Parsing Text field

[wwyngaard]

I need to parse out a 5 digit number out of a text string. It can appear anywhere within the text so it can't be a left or right word calc. Is there a way to parse a 5 digit number out? It would seem simple. Also it would be nice to be able to get the 3 letter code following the number.

Some example strings.

zJobs1 Archive:34939-oul11-2:IT - PDF Folder:

zRip2Disk Archive:35727-HOR-2(RSIdata):Default Job:Primex12.eps:Stats:

Anchor:55587-afp-02

[BobWeaver]

Try this:

Middle(TheString, Position(

Substitute(Substitute(Substitute(Substitute(Substitute(

Substitute(Substitute(Substitute(Substitute(Substitute(TheString,

"0", "~"),

"1", "~"),

"2", "~"),

"3", "~"),

"4", "~"),

"5", "~"),

"6", "~"),

"7", "~"),

"8", "~"),

"9", "~"),

"~~~~~", 1, 1), 5)

---

[wwyngaard]

Works perfectly. By increasing the last digit I can also get it to grab the 3 letter code and then strip that out by using middle words. If I wanted to just grab the 3 digit code seperately could that be done. I tried reading through the calc to duplicate it to do that and got lost.

Thank you Very Much.

Wayne

[BobWeaver]

What this formula does is replace all the numerals with the squiggle "~" character and then search for starting position of five squiggles in a row which is the starting position of the five digit number. To get the 3 character suffix, just add 6 to the position, and change the length to 3 as follows:

Middle(TheString, 6+Position(

Substitute(Substitute(Substitute(Substitute(Substitute(

Substitute(Substitute(Substitute(Substitute(Substitute(TheString,

"0", "~"),

"1", "~"),

"2", "~"),

"3", "~"),

"4", "~"),

"5", "~"),

"6", "~"),

"7", "~"),

"8", "~"),

"9", "~"),

"~~~~~", 1, 1), 3)

---

Personally, I would just use this to get the whole 9 characters and then process that separately. Considering speed issues, since this is a rather complicated formula, why use it more than once when you can grab the important chunk of 9 characters, and then parse it with much simpler formulas.

[wwyngaard]

That sounds like a good way to go. I wasn't sure if getting all the characters at once then parsing would be faster or not. It makes sense only to run the large parse for 1 calculation.

Thanks,

Wayne