Mini::CPAN mirror size

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

Venerables,

I've decided to give ~~Mini::CPAN~~ CPAN::Mini a try, given the fact that I'm often after a picky proxy that slows things down to a crawl. The first, long download left more than 400MB of stuff (when burned on CD), which resolved to occupy more than 700MB on the disk. So far, so good. I've just re-run it to get updates some minutes ago; I was scared by a lot of deletions, so I checked out - 100MB of disk occupation, i.e. 600MB less!

Now I wonder: how big should this mirror be?

Update: The problem was actually that I was working inside Linux but keeping the repository in the FAT32 partition where I keep Windows. In particular, FAT32 does not support case distinction in filenames, and Linux seems to read filenames in lowercase even when they have been registered in uppercase. I've sent to the maintainer a couple of patches I've applied to my local installation - basically, I've added a "case_insensitive" option that should hopefully work the problem around. Thank you cog and brian_d_foy!

Flavio (perl -e "print(scalar(reverse('ti.xittelop@oivalf')))")

Don't fool yourself.

Comment on Mini::CPAN mirror size

Replies are listed 'Best First'.
Re: Mini::CPAN mirror size by cog (Parson) on Apr 07, 2005 at 16:53 UTC
`$ pwd /Users/jac/tmp/perl/minicpan $ du\|tail -1 441M total` [download]	[reply] [d/l]
Re^2: Mini::CPAN mirror size by polettix (Vicar) on Apr 07, 2005 at 17:10 UTC
So, that's it! I'm suspecting something wrong in the script, in the module... or may it be that data are kept on a FAT32 partition which has no support for uppercase/lowercase distinction? Flavio (perl -e "print(scalar(reverse('ti.xittelop@oivalf')))") Don't fool yourself.	[reply]
Re: Mini::CPAN mirror size by brian_d_foy (Abbot) on Apr 07, 2005 at 20:06 UTC
Mine is about 533M right now. 100M is much too small unless you are purposedly not keeping parts of the basic minicpan. I used to keep track of The Schwartz Factor, but that machine ran out of room on which to hold a CPAN mirror (although I can take the number from http://www.cpan.org/ now. Typically that number has been around 0.2. Today the full CPAN is about 2.8G, and my minicpan is about 533M, so that sounds right. :) -- brian d foy <brian@stonehenge.com>	[reply]