Trying to decipher something 
Author Message
 Trying to decipher something

I'd appreciate any help.  I'm trying to fix a small problem with a
Perl application that the company I'm working for purchased to handle
web-based email.

Apparently attachments received in word .doc format aren't being
processed properly.  When I try to open one it wants to ask for the
server's username and password in order to be able to access it, I'm
trying to find out why.

I've looked at the link to open the URL, and it seems a little
strange, it asks for the following directory:

webmail.cgi/indiarent.doc?sid=;t=default;msgdata_did=491;do=messages-print_attach

There is a file called webmail.cgi, but it doesn't act as a directory
in any way whatsoever.  Even though it doesn't work when clicked on,
it does work when I right-click and choose to save target.

I can't get my mind around how it could work at all, can anybody tell
me what might be going on here?



Sun, 16 May 2004 08:06:46 GMT  
 Trying to decipher something


Quote:
> ...

webmail.cgi/indiarent.doc?sid=;t=default;msgdata_did=491;do=messages-print_
attach

Quote:

> There is a file called webmail.cgi, but it doesn't act as a directory
> in any way whatsoever.  Even though it doesn't work when clicked on,
> it does work when I right-click and choose to save target.

Why should it act as a directory? Is webmail.cgi the perl script you are
talking about? Nevertheless, anything behind the / will be given as
parameters to this script.

Probably you can analize this script and find out what/how it is doing? Try
to track down the problem and post relevant code here, if it's a perl
issue.

HTH,
Tom



Tue, 18 May 2004 15:40:35 GMT  
 Trying to decipher something

Quote:

>I've looked at the link to open the URL, and it seems a little
>strange, it asks for the following directory:

>webmail.cgi/indiarent.doc?sid=;t=default;msgdata_did=491;do=messages-print_attach

>There is a file called webmail.cgi, but it doesn't act as a directory
>in any way whatsoever.

From the browser's point of view, it *is* acting like a directory.
The browser thinks the document is named indiarent.doc stored in
directory "/something/webmail.cgi/", and has a bunch of CGI parameters.

Quote:
>Even though it doesn't work when clicked on,
>it does work when I right-click and choose to save target.

That implies that the CGI is not returning the right Content-type
header.  Without "Content-type: applicationi/ms-word", the browser
won't know what application to use.  But saving the file to disk
allows Windows to select the application based on file extension.

Quote:
>I can't get my mind around how it could work at all, can anybody tell
>me what might be going on here?

When webmail.cgi runs, it gets two separate pieces of information:
$ENV{QUERY_STRING} = "sid=;t=default;msgdata_did=491;do=messages-print_attach";
$ENV{PATH_INFO} = "/indiarent.doc";

The first piece is used by the CGI to select the appropriate document.
The second piece should be used to send the appropriate Content-type,
which it appears it is not doing.
        -Joe

--
See http://www.inwap.com/ for PDP-10 and "ReBoot" pages.



Thu, 20 May 2004 05:55:30 GMT  
 
 [ 3 post ] 

 Relevant Pages 

1. Trying to decipher something

2. Trying to decipher perlobj.html ...

3. Trying to decipher perlobj.html ...

4. How to Deal With Something.zoo and Something.gz

5. === Calling http:// ....something /something.cgi within another script ===

6. Deciphering Carp::carp message

7. CGI.pm: deciphering error messages

8. Deciphering Error Messages: A Lesson

9. Help deciphering Perl code

10. Need help deciphering code

11. How do I decipher Time format?

 

 
Powered by phpBB® Forum Software