FTBS rhel7 - Needs c99 cflag

[knight-of-ni]

Netatalk 3.1.16 fails to build on el7 due to missing c99 cflag:

/bin/sh ../../libtool  --tag=CC   --mode=compile gcc -DHAVE_CONFIG_H -I. -I../..   -I../../include -I../../include -I../..   -D_U_="__attribute__((unused))" -O2 -g -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector-strong --param=ssp-buffer-size=4 -grecord-gcc-switches -specs=/usr/lib/rpm/redhat/redhat-hardened-cc1  -m64 -mtune=generic -c -o libacl_la-ldap_config.lo `test -f 'ldap_config.c' || echo './'`ldap_config.c
libtool: compile:  gcc -DHAVE_CONFIG_H -I. -I../.. -I../../include -I../../include -I../.. "-D_U_=__attribute__((unused))" -O2 -g -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector-strong --param=ssp-buffer-size=4 -grecord-gcc-switches -specs=/usr/lib/rpm/redhat/redhat-hardened-cc1 -m64 -mtune=generic -c ldap_config.c  -fPIC -DPIC -o .libs/libacl_la-ldap_config.o
ldap_config.c: In function 'acl_ldap_freeconfig':
ldap_config.c:35:5: error: 'for' loop initial declarations are only allowed in C99 mode
     for (int i = 0; ldap_prefs[i].name != NULL; i++) {
     ^
ldap_config.c:35:5: note: use option -std=c99 or -std=gnu99 to compile your code
make[3]: *** [libacl_la-ldap_config.lo] Error 1
make[3]: Leaving directory `/builddir/build/BUILD/netatalk-3.1.16/libatalk/acl'
make[2]: *** [all-recursive] Error 1
make[2]: Leaving directory `/builddir/build/BUILD/netatalk-3.1.16/libatalk'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/builddir/build/BUILD/netatalk-3.1.16'
make: *** [all] Error 2
error: Bad exit status from /var/tmp/rpm-tmp.C4y0fy (%build)

I am working around this in EPEL by forcing -gnu99 in the rpm specfile.

Since el7 is such an old distro, I'm not necessarily arguing for an upstream fix. This is more of an FYI, if the maintainers wish to do anything about it.

[rdmark]

Thanks for reporting! I'm all for making the code more portable within reason.

ldap_config.c:35:5: error: 'for' loop initial declarations are only allowed in C99 mode
     for (int i = 0; ldap_prefs[i].name != NULL; i++) {
     ^

Seems to be as simple as declaring int i; before using it. That's what the subsequent function in this file does. I assume this would make the EL7 compiler happy?

[knight-of-ni]

Ah right, that would certainly fix the issue for older compilers. Thanks.

[rdmark]

Fixing in #484

For the record, netatalk2 is not affected.