[prev in list] [next in list] [prev in thread] [next in thread] 

List:       hurd-bug
Subject:    Re: gftp: FTBFS on hurd-i386: fatal error: stropts.h: No such file or directory
From:       hahawang <hahwang () yandex ! com>
Date:       2021-05-10 6:51:05
Message-ID: 1366aa15-c1c1-53b3-b75c-9dc0a7182ff3 () yandex ! com
[Download RAW message or body]

Hi, Jess

     You are right, after I pushed this patch to the upstream repository 
and the maintainer merged it.

     He gave me an reply saying the `stropts.h` now seemed meaningless 
since most major platforms

do not support it, so does GNU Hurd. These lines will be completely 
removed in the future.

     Thanks for your kind advises.

     The original reply message and the issue I have opened is here.

     https://github.com/masneyb/gftp/issues/110

---

hahawang

On 2021/5/10 上午1:37, Jessica Clarke wrote:
> On 9 May 2021, at 14:56, hahawang <hahwang@yandex.com> wrote:
> > Package: gftp
> > Severity: important
> > Version: 2.7.0b-1
> > Tags: ftbfs,patch
> > User:hahawang
> > Usertags: hurd,hurd-i386
> > X-Debbugs-CC:debian-hurd@lists.debian.org,bug-hurd@gnu.org
> > 
> > I decide to fix a broken package found at the recommended \
> > page(https://people.debian.org/~sthibault/out_of_date.txt)named `gftp`. 
> > After I download the package source and try to build without any modifications \
> > under the debian hurd running in qemu (debian-hurd-20210219.img),  I got the \
> > following error. 
> > ```
> > 
> > pty.c:64:10: fatal error: stropts.h: No such file or directory
> > 64 | #include <stropts.h>
> > > ^~~~~~~~~~~
> > compilation terminated.
> > 
> > ```
> > 
> > So that I decide to check the source code and find the following line:
> > 
> > 
> > ```
> > 
> > #if !(defined(__FreeBSD__) || defined(__NetBSD__) || defined(__APPLE__) || \
> > defined(__linux__)) #include <stropts.h>
> > #endif
> > 
> > ```
> > 
> > Obviously, the GNU Hurd also haven't the `stropt.h`, so It is better to add a \
> > check of GNU-hurd at this line. By following the document available at \
> > https://sourceforge.net/p/predef/wiki/OperatingSystems/, 
> > I have fixed the build failure, patches avaiable at the end of this email.
> > 
> > Although, IMHO it's trivial, so I decide to report this bug along with a patch \
> > directly to the debian bug tracking system, but also hopes for your advises and \
> > reviews! 
> > Thank you!
> > 
> > ---
> > 
> > hahawang
> > 
> > 
> > --- a/lib/pty.c
> > +++ b/lib/pty.c
> > @@ -60,7 +60,7 @@
> > 
> > #elif HAVE_GRANTPT
> > 
> > -#if !(defined(__FreeBSD__) || defined(__NetBSD__) || defined(__APPLE__) || \
> > defined(__linux__)) +#if !(defined(__FreeBSD__) || defined(__NetBSD__) || \
> > defined(__APPLE__) || defined(__linux__) || defined(__gnu_hurd__)) #include \
> > <stropts.h> #endif
> Why is this not just #ifdef __sun upstream? AFAIK Solaris is the only OS that
> implements that mess. Or, better yet, if it’s optional, just delete the code
> that needs this header? Or make it an autoconf check. This is hands down the
> worst way upstream could write this code.
> 
> Jess
> 


[prev in list] [next in list] [prev in thread] [next in thread]