On 2/8/1999 at 22:37, pudge@pobox.com (Chris Nandor) wrote: > >More to the point, if the preferred "fix" is to untangle > >existing files by scanning them and resolving ambiguous dates according to > >your > >own arbitrary rules rather than relying on Apple's arbitrary rules before the > >files are used for anything, then I would say that MacPerl would be the first > >choice to use for this task as far as flat ascii files go just because of its > >speed and freely distributable nature. > > Absolutely! :) I'll close this by mentioning that I've confirmed that FileMaker 4 has a serious case of the dumbs when importing files containing 2 digit years into a date field. In that context, everything is 19xx, period (tho it does use the StringToDate call when evaluating manual input). The only good news is that once a date field is populated, it is static. Anyway, that's a major application that is staggering in a major way (even now) given that in real world production work, more data is imported than input manually or created internally. So, I've set to work on my MacPerl freeware to run on text files before they are imported. I've got it reliably (?) converting 1200 dates per second on an aging 8100/100 and remain awed by MacPerl's speed when used for things like this. I'm using 50 as the "pivot point" (the buzzphrase of the y2k culture) and am thinking that a FaceSpan interface and/or hooking an AppleScript up to a drop folder in OS8.5 is the way to go with this to provide an interface and some preference options so that a user can adjust the pivot and store some other settings. Things go in and come back out so fast using MacPerl that I think the only appropriate name for the application is "Monica." Richard Gordon Gordon Consulting & Design Voice: 770-565-8267 Fax: 770-971-6887 ***** Want to unsubscribe from this list? ***** Send mail with body "unsubscribe" to mac-perl-request@iis.ee.ethz.ch