From: "Jim Jagielski" <[EMAIL PROTECTED]> Sent: Friday, February 23, 2001 6:08 PM > The best solution would be to abstract things out as much as > possible. An abstract data type that contains whatever information > we need and we pull out whatever info we want by macros or > function calls... > > apr_time_t marker; > long seconds; > long micro_seconds > marker = apr_get_time(NULL); > seconds = apr_conv_time(marker, SECONDS); > micro_seconds = apr_conv_time(marker, MICROSECS); > > as a general idea. I find the example a bit cumbersome, but following your later comments, that's what I propose - make apr_time_t an opaque structure and force the user to call [relatively simple] functions that should be implemented inline or as macros, depending on the platform. We have enough compilers that support inlined code to catch the errors that the macro user/coder would miss.
- Re: SGI Patch 10xpatch-2.0a6-2: Reduce time resolut... Bill Stoddard
- Re: SGI Patch 10xpatch-2.0a6-2: Reduce time resolut... rbb
- Re: SGI Patch 10xpatch-2.0a6-2: Reduce time re... Bill Stoddard
- Re: SGI Patch 10xpatch-2.0a6-2: Reduce tim... Roy T. Fielding
- Re: SGI Patch 10xpatch-2.0a6-2: Reduce... rbb
- Re: SGI Patch 10xpatch-2.0a6-2: Re... Roy T. Fielding
- Re: SGI Patch 10xpatch-2.0a6-... dean gaudet
- Re: SGI Patch 10xpatch-2.... Bill Stoddard
- Re: SGI Patch 10xpatch-2.... Jim Jagielski
- Re: SGI Patch 10xpatch-2.... Jim Jagielski
- Re: SGI Patch 10xpatch-2.... William A. Rowe, Jr.
- Re: SGI Patch 10xpatch-2.... Jim Jagielski
- Re: SGI Patch 10xpatch-2.... Roy T. Fielding
- Re: SGI Patch 10xpatch-2.... Greg Marr
- Re: SGI Patch 10xpatch-2.... William A. Rowe, Jr.
- Re: SGI Patch 10xpatch-2.... Greg Marr
- Re: SGI Patch 10xpatch-2.... Bill Stoddard
- Re: SGI Patch 10xpatch-2.0a6-2: Reduce tim... William A. Rowe, Jr.
- Re: SGI Patch 10xpatch-2.0a6-2: Reduce... Rodent of Unusual Size
- Re: SGI Patch 10xpatch-2.0a6-2: Reduce time re... Rodent of Unusual Size
- Re: SGI Patch 10xpatch-2.0a6-2: Reduce time re... Rodent of Unusual Size
