Question -- why is there no errhint_internal function?

Hi,

For each of the logging functions (see elog.c) there is an associated
XXX_internal function which is equivalent to its partner, but doesn't
translate the fmt message parameter.

errmsg
errmsg_internal - same as errmsg, but doesn't translate the fmt string

errdetail
errdetail_internal - same but errdetail, doesn't translate the fmt string

errhint
errhint_internal - no such thing ???

I noticed today that there is no 'errhint_internal' function partner
for the 'errhint' function.

Now, it might seem that hints are always intended for user output so
of course, you'll always want them translated.... but there are some
calls to this function (like below) where the actual hint message is
already built and translated before %s parameter substitution, so
AFAICT translation aka gettext lookup of just a "%s" format string
doesn't really achieve anything.

$ grep -r . -e 'errhint("%s"' | grep .c:
./contrib/dblink/dblink.c: message_hint ? errhint("%s", message_hint) : 0,
./contrib/postgres_fdw/connection.c: message_hint ? errhint("%s",
message_hint) : 0,
./src/backend/commands/vacuum.c: hintmsg ? errhint("%s", _(hintmsg)) : 0));
./src/backend/commands/tablecmds.c: (wentry->kind != '\0') ?
errhint("%s", _(wentry->drophint_msg)) : 0));
./src/backend/commands/tablecmds.c: errhint("%s", _(view_updatable_error))));
./src/backend/commands/view.c: errhint("%s", _(view_updatable_error))));
./src/backend/utils/misc/guc.c: hintmsg ? errhint("%s", _(hintmsg)) : 0));
./src/backend/utils/misc/guc.c: hintmsg ? errhint("%s", _(hintmsg)) : 0));
./src/backend/utils/misc/guc.c: hintmsg ? errhint("%s", _(hintmsg)) : 0));
./src/backend/utils/misc/guc.c: errhint("%s", GUC_check_errhint_string) : 0));
./src/backend/utils/misc/guc.c: errhint("%s", GUC_check_errhint_string) : 0));
./src/backend/utils/misc/guc.c: errhint("%s", GUC_check_errhint_string) : 0));
./src/backend/utils/misc/guc.c: errhint("%s", GUC_check_errhint_string) : 0));
./src/backend/utils/misc/guc.c: errhint("%s", GUC_check_errhint_string) : 0));
./src/pl/plpgsql/src/pl_exec.c: (err_hint != NULL) ? errhint("%s",
err_hint) : 0,
./src/pl/plpython/plpy_elog.c: (hint) ? errhint("%s", hint) : 0,
./src/pl/plpython/plpy_plpymodule.c: (hint != NULL) ? errhint("%s", hint) : 0,
./tmp.txt:src/backend/utils/misc/guc.c: errhint("%s",
GUC_check_errhint_string) : 0));
./tmp.txt:src/backend/utils/misc/guc.c: errhint("%s",
GUC_check_errhint_string) : 0));
./tmp.txt:src/backend/utils/misc/guc.c: errhint("%s",
GUC_check_errhint_string) : 0));
./tmp.txt:src/backend/utils/misc/guc.c: errhint("%s",
GUC_check_errhint_string) : 0));
./tmp.txt:src/backend/utils/misc/guc.c: errhint("%s",
GUC_check_errhint_string) : 0));

~

I wondered if such code as in those examples might prefer to call
errhint_internal to avoid making an unnecessary gettext lookup of
"%s".

OTOH, was an errhint_internal function deliberately omitted because
calling a superfluous gettext was not considered important enough to
bother?

======

Also, quite similar to this question --- I found a bunch of errmsg and
errdetail calls where the fmt string is just "%s". Are those correct,
or should those really be using the XXX_internal version of the
function instead?

$ grep -r . -e 'errmsg("%s"' | grep .c:
./contrib/pgcrypto/px.c: errmsg("%s", px_strerror(err))));
./src/pl/plperl/plperl.c: errmsg("%s", strip_trailing_ws(sv2cstr(ERRSV))),
./src/pl/plperl/plperl.c: errmsg("%s", strip_trailing_ws(sv2cstr(ERRSV))),
./src/pl/plperl/plperl.c: errmsg("%s", strip_trailing_ws(sv2cstr(ERRSV))),
./src/pl/plperl/plperl.c: errmsg("%s", strip_trailing_ws(sv2cstr(ERRSV))),
./src/pl/plperl/plperl.c: errmsg("%s", strip_trailing_ws(sv2cstr(ERRSV))),
./src/pl/plperl/plperl.c: errmsg("%s", strip_trailing_ws(sv2cstr(ERRSV))),
./src/pl/plperl/plperl.c: errmsg("%s", strip_trailing_ws(sv2cstr(ERRSV))),
./src/pl/plperl/plperl.c: errmsg("%s", strip_trailing_ws(sv2cstr(ERRSV)))));
./src/pl/plperl/plperl.c: errmsg("%s", strip_trailing_ws(sv2cstr(ERRSV)))));
./src/pl/plperl/plperl.c: errmsg("%s", strip_trailing_ws(sv2cstr(ERRSV)))));
./src/pl/plperl/plperl.c: errmsg("%s", strip_trailing_ws(sv2cstr(ERRSV)))));
./src/pl/tcl/pltcl.c: errmsg("%s", emsg),
./src/pl/tcl/pltcl.c: errmsg("%s", UTF_U2E(Tcl_GetString(objv[2])))));

$ grep -r . -e '\serrdetail("%s"' | grep .c:
./src/backend/executor/execExprInterp.c: errdetail("%s",
jsestate->escontext.error_data->message)));
./src/backend/executor/execExprInterp.c: errdetail("%s",
jsestate->escontext.error_data->message)));

======
Kind Regards,
Peter Smith.
Fujitsu Australia