Apparently a recent round of vandalism by the C++ standards committee
includes disallowing the standard and historic type name "unsigned". Add
a gratuitous "int" to recent changes.

C++ is really a b

Apparently a recent round of vandalism by the C++ standards committee
includes disallowing the standard and historic type name "unsigned". Add
a gratuitous "int" to recent changes.

C++ is really a blight on the world.

show more ...

u_char, u_int, u_long -> unsigned char, unsigned, unsigned long

This makes these headers work without _NETBSD_SOURCE.

Retire varargs.h support. Move machine/stdarg.h logic into MI
sys/stdarg.h and expect compiler to provide proper builtins, defaulting
to the GCC interface. lint still has a special fallback.
Reduce a

Retire varargs.h support. Move machine/stdarg.h logic into MI
sys/stdarg.h and expect compiler to provide proper builtins, defaulting
to the GCC interface. lint still has a special fallback.
Reduce abuse of _BSD_VA_LIST_ by defining __va_list by default and
derive va_list as required by standards.

show more ...

Remove clause 3 and 4 from TNF licenses

add a NULL callback initializer

sprinkle __UNCONST

de-__P -- the hack is long since useless. Discussed with christos,
matt, kleink, others. Approved by christos.

* Improve the comments in <nsswitch.h>
* Provide some more useful default ns_src arrays.

* Re-number NS_FORCEALL from 1<<7 to 1<<8, moving it out of the "public"
bitrange for nsdispatch(3) return values.
* Tweak documentation comments in nsswitch.h
* Ensure the result from the back-end

* Re-number NS_FORCEALL from 1<<7 to 1<<8, moving it out of the "public"
bitrange for nsdispatch(3) return values.
* Tweak documentation comments in nsswitch.h
* Ensure the result from the back-end method is masked with
NS_STATUSMASK before nsdispatch(3) returns it.

show more ...

Implement NS_FORCEALL, which may be set in defaults[0].flags by a caller of
nsdispatch(3) to indicate that all available database methods for a source
will be invoked. This is useful for functions s

Implement NS_FORCEALL, which may be set in defaults[0].flags by a caller of
nsdispatch(3) to indicate that all available database methods for a source
will be invoked. This is useful for functions such as endgrent(3).

show more ...

Add support for dynamically loading nsswitch modules on ELF platforms.
Adapted from FreeBSD. Maintains full backward API / ABI compatbility
with built-in-only nsdispatch().

While here, also make ns

Add support for dynamically loading nsswitch modules on ELF platforms.
Adapted from FreeBSD. Maintains full backward API / ABI compatbility
with built-in-only nsdispatch().

While here, also make nsdispatch() itself thread-safe.

show more ...

Remove variable names from prototypes.

remove declaration of _nsyylineno from here.

Declare _nsyylineno as extern since it belongs in the data section
(due to being initialised in nslexer.[cy]).

Remove 'extern' from function declarations.

change retval of _nsdbtaddsrc and _nsdbtput

use _BSD_VA_LIST_ from <machine/ansi.h> instead of using va_list from
<[stdarg,varargs}.h>. as recommended by Klaus Klein.

always add compat functions in NS_COMPAT_CB; make it the responsibility of
the methods to check that yp/hesiod is supported.

* support extra two arguments to nsdispatch():
const char *method /* method name for dynamic linking */
const ns_src defaults[] /* list of defaults */
* bring ns_src into the public arena for abo

* support extra two arguments to nsdispatch():
const char *method /* method name for dynamic linking */
const ns_src defaults[] /* list of defaults */
* bring ns_src into the public arena for above
* make NS_xxx_CB() expand to nothing instead of { src, NULL } - makes
the tables smaller
* convert 'typedef struct xyz { ... } xyz' -> 'typedef struct { ... } xyz'
(which is how it used to be before i experimented with (and ditched) TAILQs)

show more ...

actually set the cb_data field to the (,C) arg in NS_*_CB()

remove NSSRC_NISPLUS for now, NSDB_NETWORKS is now implemented

merge nsswitch into main branch