[Please ignore my previous post, sent by accident before I finished it..] In message <[EMAIL PROTECTED]>, "John E. Malmberg" <[EMAIL PROTECTED]> writes: >The file system is scheduled for a major overhaul. I can not give >details, but any program that makes any assumptions about undocumented >internal structures such as names of locks may break.
Yes, such a program may break in the future when someone eventually tries to run it under some future version of VMS. In the mean time, there are problems to be solved now where using undisclosed (they hopefully were documented somewhere by the developers) interfaces is the only practical solution. Such a program should isolate and abstract the fragile pieces as much possible. >Now I have put in requests for features to the file system, like a 64 >bit ino_t type, storing at file close the apparent UNIX size of a record >oriented file for stat to return, storing the UTC time in addition to >local time so that stat will run faster. I have also requested support >for alternate filenames, to help support ODS-2/ODS-5/ODS-? and 8.3 file >lookups for applications like Pathworks Advanced Server and SAMBA. Obviously, the first step is making a file header 8K instead of 512 bytes in size :-) For UTC times in the file headers, I think it would be more compatible with ODS-2/5 if you stored the timezone offsets to be applied to the existing local time timestamps. ----------------------------------------------------------------------------- David L. Jones | Phone: (614) 292-6929 Ohio State University | Internet: 140 W. 19th St. Rm. 231a | [EMAIL PROTECTED] Columbus, OH 43210 | [EMAIL PROTECTED] Disclaimer: I'm looking for marbles all day long. PLEASE READ THIS IMPORTANT ETIQUETTE MESSAGE BEFORE POSTING: http://www.catb.org/~esr/faqs/smart-questions.html
