<[EMAIL PROTECTED]> wrote on 01/30/2007 03:24:51 PM: > [EMAIL PROTECTED] wrote: > > > > I probably should ask my question this way: What is the *safe* method for > > 32 bit machines to build Sqlite 3.3x? > > > > "Safe" is relative. I know of a few companies that use the technique > I outlined earlier. But I have never personally tested a 32-bit build > so I cannot say what problems might come up. I would expect to find > problems if you try to insert an integer that cannot be represented in > only 32-bits, for example. But as long as you stick to smaller > integers I'm guessing everything will likely work ok. > > -- > D. Richard Hipp <[EMAIL PROTECTED]> > > > -----------------------------------------------------------------------------
> To unsubscribe, send email to [EMAIL PROTECTED] > ----------------------------------------------------------------------------- > Good News! I dropped the xlong.h, and put the defined the x64 stuff as long and unsigned long, and it compiled 1 2 3: >From sqlite3.h: #ifdef SQLITE_INT64_TYPE typedef SQLITE_INT64_TYPE sqlite_int64; typedef unsigned SQLITE_INT64_TYPE sqlite_uint64; #elif defined(_MSC_VER) || defined(__BORLANDC__) typedef __int64 sqlite_int64; typedef unsigned __int64 sqlite_uint64; #else typedef long sqlite_int64; typedef unsigned long sqlite_uint64; typedef long i64; typedef unsigned long u64; #endif I will begin testing to see what that does to the capabilities of the database....But maybe later I will get GCC version 4 built, and i can try again with the original source. Thanks again! ----------------------------------------------------------------------------- To unsubscribe, send email to [EMAIL PROTECTED] -----------------------------------------------------------------------------
