How to create a text file that has file names in a directory

[tesu]

In my "D:\work\" folder, I have two types of files, that are "8-digit-number_type.dbf" and "character_type.dbf" as follows: 00093282.dbf, 00345602.dbf, 69209340.dbf, 69806980.dbf, 92406482.dbf, 99999999.dbf,... aaa.dbf, bbb.dbf, ...

I want to take only the "8-digit-number_type.dbf" files, and exclude the character type files. (I don't care about the character type files.) Then, I want to create a text file (say, namelist.txt) that contains the list of names of the dbf files as follows:

00093282

00345602

69209340

69806980

99999999

....

How can I do that in SAS?

Thanks.

[PGStats]

Start with this:

filename toto PIPE "DIR ""D:\work\*.dbf"" " console=min;

data _null_;
length name $200;
infile toto firstobs=6 truncover;
input a$ b$ c$ file $char200.;
file "D:\work\namelist.txt";
name = trim(tranwrd(upcase(file),".DBF",""));
if anyalpha(name) = 0 then put name;
run;

PG

[tesu]

I run it, and the namelist.txt looks like the following:

35,548 00093282

17,539 00095480

12,406 00095520

38,854 00095602

10,927 00098880

18,670 00099999

29,819 00340240

12,952 00340560

569,260 00345602

198,085 00346162

3,458 00347560

25,072 00349999

18,670 00400040

3,010 00409999

65,737 00801692

...

Some numbers that I don't want in the first column. But I can remove the first column later.

Thank you so much.

[PGStats]

Take Mike's code, it's much better, mine was just an approximation. - PG

[MikeZdeb · Accepted solution]

hi ... same idea, use a pipe but with DIR/B...

filename x pipe 'dir/b "z:\*.dbf" ' ;

filename y 'z:\namelist.txt';

data _null_;

infile x dlm='.';

file y;

input name : $20.;

if ^verify(trim(name),'0123456789') then put name;

run;

[tesu]

This code produces the list that I want. Thank you so much.

[Ksharp]

If you only want 8 digit file name ,I prefer to Perl Regular Expression.

if prxmatch('/\d{8}/',name) then put name;

Ksharp