syntaxmishap
asked on
Trap Incomplete Downloads
Is there any way to trap incomplete downloads from an apache web server? For instance, if a user cancels the download, or it otherwise errors out, is there any way that I can trap this?
This information IS stored in the access log, but I am looking for a way to trap this via a php or perl script for the purpose of gathering statistical information.
This information IS stored in the access log, but I am looking for a way to trap this via a php or perl script for the purpose of gathering statistical information.
ASKER CERTIFIED SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
I agree with ahoffmann; Apache already tracks this information in the access_log/error_log by putting in a error code for each page request - normally one would run a script against the log files in order to gather statistics. What is gained by doubling the work?
ASKER
ahoffmann, periwinkle -
The benefit to trapping this in real time would be the ability to respond to a cancelled download. An example would be prompting the user with information to assist them in re-downloading, or redirecting the user to a mirror.
The benefit to trapping this in real time would be the ability to respond to a cancelled download. An example would be prompting the user with information to assist them in re-downloading, or redirecting the user to a mirror.
SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
ASKER
periwinkle -
Well, that is the idea; The files will be on the same server...
My best option is looking more and more like putting together a PHP script to serve up the files, which avails me the option of trapping the incomplete downloads that way. I was just trying to avoid that if possible, as putting together code to deal with download managers and still keep accurate statistics will require a bit more time in testing that I was looking to put in.
Well, that is the idea; The files will be on the same server...
My best option is looking more and more like putting together a PHP script to serve up the files, which avails me the option of trapping the incomplete downloads that way. I was just trying to avoid that if possible, as putting together code to deal with download managers and still keep accurate statistics will require a bit more time in testing that I was looking to put in.