我在Excel中有一些非常大的数据集需要解析 - 在数组中执行它比循环遍历工作表中的数据更快。将所有数据加载到数组中会导致内存问题(数据集很大),因此我计划将数据的子集加载到数组中,处理该数据,然后加载另一个子集。我希望使用定义LBound和UBound的数组“功能”来帮助我跟踪我在工作表中的位置。但我发现将工作表值分配给数组会更改边界。 以下代码演示了问题...
Sub myTest3()
Dim myRange As Range
Dim myArray As Variant
Dim myOffset As Long
myOffset = 10
Set myRange = Worksheets("RawData").Range("A1").CurrentRegion
ReDim myArray(myOffset To myRange.Rows.Count, myRange.Columns.Count)
MsgBox LBound(myArray, 1) & " to " & UBound(myArray)
Set myRange = myRange.Offset(myOffset, 0).Resize(myRange.Rows.Count - myOffset, myRange.Columns.Count)
myArray = myRange.Value2
MsgBox LBound(myArray, 1) & " to " & UBound(myArray)
End Sub
第一个MsgBox给了我“10到10931”。 第二个MsgBox给了我“1到10921”。
关于维护数组边界的任何想法都是我最初定义它们的吗?我知道在工作表中循环以使赋值可以做到,但它会很慢。
提前致谢。
答案 0 :(得分:4)
在这种情况下,Excel VBA无法以您希望的方式运行。执行myArray = myRange.Value2
时,myArray
的原始内容已被替换。 Redim
med数组被丢弃了。 Excel / VBA不会查看目标,它会替换它,或者更准确地说,它会创建一个新数组并使myaArray
变量指向该目标。
因此,您需要更多代码才能获得您想要的目标。我会考虑把代码放到一个单独的函数中,然后在那里进行簿记:
Function ChunkAtOffset(rng As Range, rowsInChunk As Long, colsInChunk As Long, offsetRows As Long) As Variant
' Note: doesn't cater for the case where there are fewer than 'offsetRows' in the target
Dim arr As Variant, result As Variant
Dim r As Long, c As Long
arr = rng.offset(offsetRows).Resize(rowsInChunk, colsInChunk).Value2
ReDim result(offsetRows To offsetRows + rowsInChunk - 1, 1 To colsInChunk)
For r = 1 To rowsInChunk
For c = 1 To colsInChunk
result(offsetRows - 1 + r, c) = arr(r, c)
Next
Next
ChunkAtOffset = result
End Function
如果我这样做:
Sub myTest4()
Dim curReg As Range, ary As Variant, offset As Long
With Range("A1")
Set curReg = .CurrentRegion
Do
ary = ChunkAtOffset(.CurrentRegion, 10, .CurrentRegion.Columns.Count, offset)
Debug.Print LBound(ary, 1) & " to " & UBound(ary)
offset = offset + 10
Loop Until offset >= .CurrentRegion.Rows.Count
End With
End Sub
......我现在明白了:
0 to 9
10 to 19
20 to 29