEN(INDEX($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee"),INT(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1)/4+1),MOD(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1),4)+1))-FIND(":",INDEX($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee"),INT(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1)/4+1),MOD(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1),4)+1),1)),$A$3:$Z$3,0)+1),INDEX(UglyData,MATCH(LEFT(INDEX($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee"),INT(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1)/4+1),MOD(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1),4)+1),FIND(":",INDEX($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee"),INT(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1)/4+1),MOD(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1),4)+1),1)-1),$A$3:$A$9,0),MATCH(RIGHT(INDEX($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee"),INT(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1)/4+1),MOD(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1),4)+1),LEN(INDEX($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee"),INT(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1)/4+1),MOD(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1),4)+1))-FIND(":",INDEX($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee"),INT(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1)/4+1),MOD(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1),4)+1),1)),$A$3:$Z$3,0)+2),INDEX(UglyData,MATCH(LEFT(INDEX($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee"),INT(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1)/4+1),MOD(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1),4)+1),FIND(":",INDEX($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee"),INT(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1)/4+1),MOD(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1),4)+1),1)-1),$A$3:$A$9,0),MATCH(RIGHT(INDEX($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee"),INT(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1)/4+1),MOD(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1),4)+1),LEN(INDEX($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee"),INT(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1)/4+1),MOD(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1),4)+1))-FIND(":",INDEX($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee"),INT(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1)/4+1),MOD(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1),4)+1),1)),$A$3:$Z$3,0)+3),INDEX(UglyData,MATCH(LEFT(INDEX($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee"),INT(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1)/4+1),MOD(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1),4)+1),FIND(":",INDEX($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee"),INT(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1)/4+1),MOD(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1),4)+1),1)-1),$A$3:$A$9,0),MATCH(RIGHT(INDEX($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee"),INT(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1)/4+1),MOD(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1),4)+1),LEN(INDEX($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee"),INT(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1)/4+1),MOD(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1),4)+1))-FIND(":",INDEX($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee"),INT(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1)/4+1),MOD(SEQUENCE(COUNTA($A$4:$A$9&":"&FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),1,0,1),4)+1),1)),$A$3:$Z$3,0)+4))

Aggiornare! Sopra, ho usato la funzione HTOTAL di Charles Williams per migliorare la formula di Rico. Rico si è tuffato nella documentazione per il componente aggiuntivo FastExcel e ha trovato due straordinarie funzioni: SETMEM e GETMEM. È possibile ottimizzare le formule archiviando e recuperando i risultati di espressioni ad alta intensità di calcolo utilizzate più di una volta in una formula. Questa è una funzione innovativa.

Rico lo ha usato per ridurre la sua formula di 6.866 caratteri a una formula di 593 caratteri:

=HTOTALS(CHOOSE((1,2,3,4,5,6),SETMEM(LEFT(SETMEM(LISTDISTINCTS(TRANSPOSE($A$4:$A$9)&":"&TRANSPOSE(FILTER($A$3:$Z$3,LEFT($A$3:$Z$3,8)="Employee")),,,FALSE),"A"),FIND(":",GETMEM("A"),1)-1),"B"),SETMEM(RIGHT(GETMEM("A"),LEN(GETMEM("A"))-FIND(":",GETMEM("A"),1)),"C"),INDEX(UglyData,MATCH(GETMEM("B"),$A$4:$A$9,0)+1,MATCH(GETMEM("C"),$A$3:$Z$3,0)+1),INDEX(UglyData,MATCH(GETMEM("B"),$A$4:$A$9,0)+1,MATCH(GETMEM("C"),$A$3:$Z$3,0)+2),INDEX(UglyData,MATCH(GETMEM("B"),$A$4:$A$9,0)+1,MATCH(GETMEM("C"),$A$3:$Z$3,0)+3),INDEX(UglyData,MATCH(GETMEM("B"),$A$4:$A$9,0)+1,MATCH(GETMEM("C"),$A$3:$Z$3,0)+4)))

Maggiori informazioni su SETMEM e GETMEM qui.

Charles Williams ha fatto il check-in e ha suggerito questa formula:

=HTOTALS(VSTACK(HSTACK(,A4:A9,G3,SLICES(A4:AB9,0,H4:K4)),HSTACK(,A4:A9,L3,SLICES(A4:AB9,0,M4:P4)),HSTACK(,A4:A9,Q3,SLICES(A4:AB9,0,R4:U4)),HSTACK(,A4:A9,V3,SLICES(A4:AB9,0,W4:Z4))))

Il 19 marzo 2020, dopo il debutto della funzione LET in Excel, Rico S ha inviato questa formula più breve. Ciò attualmente richiede Office 365 con Insiders Fast.

=LET( _splitChar,"~", _categories,$A$4:$A$9, _colHdrs,$A$3:$Z$3, _employees,FILTER(_colHdrs,LEFT(_colHdrs,8)="Employee"), _unique2DTable,_categories&_splitChar&_employees, _cntE,COUNTA(_employees), _cnt,COUNTA(_unique2DTable), _uniqueList,INDEX(_unique2DTable,INT(SEQUENCE(_cnt,1,0,1)/_cntE+1),MOD(SEQUENCE(_cnt,1,0,1),_cntE)+1), _category,LEFT(_uniqueList,FIND(_splitChar,_uniqueList,1)-1), _employee,RIGHT(_uniqueList,LEN(_uniqueList)-FIND(_splitChar,_uniqueList,1)), _row,MATCH(_category,_categories,0)+1, _col,MATCH(_employee,_colHdrs,0), _Q1,INDEX(UglyData,_row,_col+1), _Q2,INDEX(UglyData,_row,_col+2), _Q3,INDEX(UglyData,_row,_col+3), _Q4,INDEX(UglyData,_row,_col+4), _totalCol,INDEX(UglyData,_row,_col), _finalTable,CHOOSE((1,2,3,4,5,6,7),_category,_employee,_Q1,_Q2,_Q3,_Q4,_totalCol), SORT(_finalTable,2) )