Re: lib/43310: dirent(5) standard compliance

[Matthew Mondor <mm_lists%pulsar-zone.net@localhost>]

The following reply was made to PR lib/43310; it has been noted by GNATS.

From: Matthew Mondor <mm_lists%pulsar-zone.net@localhost>
To: gnats-bugs%NetBSD.org@localhost
Cc: 
Subject: Re: lib/43310: dirent(5) standard compliance
Date: Sat, 15 May 2010 20:41:14 -0400

 On Sat, 15 May 2010 23:45:01 +0000 (UTC)
 Matthew Mondor <mm_lists%pulsar-zone.net@localhost> wrote:
 
 >  [...]
 >  {
 >      struct dirent ent, *entp;
 >  
 >      if ((err = readdir_r(dir, &ent, &entp)) == 0) {
 
 After reading
 http://womble.decadentplace.org.uk/readdir_r-advisory.html I guess that
 indeed the problem could unfortunately be more involved, but if dirent
 was specified by the standard to always hold at least NAME_MAX in
 d_name (or that the implementation should at least take care of any
 additional required allocations transparently), and that NAME_MAX must
 at least be large enough to hold files names for all file systems, this
 would be much simpler and the OSs which can't cope with the above usage
 would remain to be fixed...
 
 At this point I'm not sure that the OG text is cleanly written or that
 I properly interpret its meaning (and I have no foot in the standards
 body to ask for clarifications or to push an agenda for a sane
 definition).
 -- 
 Matt