Re: (failed) packages
- To: debian-hurd@lists.debian.org
- Subject: Re: (failed) packages
- From: Paul Emsley <paule@chem.gla.ac.uk>
- Date: 02 Jul 2001 00:24:05 +0100
- Message-id: <[🔎] uywy9q8fd2i.fsf@chem.gla.ac.uk>
- In-reply-to: Marcus Brinkmann's message of "Sat, 30 Jun 2001 13:59:38 +0200"
- References: <m2n.s.15FrRl-000xU8@chiark.greenend.org.uk> <5b4rszdsv7.fsf@chiark.greenend.org.uk> <uywvgleps42.fsf_-_@chem.gla.ac.uk> <20010630135938.A551@212.23.136.22>
>>>>> "MB" == Marcus Brinkmann <Marcus.Brinkmann@ruhr-uni-bochum.de> writes:
MB> On Sat, Jun 30, 2001 at 04:24:29AM +0100, Paul Emsley wrote:
>> I took a look at this list and it seems to me that about 80% of
>> the "fail"s are the MAXPATHLEN, MAX_PATH, MAXHOSTNAMELEN and
>> sys_errlist problems. These have been mentioned before here.
>>
>> It may be a good idea to post a "before" and "after" example of
>> how to fix these problems in the Approved Fashion. It may also
>> be of use for pointing package maintainers at.
MB> There have been many examples in the past. There is no
MB> standard approach, you have to look at how they are used.
OK then, does one conditionalise on (e.g.) MAXHOSTNAMELEN:
#ifdef MAXHOSTNAMELEN
/* do it the traditional way */
#else
/* do it the Hurd way, typically using realloc() */
#endif
for each occurance of MAXHOSTNAMELEN (or whatever)?
MB> [snip explanation of turtle]
Thanks,
Paul.
--
Paul.Emsley@chem.gla.ac.uk
http://www.chem.gla.ac.uk/~paule
Reply to: