Ah - I see the problem. I do a regex to replace all hex encoded characters BEFORE I split on &, thusly all %26's are replaced with & before I do the split.
In terms of "why reinvent the wheel", I prefer not to use CGI.pm for simple form processing because it prevents me from doing something like print >> End_of_form ...lots of HTML code... End_of_form Unless I'm missing something there, too. Thanks very much! -John On 12/9/02 2:21 PM, "Wiggins d'Anconia" <[EMAIL PROTECTED]> wrote: > Do you know what browser(s)? Most browsers should encode the & as "%26" > and then your split would work properly. On a side note why re-invent > the wheel, when it has already been invented lots of times, see CGI > module on CPAN, or any one of the hundreds of CGI sites that give you > source code to parse query strings.... > > http://danconia.org > > > John Stokes wrote: >> I have an ongoing problem. >> >> As you know, HTML encodes GET and POST requests using the format: >> "myurl.com?name1=value1&name2=value2&..." >> >> My problem is, I have users constantly using ampersands (&) in text fields >> (ex: John & Jane Doe), so my data comes across as: >> "?name1=value1&name2=val&ue2=&name3=value3" etc... >> >> So, when I split this into a hash, the data after a user-entered & is >> effectively lost. My current solution to this is to add a Javascript that >> prevents the user from submitting the form if there's an & in the likely >> text fields, BUT... >> >> Is there an elegant solution to this? Can Perl process this effectively? >> >> Thanks. >> -- -John Stokes Computer Psychiatrist (Director of Information Technology) Church Resource Ministries [EMAIL PROTECTED] Three Pillars: Humility, Communication, Balance -- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
