Vista - Need help to split a string into text and numeric values !!

Hi,

I need a way to extract the numbers from units. I read each value into
a variable strline and then need to split the units and numbers into 2
seperate variables var1 and var2. Here's the input.txt file and part
of the code.

Thanks,
SA

Part of the code:
============
strline=objFileR.readLine

Do until objFileR.AtEndofStream
if RegExprTest("K|M|MHz|GHz", strline) then
'var1 = gets numbers
'var2 = gets units


Input.txt:
======
200K
750K
1M
915MHz
433.92MHz
802.11GHz
2.4GHz
2.45GHz

<karsagarwal@xxxxxx> wrote in message
news:b3562793-eba7-432d-ad0b-5db1d9a85e05@xxxxxxExamine the following script. The key things to note are the use of
captures "()" in the pattern and the use of the SubMatches array.

Option Explicit

Dim rgx: Set rgx = New RegExp

rgx.Pattern = "([\d\.]+)(.+)"

Dim oMatch
For Each oMatch in rgx.Execute("433.92MHz")
MsgBox oMatch.SubMatches(0) & " : " & oMatch.SubMatches(1)
Next


--
Anthony Jones - MVP ASP/ASP.NET

On Sep 2, 2:17*pm, "Anthony Jones" <AnthonyWJo...@xxxxxx>
wrote:The link does not work. CAn you please explain in detail. I can't seem
t ofigure it out.

Thanks,
SA

On Sep 2, 2:49*pm, "karsagar...@xxxxxx" <karsagar...@xxxxxx>
wrote:Anthony:

A very big thank you for your help. It works like magic. Could you
please explain the pattern search in detail.

Grateful,
SA

On Sep 2, 3:23*pm, "karsagar...@xxxxxx" <karsagar...@xxxxxx>
wrote:Please if some one can explain

rgx.Pattern = "([\d\.]+)(.+)"

This does not work with a "few changes" to my input.txt. So if I have
just a 1, 3.65 or MBRS130LT3
with no units (K ,MHz etc) I get the following output.

Output Sample
==========
200K = 200 : K (Correct)
1 = blank should be 1
3.65 = 3.6 : 5. should be 3.65
MBRS130LT3 = 130 : LT should be MBRS130LT3

Input.txt:
======
200K
750K
1
1M
915MHz
433.92MHz
802.11GHz
MBRS130LT3
3.65
2.4GHz
2.45GHz


I need to parse data with % and numbers example:

<karsagarwal@xxxxxx> wrote in message
news:25becee5-af99-423e-a921-9476e7ea6b3a@xxxxxx
On Sep 2, 3:23 pm, "karsagar...@xxxxxx" <karsagar...@xxxxxx>
wrote:Please if some one can explain

rgx.Pattern = "([\d\.]+)(.+)"

This does not work with a "few changes" to my input.txt.


===> I'm no regular expression guru, but it seems to me that Anthony
developed an expression that was based on an assumption about the nature of
the data, specifically having numeric characters followed by others the
first of which is non-numeric. Before you can arrive at a scripting
solution, you need to fully define what the scope of the problem is. If your
input might contain blank lines, binary data, data with the sequence in
reverse order (i.e. $15.25), random input, numeric only, non-numeric only,
or with numbers spelled out (two KG), this needs to be spelled out *before*
any code is written.

/Al




So if I have
just a 1, 3.65 or MBRS130LT3
with no units (K ,MHz etc) I get the following output.

Output Sample
==========
200K = 200 : K (Correct)
1 = blank should be 1
3.65 = 3.6 : 5. should be 3.65
MBRS130LT3 = 130 : LT should be MBRS130LT3

Input.txt:
======
200K
750K
1
1M
915MHz
433.92MHz
802.11GHz
MBRS130LT3
3.65
2.4GHz
2.45GHz


I need to parse data with % and numbers example:

Al Dunbar wrote:Hi Al,
it's hard to find your reply in the wealth of unnecessarily reposted parts.
Also your browser doesn't seem to indent the old messages.
I know, even OE can be trimmed to apply to usenet standards ;-)


With input data from OP and this changed Regexp
regExpr.Pattern = "^([0-9\.]*)([^0-9\.]*.*)"

I get this output:
==screen=copy===============================================================
14:20:55 C:\Test\vbs>RegExp2.vbs
(200K) 200 : K
(750K) 750 : K
(1) 1 :
(1M) 1 : M
(915MHz) 915 : MHz
(433.92MHz) 433.92 : MHz
(802.11GHz) 802.11 : GHz
(MBRS130LT3) : MBRS130LT3
(3.65) 3.65 :
(2.4GHz) 2.4 : GHz
(2.45GHz) 2.45 :
GHz==screen=copy===============================================================

The problem in the first pattern was the + which requires at least one
occurence of the tagged part.

As either could be absent and the definition was changed underway it
couldn't work.

The new pattern takes all digits and dots anchored from start(^) upto any
number of non digit or dot characters (negated class with ^ inside
brackets) followed by the remainder of the line =.* you could also add a
$sign but this is redundant .

Here my whole RegExp2.vbs________________________________________________
Option Explicit
Dim fiiLename, fso, txs, text, regExpr, oMatch, oMatches, retVal
Const ForReading = 1

fiiLename = "C:\Test\Input.txt"
set fso = createobject("scripting.filesystemobject")
set txs = fso.opentextfile(fiiLename, ForReading, False)

Set regExpr = New RegExp
regExpr.Pattern = "^([0-9\.]*)([^0-9\.]*.*)"

Do While txs.AtEndOfLine <> True
text = txs.readline
Set oMatches = regExpr.Execute(text)
For Each oMatch in oMatches
RetVal = oMatch.SubMatches(0) & vbTab & ": " & oMatch.SubMatches(1)
Next
wscript.echo "(" & LJust(text & ")", 15 ) & retVal
Loop

Function LJust(Str, iLen)
If Len (Str) < iLen then
LJust=Left(Str & Space(iLen), iLen)
Else
LJust=Str
End If
end Function


--
HTH
Matthias

<karsagarwal@xxxxxx> wrote in message
news:25becee5-af99-423e-a921-9476e7ea6b3a@xxxxxx
On Sep 2, 3:23 pm, "karsagar...@xxxxxx" <karsagar...@xxxxxx>
wrote:Please if some one can explain

rgx.Pattern = "([\d\.]+)(.+)"

This does not work with a "few changes" to my input.txt. So if I have
just a 1, 3.65 or MBRS130LT3
with no units (K ,MHz etc) I get the following output.

Output Sample
==========
200K = 200 : K (Correct)
1 = blank should be 1
3.65 = 3.6 : 5. should be 3.65
MBRS130LT3 = 130 : LT should be MBRS130LT3

Input.txt:
======
200K
750K
1
1M
915MHz
433.92MHz
802.11GHz
MBRS130LT3
3.65
2.4GHz
2.45GHz


I need to parse data with % and numbers example:

--------------------------------------------------------------------------
There are many flavors of Regular Expression engines, and even if there were
only one universal RE engine, regular expressions would still be a complex
and confusing subject. It would be nice if there were a regular expression
interpreter, to convert something like:
"([\d\.]+)(.+)"
into:
Capture
Any character in "\d\."
+ (one or more times)
End Capture
Capture
. (any character)
+ (one or more times)
End Capture

or convert :
"^([0-9\.]*)([^0-9\.]*.*)"
into:
^ (anchor to start of string)
Capture
Any character in "0-9\."
* (zero or more times)
End Capture
Capture
Any character not in "0-9\."
* (zero or more times)
. (any character)
* (zero or more times)
End Capture

The interpretations above are how the dot net RE engine would execute the
sample regular expressions. VBScript's RE engine might interpret them
differently.

I have found only one free RE interpreter, called Regular Expression
Workbench, which is dot net specific, but correctly interprets many VBScript
regular expressions. I'm not sure if it is still available -- Microsoft
rearranges (maybe scrambles is a better word) its web sites pretty often.
Last year the following URL took you to the download page:
http://www.gotdotnet.com/Community/U...1-4EE2729D7322

You might search for something like:
Regular Expression Workbench
Eric Gunnerson (EricGu@xxxxxx)

-Paul Randall

Paul Randall wrote:
I developed my version inside Textpad editor, but there are differences in
the regex derivates.

Also you will IMHO have to learn by experience, cause some not so obvious
things like the greediness without which my regex wouldn't work.
Here are some other/more links:
<http://www.ultrapico.com/Resources.htm>

--
HTH
Matthias

"Matthias Tacke" <Matthias@xxxxxx> wrote in message
news:6i7pgiFphrqfU1@xxxxxx
Thanks for the link. I think it will help me learn more about creating my
own regular expressions and about analyzing RE's I find on the net.

-Paul Randall