hello
pls how can i achieve the unix function tr -cs 'A-Za-z' '\n' < p.txt  in perl.
such that if i have a file called p.txt containing the following

<DOC>
<PER> peter </PER> asked,  has the bus arrived?. but she replied  with anger 'NO!!!!!!'.
</DOC>

i want the program to make a newline anytime it sees any word ^ with< and $ with >  and anytime it sees white space or non consecutive alphanumeric characters such as .,?!  
finally It should be able to squeeze spaces caused by the '\012' .

thus the output should look like this-:
peter
asked
has
the
bus
arrived
but
she
replied
with
anger
no

cheers
baby perl

Your Unix tr will translate straight into Perl but it doesn't do what you're looking for (even in Unix):

Code:

earth-wind-and-fire:~/feb05 grahamellis$ perl -pe 'tr /A-Za-z/\n/cs' p.txt DOC PER peter PER asked has the bus arrived but she replied with anger NO DOC earth-wind-and-fire:~/feb05 grahamellis$

We're primarily here to help with problems that you're having in writing Perl programs and not to provide complete solutions (people get paid for that and I would hate to put them out of business   .   How far have you got?   Can you post up the Perl code that you're working on so that we can offer help?

Personally, I wouldn't look to doing a direct translation and using Perl's tr function;  I would probably write something like:

Code:

open (FH,"p.txt"); read (FH,$in,-s "p.txt"); $in =~ s/<.*?>//sg; @words = grep(/./,("\n",split(/[^[:alnum:]]+/s,$in))); print (join("\n",@words),"\n");

---