On Wed, Feb 20, 2002 at 11:17:24AM -0800, you [Martin Pool] wrote: > > OK, already fixed. --no-fork would be good to add in the future -- it > can be handy for debugging.
As said, I only added the --dont-fork==--no-detach mostly because I use it in my scripts. (The original --dont-fork did disable just the fork() call, not the whole become_daemon() call, so it was slightly different from --no-detach.) > > O_TEXT and O_BINARY > > Good. > > It might be cleaner to #ifdef on O_BINARY or something that will also > work on MSVC++. O_TEXT and O_BINARY also work on MSVC (and propably on most other windows compilers.) So perhaps #if defined(_WIN32) || defined(__CYGWIN__) #define TEXT O_TEXT #define BINARY O_BINARY #else #define TEXT 0 #define BINARY 0 #endif and fd = open(fname, O_RDONLY | TEXT) (not sure what to call those -- "TEXT"/"BINARY" might collide with something?) Or is #if !defined(O_TEXT) #define O_TEXT 0 #endif #if !defined(O_BINARY) #define O_BINARY 0 #endif better? > Can't you say "rt" as an fopen mode? That works on all C platforms > that I know, and avoids the ifdef. I'm not sure what I was thinking back then :) (This dates back to few years ago). Perhaps I tried "rt", but it didn't work or something. Or perhaps it was just my thinko. It should work now with cygwin, you're right. > > (3) avoid perm check on password file > > This seems good, though again perhaps ideally we would ifdef for "the > windows platform" rather than "cygwin". Ok, how about #if defined(_WIN32) || defined(__CYGWIN__) ? -- v -- [EMAIL PROTECTED] ---- To unsubscribe or change options: http://lists.samba.org/mailman/listinfo/rsync Before posting, read: http://www.tuxedo.org/~esr/faqs/smart-questions.html