How to extract a sub string

[chintu red]

Hi all,

I wanted to extract a sub string from a string,
The string in my data set is as follows
222222BTTTTT000000000TTTTTT333333
TTTT0000000000TTTTT22222222222222
00000TTTTTT00000000000000000000000

like wise I have almost 50,000 observations in a variable. I wanted to extract the sub string between two "T"'s in each string or string after "T". So please do help me in this regard.

Thanks in advance.

[Wouter Wakker]

If I understand you well you want something like this.

* Example generated by -dataex-. To install: ssc install dataex
clear
input str34 var
"222222BTTTTT000000000TTTTTT333333"
"TTTT0000000000TTTTT22222222222222"
"00000TTTTTT00000000000000000000000"
end

replace var = subinstr(var, "TTTTTT", "TTTTT", .) // start from longest serie of T's
replace var = subinstr(var, "TTTTT", "TTTT", .) // repeat until they all have same length
gen wanted = substr(var, strpos(var, "TTTT")+4, .)
replace wanted = substr(wanted, 1, strpos(wanted, "TTTT")-1) if strpos(wanted, "TTTT") > 0
list, noobs

. list, noobs

  +------------------------------------------------------------+
  |                              var                    wanted |
  |------------------------------------------------------------|
  |   222222BTTTT000000000TTTT333333                 000000000 |
  | TTTT0000000000TTTT22222222222222                0000000000 |
  | 00000TTTT00000000000000000000000   00000000000000000000000 |
  +------------------------------------------------------------+

[Bjarte Aagnes]

gen tokeep = regexs(1) if regexm(s,"T([0-9]+)[^T]")

[chintu red]

Thanks Bjarte Aagnes it worked. Thanks a lot