From: <ir...@ms...> - 2002-09-04 19:41:38
|
On Wed, Sep 04, 2002 at 02:52:59PM -0400, Steve Howell wrote: > > For date/time parsing, if the timestamp type is not removed, I want an > > option to disable it, or to disable it globally and reenable it only for > > certain fields (dictionary keys). > > > > Well, one proposal on the table is to make the timestamp type have a very > particular regex that innocent users are not likely to collide with. Not my > favorite solution, but a step in the right direction. > > Clark is actually the guy who did much of the querying code, and he's making the > interesting argument that we need some kind of YAML stance on dates, because of > all these future YAML applications. I could live with a precise timestamp regex that my datetimes wouldn't collide with (since they don't have seconds, hundredths or the timezone suffix), if the docs say that timestamp is a specific type for a specific use (YAML's querying code), and that it can be used by user code that accepts timestamps' syntax/precision restrictions, but it's not a general date/time type. Otherwise I feel like a dog seeing a bone behind a piece of glass. "Wow, this wonderful timestamp type but I can't use it because it's too rigid for many uses." More kudos if we provide examples in the docs or in an examples/ directory that show date/time parsing of non-timestamps using hooks or application code, so that every date-using user doesn't have to reinvent the wheel. -- -Mike (Iron) Orr, ir...@ms... (if mail problems: ms...@oz...) http://iron.cx/ English * Esperanto * Russkiy * Deutsch * Espan~ol |