"Yu, Ken" wrote: > The creation time is a 32-bit value. Have you taken account the word > ordering difference between Palm and Intel machines? (That's if you are > using a PC) Yep, took that into account. Just to make sure I'm reading it properly, I've created a field in my record structure that is UInt32, that I read-back correctly.
"David Fedor" wrote: > Save yourself some effort and just read the docs of the PRC, PDB and PQA > file formats. It is available at http://www.palmos.com/dev/tech/docs/ *doh! slap on my head* After reading that the pdb structure will not be officially documented in the knowledge base, I tried to figure out the structure myself. Then, I searched thru the newsgroup and saw the roadcoder post. "Keith" wrote: > Really? You mean you've managed to eliminate the only correct answer? :-) Well, I was really hoping that it is zero-based on 1/1/1904, and tried to massage the numbers to get that , but nothing sensible came out. So, I created a totally new database, with just one record, containing a UInt32 value obtained with UInt32 t=TimGetSeconds(), and a Char c[10] = "ZzZzZzZzZz" and closed it immediately. ("ZzZzZzZzZz" as a beacon to tell me where t is). My guess would be that the creation_date would be very close to t. Turns out creation_date and t were way off. Seems like creation_date is zero-based on 31/12/1969 1600hrs. I hope I am wrong, 'cause if I'm not, we would all have to check the OS version that created the pdb. Please help. I used "PAR.EXE" to decompile my pdb, it returns with invalid dates for both creation_date and modification_date. "PDB.EXE" by Weiske reads the dates properly. -- For information on using the ACCESS Developer Forums, or to unsubscribe, please see http://www.access-company.com/developers/forums/
