Why am I getting a Divide by Zero error?

agent00013 has asked for the wisdom of the Perl Monks concerning the following question:

I'm grabbing a link from HTML and fixing it to take out the /../ and replace it with the absolute address. Here's the regex part of the code and some example variables.

$base_url =
"http://www.something.com/angels/devils";
$base_url =~ s!/[^/]*?$!!; #strip off the subdirectory
print "$base_url \n";
$address =
"http://www.something.com/angels/devils/../faries/";
$address =~ s!.*/\.\./(.*)!$base_url/$1!eg;
### stick this in &grabLinks ###
print "$address \n";
[download]

Output is:

http://leto.advtech.uswest.com/SERVICES
Illegal division by zero at line 14.    #it's line 7 here...

I assume I'm doing something wrong with my formatting, but I can't figure it out. I tried adding {} brackets around the variable names but it wasn't working. Thanx for your help.

Comment on Why am I getting a Divide by Zero error? Download Code

Replies are listed 'Best First'.
Re: Why am I getting a Divide by Zero error? by davorg (Chancellor) on Jun 20, 2001 at 20:15 UTC
The problem line is this one: `$address =~ s!./\.\./(.)!$base_url/$1!eg;` [download] The `/e` option askes Perl to execute the replacement string before using it. Perl is therefore trying to divide $base_url by $1. Presumably, when you evaluate $1 as a number it becomes zero and causes your error. You don't need the `/e` and can safely remove it.	[reply] [d/l]
Re: Why am I getting a Divide by Zero error? by dimmesdale (Friar) on Jun 20, 2001 at 21:57 UTC
`$base_url =~ s!/[^/]?$!!; #strip off the subdirectory` [download] Why are you using `?` here? The greedy quantifier would seem to do what you want. But there's a better choice. A regex is a little bit of overkill here; you might want to check out rindex on the perl manpages, or this site. Your second regex could use a little improving too, though its hard to say what exactly could be done without knowing what $address is holding; I had some advice, but I noticed the /g on the regex. That leads to the idea that there are many addresses there(and that you should be careful with .* then, unless they are separated by \n's).	[reply] [d/l] [select]