W
Walter Briscoe
I am using Excel 2003 on Windows Vista
I have a file of London fare data. (Shows fares between origin and
destination stations.)
It consists of about 600 rows and columns.
Because Excel 2003 only supports 256 columns in a sheet, I need to split
my data over more than one sheet.
The data ought to be symmetrical and mostly is. However, I have found
one pair of stations where the AB price and BA price are different.
It takes several elapsed days to assemble that file from the Internet.
I would like to identify asymmetric data to assess the extent of that
asymmetry. i.e. I assume AB price = BA price is probably correct.
I want to assemble an "array" in memory.
I can read data from a single sheet with code like this:
Dim PricesA_D As Variant
PricesA_D = Sheets("A-D").Range("B2", _
Sheets("A-D").Cells.SpecialCells(xlCellTypeLastCell).Address)
' The first row and column contain station names
I think I need a picture.
Sheets("A-D") might start
Abbey Wood Acton Central Acton Main Line ...
Abbey Wood X 1.23 2.34
Acton Central 3.21 X 3.45
Acton Main Line 2.34 3.45 X
....
I have shown asymmetric data between Abbey Wood and Acton Central.
Sheets("E-K") station names might start
Ealing Broadway Ealing Common Earls Court ...
Abbey Wood
Acton Central
Acton Main Line
....
I want an "array" in which the E-K data is appended to the A-D data.
I tried
Dim Prices As Variant
Prices = Application.Union( _
Sheets("A-D").Range(
"B2", _
Sheets("A-D").Cells.SpecialCells(xlCellTypeLastCell).Address), _
Sheets("E-K").Range(
"B2",
Sheets("E-K").Cells.SpecialCells(xlCellTypeLastCell).Address), _
Sheets("L-R").Range(
"B2",
Sheets("L-R").Cells.SpecialCells(xlCellTypeLastCell).Address), _
Sheets("S-Z").Range(
"B2",
Sheets("S-Z").Cells.SpecialCells(xlCellTypeLastCell).Address))
I got Run-time error '1004':
Method 'Union' of object '_Application' failed
I assumed that was because Union tries to create a sheet with more than
256 columns.
To test that assumption, I tried
Prices = Application.Union( _
Sheets("A-D").Range("B2", "T20"), _
Sheets("E-K").Range("B2", "T20"))
but got the same error.
1) Why does Union get 1004?
2) How do I construct a large "array" with minimal coding?
3) After
Dim PricesA_D As Variant
PricesA_D = Sheets("A-D").Range( _
"B2", _
Sheets("A-D").Cells.SpecialCells(xlCellTypeLastCell).Address)
Locals shows StationsA_D has Type Variant/Variant(1 to 623, 1 to 151)
What code will find those values 623 and 151 in StationsA_D?
My original problem of finding asymmetric data can easily be solved by
comparing cells on my 4 sheets.
I want to be able to process copies of sheet data in memory.
I have a file of London fare data. (Shows fares between origin and
destination stations.)
It consists of about 600 rows and columns.
Because Excel 2003 only supports 256 columns in a sheet, I need to split
my data over more than one sheet.
The data ought to be symmetrical and mostly is. However, I have found
one pair of stations where the AB price and BA price are different.
It takes several elapsed days to assemble that file from the Internet.
I would like to identify asymmetric data to assess the extent of that
asymmetry. i.e. I assume AB price = BA price is probably correct.
I want to assemble an "array" in memory.
I can read data from a single sheet with code like this:
Dim PricesA_D As Variant
PricesA_D = Sheets("A-D").Range("B2", _
Sheets("A-D").Cells.SpecialCells(xlCellTypeLastCell).Address)
' The first row and column contain station names
I think I need a picture.
Sheets("A-D") might start
Abbey Wood Acton Central Acton Main Line ...
Abbey Wood X 1.23 2.34
Acton Central 3.21 X 3.45
Acton Main Line 2.34 3.45 X
....
I have shown asymmetric data between Abbey Wood and Acton Central.
Sheets("E-K") station names might start
Ealing Broadway Ealing Common Earls Court ...
Abbey Wood
Acton Central
Acton Main Line
....
I want an "array" in which the E-K data is appended to the A-D data.
I tried
Dim Prices As Variant
Prices = Application.Union( _
Sheets("A-D").Range(
"B2", _
Sheets("A-D").Cells.SpecialCells(xlCellTypeLastCell).Address), _
Sheets("E-K").Range(
"B2",
Sheets("E-K").Cells.SpecialCells(xlCellTypeLastCell).Address), _
Sheets("L-R").Range(
"B2",
Sheets("L-R").Cells.SpecialCells(xlCellTypeLastCell).Address), _
Sheets("S-Z").Range(
"B2",
Sheets("S-Z").Cells.SpecialCells(xlCellTypeLastCell).Address))
I got Run-time error '1004':
Method 'Union' of object '_Application' failed
I assumed that was because Union tries to create a sheet with more than
256 columns.
To test that assumption, I tried
Prices = Application.Union( _
Sheets("A-D").Range("B2", "T20"), _
Sheets("E-K").Range("B2", "T20"))
but got the same error.
1) Why does Union get 1004?
2) How do I construct a large "array" with minimal coding?
3) After
Dim PricesA_D As Variant
PricesA_D = Sheets("A-D").Range( _
"B2", _
Sheets("A-D").Cells.SpecialCells(xlCellTypeLastCell).Address)
Locals shows StationsA_D has Type Variant/Variant(1 to 623, 1 to 151)
What code will find those values 623 and 151 in StationsA_D?
My original problem of finding asymmetric data can easily be solved by
comparing cells on my 4 sheets.
I want to be able to process copies of sheet data in memory.