At 10:24 PM 11/9/2001, Chris Winters wrote: >* Matt Sergeant ([EMAIL PROTECTED]) [011109 02:37]: > > We may wish to consider either simple INI files (which is what I'm > > using to bootstrap XML::SAX, see the perl-xml list for details), or > >I think INI files are an excellent middle ground between >expressiveness and simplicity, and IME they're easily understood by >non-hackers as well.
INI... Shudder. :) Actually INIs are reasonable. I prefer calling them property files (in terms of mapping to an API for also reading those files and getting the same basic instinct out of them. >I've also found an interesting side-effect of using them -- because >it's more difficult to make structures of arbitrary depth in an INI >configuration, you force yourself to be clear rather than rely on the >fact that you can just autovivify your way out of a sticky >situation. (OpenInteract will have the option to use an INI-style file >for the main server configuration with the next version.) > > > something like Ingy's XML-come-Data-Dumper-come-Python config > > files. I forget what it's called right now (MinML?). > >Data::Denter -- it's quite readable. > >As much as I love them and use them everywhere, the main issue I have >with Perl-config files are comments -- eval'ing a data structure and >then re-serializing it loses all the comments, unless the comments >themselves are part of the data structure which is just evil. I >haven't seen a way around this but would love to be surprised :-) That's interesting. Not that I am on a particular XML vs Perl kick but are comments also part of typical XML structures? I thought they were ignored when reading them.
