Re: RFC: Automatic logger module

[afoken]

I have written a Perl module that redirects the the output of STDOUT and STDERR to a file very easily

In one word: Why?

I don't understand why this should ever be useful.

From a shell environment (i.e. during development), I/O redirection can be done at the shell level very easily. perl myscript.pl < source.txt > output.txt 2> errors.txt or perhaps generator foo bar | perl myscript.pl 2> errors.txt | tee output.txt | grep waddawadda | sort. Or very simple: perl myscript.pl 2>&1 | less

For "daemonized" environments, where a perl script runs "in background", the daemonizing environment handles I/O redirection and especially logging in any way you can think of, without changing a single bit of the source code. Syslog, simple logfile, rotated logfile, database, whatever.

Started from inetd and tools like tcpserver, STDOUT is used to answer requests coming in via STDIN. Moving output from STDOUT to a logfile instantly makes the program useless.

For use in interactive programs, redirecting both STDOUT and STDERR to a logfile is almost always wrong. STDOUT is for results, STDERR is for warnings and errors. When STDOUT is redirected to a logfile and mixed with warnings and errors, it is impossible to post-process the results. Things like perl myscript.pl foo bar | grep waddawadda | sort simply won't work any more.

Also, please do give some suggestions for the module name. Do you like the name 'Auto::Log'?

No, I don't like it. The primary purpose is logging, not "Auto". So I think the name should start with "Log". "Auto" does not describe at all what this module does. "CaptureAllOutput" does. So perhaps "Log::CaptureAllOutput"?

Or "Log::I::don't::like::shell::redirections"? ;-)

Alexander