In support of these Threads
' https://www.ozgrid.com/forum/index.p...ng-a-2d-array/
' https://www.excelforum.com/excel-pro...dim-array.html
https://www.excelforum.com/tips-and-...ml#post5408376





I am not totally sure what the OP is asking.
Is the OP asking
(i) _ to put values into an existing array where that existing array already has values in it
or
(ii)_ changing the array dimension and positioning of elements in an array
or
(iii)_ maybe its lost in the translation and/ or the OP is not sure him/herself.
The initial answer to (i)_ I think we seem clear about:- It will likely in VBA require a code line for each element to be “moved” from one array to the other , so likely looping will be involved for a multi element array.
The Thread title and OPs first question infers to me converting a 1 D array to a 2 D array, without looping.
If the existing array with values already in it is a dynamic array, then overwriting along with re dimensioning means that those (i)_ and (ii)_ are somewhat merged in meaning anyway.
So I am not totally clear what is going on here, but I think it there is a discussion of generally … …”1 D arrays to 2 D arrays”
So lets say we are talking generally about …”1 D arrays to 2 D arrays” and leave it loosely defined for now and go with that…

Frederick has shown in his second code line that a characteristic of the Transpose function is that if a 1 D array is given to the Transpose function then the transposed array becomes a 2 D array , all be it a quasi “1 column array” ***
Transpose does that, as it does the opposite way converting a single column 2D array to a 1D array.
I think most of us are not quite sure why it has been wired to do that. Some other things seem to default to making a “one row” thing be a 1D array rather than a 2D array, even when the thing it may have been given to work on was a 2D array. ( It does not screw things up to badly when playing with spreadsheets since that transposed in its final 1 D form will be “seen” by Excel as if it was a single row 2 Dimensional array when applied to a spreadsheet range. So usually a “row” becomes a row, if you catch my drift).
We can go the other way. ( If we do that with Rick’s example , we will see a small difference, the 1 D array returned will have indices of 1 2 3 4 5 as opposed to the 0 1 2 3 4 , (since the Split function Rick used returns those starting a base 0 ) . I am not sure why Excel chooses to start a t 1 in this case: Possibly it was just made that way because its more often to do with worksheet/spreadsheet stuff, and we think about rows and columns starting at 1, and something like a row of 1 is a bit stupid. )

Index with arrays as co ordinate arguments
This stuff is worth knowing about:
A further function that can be very helpful in doing this sort of manipulation of arrays without looping is the Index Function. It becomes so useful because it will accept arrays in place of the more conventional single value indices in its second ( row ) and third ( column ) arguments. The evaluation is then done in the conventional Excel way, “along the columns of a row” , then down to repeat at the next row: along the columns of that row, then down to repeat at the next row: along the columns of that row, then down to repeat at the next row: along the columns of that row , ….etc. Usually VBA will do its best to give out the results in an array dimensioned appropriate for the array dimensions supplied in those second and third arguments, following the conventional “along the columns of a row” , then down to repeat at the next row: along the columns of that row, ………

As example we can do that Transpose code line in this pseudo way
Code:
'                                        Index(OneDimensionalArray(),  1                           ,       1
'                                                                      1                                   2
'                                                                      1                                   3
'                                                                      1                                   4
'                                                                      1                                   5               )
We are doing 5 calculations there, talking each time the first row and consecutive columns, the result coming out in a form that the Excel calculations are done - .. “along the columns of a row” , then down to repeat at the next row… but we only have one column in this case, so that is actually just going down the rows, 5 times. Hence our output is the 90degree transpose of OneDimensionalArray()

That was just one example, but the important point is that you can supply different arrays in the Index second ( “row” ) and third ( “column” ) arguments. So you can pretty well take any1 or 2 D array in the Index first argument, and in one code line, without looping , put all or some of the values from that array in some other order in any other 1 or 2 D array. That could be what the OP was asking for ….
Dim Array1(2, 2) As Integer
Dim Array2(2) As Integer
…………… way to copy the values from Array2 into Array1?
The restriction is that we can’t make use of this to put values into Array1( ) if it already existed. You would have to be in like having
Dim Array1() As Variant
Dim Array2(2) As Integer
-……..
Array1()= Index ( Array2(2) , { _.... } , { _... } )
( Variant is needed in the first declaration as the index chucks its output values housed in Variant types. AFAIK the first argument can be any sort of 1 D or 2 D array, ( or it can be any range object ) )

Another not looping option to assist in a conversion could be to remove rows or columns of a 2 D array with a single code line. Best look at some posts of Rick ( Frederick Rothstein ‘s ) , stuff for that ( https://excelfox.com/forum/showthrea...-Variant-Array )


One last curiosity , a weird thing I only recently came across. An array of arrays, sometimes refereed as a “jagged array”, is peculiarly treated in some cases by Index as a 2 D array. This gives us some interesting further one liner code line possibilities.
Example, If I had a 1 D array of 1 D arrays, something of this sort of form
{ { “Head1” , 2, 3 } , {“Head3”, 4, 5 } , {“Haed2”, 7, 9} }
then I can convert that, for example, to re ordered in data columns like this
Code:
'  Head1  ,  Haed2  ,  Head3
'    2    ,    7    ,  4
'    3    ,    9    ,  5
I can do that using like a Index one code liner pseudo
Code:
'                                Index(  Head1 , 2, 3          1  ,    3   ,   2                  1  ,  1   ,  1
'                                        Head3 , 4, 5          1  ,    3   ,   2                  2  ,  2   ,  2
'                                        Haed2 , 7, 9          1  ,    3   ,   2                  3  ,  3   ,  3          )


I put some more details of all I have been saying , in a macro in the uploaded file. Probably its best to step through the macro in Debug mode ( do that by hitting Key F8 after clicking anywhere in the macro )



Quote Originally Posted by vba_php View Post
....to be honest with you I've never seen your type of question asked in 20 years of writing code my lifetime. ....
Hello Adam.
I expect you are referring specifically to the idea of putting existing values from an array into another existing array, although I am not fully clear if the OP wanted that: Possibly the language barrier prevented the OP getting anything out of the links you gave him…. The best thing probably, as Rory asked for, was an example from the OP of what he wanted to do…
Anyway, you probably know all the following, but I thought I’d add it to the Thread, while I am in the mood…
Generally questions along the lines of “1 D array to 2 D array” or visa versa are quite common in Excel VBA. I expect this is because
_ a) a lot of things done “internally” in coding involve 1 D arrays,
but/ and
_ b) a range from a spreadsheet will often likely end up in an array of 2 Dimensions, I think Excel does this so that we can make the distinction what is a row and what is a column.***
So things might not always work as we wanted, for example a problem might occur when a 1 D array appears when a 2 D array was expected/ wanted, and visa versa. To solve the problem a conversion from a 1D to 2D or visa versa might get us out of trouble.
Example: we got a Join function that is something like the reverse of the Split function mentioned in this Thread . Basically you can use it to join the contents of an array into a string. The bummer is that it only accepts a 1 D array. So if I give it a column or row of data to join it will error. You’ll need to change the 2D array got from a spreadsheet single row or a spreadsheet single column to a 1D array for join to work on it. ( One way you can do that is with some of the one liner codings I been talking about – I added a example for you in the uploaded macro ### )

***I suppose a 2 D array does not really have “rows” and “columns”, it simply has 2 dimensions. But Excel conventionally puts a spreadsheet row into the fist dimension, and a spreadsheet column into the second dimension. So after using Excel VBA arrays a lot you often get to think of a 2 D array in terms of like arr(row, column) or in terms of orientation like arr(horizontal, vertical). Its just a convenient frame of reference perception.
A 1 D array has no orientation. I can’t really perceive that unless I have drunk a lot of Jack Daniels, as the world starts spinning around, then it becomes very clear, relatively speaking. I suppose Excel can’t get drunk, and as mentioned, a 1 D array seems to be often regarded as like a 2 D array of first dimension size of 1, or pseudo 1 “row” 2 D array.

Molly







Adam, I have definitely had random occurrences of an error like you mentioned, all be it very rarely. When it has happened , I was pretty damm sure it shouldn’t have happened.
I think we all agree that Activateing and Selecting when dealing with worksheet ranges via VBA is rarely needed and is usually a bad idea as the interaction with a spreadsheet slams the brakes on.

I will usually optimise a macro first, with no Activateing and Selecting , ignoring the odd error of that sort you mentioned.
After that I will often see if I don’t compromise the performance much if I add an occasional code line pair of something like
Worksheet("x").Activate: Worksheet("x").Range("A1").Select
Or, if dealing with multiple open workbooks,
Workbooks("x“).Activate: Worksheet("x").Activate: Worksheet("x").Range("A1").Select
at some strategic points.

A typical point would be just before I start doing things to ranges in Worksheet("x") via VBA. I know those two ( three ) code lines should be unnecessary. But it’s been my experience that they help stop that occasional error.
I have no idea what causes the occasional error when all suggest it should not error. I think possibly Excel has some memories of what was last active. Possibly that can become corrupted, and doing a quick Worksheet("x").Activate: Worksheet("x").Range("A1").Select refreshes it.

One thing that has already been touched on here in the Thread a couple of times, which has caught me out a few times: Selecting a range does not activate the worksheet of the range you select.
If the worksheet is not active and you try to select that range then you will get that error.
But selecting a worksheet does activate that worksheet. (Activateing and Selecting a worksheet do something similar, - I think the main difference being that you can select things, but only activate a thing. I have not explored that much yet… )

Quote Originally Posted by vba_php View Post
…but based on the millions of tests that I ran, it became evident that this line of code automatically made the book active:
Code:
wbDrawings.SaveAs (ThisWorkbook.Path & Application.PathSeparator & "temp.csv")
.....
I would hazard a guess that that might be version dependent and possibly unreliable, as Rory suggested. That dose not consistently activate the workbook being saved, for me.

Molly