|
code
newsgroups
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
Stupid Format$ QuestionSomething like: 12.3456789 1234.56789 I'm using a Double for the calculations, and have thus far done this: '(n is a Double) Label1.Text = Left$(CStr(n), 10) That works okay except for two cases. First, it doesn't ROUND to those 10 digits, it truncates the text. So 0.499999999999999 reads as 0.49999999 instead of 0.50000000. Second, at some point with small and large numbers, I'll probably want to switch to Scientific notation, in order to have reasonable accuracy, or display very large numbers. So a Double value of 1.23456789E-7 would show like that, instead of 0.00000012. And numbers larger than 9999999999 could still be displayed. I realize I can make a series of If-Then statements, or Select Case, to cover all the bases. But I wondered if there was a simpler way of at least doing the first item. The second is easily covered by two statements (greater than a given number and less than a different given number). -- Regards, Rick Raisley heavymetal-A-T-bellsouth-D-O-T-net Try:
round(0.499999999999999 ,10) You might have to tweak the second parameter. Saga -- Show quoteHide quote"Rick Raisley" <heavymetal-A-T-bellsouth-D-O-Tnet> wrote in message news:u94X4hrlJHA.1388@TK2MSFTNGP06.phx.gbl... >I want to show a number as precisely as I can, using, say, 10 characters. Something like: > > 12.3456789 > 1234.56789 > > I'm using a Double for the calculations, and have thus far done this: > > '(n is a Double) > > Label1.Text = Left$(CStr(n), 10) > > That works okay except for two cases. First, it doesn't ROUND to those 10 digits, it truncates the > text. So 0.499999999999999 reads as 0.49999999 instead of 0.50000000. > > Second, at some point with small and large numbers, I'll probably want to switch to Scientific > notation, in order to have reasonable accuracy, or display very large numbers. So a Double value > of 1.23456789E-7 would show like that, instead of 0.00000012. And numbers larger than 9999999999 > could still be displayed. > > I realize I can make a series of If-Then statements, or Select Case, to cover all the bases. But I > wondered if there was a simpler way of at least doing the first item. The second is easily covered > by two statements (greater than a given number and less than a different given number). > > -- > Regards, > > Rick Raisley > heavymetal-A-T-bellsouth-D-O-T-net > > Rick Raisley wrote:
> I want to show a number as precisely as I can, using, say, 10 characters. Sure, use Fortran G format... <vbg>> Something like: .... > Second, at some point with small and large numbers, I'll probably want to > switch to Scientific notation, in order to have reasonable accuracy, or > display very large numbers. ... > ...I wondered if there was a simpler way ... (While _mostly_ "just joshin'" I have in the past done similar manipulations by using Fortran in a DLL and returning the string to VB for printing because it is so much simpler to let the Fortran runtime take care of it via internal writes. I suppose it could be done w/ the C runtime as well if one is more familiar w/ C than Fortran but I'm not conversant enough w/ the C formats to know the characteristic behavior of g in C.) -- A solution that does not handle scientific notation:
FormatNumber(strValue, 10 - Int((Log(strValue)/Log(10)) + 1)) Show quoteHide quote "Rick Raisley" <heavymetal-A-T-bellsouth-D-O-Tnet> wrote in message news:u94X4hrlJHA.1388@TK2MSFTNGP06.phx.gbl... >I want to show a number as precisely as I can, using, say, 10 characters. >Something like: > > 12.3456789 > 1234.56789 > > I'm using a Double for the calculations, and have thus far done this: > > '(n is a Double) > > Label1.Text = Left$(CStr(n), 10) > > That works okay except for two cases. First, it doesn't ROUND to those 10 > digits, it truncates the text. So 0.499999999999999 reads as 0.49999999 > instead of 0.50000000. > > Second, at some point with small and large numbers, I'll probably want to > switch to Scientific notation, in order to have reasonable accuracy, or > display very large numbers. So a Double value of 1.23456789E-7 would show > like that, instead of 0.00000012. And numbers larger than 9999999999 could > still be displayed. > > I realize I can make a series of If-Then statements, or Select Case, to > cover all the bases. But I wondered if there was a simpler way of at least > doing the first item. The second is easily covered by two statements > (greater than a given number and less than a different given number). > > -- > Regards, > > Rick Raisley > heavymetal-A-T-bellsouth-D-O-T-net > > It also doesn't work well with negative numbers, but thanks.
-- Show quoteHide quoteRegards, Rick Raisley heavymetal-A-T-bellsouth-D-O-T-net "Richard Mueller [MVP]" <rlmueller-nospam@ameritech.nospam.net> wrote in message news:uUAub4rlJHA.5980@TK2MSFTNGP06.phx.gbl... >A solution that does not handle scientific notation: > > FormatNumber(strValue, 10 - Int((Log(strValue)/Log(10)) + 1)) > > -- > Richard Mueller > MVP Directory Services > Hilltop Lab - http://www.rlmueller.net > -- > > "Rick Raisley" <heavymetal-A-T-bellsouth-D-O-Tnet> wrote in message > news:u94X4hrlJHA.1388@TK2MSFTNGP06.phx.gbl... >>I want to show a number as precisely as I can, using, say, 10 characters. >>Something like: >> >> 12.3456789 >> 1234.56789 >> >> I'm using a Double for the calculations, and have thus far done this: >> >> '(n is a Double) >> >> Label1.Text = Left$(CStr(n), 10) >> >> That works okay except for two cases. First, it doesn't ROUND to those 10 >> digits, it truncates the text. So 0.499999999999999 reads as 0.49999999 >> instead of 0.50000000. >> >> Second, at some point with small and large numbers, I'll probably want to >> switch to Scientific notation, in order to have reasonable accuracy, or >> display very large numbers. So a Double value of 1.23456789E-7 would show >> like that, instead of 0.00000012. And numbers larger than 9999999999 >> could still be displayed. >> >> I realize I can make a series of If-Then statements, or Select Case, to >> cover all the bases. But I wondered if there was a simpler way of at >> least doing the first item. The second is easily covered by two >> statements (greater than a given number and less than a different given >> number). >> >> -- >> Regards, >> >> Rick Raisley >> heavymetal-A-T-bellsouth-D-O-T-net >> >> > > I'm sure this is more complex than necessary, but I've written the following
routine to handle my requirement: Function RoundLen(ByVal a As Double, ByVal length As Integer) As String Dim g As String, x As Integer, ohs As String, neg As String If a = 0 Then 'Don't pad zeros. Makes them look more obvious. RoundLen = "0" Exit Function End If If a < 0 Then neg = "-" 'I prefer constant length numbers with negative sign added. If total length must be constant, use length = length - 1 a = Abs(a) ohs = String(length, "0") x = Int(Log(a) / Log(10#)) If a > CDbl(String(length, "9")) Then 'For very large numbers, revert to scientific notation g = Format$(a, "0." & Left$(ohs, length - 7) & " E+00") ElseIf x / length < -0.5 Then 'For very small numbers, revert to scientific notation. x/length seems to optimize when to use it for most displayed precision g = Format$(a, "0." & Left$(ohs, length - 7) & " E+00") ElseIf a < 1 Then 'Handles numbers less than 1 g = Format$(a, "0." & Left$(ohs, length - 2)) ElseIf x > length - 2 Then 'Handles numbers of maximum length only, with decimal g = Format$(a, Left$(ohs, length)) Else 'Handles "normal" numbers, with 1 or more places on each side of decimal. g = Format$(a, "0." & Left$(ohs, length - x - 2)) End If RoundLen = neg & g End Function Any simplification suggestions would be appreciated. ;-) -- Show quoteHide quoteRegards, Rick Raisley heavymetal-A-T-bellsouth-D-O-T-net "Rick Raisley" <heavymetal-A-T-bellsouth-D-O-Tnet> wrote in message news:eal4vj0lJHA.3888@TK2MSFTNGP02.phx.gbl... > It also doesn't work well with negative numbers, but thanks. > > -- > Regards, > > Rick Raisley > heavymetal-A-T-bellsouth-D-O-T-net > > "Richard Mueller [MVP]" <rlmueller-nospam@ameritech.nospam.net> wrote in > message news:uUAub4rlJHA.5980@TK2MSFTNGP06.phx.gbl... >>A solution that does not handle scientific notation: >> >> FormatNumber(strValue, 10 - Int((Log(strValue)/Log(10)) + 1)) >> >> -- >> Richard Mueller >> MVP Directory Services >> Hilltop Lab - http://www.rlmueller.net >> -- >> >> "Rick Raisley" <heavymetal-A-T-bellsouth-D-O-Tnet> wrote in message >> news:u94X4hrlJHA.1388@TK2MSFTNGP06.phx.gbl... >>>I want to show a number as precisely as I can, using, say, 10 characters. >>>Something like: >>> >>> 12.3456789 >>> 1234.56789 >>> >>> I'm using a Double for the calculations, and have thus far done this: >>> >>> '(n is a Double) >>> >>> Label1.Text = Left$(CStr(n), 10) >>> >>> That works okay except for two cases. First, it doesn't ROUND to those >>> 10 digits, it truncates the text. So 0.499999999999999 reads as >>> 0.49999999 instead of 0.50000000. >>> >>> Second, at some point with small and large numbers, I'll probably want >>> to switch to Scientific notation, in order to have reasonable accuracy, >>> or display very large numbers. So a Double value of 1.23456789E-7 would >>> show like that, instead of 0.00000012. And numbers larger than >>> 9999999999 could still be displayed. >>> >>> I realize I can make a series of If-Then statements, or Select Case, to >>> cover all the bases. But I wondered if there was a simpler way of at >>> least doing the first item. The second is easily covered by two >>> statements (greater than a given number and less than a different given >>> number). >>> >>> -- >>> Regards, >>> >>> Rick Raisley >>> heavymetal-A-T-bellsouth-D-O-T-net >>> >>> >> >> > > Sorry about my other post... I misread what you wanted to do. Give the
following function a try (it is a little shorter and does less work than your function)... Function RoundLen(Value As Double, Length As Integer) As Double Dim Num As Double Dim Parts() As String Num = CDbl(Format(Value, "0.##############################;;0")) If InStr(Num, "E") Then Parts = Split(CStr(Num), "E") RoundLen = Format(Parts(0), "0." & String(Length, "0") & "E" & Parts(1)) ElseIf Num <> 0 Then RoundLen = Format(Num, "0." & String(Length, "0")) End If End Function -- Show quoteHide quoteRick (MVP - Excel) "Rick Raisley" <heavymetal-A-T-bellsouth-D-O-Tnet> wrote in message news:%23B4vby0lJHA.1340@TK2MSFTNGP06.phx.gbl... > I'm sure this is more complex than necessary, but I've written the > following routine to handle my requirement: > > Function RoundLen(ByVal a As Double, ByVal length As Integer) As String > Dim g As String, x As Integer, ohs As String, neg As String > > If a = 0 Then > 'Don't pad zeros. Makes them look more obvious. > RoundLen = "0" > Exit Function > End If > If a < 0 Then neg = "-" > 'I prefer constant length numbers with negative sign added. If total > length must be constant, use length = length - 1 > a = Abs(a) > ohs = String(length, "0") > x = Int(Log(a) / Log(10#)) > If a > CDbl(String(length, "9")) Then > 'For very large numbers, revert to scientific notation > g = Format$(a, "0." & Left$(ohs, length - 7) & " E+00") > ElseIf x / length < -0.5 Then > 'For very small numbers, revert to scientific notation. x/length > seems to optimize when to use it for most displayed precision > g = Format$(a, "0." & Left$(ohs, length - 7) & " E+00") > ElseIf a < 1 Then > 'Handles numbers less than 1 > g = Format$(a, "0." & Left$(ohs, length - 2)) > ElseIf x > length - 2 Then > 'Handles numbers of maximum length only, with decimal > g = Format$(a, Left$(ohs, length)) > Else > 'Handles "normal" numbers, with 1 or more places on each side of > decimal. > g = Format$(a, "0." & Left$(ohs, length - x - 2)) > End If > RoundLen = neg & g > End Function > > Any simplification suggestions would be appreciated. ;-) > > -- > Regards, > > Rick Raisley > heavymetal-A-T-bellsouth-D-O-T-net > > "Rick Raisley" <heavymetal-A-T-bellsouth-D-O-Tnet> wrote in message > news:eal4vj0lJHA.3888@TK2MSFTNGP02.phx.gbl... >> It also doesn't work well with negative numbers, but thanks. >> >> -- >> Regards, >> >> Rick Raisley >> heavymetal-A-T-bellsouth-D-O-T-net >> >> "Richard Mueller [MVP]" <rlmueller-nospam@ameritech.nospam.net> wrote in >> message news:uUAub4rlJHA.5980@TK2MSFTNGP06.phx.gbl... >>>A solution that does not handle scientific notation: >>> >>> FormatNumber(strValue, 10 - Int((Log(strValue)/Log(10)) + 1)) >>> >>> -- >>> Richard Mueller >>> MVP Directory Services >>> Hilltop Lab - http://www.rlmueller.net >>> -- >>> >>> "Rick Raisley" <heavymetal-A-T-bellsouth-D-O-Tnet> wrote in message >>> news:u94X4hrlJHA.1388@TK2MSFTNGP06.phx.gbl... >>>>I want to show a number as precisely as I can, using, say, 10 >>>>characters. Something like: >>>> >>>> 12.3456789 >>>> 1234.56789 >>>> >>>> I'm using a Double for the calculations, and have thus far done this: >>>> >>>> '(n is a Double) >>>> >>>> Label1.Text = Left$(CStr(n), 10) >>>> >>>> That works okay except for two cases. First, it doesn't ROUND to those >>>> 10 digits, it truncates the text. So 0.499999999999999 reads as >>>> 0.49999999 instead of 0.50000000. >>>> >>>> Second, at some point with small and large numbers, I'll probably want >>>> to switch to Scientific notation, in order to have reasonable accuracy, >>>> or display very large numbers. So a Double value of 1.23456789E-7 would >>>> show like that, instead of 0.00000012. And numbers larger than >>>> 9999999999 could still be displayed. >>>> >>>> I realize I can make a series of If-Then statements, or Select Case, to >>>> cover all the bases. But I wondered if there was a simpler way of at >>>> least doing the first item. The second is easily covered by two >>>> statements (greater than a given number and less than a different given >>>> number). >>>> >>>> -- >>>> Regards, >>>> >>>> Rick Raisley >>>> heavymetal-A-T-bellsouth-D-O-T-net >>>> >>>> >>> >>> >> >> > > Slight modification to return a String instead of a Double and to suppress
the decimal point from values rounded to Length = 0... Function RoundLen(Value As Double, Length As Integer) As String Dim Num As Double Dim Parts() As String Num = CDbl(Format(Value, "0.##############################;;0")) If InStr(Num, "E") Then Parts = Split(CStr(Num), "E") RoundLen = Format(Parts(0), "0" & Left(".", -(Length <> 0)) & _ String(Length, "0") & "E" & Parts(1)) ElseIf Num <> 0 Then RoundLen = Format(Num, "0" & Left(".", _ -(Length <> 0)) & String(Length, "0")) Else RoundLen = "0" End If End Function -- Show quoteHide quoteRick (MVP - Excel) "Rick Rothstein" <rick.newsNO.SPAM@NO.SPAMverizon.net> wrote in message news:eZlwp71lJHA.4520@TK2MSFTNGP03.phx.gbl... > Sorry about my other post... I misread what you wanted to do. Give the > following function a try (it is a little shorter and does less work than > your function)... > > Function RoundLen(Value As Double, Length As Integer) As Double > Dim Num As Double > Dim Parts() As String > Num = CDbl(Format(Value, "0.##############################;;0")) > If InStr(Num, "E") Then > Parts = Split(CStr(Num), "E") > RoundLen = Format(Parts(0), "0." & String(Length, "0") & "E" & > Parts(1)) > ElseIf Num <> 0 Then > RoundLen = Format(Num, "0." & String(Length, "0")) > End If > End Function > > -- > Rick (MVP - Excel) > > > "Rick Raisley" <heavymetal-A-T-bellsouth-D-O-Tnet> wrote in message > news:%23B4vby0lJHA.1340@TK2MSFTNGP06.phx.gbl... >> I'm sure this is more complex than necessary, but I've written the >> following routine to handle my requirement: >> >> Function RoundLen(ByVal a As Double, ByVal length As Integer) As String >> Dim g As String, x As Integer, ohs As String, neg As String >> >> If a = 0 Then >> 'Don't pad zeros. Makes them look more obvious. >> RoundLen = "0" >> Exit Function >> End If >> If a < 0 Then neg = "-" >> 'I prefer constant length numbers with negative sign added. If total >> length must be constant, use length = length - 1 >> a = Abs(a) >> ohs = String(length, "0") >> x = Int(Log(a) / Log(10#)) >> If a > CDbl(String(length, "9")) Then >> 'For very large numbers, revert to scientific notation >> g = Format$(a, "0." & Left$(ohs, length - 7) & " E+00") >> ElseIf x / length < -0.5 Then >> 'For very small numbers, revert to scientific notation. x/length >> seems to optimize when to use it for most displayed precision >> g = Format$(a, "0." & Left$(ohs, length - 7) & " E+00") >> ElseIf a < 1 Then >> 'Handles numbers less than 1 >> g = Format$(a, "0." & Left$(ohs, length - 2)) >> ElseIf x > length - 2 Then >> 'Handles numbers of maximum length only, with decimal >> g = Format$(a, Left$(ohs, length)) >> Else >> 'Handles "normal" numbers, with 1 or more places on each side of >> decimal. >> g = Format$(a, "0." & Left$(ohs, length - x - 2)) >> End If >> RoundLen = neg & g >> End Function >> >> Any simplification suggestions would be appreciated. ;-) >> >> -- >> Regards, >> >> Rick Raisley >> heavymetal-A-T-bellsouth-D-O-T-net >> >> "Rick Raisley" <heavymetal-A-T-bellsouth-D-O-Tnet> wrote in message >> news:eal4vj0lJHA.3888@TK2MSFTNGP02.phx.gbl... >>> It also doesn't work well with negative numbers, but thanks. >>> >>> -- >>> Regards, >>> >>> Rick Raisley >>> heavymetal-A-T-bellsouth-D-O-T-net >>> >>> "Richard Mueller [MVP]" <rlmueller-nospam@ameritech.nospam.net> wrote in >>> message news:uUAub4rlJHA.5980@TK2MSFTNGP06.phx.gbl... >>>>A solution that does not handle scientific notation: >>>> >>>> FormatNumber(strValue, 10 - Int((Log(strValue)/Log(10)) + 1)) >>>> >>>> -- >>>> Richard Mueller >>>> MVP Directory Services >>>> Hilltop Lab - http://www.rlmueller.net >>>> -- >>>> >>>> "Rick Raisley" <heavymetal-A-T-bellsouth-D-O-Tnet> wrote in message >>>> news:u94X4hrlJHA.1388@TK2MSFTNGP06.phx.gbl... >>>>>I want to show a number as precisely as I can, using, say, 10 >>>>>characters. Something like: >>>>> >>>>> 12.3456789 >>>>> 1234.56789 >>>>> >>>>> I'm using a Double for the calculations, and have thus far done this: >>>>> >>>>> '(n is a Double) >>>>> >>>>> Label1.Text = Left$(CStr(n), 10) >>>>> >>>>> That works okay except for two cases. First, it doesn't ROUND to those >>>>> 10 digits, it truncates the text. So 0.499999999999999 reads as >>>>> 0.49999999 instead of 0.50000000. >>>>> >>>>> Second, at some point with small and large numbers, I'll probably want >>>>> to switch to Scientific notation, in order to have reasonable >>>>> accuracy, or display very large numbers. So a Double value of >>>>> 1.23456789E-7 would show like that, instead of 0.00000012. And numbers >>>>> larger than 9999999999 could still be displayed. >>>>> >>>>> I realize I can make a series of If-Then statements, or Select Case, >>>>> to cover all the bases. But I wondered if there was a simpler way of >>>>> at least doing the first item. The second is easily covered by two >>>>> statements (greater than a given number and less than a different >>>>> given number). >>>>> >>>>> -- >>>>> Regards, >>>>> >>>>> Rick Raisley >>>>> heavymetal-A-T-bellsouth-D-O-T-net >>>>> >>>>> >>>> >>>> >>> >>> >> >> > > >>>>> -- Nice Rick, and now a one liner ;)> >>>>> Regards, > > >>>>> Rick Raisley > >>>>> heavymetal-A-T-bellsouth-D-O-T-net- Dölj citerad text - > > - Visa citerad text - just tried something but didnt go with the E notations... stole the idea from all above (well not the E notations), just wanted to see if it works.. and in XL (VBA), as I have no VB6 at home :( //CY Public Sub test() Debug.Print x(0.5) Debug.Print x(0.49999) Debug.Print x(123456.49999) End Sub Private Function x(y As Double) As String x = Left(Format(y, "0.##########") & "0000000000", 10) End Function Just to clarify, the Length argument in my function sets the number of
decimal places to round to, NOT the total number of significant digits to be returned independent of the decimal point's location. -- Show quoteHide quoteRick (MVP - Excel) "Rick Rothstein" <rick.newsNO.SPAM@NO.SPAMverizon.net> wrote in message news:%23Rvo4m2lJHA.3760@TK2MSFTNGP03.phx.gbl... > Slight modification to return a String instead of a Double and to suppress > the decimal point from values rounded to Length = 0... > > Function RoundLen(Value As Double, Length As Integer) As String > Dim Num As Double > Dim Parts() As String > Num = CDbl(Format(Value, "0.##############################;;0")) > If InStr(Num, "E") Then > Parts = Split(CStr(Num), "E") > RoundLen = Format(Parts(0), "0" & Left(".", -(Length <> 0)) & _ > String(Length, "0") & "E" & Parts(1)) > ElseIf Num <> 0 Then > RoundLen = Format(Num, "0" & Left(".", _ > -(Length <> 0)) & String(Length, "0")) > Else > RoundLen = "0" > End If > End Function > > -- > Rick (MVP - Excel) > > > "Rick Rothstein" <rick.newsNO.SPAM@NO.SPAMverizon.net> wrote in message > news:eZlwp71lJHA.4520@TK2MSFTNGP03.phx.gbl... >> Sorry about my other post... I misread what you wanted to do. Give the >> following function a try (it is a little shorter and does less work than >> your function)... >> >> Function RoundLen(Value As Double, Length As Integer) As Double >> Dim Num As Double >> Dim Parts() As String >> Num = CDbl(Format(Value, "0.##############################;;0")) >> If InStr(Num, "E") Then >> Parts = Split(CStr(Num), "E") >> RoundLen = Format(Parts(0), "0." & String(Length, "0") & "E" & >> Parts(1)) >> ElseIf Num <> 0 Then >> RoundLen = Format(Num, "0." & String(Length, "0")) >> End If >> End Function >> >> -- >> Rick (MVP - Excel) >> >> >> "Rick Raisley" <heavymetal-A-T-bellsouth-D-O-Tnet> wrote in message >> news:%23B4vby0lJHA.1340@TK2MSFTNGP06.phx.gbl... >>> I'm sure this is more complex than necessary, but I've written the >>> following routine to handle my requirement: >>> >>> Function RoundLen(ByVal a As Double, ByVal length As Integer) As String >>> Dim g As String, x As Integer, ohs As String, neg As String >>> >>> If a = 0 Then >>> 'Don't pad zeros. Makes them look more obvious. >>> RoundLen = "0" >>> Exit Function >>> End If >>> If a < 0 Then neg = "-" >>> 'I prefer constant length numbers with negative sign added. If total >>> length must be constant, use length = length - 1 >>> a = Abs(a) >>> ohs = String(length, "0") >>> x = Int(Log(a) / Log(10#)) >>> If a > CDbl(String(length, "9")) Then >>> 'For very large numbers, revert to scientific notation >>> g = Format$(a, "0." & Left$(ohs, length - 7) & " E+00") >>> ElseIf x / length < -0.5 Then >>> 'For very small numbers, revert to scientific notation. x/length >>> seems to optimize when to use it for most displayed precision >>> g = Format$(a, "0." & Left$(ohs, length - 7) & " E+00") >>> ElseIf a < 1 Then >>> 'Handles numbers less than 1 >>> g = Format$(a, "0." & Left$(ohs, length - 2)) >>> ElseIf x > length - 2 Then >>> 'Handles numbers of maximum length only, with decimal >>> g = Format$(a, Left$(ohs, length)) >>> Else >>> 'Handles "normal" numbers, with 1 or more places on each side of >>> decimal. >>> g = Format$(a, "0." & Left$(ohs, length - x - 2)) >>> End If >>> RoundLen = neg & g >>> End Function >>> >>> Any simplification suggestions would be appreciated. ;-) >>> >>> -- >>> Regards, >>> >>> Rick Raisley >>> heavymetal-A-T-bellsouth-D-O-T-net >>> >>> "Rick Raisley" <heavymetal-A-T-bellsouth-D-O-Tnet> wrote in message >>> news:eal4vj0lJHA.3888@TK2MSFTNGP02.phx.gbl... >>>> It also doesn't work well with negative numbers, but thanks. >>>> >>>> -- >>>> Regards, >>>> >>>> Rick Raisley >>>> heavymetal-A-T-bellsouth-D-O-T-net >>>> >>>> "Richard Mueller [MVP]" <rlmueller-nospam@ameritech.nospam.net> wrote >>>> in message news:uUAub4rlJHA.5980@TK2MSFTNGP06.phx.gbl... >>>>>A solution that does not handle scientific notation: >>>>> >>>>> FormatNumber(strValue, 10 - Int((Log(strValue)/Log(10)) + 1)) >>>>> >>>>> -- >>>>> Richard Mueller >>>>> MVP Directory Services >>>>> Hilltop Lab - http://www.rlmueller.net >>>>> -- >>>>> >>>>> "Rick Raisley" <heavymetal-A-T-bellsouth-D-O-Tnet> wrote in message >>>>> news:u94X4hrlJHA.1388@TK2MSFTNGP06.phx.gbl... >>>>>>I want to show a number as precisely as I can, using, say, 10 >>>>>>characters. Something like: >>>>>> >>>>>> 12.3456789 >>>>>> 1234.56789 >>>>>> >>>>>> I'm using a Double for the calculations, and have thus far done this: >>>>>> >>>>>> '(n is a Double) >>>>>> >>>>>> Label1.Text = Left$(CStr(n), 10) >>>>>> >>>>>> That works okay except for two cases. First, it doesn't ROUND to >>>>>> those 10 digits, it truncates the text. So 0.499999999999999 reads as >>>>>> 0.49999999 instead of 0.50000000. >>>>>> >>>>>> Second, at some point with small and large numbers, I'll probably >>>>>> want to switch to Scientific notation, in order to have reasonable >>>>>> accuracy, or display very large numbers. So a Double value of >>>>>> 1.23456789E-7 would show like that, instead of 0.00000012. And >>>>>> numbers larger than 9999999999 could still be displayed. >>>>>> >>>>>> I realize I can make a series of If-Then statements, or Select Case, >>>>>> to cover all the bases. But I wondered if there was a simpler way of >>>>>> at least doing the first item. The second is easily covered by two >>>>>> statements (greater than a given number and less than a different >>>>>> given number). >>>>>> >>>>>> -- >>>>>> Regards, >>>>>> >>>>>> Rick Raisley >>>>>> heavymetal-A-T-bellsouth-D-O-T-net >>>>>> >>>>>> >>>>> >>>>> >>>> >>>> >>> >>> >> > Yeah, I noticed. I want the total string length to always be a set maximum
value, except I've decided that it looks best (and there's room) for the negative sign to be in addition to the number of desired places. My code does all that, but it's messy (hidden in a function, though, at least). ;-) -- Show quoteHide quoteRegards, Rick Raisley heavymetal-A-T-bellsouth-D-O-T-net "Rick Rothstein" <rick.newsNO.SPAM@NO.SPAMverizon.net> wrote in message news:Obcd1V4lJHA.2384@TK2MSFTNGP04.phx.gbl... > Just to clarify, the Length argument in my function sets the number of > decimal places to round to, NOT the total number of significant digits to > be returned independent of the decimal point's location. > > -- > Rick (MVP - Excel) > > > "Rick Rothstein" <rick.newsNO.SPAM@NO.SPAMverizon.net> wrote in message > news:%23Rvo4m2lJHA.3760@TK2MSFTNGP03.phx.gbl... >> Slight modification to return a String instead of a Double and to >> suppress the decimal point from values rounded to Length = 0... >> >> Function RoundLen(Value As Double, Length As Integer) As String >> Dim Num As Double >> Dim Parts() As String >> Num = CDbl(Format(Value, "0.##############################;;0")) >> If InStr(Num, "E") Then >> Parts = Split(CStr(Num), "E") >> RoundLen = Format(Parts(0), "0" & Left(".", -(Length <> 0)) & _ >> String(Length, "0") & "E" & Parts(1)) >> ElseIf Num <> 0 Then >> RoundLen = Format(Num, "0" & Left(".", _ >> -(Length <> 0)) & String(Length, "0")) >> Else >> RoundLen = "0" >> End If >> End Function >> >> -- >> Rick (MVP - Excel) >> >> >> "Rick Rothstein" <rick.newsNO.SPAM@NO.SPAMverizon.net> wrote in message >> news:eZlwp71lJHA.4520@TK2MSFTNGP03.phx.gbl... >>> Sorry about my other post... I misread what you wanted to do. Give the >>> following function a try (it is a little shorter and does less work than >>> your function)... >>> >>> Function RoundLen(Value As Double, Length As Integer) As Double >>> Dim Num As Double >>> Dim Parts() As String >>> Num = CDbl(Format(Value, "0.##############################;;0")) >>> If InStr(Num, "E") Then >>> Parts = Split(CStr(Num), "E") >>> RoundLen = Format(Parts(0), "0." & String(Length, "0") & "E" & >>> Parts(1)) >>> ElseIf Num <> 0 Then >>> RoundLen = Format(Num, "0." & String(Length, "0")) >>> End If >>> End Function >>> >>> -- >>> Rick (MVP - Excel) >>> >>> >>> "Rick Raisley" <heavymetal-A-T-bellsouth-D-O-Tnet> wrote in message >>> news:%23B4vby0lJHA.1340@TK2MSFTNGP06.phx.gbl... >>>> I'm sure this is more complex than necessary, but I've written the >>>> following routine to handle my requirement: >>>> >>>> Function RoundLen(ByVal a As Double, ByVal length As Integer) As String >>>> Dim g As String, x As Integer, ohs As String, neg As String >>>> >>>> If a = 0 Then >>>> 'Don't pad zeros. Makes them look more obvious. >>>> RoundLen = "0" >>>> Exit Function >>>> End If >>>> If a < 0 Then neg = "-" >>>> 'I prefer constant length numbers with negative sign added. If total >>>> length must be constant, use length = length - 1 >>>> a = Abs(a) >>>> ohs = String(length, "0") >>>> x = Int(Log(a) / Log(10#)) >>>> If a > CDbl(String(length, "9")) Then >>>> 'For very large numbers, revert to scientific notation >>>> g = Format$(a, "0." & Left$(ohs, length - 7) & " E+00") >>>> ElseIf x / length < -0.5 Then >>>> 'For very small numbers, revert to scientific notation. x/length >>>> seems to optimize when to use it for most displayed precision >>>> g = Format$(a, "0." & Left$(ohs, length - 7) & " E+00") >>>> ElseIf a < 1 Then >>>> 'Handles numbers less than 1 >>>> g = Format$(a, "0." & Left$(ohs, length - 2)) >>>> ElseIf x > length - 2 Then >>>> 'Handles numbers of maximum length only, with decimal >>>> g = Format$(a, Left$(ohs, length)) >>>> Else >>>> 'Handles "normal" numbers, with 1 or more places on each side of >>>> decimal. >>>> g = Format$(a, "0." & Left$(ohs, length - x - 2)) >>>> End If >>>> RoundLen = neg & g >>>> End Function >>>> >>>> Any simplification suggestions would be appreciated. ;-) >>>> >>>> -- >>>> Regards, >>>> >>>> Rick Raisley >>>> heavymetal-A-T-bellsouth-D-O-T-net >>>> >>>> "Rick Raisley" <heavymetal-A-T-bellsouth-D-O-Tnet> wrote in message >>>> news:eal4vj0lJHA.3888@TK2MSFTNGP02.phx.gbl... >>>>> It also doesn't work well with negative numbers, but thanks. >>>>> >>>>> -- >>>>> Regards, >>>>> >>>>> Rick Raisley >>>>> heavymetal-A-T-bellsouth-D-O-T-net >>>>> >>>>> "Richard Mueller [MVP]" <rlmueller-nospam@ameritech.nospam.net> wrote >>>>> in message news:uUAub4rlJHA.5980@TK2MSFTNGP06.phx.gbl... >>>>>>A solution that does not handle scientific notation: >>>>>> >>>>>> FormatNumber(strValue, 10 - Int((Log(strValue)/Log(10)) + 1)) >>>>>> >>>>>> -- >>>>>> Richard Mueller >>>>>> MVP Directory Services >>>>>> Hilltop Lab - http://www.rlmueller.net >>>>>> -- >>>>>> >>>>>> "Rick Raisley" <heavymetal-A-T-bellsouth-D-O-Tnet> wrote in message >>>>>> news:u94X4hrlJHA.1388@TK2MSFTNGP06.phx.gbl... >>>>>>>I want to show a number as precisely as I can, using, say, 10 >>>>>>>characters. Something like: >>>>>>> >>>>>>> 12.3456789 >>>>>>> 1234.56789 >>>>>>> >>>>>>> I'm using a Double for the calculations, and have thus far done >>>>>>> this: >>>>>>> >>>>>>> '(n is a Double) >>>>>>> >>>>>>> Label1.Text = Left$(CStr(n), 10) >>>>>>> >>>>>>> That works okay except for two cases. First, it doesn't ROUND to >>>>>>> those 10 digits, it truncates the text. So 0.499999999999999 reads >>>>>>> as 0.49999999 instead of 0.50000000. >>>>>>> >>>>>>> Second, at some point with small and large numbers, I'll probably >>>>>>> want to switch to Scientific notation, in order to have reasonable >>>>>>> accuracy, or display very large numbers. So a Double value of >>>>>>> 1.23456789E-7 would show like that, instead of 0.00000012. And >>>>>>> numbers larger than 9999999999 could still be displayed. >>>>>>> >>>>>>> I realize I can make a series of If-Then statements, or Select Case, >>>>>>> to cover all the bases. But I wondered if there was a simpler way of >>>>>>> at least doing the first item. The second is easily covered by two >>>>>>> statements (greater than a given number and less than a different >>>>>>> given number). >>>>>>> >>>>>>> -- >>>>>>> Regards, >>>>>>> >>>>>>> Rick Raisley >>>>>>> heavymetal-A-T-bellsouth-D-O-T-net >>>>>>> >>>>>>> >>>>>> >>>>>> >>>>> >>>>> >>>> >>>> >>> >> > I guess I am not sure of your requirements, but I got your function to fail
with these two test values... Debug.Print RoundLen(12345, 4) Debug.Print RoundLen(123456.49999, 4) Is there some required length for you values? How did you want to handle values with E-notation... is the E## (where each # is a digit) part of the desired length or is the numeric part to be of the specified length and the E## concatenated onto it afterwards? -- Show quoteHide quoteRick (MVP - Excel) "Rick Raisley" <heavymetal-A-T-bellsouth-D-O-Tnet> wrote in message news:u8GkLNBmJHA.4760@TK2MSFTNGP04.phx.gbl... > Yeah, I noticed. I want the total string length to always be a set maximum > value, except I've decided that it looks best (and there's room) for the > negative sign to be in addition to the number of desired places. My code > does all that, but it's messy (hidden in a function, though, at least). > ;-) > > -- > Regards, > > Rick Raisley > heavymetal-A-T-bellsouth-D-O-T-net > > "Rick Rothstein" <rick.newsNO.SPAM@NO.SPAMverizon.net> wrote in message > news:Obcd1V4lJHA.2384@TK2MSFTNGP04.phx.gbl... >> Just to clarify, the Length argument in my function sets the number of >> decimal places to round to, NOT the total number of significant digits to >> be returned independent of the decimal point's location. >> >> -- >> Rick (MVP - Excel) >> >> >> "Rick Rothstein" <rick.newsNO.SPAM@NO.SPAMverizon.net> wrote in message >> news:%23Rvo4m2lJHA.3760@TK2MSFTNGP03.phx.gbl... >>> Slight modification to return a String instead of a Double and to >>> suppress the decimal point from values rounded to Length = 0... >>> >>> Function RoundLen(Value As Double, Length As Integer) As String >>> Dim Num As Double >>> Dim Parts() As String >>> Num = CDbl(Format(Value, "0.##############################;;0")) >>> If InStr(Num, "E") Then >>> Parts = Split(CStr(Num), "E") >>> RoundLen = Format(Parts(0), "0" & Left(".", -(Length <> 0)) & _ >>> String(Length, "0") & "E" & Parts(1)) >>> ElseIf Num <> 0 Then >>> RoundLen = Format(Num, "0" & Left(".", _ >>> -(Length <> 0)) & String(Length, "0")) >>> Else >>> RoundLen = "0" >>> End If >>> End Function >>> >>> -- >>> Rick (MVP - Excel) >>> >>> >>> "Rick Rothstein" <rick.newsNO.SPAM@NO.SPAMverizon.net> wrote in message >>> news:eZlwp71lJHA.4520@TK2MSFTNGP03.phx.gbl... >>>> Sorry about my other post... I misread what you wanted to do. Give the >>>> following function a try (it is a little shorter and does less work >>>> than your function)... >>>> >>>> Function RoundLen(Value As Double, Length As Integer) As Double >>>> Dim Num As Double >>>> Dim Parts() As String >>>> Num = CDbl(Format(Value, "0.##############################;;0")) >>>> If InStr(Num, "E") Then >>>> Parts = Split(CStr(Num), "E") >>>> RoundLen = Format(Parts(0), "0." & String(Length, "0") & "E" & >>>> Parts(1)) >>>> ElseIf Num <> 0 Then >>>> RoundLen = Format(Num, "0." & String(Length, "0")) >>>> End If >>>> End Function >>>> >>>> -- >>>> Rick (MVP - Excel) >>>> >>>> >>>> "Rick Raisley" <heavymetal-A-T-bellsouth-D-O-Tnet> wrote in message >>>> news:%23B4vby0lJHA.1340@TK2MSFTNGP06.phx.gbl... >>>>> I'm sure this is more complex than necessary, but I've written the >>>>> following routine to handle my requirement: >>>>> >>>>> Function RoundLen(ByVal a As Double, ByVal length As Integer) As >>>>> String >>>>> Dim g As String, x As Integer, ohs As String, neg As String >>>>> >>>>> If a = 0 Then >>>>> 'Don't pad zeros. Makes them look more obvious. >>>>> RoundLen = "0" >>>>> Exit Function >>>>> End If >>>>> If a < 0 Then neg = "-" >>>>> 'I prefer constant length numbers with negative sign added. If total >>>>> length must be constant, use length = length - 1 >>>>> a = Abs(a) >>>>> ohs = String(length, "0") >>>>> x = Int(Log(a) / Log(10#)) >>>>> If a > CDbl(String(length, "9")) Then >>>>> 'For very large numbers, revert to scientific notation >>>>> g = Format$(a, "0." & Left$(ohs, length - 7) & " E+00") >>>>> ElseIf x / length < -0.5 Then >>>>> 'For very small numbers, revert to scientific notation. x/length >>>>> seems to optimize when to use it for most displayed precision >>>>> g = Format$(a, "0." & Left$(ohs, length - 7) & " E+00") >>>>> ElseIf a < 1 Then >>>>> 'Handles numbers less than 1 >>>>> g = Format$(a, "0." & Left$(ohs, length - 2)) >>>>> ElseIf x > length - 2 Then >>>>> 'Handles numbers of maximum length only, with decimal >>>>> g = Format$(a, Left$(ohs, length)) >>>>> Else >>>>> 'Handles "normal" numbers, with 1 or more places on each side of >>>>> decimal. >>>>> g = Format$(a, "0." & Left$(ohs, length - x - 2)) >>>>> End If >>>>> RoundLen = neg & g >>>>> End Function >>>>> >>>>> Any simplification suggestions would be appreciated. ;-) >>>>> >>>>> -- >>>>> Regards, >>>>> >>>>> Rick Raisley >>>>> heavymetal-A-T-bellsouth-D-O-T-net >>>>> >>>>> "Rick Raisley" <heavymetal-A-T-bellsouth-D-O-Tnet> wrote in message >>>>> news:eal4vj0lJHA.3888@TK2MSFTNGP02.phx.gbl... >>>>>> It also doesn't work well with negative numbers, but thanks. >>>>>> >>>>>> -- >>>>>> Regards, >>>>>> >>>>>> Rick Raisley >>>>>> heavymetal-A-T-bellsouth-D-O-T-net >>>>>> >>>>>> "Richard Mueller [MVP]" <rlmueller-nospam@ameritech.nospam.net> wrote >>>>>> in message news:uUAub4rlJHA.5980@TK2MSFTNGP06.phx.gbl... >>>>>>>A solution that does not handle scientific notation: >>>>>>> >>>>>>> FormatNumber(strValue, 10 - Int((Log(strValue)/Log(10)) + 1)) >>>>>>> >>>>>>> -- >>>>>>> Richard Mueller >>>>>>> MVP Directory Services >>>>>>> Hilltop Lab - http://www.rlmueller.net >>>>>>> -- >>>>>>> >>>>>>> "Rick Raisley" <heavymetal-A-T-bellsouth-D-O-Tnet> wrote in message >>>>>>> news:u94X4hrlJHA.1388@TK2MSFTNGP06.phx.gbl... >>>>>>>>I want to show a number as precisely as I can, using, say, 10 >>>>>>>>characters. Something like: >>>>>>>> >>>>>>>> 12.3456789 >>>>>>>> 1234.56789 >>>>>>>> >>>>>>>> I'm using a Double for the calculations, and have thus far done >>>>>>>> this: >>>>>>>> >>>>>>>> '(n is a Double) >>>>>>>> >>>>>>>> Label1.Text = Left$(CStr(n), 10) >>>>>>>> >>>>>>>> That works okay except for two cases. First, it doesn't ROUND to >>>>>>>> those 10 digits, it truncates the text. So 0.499999999999999 reads >>>>>>>> as 0.49999999 instead of 0.50000000. >>>>>>>> >>>>>>>> Second, at some point with small and large numbers, I'll probably >>>>>>>> want to switch to Scientific notation, in order to have reasonable >>>>>>>> accuracy, or display very large numbers. So a Double value of >>>>>>>> 1.23456789E-7 would show like that, instead of 0.00000012. And >>>>>>>> numbers larger than 9999999999 could still be displayed. >>>>>>>> >>>>>>>> I realize I can make a series of If-Then statements, or Select >>>>>>>> Case, to cover all the bases. But I wondered if there was a simpler >>>>>>>> way of at least doing the first item. The second is easily covered >>>>>>>> by two statements (greater than a given number and less than a >>>>>>>> different given number). >>>>>>>> >>>>>>>> -- >>>>>>>> Regards, >>>>>>>> >>>>>>>> Rick Raisley >>>>>>>> heavymetal-A-T-bellsouth-D-O-T-net >>>>>>>> >>>>>>>> >>>>>>> >>>>>>> >>>>>> >>>>>> >>>>> >>>>> >>>> >>> >> > > Yeah, my function wouldn't work for only 4 length. For my purposes, I'd
never have anything less than 8. For consistency in notation, when using Scientific notation, the format would be: 1.2 E-06 That in itself is 8 characters long, and is the very shortest I'd consider for this function (for my purposes - not a general Format function, of course). As to my requirements, they are simply to display the maximum number of significant digits possible, using all of the available places. For 9 digits, for example, examples of this would be: 1.234 E+10 1.234 E+09 1.234 E+08 123456789 12345678. (don't like this one, but I prefer them all the same length) 1234567.8 123456.78 12345.678 1234.5678 123.45678 12.345678 1.2345678 0.1234567 0.0123456 0.0012345 0.0001234 1.234 E-05 1.234 E-06 etc. I believe my function, as convoluted as it is, handles this. -- Show quoteHide quoteRegards, Rick Raisley heavymetal-A-T-bellsouth-D-O-T-net "Rick Rothstein" <rick.newsNO.SPAM@NO.SPAMverizon.net> wrote in message news:e$Chx4BmJHA.5652@TK2MSFTNGP02.phx.gbl... >I guess I am not sure of your requirements, but I got your function to fail >with these two test values... > > Debug.Print RoundLen(12345, 4) > Debug.Print RoundLen(123456.49999, 4) > > Is there some required length for you values? How did you want to handle > values with E-notation... is the E## (where each # is a digit) part of the > desired length or is the numeric part to be of the specified length and > the E## concatenated onto it afterwards? > > -- > Rick (MVP - Excel) > > > "Rick Raisley" <heavymetal-A-T-bellsouth-D-O-Tnet> wrote in message > news:u8GkLNBmJHA.4760@TK2MSFTNGP04.phx.gbl... >> Yeah, I noticed. I want the total string length to always be a set >> maximum value, except I've decided that it looks best (and there's room) >> for the negative sign to be in addition to the number of desired places. >> My code does all that, but it's messy (hidden in a function, though, at >> least). ;-) >> >> -- >> Regards, >> >> Rick Raisley >> heavymetal-A-T-bellsouth-D-O-T-net >> >> "Rick Rothstein" <rick.newsNO.SPAM@NO.SPAMverizon.net> wrote in message >> news:Obcd1V4lJHA.2384@TK2MSFTNGP04.phx.gbl... >>> Just to clarify, the Length argument in my function sets the number of >>> decimal places to round to, NOT the total number of significant digits >>> to be returned independent of the decimal point's location. >>> >>> -- >>> Rick (MVP - Excel) >>> >>> >>> "Rick Rothstein" <rick.newsNO.SPAM@NO.SPAMverizon.net> wrote in message >>> news:%23Rvo4m2lJHA.3760@TK2MSFTNGP03.phx.gbl... >>>> Slight modification to return a String instead of a Double and to >>>> suppress the decimal point from values rounded to Length = 0... >>>> >>>> Function RoundLen(Value As Double, Length As Integer) As String >>>> Dim Num As Double >>>> Dim Parts() As String >>>> Num = CDbl(Format(Value, "0.##############################;;0")) >>>> If InStr(Num, "E") Then >>>> Parts = Split(CStr(Num), "E") >>>> RoundLen = Format(Parts(0), "0" & Left(".", -(Length <> 0)) & _ >>>> String(Length, "0") & "E" & Parts(1)) >>>> ElseIf Num <> 0 Then >>>> RoundLen = Format(Num, "0" & Left(".", _ >>>> -(Length <> 0)) & String(Length, "0")) >>>> Else >>>> RoundLen = "0" >>>> End If >>>> End Function >>>> >>>> -- >>>> Rick (MVP - Excel) >>>> >>>> >>>> "Rick Rothstein" <rick.newsNO.SPAM@NO.SPAMverizon.net> wrote in message >>>> news:eZlwp71lJHA.4520@TK2MSFTNGP03.phx.gbl... >>>>> Sorry about my other post... I misread what you wanted to do. Give the >>>>> following function a try (it is a little shorter and does less work >>>>> than your function)... >>>>> >>>>> Function RoundLen(Value As Double, Length As Integer) As Double >>>>> Dim Num As Double >>>>> Dim Parts() As String >>>>> Num = CDbl(Format(Value, "0.##############################;;0")) >>>>> If InStr(Num, "E") Then >>>>> Parts = Split(CStr(Num), "E") >>>>> RoundLen = Format(Parts(0), "0." & String(Length, "0") & "E" & >>>>> Parts(1)) >>>>> ElseIf Num <> 0 Then >>>>> RoundLen = Format(Num, "0." & String(Length, "0")) >>>>> End If >>>>> End Function >>>>> >>>>> -- >>>>> Rick (MVP - Excel) >>>>> >>>>> >>>>> "Rick Raisley" <heavymetal-A-T-bellsouth-D-O-Tnet> wrote in message >>>>> news:%23B4vby0lJHA.1340@TK2MSFTNGP06.phx.gbl... >>>>>> I'm sure this is more complex than necessary, but I've written the >>>>>> following routine to handle my requirement: >>>>>> >>>>>> Function RoundLen(ByVal a As Double, ByVal length As Integer) As >>>>>> String >>>>>> Dim g As String, x As Integer, ohs As String, neg As String >>>>>> >>>>>> If a = 0 Then >>>>>> 'Don't pad zeros. Makes them look more obvious. >>>>>> RoundLen = "0" >>>>>> Exit Function >>>>>> End If >>>>>> If a < 0 Then neg = "-" >>>>>> 'I prefer constant length numbers with negative sign added. If >>>>>> total length must be constant, use length = length - 1 >>>>>> a = Abs(a) >>>>>> ohs = String(length, "0") >>>>>> x = Int(Log(a) / Log(10#)) >>>>>> If a > CDbl(String(length, "9")) Then >>>>>> 'For very large numbers, revert to scientific notation >>>>>> g = Format$(a, "0." & Left$(ohs, length - 7) & " E+00") >>>>>> ElseIf x / length < -0.5 Then >>>>>> 'For very small numbers, revert to scientific notation. x/length >>>>>> seems to optimize when to use it for most displayed precision >>>>>> g = Format$(a, "0." & Left$(ohs, length - 7) & " E+00") >>>>>> ElseIf a < 1 Then >>>>>> 'Handles numbers less than 1 >>>>>> g = Format$(a, "0." & Left$(ohs, length - 2)) >>>>>> ElseIf x > length - 2 Then >>>>>> 'Handles numbers of maximum length only, with decimal >>>>>> g = Format$(a, Left$(ohs, length)) >>>>>> Else >>>>>> 'Handles "normal" numbers, with 1 or more places on each side of >>>>>> decimal. >>>>>> g = Format$(a, "0." & Left$(ohs, length - x - 2)) >>>>>> End If >>>>>> RoundLen = neg & g >>>>>> End Function >>>>>> >>>>>> Any simplification suggestions would be appreciated. ;-) >>>>>> >>>>>> -- >>>>>> Regards, >>>>>> >>>>>> Rick Raisley >>>>>> heavymetal-A-T-bellsouth-D-O-T-net >>>>>> >>>>>> "Rick Raisley" <heavymetal-A-T-bellsouth-D-O-Tnet> wrote in message >>>>>> news:eal4vj0lJHA.3888@TK2MSFTNGP02.phx.gbl... >>>>>>> It also doesn't work well with negative numbers, but thanks. >>>>>>> >>>>>>> -- >>>>>>> Regards, >>>>>>> >>>>>>> Rick Raisley >>>>>>> heavymetal-A-T-bellsouth-D-O-T-net >>>>>>> >>>>>>> "Richard Mueller [MVP]" <rlmueller-nospam@ameritech.nospam.net> >>>>>>> wrote in message news:uUAub4rlJHA.5980@TK2MSFTNGP06.phx.gbl... >>>>>>>>A solution that does not handle scientific notation: >>>>>>>> >>>>>>>> FormatNumber(strValue, 10 - Int((Log(strValue)/Log(10)) + 1)) >>>>>>>> >>>>>>>> -- >>>>>>>> Richard Mueller >>>>>>>> MVP Directory Services >>>>>>>> Hilltop Lab - http://www.rlmueller.net >>>>>>>> -- >>>>>>>> >>>>>>>> "Rick Raisley" <heavymetal-A-T-bellsouth-D-O-Tnet> wrote in message >>>>>>>> news:u94X4hrlJHA.1388@TK2MSFTNGP06.phx.gbl... >>>>>>>>>I want to show a number as precisely as I can, using, say, 10 >>>>>>>>>characters. Something like: >>>>>>>>> >>>>>>>>> 12.3456789 >>>>>>>>> 1234.56789 >>>>>>>>> >>>>>>>>> I'm using a Double for the calculations, and have thus far done >>>>>>>>> this: >>>>>>>>> >>>>>>>>> '(n is a Double) >>>>>>>>> >>>>>>>>> Label1.Text = Left$(CStr(n), 10) >>>>>>>>> >>>>>>>>> That works okay except for two cases. First, it doesn't ROUND to >>>>>>>>> those 10 digits, it truncates the text. So 0.499999999999999 reads >>>>>>>>> as 0.49999999 instead of 0.50000000. >>>>>>>>> >>>>>>>>> Second, at some point with small and large numbers, I'll probably >>>>>>>>> want to switch to Scientific notation, in order to have reasonable >>>>>>>>> accuracy, or display very large numbers. So a Double value of >>>>>>>>> 1.23456789E-7 would show like that, instead of 0.00000012. And >>>>>>>>> numbers larger than 9999999999 could still be displayed. >>>>>>>>> >>>>>>>>> I realize I can make a series of If-Then statements, or Select >>>>>>>>> Case, to cover all the bases. But I wondered if there was a >>>>>>>>> simpler way of at least doing the first item. The second is easily >>>>>>>>> covered by two statements (greater than a given number and less >>>>>>>>> than a different given number). >>>>>>>>> >>>>>>>>> -- >>>>>>>>> Regards, >>>>>>>>> >>>>>>>>> Rick Raisley >>>>>>>>> heavymetal-A-T-bellsouth-D-O-T-net >>>>>>>>> >>>>>>>>> >>>>>>>> >>>>>>>> >>>>>>> >>>>>>> >>>>>> >>>>>> >>>>> >>>> >>> >> >> > Give something like this a try...
Label1.Text = CDbl(Format(n, "0.##############################")) -- Show quoteHide quoteRick (MVP - Excel) "Rick Raisley" <heavymetal-A-T-bellsouth-D-O-Tnet> wrote in message news:u94X4hrlJHA.1388@TK2MSFTNGP06.phx.gbl... >I want to show a number as precisely as I can, using, say, 10 characters. > Something like: > > 12.3456789 > 1234.56789 > > I'm using a Double for the calculations, and have thus far done this: > > '(n is a Double) > > Label1.Text = Left$(CStr(n), 10) > > That works okay except for two cases. First, it doesn't ROUND to those 10 > digits, it truncates the text. So 0.499999999999999 reads as 0.49999999 > instead of 0.50000000. > > Second, at some point with small and large numbers, I'll probably want to > switch to Scientific notation, in order to have reasonable accuracy, or > display very large numbers. So a Double value of 1.23456789E-7 would show > like that, instead of 0.00000012. And numbers larger than 9999999999 could > still be displayed. > > I realize I can make a series of If-Then statements, or Select Case, to > cover all the bases. But I wondered if there was a simpler way of at least > doing the first item. The second is easily covered by two statements > (greater than a given number and less than a different given number). > > -- > Regards, > > Rick Raisley > heavymetal-A-T-bellsouth-D-O-T-net > > VB6 would be a bad program language as it was doing it with floating point
Value types in another way then you describe. Simple don't use floating point types as you want to round. Cor Show quoteHide quote "Rick Raisley" <heavymetal-A-T-bellsouth-D-O-Tnet> wrote in message news:u94X4hrlJHA.1388@TK2MSFTNGP06.phx.gbl... >I want to show a number as precisely as I can, using, say, 10 characters. >Something like: > > 12.3456789 > 1234.56789 > > I'm using a Double for the calculations, and have thus far done this: > > '(n is a Double) > > Label1.Text = Left$(CStr(n), 10) > > That works okay except for two cases. First, it doesn't ROUND to those 10 > digits, it truncates the text. So 0.499999999999999 reads as 0.49999999 > instead of 0.50000000. > > Second, at some point with small and large numbers, I'll probably want to > switch to Scientific notation, in order to have reasonable accuracy, or > display very large numbers. So a Double value of 1.23456789E-7 would show > like that, instead of 0.00000012. And numbers larger than 9999999999 could > still be displayed. > > I realize I can make a series of If-Then statements, or Select Case, to > cover all the bases. But I wondered if there was a simpler way of at least > doing the first item. The second is easily covered by two statements > (greater than a given number and less than a different given number). > > -- > Regards, > > Rick Raisley > heavymetal-A-T-bellsouth-D-O-T-net > > 
Other interesting topics
OT: Win 3.1
optional args to a class.Init method Save Picture Q IE menu extension Should Reg-free COM still utilise an installation procedure? Office 2003 installer pops up when loading a project Gradient Windows Update KB960715 blocks MSFLXGRD.OCX!!!!! Any solution ? Unsigned C long to signed VB5 Long query. Search a Combobox |
|||||||||||||||||||||||
