sub-problem with regex

[Alexander Stohr]

i was doing this for several times:
  sub("[A-Z]*$", "", text);
but came to the conclusion that it is broken
since awk version 3.1.1 - i am using a RedHet beta of it.

strings like "something124abc" were converted to "something124"
which is really not what the above code should do.

luckily i came to the following work around:
  sub("[[:upper:]*$", "", text);

but in fact the awk program itself is now under heavy suspect
of beeing buggy. if you think you cant reproduce, then i will
take my time to look after your test cases for verification.

-Alex.

PS: please CC me on answers because i am not a list member.