URL https://opencores.org/ocsvn/openrisc/openrisc/trunk
Subversion Repositories openrisc

[/] [openrisc/] [trunk/] [rtos/] [ecos-2.0/] [packages/] [net/] [snmp/] [lib/] [v2_0/] [src/] [mib.c] - Blame information for rev 174

Details | Compare with Previous | View Log

//==========================================================================
//
//      ./lib/current/src/mib.c
//
//
//==========================================================================
//####ECOSGPLCOPYRIGHTBEGIN####
// -------------------------------------------
// This file is part of eCos, the Embedded Configurable Operating System.
// Copyright (C) 1998, 1999, 2000, 2001, 2002 Red Hat, Inc.
//
// eCos is free software; you can redistribute it and/or modify it under
// the terms of the GNU General Public License as published by the Free
// Software Foundation; either version 2 or (at your option) any later version.
//
// eCos is distributed in the hope that it will be useful, but WITHOUT ANY
// WARRANTY; without even the implied warranty of MERCHANTABILITY or
// FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
// for more details.
//
// You should have received a copy of the GNU General Public License along
// with eCos; if not, write to the Free Software Foundation, Inc.,
// 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA.
//
// As a special exception, if other files instantiate templates or use macros
// or inline functions from this file, or you compile this file and link it
// with other works to produce a work based on this file, this file does not
// by itself cause the resulting work to be covered by the GNU General Public
// License. However the source code for this file must still be made available
// in accordance with section (3) of the GNU General Public License.
//
// This exception does not invalidate any other reasons why a work based on
// this file might be covered by the GNU General Public License.
//
// Alternative licenses for eCos may be arranged by contacting Red Hat, Inc.
// at http://sources.redhat.com/ecos/ecos-license/
// -------------------------------------------
//####ECOSGPLCOPYRIGHTEND####
//####UCDSNMPCOPYRIGHTBEGIN####
//
// -------------------------------------------
//
// Portions of this software may have been derived from the UCD-SNMP
// project,  <http://ucd-snmp.ucdavis.edu/>  from the University of
// California at Davis, which was originally based on the Carnegie Mellon
// University SNMP implementation.  Portions of this software are therefore
// covered by the appropriate copyright disclaimers included herein.
//
// The release used was version 4.1.2 of May 2000.  "ucd-snmp-4.1.2"
// -------------------------------------------
//
//####UCDSNMPCOPYRIGHTEND####
//==========================================================================
//#####DESCRIPTIONBEGIN####
//
// Author(s):    hmt
// Contributors: hmt
// Date:         2000-05-30
// Purpose:      Port of UCD-SNMP distribution to eCos.
// Description:  
//              
//
//####DESCRIPTIONEND####
//
//==========================================================================
/********************************************************************
       Copyright 1989, 1991, 1992 by Carnegie Mellon University
 
                          Derivative Work -
Copyright 1996, 1998, 1999, 2000 The Regents of the University of California
 
                         All Rights Reserved
 
Permission to use, copy, modify and distribute this software and its
documentation for any purpose and without fee is hereby granted,
provided that the above copyright notice appears in all copies and
that both that copyright notice and this permission notice appear in
supporting documentation, and that the name of CMU and The Regents of
the University of California not be used in advertising or publicity
pertaining to distribution of the software without specific written
permission.
 
CMU AND THE REGENTS OF THE UNIVERSITY OF CALIFORNIA DISCLAIM ALL
WARRANTIES WITH REGARD TO THIS SOFTWARE, INCLUDING ALL IMPLIED
WARRANTIES OF MERCHANTABILITY AND FITNESS.  IN NO EVENT SHALL CMU OR
THE REGENTS OF THE UNIVERSITY OF CALIFORNIA BE LIABLE FOR ANY SPECIAL,
INDIRECT OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING
FROM THE LOSS OF USE, DATA OR PROFITS, WHETHER IN AN ACTION OF
CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF OR IN
CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
*********************************************************************/
/*
 * mib.c
 *
 * Update: 1998-07-17 <jhy@gsu.edu>
 * Added print_oid_report* functions.
 *
 */
/**********************************************************************
        Copyright 1988, 1989, 1991, 1992 by Carnegie Mellon University
 
                      All Rights Reserved
 
Permission to use, copy, modify, and distribute this software and its
documentation for any purpose and without fee is hereby granted,
provided that the above copyright notice appear in all copies and that
both that copyright notice and this permission notice appear in
supporting documentation, and that the name of CMU not be
used in advertising or publicity pertaining to distribution of the
software without specific, written prior permission.
 
CMU DISCLAIMS ALL WARRANTIES WITH REGARD TO THIS SOFTWARE, INCLUDING
ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS, IN NO EVENT SHALL
CMU BE LIABLE FOR ANY SPECIAL, INDIRECT OR CONSEQUENTIAL DAMAGES OR
ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS,
WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION,
ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS
SOFTWARE.
******************************************************************/
#include <config.h>
 
#include <stdio.h>
#include <ctype.h>
#include <sys/types.h>
#if HAVE_NETINET_IN_H
#include <netinet/in.h>
#endif
#if TIME_WITH_SYS_TIME
# ifdef WIN32
#  include <sys/timeb.h>
# else
#  include <sys/time.h>
# endif
# include <time.h>
#else
# if HAVE_SYS_TIME_H
#  include <sys/time.h>
# else
#  include <time.h>
# endif
#endif
#if HAVE_STRING_H
#include <string.h>
#else
#include <strings.h>
#endif
#if HAVE_STDLIB_H
#include <stdlib.h>
#endif
#if HAVE_SYS_SELECT_H
#include <sys/select.h>
#endif
 
#if HAVE_WINSOCK_H
#include <winsock.h>
#endif
 
#if HAVE_DMALLOC_H
#include <dmalloc.h>
#endif
 
#include "asn1.h"
#include "snmp_api.h"
#include "mib.h"
#include "snmp.h"
#include "snmp_impl.h"
#include "parse.h"
#include "int64.h"
#include "system.h"
#include "read_config.h"
#include "snmp_debug.h"
#include "default_store.h"
 
static void sprint_by_type (char *, struct variable_list *, struct enum_list *, const char *, const char *);
static int parse_subtree (struct tree *, const char *, oid *, size_t *);
static struct tree * _sprint_objid(char *buf, oid *objid, size_t objidlen);
static char *uptimeString (u_long, char *);
static void sprint_octet_string (char *, struct variable_list *, struct enum_list *, const char *, const char *);
static void sprint_opaque (char *, struct variable_list *, struct enum_list *, const char *, const char *);
static void sprint_object_identifier (char *, struct variable_list *, struct enum_list *, const char *, const char *);
static void sprint_timeticks (char *, struct variable_list *, struct enum_list *, const char *, const char *);
static void sprint_hinted_integer (char *, long, const char *, const char *);
static void sprint_integer (char *, struct variable_list *, struct enum_list *, const char *, const char *);
static void sprint_uinteger (char *, struct variable_list *, struct enum_list *, const char *, const char *);
static void sprint_gauge (char *, struct variable_list *, struct enum_list *, const char *, const char *);
static void sprint_counter (char *, struct variable_list *, struct enum_list *, const char *, const char *);
static void sprint_networkaddress (char *, struct variable_list *, struct enum_list *, const char *, const char *);
static void sprint_ipaddress (char *, struct variable_list *, struct enum_list *, const char *, const char *);
static void sprint_null (char *, struct variable_list *, struct enum_list *, const char *, const char *);
static void sprint_bitstring (char *, struct variable_list *, struct enum_list *, const char *, const char *);
static void sprint_nsapaddress (char *, struct variable_list *, struct enum_list *, const char *, const char *);
static void sprint_counter64 (char *, struct variable_list *, struct enum_list *, const char *, const char *);
static void sprint_unknowntype (char *, struct variable_list *, struct enum_list *, const char *, const char *);
static void sprint_badtype (char *, struct variable_list *, struct enum_list *, const char *, const char *);
struct tree *_get_symbol(oid *objid, size_t objidlen, struct tree *subtree, char *buf, struct index_list *in_dices, char **end_of_known);
 
#ifdef OPAQUE_SPECIAL_TYPES
static void sprint_float (char *, struct variable_list *, struct enum_list *, const char *, const char *);
static void sprint_double (char *, struct variable_list *, struct enum_list *, const char *, const char *);
#endif
void print_tree_node (FILE *f, struct tree *tp);
 
/* helper functions for get_module_node */
int node_to_oid(struct tree *, oid *, size_t *);
static int _add_strings_to_oid(struct tree *, char *,
             oid *, size_t *, size_t);
 
extern struct tree *tree_head;
 
struct tree *Mib;             /* Backwards compatibility */
 
oid RFC1213_MIB[] = { 1, 3, 6, 1, 2, 1 };
static char Standard_Prefix[] = ".1.3.6.1.2.1";
 
/* Set default here as some uses of read_objid require valid pointer. */
static char *Prefix = &Standard_Prefix[0];
typedef struct _PrefixList {
        const char *str;
        int len;
} *PrefixListPtr, PrefixList;
 
/*
 * Here are the prefix strings.
 * Note that the first one finds the value of Prefix or Standard_Prefix.
 * Any of these MAY start with period; all will NOT end with period.
 * Period is added where needed.  See use of Prefix in this module.
 */
PrefixList mib_prefixes[] = {
        { &Standard_Prefix[0] }, /* placeholder for Prefix data */
        { ".iso.org.dod.internet.mgmt.mib-2" },
        { ".iso.org.dod.internet.experimental" },
        { ".iso.org.dod.internet.private" },
        { ".iso.org.dod.internet.snmpParties" },
        { ".iso.org.dod.internet.snmpSecrets" },
        { NULL, 0 }  /* end of list */
};
 
static char *
uptimeString(u_long timeticks,
             char *buf)
{
    int centisecs, seconds, minutes, hours, days;
 
    centisecs = timeticks % 100;
    timeticks /= 100;
    days = timeticks / (60 * 60 * 24);
    timeticks %= (60 * 60 * 24);
 
    hours = timeticks / (60 * 60);
    timeticks %= (60 * 60);
 
    minutes = timeticks / 60;
    seconds = timeticks % 60;
 
    if (ds_get_boolean(DS_LIBRARY_ID, DS_LIB_QUICK_PRINT))
        sprintf(buf, "%d:%d:%02d:%02d.%02d",
                days, hours, minutes, seconds, centisecs);
    else {
        if (days == 0){
            sprintf(buf, "%d:%02d:%02d.%02d",
                hours, minutes, seconds, centisecs);
        } else if (days == 1) {
            sprintf(buf, "%d day, %d:%02d:%02d.%02d",
                days, hours, minutes, seconds, centisecs);
        } else {
            sprintf(buf, "%d days, %d:%02d:%02d.%02d",
                days, hours, minutes, seconds, centisecs);
        }
    }
    return buf;
}
 
 
 
void sprint_hexstring(char *buf,
                      const u_char *cp,
                      size_t len)
{
 
    for(; len >= 16; len -= 16){
        sprintf(buf, "%02X %02X %02X %02X %02X %02X %02X %02X ", cp[0], cp[1], cp[2], cp[3], cp[4], cp[5], cp[6], cp[7]);
        buf += strlen(buf);
        cp += 8;
        sprintf(buf, "%02X %02X %02X %02X %02X %02X %02X %02X", cp[0], cp[1], cp[2], cp[3], cp[4], cp[5], cp[6], cp[7]);
        buf += strlen(buf);
        if (len > 16) { *buf++ = '\n'; *buf = 0; }
        cp += 8;
    }
    for(; len > 0; len--){
        sprintf(buf, "%02X ", *cp++);
        buf += strlen(buf);
    }
    *buf = '\0';
}
 
void sprint_asciistring(char *buf,
                               u_char  *cp,
                               size_t       len)
{
    int x;
 
    for(x = 0; x < (int)len; x++){
        if (isprint(*cp)){
            *buf++ = *cp++;
        } else {
            *buf++ = '.';
            cp++;
        }
#if 0
        if ((x % 48) == 47)
            *buf++ = '\n';
#endif
    }
    *buf = '\0';
}
 
 
/*
 
  < 4
  hex
 
 
  < 4 hex Hex: oo oo oo
  < 4     "fgh" Hex: oo oo oo
  > 4 hex Hex: oo oo oo oo oo oo oo oo
  > 4     "this is a test"
 
  */
 
static void
sprint_octet_string(char *buf,
                    struct variable_list *var,
                    struct enum_list *enums,
                    const char *hint,
                    const char *units)
{
    int hex, x;
    u_char *cp;
    const char *saved_hint = hint;
    char *saved_buf = buf;
 
    if (var->type != ASN_OCTET_STR){
        sprintf(buf, "Wrong Type (should be OCTET STRING): ");
        buf += strlen(buf);
        sprint_by_type(buf, var, NULL, NULL, NULL);
        return;
    }
 
    if (hint) {
        int repeat, width = 1;
        long value;
        char code = 'd', separ = 0, term = 0, ch;
        u_char *ecp;
 
        *buf = 0;
        cp = var->val.string;
        ecp = cp + var->val_len;
        while (cp < ecp) {
            repeat = 1;
            if (*hint) {
                if (*hint == '*') {
                    repeat = *cp++;
                    hint++;
                }
                width = 0;
                while ('0' <= *hint && *hint <= '9')
                    width = width * 10 + *hint++ - '0';
                code = *hint++;
                if ((ch = *hint) && ch != '*' && (ch < '0' || ch > '9')
                    && (width != 0 || (ch != 'x' && ch != 'd' && ch != 'o')))
                    separ = *hint++;
                else separ = 0;
                if ((ch = *hint) && ch != '*' && (ch < '0' || ch > '9')
                    && (width != 0 || (ch != 'x' && ch != 'd' && ch != 'o')))
                    term = *hint++;
                else term = 0;
                if (width == 0) width = 1;
            }
            while (repeat && cp < ecp) {
                value = 0;
                if (code != 'a')
                    for (x = 0; x < width; x++) value = value * 256 + *cp++;
                switch (code) {
                case 'x':
                    sprintf (buf, "%lx", value); break;
                case 'd':
                    sprintf (buf, "%ld", value); break;
                case 'o':
                    sprintf (buf, "%lo", value); break;
                case 'a':
                    for (x = 0; x < width && cp < ecp; x++)
                        *buf++ = *cp++;
                    *buf = 0;
                    break;
                default:
                    sprintf(saved_buf, "(Bad hint ignored: %s) ", saved_hint);
                    sprint_octet_string(saved_buf+strlen(saved_buf),
                                        var, enums, NULL, NULL);
                    return;
                }
                buf += strlen (buf);
                if (cp < ecp && separ) *buf++ = separ;
                repeat--;
            }
            if (term && cp < ecp) *buf++ = term;
        }
        if (units) sprintf (buf, " %s", units);
        return;
    }
 
    hex = 0;
    for(cp = var->val.string, x = 0; x < (int)var->val_len; x++, cp++){
        if (!(isprint(*cp) || isspace(*cp)))
            hex = 1;
    }
    if (var->val_len == 0){
        strcpy(buf, "\"\"");
        return;
    }
    if (!hex){
        *buf++ = '"';
        sprint_asciistring(buf, var->val.string, var->val_len);
        buf += strlen(buf);
        *buf++ = '"';
        *buf = '\0';
    }
    if (hex || ((var->val_len <= 4) && !ds_get_boolean(DS_LIBRARY_ID, DS_LIB_QUICK_PRINT))){
        if (ds_get_boolean(DS_LIBRARY_ID, DS_LIB_QUICK_PRINT)){
            *buf++ = '"';
            *buf = '\0';
        } else {
            sprintf(buf, " Hex: ");
            buf += strlen(buf);
        }
        sprint_hexstring(buf, var->val.string, var->val_len);
        if (ds_get_boolean(DS_LIBRARY_ID, DS_LIB_QUICK_PRINT)){
            buf += strlen(buf);
            *buf++ = '"';
            *buf = '\0';
        }
    }
    if (units) sprintf (buf, " %s", units);
}
 
#ifdef OPAQUE_SPECIAL_TYPES
 
static void
sprint_float(char *buf,
             struct variable_list *var,
             struct enum_list *enums,
             const char *hint,
             const char *units)
{
  if (var->type != ASN_OPAQUE_FLOAT) {
        sprintf(buf, "Wrong Type (should be Float): ");
        buf += strlen(buf);
        sprint_by_type(buf, var, NULL, NULL, NULL);
        return;
    }
    if (!ds_get_boolean(DS_LIBRARY_ID, DS_LIB_QUICK_PRINT)){
        sprintf(buf, "Opaque: Float:");
        buf += strlen(buf);
    }
    sprintf(buf, " %f", *var->val.floatVal);
    buf += strlen (buf);
    if (units) sprintf (buf, " %s", units);
}
 
static void
sprint_double(char *buf,
              struct variable_list *var,
              struct enum_list *enums,
              const char *hint,
              const char *units)
{
  if (var->type != ASN_OPAQUE_DOUBLE) {
        sprintf(buf, "Wrong Type (should be Double): ");
        buf += strlen(buf);
        sprint_by_type(buf, var, NULL, NULL, NULL);
        return;
    }
    if (!ds_get_boolean(DS_LIBRARY_ID, DS_LIB_QUICK_PRINT)){
        sprintf(buf, "Opaque: Double:");
        buf += strlen(buf);
    }
    sprintf(buf, " %f", *var->val.doubleVal);
    buf += strlen (buf);
    if (units) sprintf (buf, " %s", units);
}
 
#endif /* OPAQUE_SPECIAL_TYPES */
 
static void
sprint_opaque(char *buf,
              struct variable_list *var,
              struct enum_list *enums,
              const char *hint,
              const char *units)
{
 
    if (var->type != ASN_OPAQUE
#ifdef OPAQUE_SPECIAL_TYPES
        && var->type != ASN_OPAQUE_COUNTER64
        && var->type != ASN_OPAQUE_U64
        && var->type != ASN_OPAQUE_I64
        && var->type != ASN_OPAQUE_FLOAT
        && var->type != ASN_OPAQUE_DOUBLE
#endif /* OPAQUE_SPECIAL_TYPES */
      ){
        sprintf(buf, "Wrong Type (should be Opaque): ");
        buf += strlen(buf);
        sprint_by_type(buf, var, NULL, NULL, NULL);
        return;
    }
#ifdef OPAQUE_SPECIAL_TYPES
    switch(var->type) {
      case ASN_OPAQUE_COUNTER64:
      case ASN_OPAQUE_U64:
      case ASN_OPAQUE_I64:
        sprint_counter64(buf, var, enums, hint, units);
        break;
 
      case ASN_OPAQUE_FLOAT:
        sprint_float(buf, var, enums, hint, units);
        break;
 
      case ASN_OPAQUE_DOUBLE:
        sprint_double(buf, var, enums, hint, units);
        break;
 
      case ASN_OPAQUE:
#endif
    if (!ds_get_boolean(DS_LIBRARY_ID, DS_LIB_QUICK_PRINT)){
        sprintf(buf, "OPAQUE: ");
        buf += strlen(buf);
    }
    sprint_hexstring(buf, var->val.string, var->val_len);
    buf += strlen (buf);
#ifdef OPAQUE_SPECIAL_TYPES
    }
#endif
    if (units) sprintf (buf, " %s", units);
}
 
static void
sprint_object_identifier(char *buf,
                         struct variable_list *var,
                         struct enum_list *enums,
                         const char *hint,
                         const char *units)
{
    if (var->type != ASN_OBJECT_ID){
        sprintf(buf, "Wrong Type (should be OBJECT IDENTIFIER): ");
        buf += strlen(buf);
        sprint_by_type(buf, var, NULL, NULL, NULL);
        return;
    }
    if (!ds_get_boolean(DS_LIBRARY_ID, DS_LIB_QUICK_PRINT)){
        sprintf(buf, "OID: ");
        buf += strlen(buf);
    }
    _sprint_objid(buf, (oid *)(var->val.objid), var->val_len / sizeof(oid));
    buf += strlen (buf);
    if (units) sprintf (buf, " %s", units);
}
 
static void
sprint_timeticks(char *buf,
                 struct variable_list *var,
                 struct enum_list *enums,
                 const char *hint,
                 const char *units)
{
    char timebuf[32];
 
    if (var->type != ASN_TIMETICKS){
        sprintf(buf, "Wrong Type (should be Timeticks): ");
        buf += strlen(buf);
        sprint_by_type(buf, var, NULL, NULL, NULL);
        return;
    }
    if (!ds_get_boolean(DS_LIBRARY_ID, DS_LIB_QUICK_PRINT)){
        sprintf(buf, "Timeticks: (%lu) ", *(u_long *)(var->val.integer));
        buf += strlen(buf);
    }
    sprintf(buf, "%s", uptimeString(*(u_long *)(var->val.integer), timebuf));
    buf += strlen (buf);
    if (units) sprintf (buf, " %s", units);
}
 
static void
sprint_hinted_integer (char *buf,
                       long val,
                       const char *hint,
                       const char *units)
{
    char code;
    int shift, len;
    char tmp[256];
    char fmt[10];
 
    code = hint[0];
    if (hint [1] == '-') {
        shift = atoi (hint+2);
    }
    else shift = 0;
    fmt[0] = '%';
    fmt[1] = 'l';
    fmt[2] = code;
    fmt[3] = 0;
    sprintf (tmp, fmt, val);
    if (shift != 0) {
        len = strlen (tmp);
        if (shift <= len) {
            tmp[len+1] = 0;
            while (shift--) {
                tmp[len] = tmp[len-1];
                len--;
            }
            tmp[len] = '.';
        }
        else {
            tmp[shift+1] = 0;
            while (shift) {
                if (len-- > 0) tmp [shift] = tmp [len];
                else tmp[shift] = '0';
                shift--;
            }
            tmp[0] = '.';
        }
    }
    strcpy (buf, tmp);
}
 
static void
sprint_integer(char *buf,
               struct variable_list *var,
               struct enum_list *enums,
               const char *hint,
               const char *units)
{
    char    *enum_string = NULL;
 
    if (var->type != ASN_INTEGER){
        sprintf(buf, "Wrong Type (should be INTEGER): ");
        buf += strlen(buf);
        sprint_by_type(buf, var, NULL, NULL, NULL);
        return;
    }
    for (; enums; enums = enums->next)
        if (enums->value == *var->val.integer){
            enum_string = enums->label;
            break;
        }
    if (enum_string == NULL ||
        ds_get_boolean(DS_LIBRARY_ID,DS_LIB_PRINT_NUMERIC_ENUM)) {
        if (hint) sprint_hinted_integer(buf, *var->val.integer, hint, units);
        else sprintf(buf, "%ld", *var->val.integer);
    }
    else if (ds_get_boolean(DS_LIBRARY_ID, DS_LIB_QUICK_PRINT))
        sprintf(buf, "%s", enum_string);
    else
        sprintf(buf, "%s(%ld)", enum_string, *var->val.integer);
    buf += strlen (buf);
    if (units) sprintf (buf, " %s", units);
}
 
static void
sprint_uinteger(char *buf,
                struct variable_list *var,
                struct enum_list *enums,
                const char *hint,
                const char *units)
{
    char    *enum_string = NULL;
 
    if (var->type != ASN_UINTEGER){
        sprintf(buf, "Wrong Type (should be UInteger32): ");
        buf += strlen(buf);
        sprint_by_type(buf, var, NULL, NULL, NULL);
        return;
    }
    for (; enums; enums = enums->next)
        if (enums->value == *var->val.integer){
            enum_string = enums->label;
            break;
        }
    if (enum_string == NULL ||
        ds_get_boolean(DS_LIBRARY_ID,DS_LIB_PRINT_NUMERIC_ENUM))
        sprintf(buf, "%lu", *var->val.integer);
    else if (ds_get_boolean(DS_LIBRARY_ID, DS_LIB_QUICK_PRINT))
        sprintf(buf, "%s", enum_string);
    else
        sprintf(buf, "%s(%lu)", enum_string, *var->val.integer);
    buf += strlen (buf);
    if (units) sprintf (buf, " %s", units);
}
 
static void
sprint_gauge(char *buf,
             struct variable_list *var,
             struct enum_list *enums,
             const char *hint,
             const char *units)
{
    if (var->type != ASN_GAUGE){
        sprintf(buf, "Wrong Type (should be Gauge): ");
        buf += strlen(buf);
        sprint_by_type(buf, var, NULL, NULL, NULL);
        return;
    }
    if (ds_get_boolean(DS_LIBRARY_ID, DS_LIB_QUICK_PRINT))
        sprintf(buf, "%lu", *var->val.integer);
    else
        sprintf(buf, "Gauge: %lu", *var->val.integer);
    buf += strlen (buf);
    if (units) sprintf (buf, " %s", units);
}
 
static void
sprint_counter(char *buf,
               struct variable_list *var,
               struct enum_list *enums,
               const char *hint,
               const char *units)
{
    if (var->type != ASN_COUNTER){
        sprintf(buf, "Wrong Type (should be Counter): ");
        buf += strlen(buf);
        sprint_by_type(buf, var, NULL, NULL, NULL);
        return;
    }
    sprintf(buf, "%lu", *var->val.integer);
    buf += strlen (buf);
    if (units) sprintf (buf, " %s", units);
}
 
static void
sprint_networkaddress(char *buf,
                      struct variable_list *var,
                      struct enum_list *enums,
                      const char *hint,
                      const char *units)
{
    int x, len;
    u_char *cp;
 
    if (!ds_get_boolean(DS_LIBRARY_ID, DS_LIB_QUICK_PRINT)){
        sprintf(buf, "Network Address: ");
        buf += strlen(buf);
    }
    cp = var->val.string;
    len = var->val_len;
    for(x = 0; x < len; x++){
        sprintf(buf, "%02X", *cp++);
        buf += strlen(buf);
        if (x < (len - 1))
            *buf++ = ':';
    }
}
 
static void
sprint_ipaddress(char *buf,
                 struct variable_list *var,
                 struct enum_list *enums,
                 const char *hint,
                 const char *units)
{
    u_char *ip;
 
    if (var->type != ASN_IPADDRESS){
        sprintf(buf, "Wrong Type (should be Ipaddress): ");
        buf += strlen(buf);
        sprint_by_type(buf, var, NULL, NULL, NULL);
        return;
    }
    ip = var->val.string;
    if (ds_get_boolean(DS_LIBRARY_ID, DS_LIB_QUICK_PRINT))
        sprintf(buf, "%d.%d.%d.%d",ip[0], ip[1], ip[2], ip[3]);
    else
        sprintf(buf, "IpAddress: %d.%d.%d.%d",ip[0], ip[1], ip[2], ip[3]);
}
 
static void
sprint_null(char *buf,
            struct variable_list *var,
            struct enum_list *enums,
            const char *hint,
            const char *units)
{
    if (var->type != ASN_NULL){
        sprintf(buf, "Wrong Type (should be NULL): ");
        buf += strlen(buf);
        sprint_by_type(buf, var, NULL, NULL, NULL);
        return;
    }
    sprintf(buf, "NULL");
}
 
static void
sprint_bitstring(char *buf,
                 struct variable_list *var,
                 struct enum_list *enums,
                 const char *hint,
                 const char *units)
{
    int len, bit;
    u_char *cp;
    char *enum_string;
 
    if (var->type != ASN_BIT_STR && var->type != ASN_OCTET_STR){
        sprintf(buf, "Wrong Type (should be BIT STRING): ");
        buf += strlen(buf);
        sprint_by_type(buf, var, NULL, NULL, NULL);
        return;
    }
    if (ds_get_boolean(DS_LIBRARY_ID, DS_LIB_QUICK_PRINT)){
        *buf++ = '"';
        *buf = '\0';
    } else {
        sprintf(buf, "BITS: ");
        buf += strlen(buf);
    }
    sprint_hexstring(buf, var->val.bitstring, var->val_len);
    buf += strlen(buf);
 
    if (ds_get_boolean(DS_LIBRARY_ID, DS_LIB_QUICK_PRINT)){
        buf += strlen(buf);
        *buf++ = '"';
        *buf = '\0';
    } else {
        cp = var->val.bitstring;
        for(len = 0; len < (int)var->val_len; len++){
            for(bit = 0; bit < 8; bit++){
                if (*cp & (0x80 >> bit)){
                    enum_string = NULL;
                    for (; enums; enums = enums->next)
                        if (enums->value == (len * 8) + bit){
                            enum_string = enums->label;
                            break;
                        }
                    if (enum_string == NULL ||
                        ds_get_boolean(DS_LIBRARY_ID,DS_LIB_PRINT_NUMERIC_ENUM))
                        sprintf(buf, "%d ", (len * 8) + bit);
                    else
                        sprintf(buf, "%s(%d) ", enum_string, (len * 8) + bit);
                    buf += strlen(buf);
                }
            }
            cp ++;
        }
    }
}
 
static void
sprint_nsapaddress(char *buf,
                   struct variable_list *var,
                   struct enum_list *enums,
                   const char *hint,
                   const char *units)
{
    if (var->type != ASN_NSAP){
        sprintf(buf, "Wrong Type (should be NsapAddress): ");
        buf += strlen(buf);
        sprint_by_type(buf, var, NULL, NULL, NULL);
        return;
    }
    if (!ds_get_boolean(DS_LIBRARY_ID, DS_LIB_QUICK_PRINT)){
        sprintf(buf, "NsapAddress: ");
        buf += strlen(buf);
    }
    sprint_hexstring(buf, var->val.string, var->val_len);
}
 
static void
sprint_counter64(char *buf,
                 struct variable_list *var,
                 struct enum_list *enums,
                 const char *hint,
                 const char *units)
{
    char a64buf[I64CHARSZ+1];
 
  if (var->type != ASN_COUNTER64
#ifdef OPAQUE_SPECIAL_TYPES
      && var->type != ASN_OPAQUE_COUNTER64
      && var->type != ASN_OPAQUE_I64
      && var->type != ASN_OPAQUE_U64
#endif
    ){
        sprintf(buf, "Wrong Type (should be Counter64): ");
        buf += strlen(buf);
        sprint_by_type(buf, var, NULL, NULL, NULL);
        return;
    }
    if (!ds_get_boolean(DS_LIBRARY_ID, DS_LIB_QUICK_PRINT)){
#ifdef OPAQUE_SPECIAL_TYPES
      if (var->type != ASN_COUNTER64) {
        sprintf(buf, "Opaque: ");
        buf += strlen(buf);
      }
#endif
#ifdef OPAQUE_SPECIAL_TYPES
        switch(var->type) {
          case ASN_OPAQUE_U64:
            sprintf(buf, "UInt64: ");
            break;
          case ASN_OPAQUE_I64:
            sprintf(buf, "Int64: ");
            break;
          case ASN_COUNTER64:
          case ASN_OPAQUE_COUNTER64:
#endif
            sprintf(buf, "Counter64: ");
#ifdef OPAQUE_SPECIAL_TYPES
        }
#endif
        buf += strlen(buf);
    }
#ifdef OPAQUE_SPECIAL_TYPES
    if (var->type == ASN_OPAQUE_I64)
    {
      printI64(a64buf, var->val.counter64);
      sprintf(buf, a64buf);
    }
    else
#endif
    {
      printU64(a64buf, var->val.counter64);
      sprintf(buf, a64buf);
    }
    buf += strlen (buf);
    if (units) sprintf (buf, " %s", units);
}
 
static void
sprint_unknowntype(char *buf,
                   struct variable_list *var,
                   struct enum_list *enums,
                   const char *hint,
                   const char *units)
{
/*    sprintf(buf, "Variable has bad type"); */
    sprint_by_type(buf, var, NULL, NULL, NULL);
}
 
static void
sprint_badtype(char *buf,
               struct variable_list *var,
               struct enum_list *enums,
               const char *hint,
               const char *units)
{
    sprintf(buf, "Variable has bad type");
}
 
static void
sprint_by_type(char *buf,
               struct variable_list *var,
               struct enum_list *enums,
               const char *hint,
               const char *units)
{
    switch (var->type){
        case ASN_INTEGER:
            sprint_integer(buf, var, enums, hint, units);
            break;
        case ASN_OCTET_STR:
            sprint_octet_string(buf, var, enums, hint, units);
            break;
        case ASN_BIT_STR:
            sprint_bitstring(buf, var, enums, hint, units);
            break;
        case ASN_OPAQUE:
            sprint_opaque(buf, var, enums, hint, units);
            break;
        case ASN_OBJECT_ID:
            sprint_object_identifier(buf, var, enums, hint, units);
            break;
        case ASN_TIMETICKS:
            sprint_timeticks(buf, var, enums, hint, units);
            break;
        case ASN_GAUGE:
            sprint_gauge(buf, var, enums, hint, units);
            break;
        case ASN_COUNTER:
            sprint_counter(buf, var, enums, hint, units);
            break;
        case ASN_IPADDRESS:
            sprint_ipaddress(buf, var, enums, hint, units);
            break;
        case ASN_NULL:
            sprint_null(buf, var, enums, hint, units);
            break;
        case ASN_UINTEGER:
            sprint_uinteger(buf, var, enums, hint, units);
            break;
        case ASN_COUNTER64:
#ifdef OPAQUE_SPECIAL_TYPES
        case ASN_OPAQUE_U64:
        case ASN_OPAQUE_I64:
        case ASN_OPAQUE_COUNTER64:
#endif /* OPAQUE_SPECIAL_TYPES */
            sprint_counter64(buf, var, enums, hint, units);
            break;
#ifdef OPAQUE_SPECIAL_TYPES
        case ASN_OPAQUE_FLOAT:
            sprint_float(buf, var, enums, hint, units);
            break;
        case ASN_OPAQUE_DOUBLE:
            sprint_double(buf, var, enums, hint, units);
            break;
#endif /* OPAQUE_SPECIAL_TYPES */
        default:
            DEBUGMSGTL(("sprint_by_type", "bad type: %d\n", var->type));
            sprint_badtype(buf, var, enums, hint, units);
            break;
    }
}
 
 
struct tree *get_tree_head(void)
{
   return(tree_head);
}
 
static char *confmibdir=NULL;
static char *confmibs=NULL;
 
void
handle_mibdirs_conf(const char *token,
                    char *line)
{
    char *ctmp;
 
    if (confmibdir) {
        ctmp = (char *)malloc(strlen(confmibdir) + strlen(line) + 1);
        if (*line == '+')
            line++;
        sprintf(ctmp,"%s%c%s",confmibdir, ENV_SEPARATOR_CHAR, line);
        free(confmibdir);
        confmibdir = ctmp;
    } else {
        confmibdir=strdup(line);
    }
    DEBUGMSGTL(("read_config:initmib", "using mibdirs: %s\n", confmibdir));
}
 
void
handle_mibs_conf(const char *token,
                 char *line)
{
    char *ctmp;
 
    if (confmibs) {
        ctmp = (char *)malloc(strlen(confmibs) + strlen(line) + 1);
        if (*line == '+')
            line++;
        sprintf(ctmp,"%s%c%s",confmibs, ENV_SEPARATOR_CHAR, line);
        free(confmibs);
        confmibs = ctmp;
    } else {
        confmibs=strdup(line);
    }
    DEBUGMSGTL(("read_config:initmib", "using mibs: %s\n", confmibs));
}
 
void
handle_mibfile_conf(const char *token,
                    char *line)
{
  DEBUGMSGTL(("read_config:initmib", "reading mibfile: %s\n", line));
  read_mib(line);
}
 
char *
snmp_out_toggle_options(char *options)
{
    while(*options) {
        switch(*options++) {
        case 'n':
            ds_toggle_boolean(DS_LIBRARY_ID, DS_LIB_PRINT_NUMERIC_OIDS);
            break;
        case 'e':
            ds_toggle_boolean(DS_LIBRARY_ID, DS_LIB_PRINT_NUMERIC_ENUM);
            break;
        case 'b':
            ds_toggle_boolean(DS_LIBRARY_ID, DS_LIB_DONT_BREAKDOWN_OIDS);
            break;
        case 'q':
            ds_toggle_boolean(DS_LIBRARY_ID, DS_LIB_QUICK_PRINT);
            break;
        case 'f':
            ds_toggle_boolean(DS_LIBRARY_ID, DS_LIB_PRINT_FULL_OID);
            break;
        case 's':
            snmp_set_suffix_only(1);
            break;
        case 'S':
            snmp_set_suffix_only(2);
            break;
        default:
            return options-1;
        }
    }
    return NULL;
}
 
void snmp_out_toggle_options_usage(const char *lead, FILE *outf)
{
  fprintf(outf, "%sOUTOPTS values:\n", lead);
  fprintf(outf, "%s    n: Print oids numerically.\n", lead);
  fprintf(outf, "%s    e: Print enums numerically.\n", lead);
  fprintf(outf, "%s    b: Dont break oid indexes down.\n", lead);
  fprintf(outf, "%s    q: Quick print for easier parsing.\n", lead);
  fprintf(outf, "%s    f: Print full oids on output.\n", lead);
  fprintf(outf, "%s    s: Print only last symbolic element of oid.\n", lead);
  fprintf(outf, "%s    S: Print MIB module-id plus last element.\n", lead);
}
 
char *
snmp_in_toggle_options(char *options)
{
    while(*options) {
        switch(*options++) {
        case 'R':
            ds_toggle_boolean(DS_LIBRARY_ID, DS_LIB_RANDOM_ACCESS);
            break;
        case 'b':
            ds_toggle_boolean(DS_LIBRARY_ID, DS_LIB_REGEX_ACCESS);
            break;
        default:
            return options-1;
        }
    }
    return NULL;
}
 
void snmp_in_toggle_options_usage(const char *lead, FILE *outf)
{
  fprintf(outf, "%sINOPTS values:\n", lead);
  fprintf(outf, "%s    R: Do random access to oid labels.\n", lead);
  fprintf(outf, "%s    b: Do best/regex matching to find a MIB node.\n", lead);
}
 
void
register_mib_handlers (void)
{
    register_premib_handler("snmp","mibdirs",
                            handle_mibdirs_conf, NULL,
                            "[mib-dirs|+mib-dirs]");
    register_premib_handler("snmp","mibs",
                            handle_mibs_conf,NULL,
                            "[mib-tokens|+mib-tokens]");
    register_config_handler("snmp","mibfile",
                            handle_mibfile_conf, NULL,
                            "mibfile-to-read");
 
    /* register the snmp.conf configuration handlers for default
       parsing behaviour */
 
    ds_register_premib(ASN_BOOLEAN, "snmp","showMibErrors",
                       DS_LIBRARY_ID, DS_LIB_MIB_ERRORS);
    ds_register_premib(ASN_BOOLEAN, "snmp","strictCommentTerm",
                       DS_LIBRARY_ID, DS_LIB_MIB_COMMENT_TERM);
    ds_register_premib(ASN_BOOLEAN, "snmp","mibAllowUnderline",
                       DS_LIBRARY_ID, DS_LIB_MIB_PARSE_LABEL);
    ds_register_premib(ASN_INTEGER, "snmp","mibWarningLevel",
                       DS_LIBRARY_ID, DS_LIB_MIB_WARNINGS);
    ds_register_premib(ASN_BOOLEAN, "snmp","mibReplaceWithLatest",
                       DS_LIBRARY_ID, DS_LIB_MIB_REPLACE);
 
    ds_register_config(ASN_BOOLEAN, "snmp","printNumericEnums",
                       DS_LIBRARY_ID, DS_LIB_PRINT_NUMERIC_ENUM);
    ds_register_config(ASN_BOOLEAN, "snmp","printNumericOids",
                       DS_LIBRARY_ID, DS_LIB_PRINT_NUMERIC_OIDS);
    ds_register_config(ASN_BOOLEAN, "snmp","dontBreakdownOids",
                       DS_LIBRARY_ID, DS_LIB_DONT_BREAKDOWN_OIDS);
    ds_register_config(ASN_BOOLEAN, "snmp","quickPrinting",
                       DS_LIBRARY_ID, DS_LIB_QUICK_PRINT);
    ds_register_config(ASN_INTEGER, "snmp","suffixPrinting",
                       DS_LIBRARY_ID, DS_LIB_PRINT_SUFFIX_ONLY);
 
    /* setup the default parser configurations, as specified by configure */
#ifdef MIB_COMMENT_IS_EOL_TERMINATED
    ds_set_boolean(DS_LIBRARY_ID, DS_LIB_MIB_COMMENT_TERM, 1);
#else  /* !MIB_COMMENT_IS_EOL_TERMINATED */
    ds_set_boolean(DS_LIBRARY_ID, DS_LIB_MIB_COMMENT_TERM, 0);
#endif /* !MIB_COMMENT_IS_EOL_TERMINATED */
}
 
void
init_mib (void)
{
    const char *prefix;
    char  *env_var, *entry;
    PrefixListPtr pp = &mib_prefixes[0];
    char *new_mibdirs, *homepath, *cp_home;
 
    if (Mib) return;
 
    /* Initialise the MIB directory/ies */
 
    /* we can't use the environment variable directly, because strtok
       will modify it. */
 
    env_var = getenv("MIBDIRS");
    if ( env_var == NULL ) {
        if (confmibdir != NULL)
            env_var = strdup(confmibdir);
        else
            env_var = strdup(DEFAULT_MIBDIRS);
    } else {
        env_var = strdup(env_var);
    }
    if (*env_var == '+') {
        entry = (char *)malloc(strlen(DEFAULT_MIBDIRS)+strlen(env_var)+2);
        sprintf(entry, "%s%c%s", DEFAULT_MIBDIRS, ENV_SEPARATOR_CHAR, env_var+1);
        free(env_var);
        env_var = entry;
    }
 
    /* replace $HOME in the path with the users home directory */
    homepath=getenv("HOME");
 
    if (homepath) {
      while((cp_home = strstr(env_var, "$HOME"))) {
        new_mibdirs = (char *) malloc(strlen(env_var) - strlen("$HOME") +
                                      strlen(homepath)+1);
        *cp_home = 0; /* null out the spot where we stop copying */
        sprintf(new_mibdirs, "%s%s%s", env_var, homepath,
                cp_home + strlen("$HOME"));
        /* swap in the new value and repeat */
        free(env_var);
        env_var = new_mibdirs;
      }
    }
 
    DEBUGMSGTL(("init_mib","Seen MIBDIRS: Looking in '%s' for mib dirs ...\n",env_var));
 
    entry = strtok( env_var, ENV_SEPARATOR );
    while ( entry ) {
        add_mibdir(entry);
        entry = strtok( NULL, ENV_SEPARATOR);
    }
    free(env_var);
 
    init_mib_internals();
 
    /* Read in any modules or mibs requested */
 
    env_var = getenv("MIBS");
    if ( env_var == NULL ) {
        if (confmibs != NULL)
        env_var = strdup(confmibs);
        else
            env_var = strdup(DEFAULT_MIBS);
    } else {
        env_var = strdup(env_var);
    }
    if (*env_var == '+') {
        entry = (char *)malloc(strlen(DEFAULT_MIBS)+strlen(env_var)+2);
        sprintf(entry, "%s%c%s", DEFAULT_MIBS, ENV_SEPARATOR_CHAR, env_var+1);
        free(env_var);
        env_var = entry;
    }
 
    DEBUGMSGTL(("init_mib","Seen MIBS: Looking in '%s' for mib files ...\n",env_var));
    entry = strtok( env_var, ENV_SEPARATOR );
    while ( entry ) {
        if (strcasecmp(entry, DEBUG_ALWAYS_TOKEN) == 0) {
            read_all_mibs();
        }
        else if (strstr (entry, "/") != 0) {
            read_mib(entry);
        }
        else {
            read_module(entry);
        }
            entry = strtok( NULL, ENV_SEPARATOR);
    }
    adopt_orphans();
    free(env_var);
 
    env_var = getenv("MIBFILES");
    if ( env_var != NULL ) {
        if (*env_var == '+') {
#ifdef DEFAULT_MIBFILES
            entry = (char *)malloc(strlen(DEFAULT_MIBFILES)+strlen(env_var)+2);
            sprintf(entry, "%s%c%s", DEFAULT_MIBFILES, ENV_SEPARATOR_CHAR,
                    env_var+1);
            free(env_var);
            env_var = entry;
#else
            env_var = strdup(env_var+1);
#endif
        } else {
            env_var = strdup(env_var);
        }
    } else {
#ifdef DEFAULT_MIBFILES
        env_var = strdup(DEFAULT_MIBFILES);
#endif
    }
 
    if ( env_var != 0 ) {
        DEBUGMSGTL(("init_mib","Seen MIBFILES: Looking in '%s' for mib files ...\n",env_var));
        entry = strtok( env_var, ENV_SEPARATOR );
        while ( entry ) {
            read_mib(entry);
            entry = strtok( NULL, ENV_SEPARATOR);
        }
        free(env_var);
    }
 
    prefix = getenv("PREFIX");
 
    if (!prefix)
        prefix = Standard_Prefix;
 
    Prefix = (char*)malloc(strlen(prefix)+2);
    strcpy(Prefix, prefix);
 
    DEBUGMSGTL(("init_mib","Seen PREFIX: Looking in '%s' for prefix ...\n", Prefix));
 
    /* remove trailing dot */
    env_var = &Prefix[strlen(Prefix) - 1];
    if (*env_var == '.') *env_var = '\0';
 
    pp->str = Prefix;   /* fixup first mib_prefix entry */
    /* now that the list of prefixes is built, save each string length. */
    while (pp->str) {
        pp->len = strlen(pp->str);
        pp++;
    }
 
    if (getenv("SUFFIX"))
        ds_set_boolean(DS_LIBRARY_ID, DS_LIB_PRINT_SUFFIX_ONLY, 1);
 
    Mib = tree_head;          /* Backwards compatibility */
}
 
void
print_mib (FILE *fp)
{
    print_subtree (fp, tree_head, 0);
}
 
void
print_ascii_dump (FILE *fp)
{
  fprintf(fp, "dump DEFINITIONS ::= BEGIN\n");
  print_ascii_dump_tree (fp, tree_head, 0);
  fprintf(fp, "END\n");
}
 
void
set_function(struct tree *subtree)
{
    switch(subtree->type){
        case TYPE_OBJID:
            subtree->printer = sprint_object_identifier;
            break;
            case TYPE_OCTETSTR:
                subtree->printer = sprint_octet_string;
                break;
            case TYPE_INTEGER:
                subtree->printer = sprint_integer;
                break;
            case TYPE_NETADDR:
                subtree->printer = sprint_networkaddress;
                break;
            case TYPE_IPADDR:
                subtree->printer = sprint_ipaddress;
                break;
            case TYPE_COUNTER:
                subtree->printer = sprint_counter;
                break;
            case TYPE_GAUGE:
                subtree->printer = sprint_gauge;
                break;
            case TYPE_TIMETICKS:
                subtree->printer = sprint_timeticks;
                break;
            case TYPE_OPAQUE:
                subtree->printer = sprint_opaque;
                break;
            case TYPE_NULL:
                subtree->printer = sprint_null;
                break;
            case TYPE_BITSTRING:
                subtree->printer = sprint_bitstring;
                break;
            case TYPE_NSAPADDRESS:
                subtree->printer = sprint_nsapaddress;
                break;
            case TYPE_COUNTER64:
                subtree->printer = sprint_counter64;
                break;
            case TYPE_UINTEGER:
                subtree->printer = sprint_uinteger;
                break;
            case TYPE_OTHER:
            default:
                subtree->printer = sprint_unknowntype;
                break;
        }
}
 
/*
 * Read an object identifier from input string into internal OID form.
 * Returns 1 if successful.
 * If an error occurs, this function returns 0 and MAY set snmp_errno.
 * snmp_errno is NOT set if SET_SNMP_ERROR evaluates to nothing.
 * This can make multi-threaded use a tiny bit more robust.
 */
int read_objid(const char *input,
               oid *output,
               size_t *out_len)   /* number of subid's in "output" */
{
    struct tree *root = tree_head;
    char buf[SPRINT_MAX_LEN];
    int ret;
 
    if (strchr(input, ':')) {
        return get_node(input, output, out_len);
    }
 
    if (*input == '.')
        input++;
    else {
    /* get past leading '.', append '.' to Prefix. */
        if (*Prefix == '.')
            strcpy(buf, Prefix+1);
        else
            strcpy(buf, Prefix);
        strcat(buf, ".");
        strcat(buf, input);
        input = buf;
    }
 
    if (root == NULL){
        SET_SNMP_ERROR(SNMPERR_NOMIB);
        *out_len = 0;
        return(0);
    }
    if ((ret = parse_subtree(root, input, output, out_len)) <= 0)
    {
        int errc = (ret ? ret : SNMPERR_UNKNOWN_OBJID);
        SET_SNMP_ERROR(errc);
        return (0);
    }
    *out_len = ret;
 
    return (1);
}
 
 
/*
 * RECURSIVE helper methods for read_objid
 * Returns:
 * < 0  the SNMPERR_ errorcode
 * = 0  input string is empty.
 * > 0  the number of sub-identifiers found in the input string.
 */
static int
parse_subtree(struct tree *subtree,
              const char *input,
              oid *output,
              size_t *out_len)   /* number of subid's */
{
    char buf[SPRINT_MAX_LEN], *to = buf, *cp;
    u_long subid = 0;
    struct tree *tp;
    int ret, len;
 
    /*
     * No empty strings.  Can happen if there is a trailing '.' or two '.'s
     * in a row, i.e. "..".
     */
    if ((*input == '\0') ||
        (*input == '.'))
        return (0);
 
    if (*input == '"' || *input == '\'') {
      /*
       * This is a string that should be converted into an OID
       *  Note:  assumes variable length index is required, and prepends
       *         the string length.
       */
      if ((cp = strchr(input+1, *input)) == NULL) {
        /* error.  Should be a matching quote somewhere. */
        return (0);
      }
 
      /* is there room enough for the string in question plus its length */
      len = cp-input-1;
      if ((int)*out_len <= len){
        return (SNMPERR_LONG_OID);
      }
 
      /* copy everything in */
      if (*input++ == '"') {
        /* add the length for " quoted objects */
        *output++ = len++;
      }
 
      *out_len -= len;
      while (input < cp) {
        *output++ = *input++;
      }
 
      /* Now, we assume that nothing beyond this exists in the parse
         tree, which should always be true (or else we have a really wacked
         mib designer somewhere. */
      input = cp + 1; /* past  the quote */
 
      if (*input != '.')
        return (len);
 
      ret = parse_subtree(NULL, ++input, output, out_len);
      if (ret <= 0)
        return (ret);
      return ret+len;
 
    } else if (isdigit(*input)) {
        /*
         * Read the number, then try to find it in the subtree.
         */
        while (isdigit(*input)) {
            *to++ = *input;
            subid *= 10;
            subid += *input++ - '0';
        }
        if (*input != '.' && *input != 0) {
            while (*input != 0 && *input != '.') *to++ = *input++;
            *to = 0;
            snmp_set_detail(buf);
            return SNMPERR_BAD_SUBID;
        }
        *to = '\0';
 
        for (tp = subtree; tp; tp = tp->next_peer) {
            if (tp->subid == subid)
                goto found;
        }
    }
    else {
        /*
         * Read the name into a buffer.
         */
        while ((*input != '\0') &&
               (*input != '.')) {
            *to++ = *input++;
        }
        *to = '\0';
 
        /*
         * Find the name in the subtree;
         */
        for (tp = subtree; tp; tp = tp->next_peer) {
            if (strcasecmp(tp->label, buf) == 0) {
                subid = tp->subid;
                goto found;
            }
        }
 
        /*
         * If we didn't find the entry, punt...
         */
        if (tp == NULL) {
            snmp_set_detail(buf);
            return (SNMPERR_BAD_SUBID);
        }
    }
 
found:
    if(subid > (u_long)MAX_SUBID){
        snmp_set_detail(buf);
        return (SNMPERR_MAX_SUBID);
    }
 
    if ((int)*out_len <= 0){
        return (SNMPERR_LONG_OID);
    }
 
    (*out_len)--;
    *output++ = subid;
 
    if (*input != '.')
        return (1);
 
    ret = parse_subtree(tp ? tp->child_list : NULL,
                             ++input, output, out_len);
    if (ret <= 0)
        return (ret);
    return ret+1;
}
 
static struct tree *
_sprint_objid(char *buf,
             oid *objid,
             size_t objidlen)   /* number of subidentifiers */
{
    char    tempbuf[SPRINT_MAX_LEN], *cp;
    struct tree    *subtree = tree_head;
    char *midpoint = 0;
 
    *tempbuf = '.';     /* this is a fully qualified name */
    subtree = _get_symbol(objid, objidlen, subtree, tempbuf + 1, 0, &midpoint);
    if (ds_get_boolean(DS_LIBRARY_ID,DS_LIB_PRINT_NUMERIC_OIDS)) {
        cp = tempbuf;
    } else if (ds_get_int(DS_LIBRARY_ID, DS_LIB_PRINT_SUFFIX_ONLY)){
        for(cp = tempbuf; *cp; cp++)
            ;
        if (midpoint)
            cp = midpoint-2; /* beyond the '.' */
        else {
            while(cp >= tempbuf){
                if (isalpha(*cp))
                    break;
                cp--;
            }
        }
        while(cp >= tempbuf){
            if (*cp == '.')
                break;
            cp--;
        }
        cp++;
        if (ds_get_int(DS_LIBRARY_ID, DS_LIB_PRINT_SUFFIX_ONLY) == 2 && cp > tempbuf) {
            char modbuf[256];
            char *mod = module_name(subtree->modid, modbuf);
            size_t len = strlen(mod);
            if ((int)len+1 >= cp-tempbuf) {
                memmove(tempbuf+len+2, cp, strlen(cp)+1);
                cp = tempbuf+len+2;
            }
            cp -= len+2;
            memcpy(cp, mod, len);
            cp[len] = ':';
            cp[len+1] = ':';
        }
    }
    else if (!ds_get_boolean(DS_LIBRARY_ID, DS_LIB_PRINT_FULL_OID)) {
        PrefixListPtr pp = &mib_prefixes[0];
        int ii;
        size_t ilen, tlen;
        const char *testcp;
        cp = tempbuf; tlen = strlen(tempbuf);
        ii = 0;
        while (pp->str) {
            ilen = pp->len; testcp = pp->str;
            if ((tlen > ilen) && !memcmp(tempbuf, testcp, ilen)) {
                cp += (ilen + 1);
                break;
            }
            pp++;
        }
    }
    else cp = tempbuf;
    strcpy(buf, cp);
    return subtree;
}
 
char * sprint_objid(char *buf, oid *objid, size_t objidlen)
{
    _sprint_objid(buf,objid,objidlen);
    return buf;
}
 
void
print_objid(oid *objid,
            size_t objidlen)    /* number of subidentifiers */
{
  fprint_objid(stdout, objid, objidlen);
}
 
void
fprint_objid(FILE *f,
             oid *objid,
             size_t objidlen)   /* number of subidentifiers */
{
    char    buf[SPRINT_MAX_LEN];
 
    _sprint_objid(buf, objid, objidlen);
    fprintf(f, "%s\n", buf);
}
 
void
sprint_variable(char *buf,
                oid *objid,
                size_t objidlen,
                struct variable_list *variable)
{
    struct tree    *subtree;
 
    subtree = _sprint_objid(buf, objid, objidlen);
    buf += strlen(buf);
    if (ds_get_boolean(DS_LIBRARY_ID, DS_LIB_QUICK_PRINT))
        strcat(buf, " ");
    else
        strcat(buf, " = ");
    buf += strlen(buf);
 
    if (variable->type == SNMP_NOSUCHOBJECT)
        strcpy(buf, "No Such Object available on this agent");
    else if (variable->type == SNMP_NOSUCHINSTANCE)
        strcpy(buf, "No Such Instance currently exists");
    else if (variable->type == SNMP_ENDOFMIBVIEW)
        strcpy(buf, "No more variables left in this MIB View");
    else if (subtree) {
        if (subtree->printer)
            (*subtree->printer)(buf, variable, subtree->enums, subtree->hint, subtree->units);
          else {
            sprint_by_type(buf, variable, subtree->enums, subtree->hint, subtree->units);
          }
    }
    else { /* handle rare case where tree is empty */
        sprint_by_type(buf, variable, 0, 0, 0);
    }
}
 
void
print_variable(oid *objid,
               size_t objidlen,
               struct variable_list *variable)
{
    fprint_variable(stdout, objid, objidlen, variable);
}
 
void
fprint_variable(FILE *f,
                oid *objid,
                size_t objidlen,
                struct variable_list *variable)
{
    char    buf[SPRINT_MAX_LEN];
 
    sprint_variable(buf, objid, objidlen, variable);
    fprintf(f, "%s\n", buf);
}
 
void
sprint_value(char *buf,
             oid *objid,
             size_t objidlen,
             struct variable_list *variable)
{
    char    tempbuf[SPRINT_MAX_LEN];
    struct tree    *subtree = tree_head;
 
    if (variable->type == SNMP_NOSUCHOBJECT)
        sprintf(buf, "No Such Object available on this agent");
    else if (variable->type == SNMP_NOSUCHINSTANCE)
        sprintf(buf, "No Such Instance currently exists");
    else if (variable->type == SNMP_ENDOFMIBVIEW)
        sprintf(buf, "No more variables left in this MIB View");
    else {
        subtree = get_symbol(objid, objidlen, subtree, tempbuf);
        if (subtree->printer)
            (*subtree->printer)(buf, variable, subtree->enums, subtree->hint, subtree->units);
        else {
            sprint_by_type(buf, variable, subtree->enums, subtree->hint, subtree->units);
        }
    }
}
 
void
print_value(oid *objid,
            size_t objidlen,
            struct variable_list *variable)
{
    fprint_value(stdout, objid, objidlen, variable);
}
 
void
fprint_value(FILE *f,
             oid *objid,
             size_t objidlen,
             struct variable_list *variable)
{
    char    tempbuf[SPRINT_MAX_LEN];
 
    sprint_value(tempbuf, objid, objidlen, variable);
    fprintf(f, "%s\n", tempbuf);
}
 
 
/*
 * Append a quoted printable string to buffer "buf"
 * that represents a range of sub-identifiers "objid".
 *
 * Display '.' for all non-printable sub-identifiers.
 * If successful, "buf" points past the appended string.
 */
char *
dump_oid_to_string(oid *objid,
                   size_t objidlen,
                   char *buf,
                   char quotechar)
{
  if (buf)
  { int ii, alen;
    char *scp;
    char *cp = buf + (strlen(buf));
    scp = cp;
    for (ii= 0, alen = 0; ii < (int)objidlen; ii++)
    {
        oid tst = objid[ii];
        if ((tst > 254) || (!isprint(tst)))
            tst = (oid)'.';
 
        if (alen == 0) *cp++ = quotechar;
        *cp++ = (char)tst;
        alen++;
    }
    if (alen) *cp++ = quotechar;
    *cp = '\0';
    buf = cp;
  }
 
  return buf;
}
 
struct tree *
_get_symbol(oid *objid,
           size_t objidlen,
           struct tree *subtree,
           char *buf,
           struct index_list *in_dices,
           char **end_of_known)
{
    struct tree    *return_tree = NULL;
 
    if (!objid || !buf)
        return NULL;
 
    for(; subtree; subtree = subtree->next_peer){
        if (*objid == subtree->subid){
            if (subtree->indexes)
                in_dices = subtree->indexes;
            if (!strncmp( subtree->label, ANON, ANON_LEN) ||
                ds_get_boolean(DS_LIBRARY_ID,DS_LIB_PRINT_NUMERIC_OIDS))
                sprintf(buf, "%lu", subtree->subid);
            else
                strcpy(buf, subtree->label);
            goto found;
        }
    }
 
    if (end_of_known)
        *end_of_known = buf;
 
    /* subtree not found */
 
    while (in_dices && (objidlen > 0) &&
           !ds_get_boolean(DS_LIBRARY_ID,DS_LIB_PRINT_NUMERIC_OIDS) &&
           !ds_get_boolean(DS_LIBRARY_ID,DS_LIB_DONT_BREAKDOWN_OIDS)) {
        size_t numids;
        struct tree *tp;
        tp = find_tree_node(in_dices->ilabel, -1);
        if (0 == tp) {
            /* ack.  Can't find an index in the mib tree.  bail */
            goto finish_it;
        }
        switch(tp->type) {
        case TYPE_OCTETSTR:
            if (in_dices->isimplied) {
                numids = objidlen;
                buf = dump_oid_to_string(objid, numids, buf, '\'');
            } else {
                numids = (size_t)*objid+1;
                if (numids > objidlen)
                    goto finish_it;
                if (numids == 1) {
                    *buf++ = '"'; *buf++ = '"';
                }
                else
                    buf = dump_oid_to_string(objid+1, numids-1, buf, '"');
            }
            objid += (numids);
            objidlen -= (numids);
            *buf++ = '.';
            *buf = '\0';
            break;
        case TYPE_INTEGER:
            sprintf(buf, "%lu.", *objid++);
            while(*buf)
                buf++;
            objidlen--;
            break;
        case TYPE_OBJID:
            if (in_dices->isimplied) {
                numids = objidlen;
            } else {
                numids = (size_t)*objid+1;
            }
            if ( numids > objidlen)
                goto finish_it;
            _get_symbol(objid, numids, NULL, buf, NULL, NULL);
            objid += (numids);
            objidlen -= (numids);
            buf += strlen(buf);
            *buf++ = '.';
            *buf = '\0';
            break;
        default:
            goto finish_it;
            break;
        }
        in_dices = in_dices->next;
    }
 
finish_it:
 
    while(objidlen-- > 0){       /* output rest of name, uninterpreted */
        sprintf(buf, "%lu.", *objid++);
        while(*buf)
            buf++;
    }
    *(buf - 1) = '\0'; /* remove trailing dot */
    return NULL;
 
found:
    if (objidlen > 1){
        while(*buf)
            buf++;
        *buf++ = '.';
        *buf = '\0';
 
        return_tree = _get_symbol(objid + 1, objidlen - 1, subtree->child_list,
                                 buf, in_dices, end_of_known);
    }
    if (return_tree != NULL)
        return return_tree;
    else
        return subtree;
}
 
struct tree *
get_symbol(oid *objid,
           size_t objidlen,
           struct tree *subtree,
           char *buf)
{
   return _get_symbol(objid,objidlen,subtree,buf,0,0);
}
 
/*
 * Clone of get_symbol that doesn't take a buffer argument
 */
struct tree *
get_tree(oid *objid,
         size_t objidlen,
         struct tree *subtree)
{
    struct tree    *return_tree = NULL;
 
    for(; subtree; subtree = subtree->next_peer){
        if (*objid == subtree->subid)
            goto found;
    }
 
    return NULL;
 
found:
    if (objidlen > 1)
        return_tree = get_tree(objid + 1, objidlen - 1, subtree->child_list);
    if (return_tree != NULL)
        return return_tree;
    else
        return subtree;
}
 
void
print_description(oid *objid,
                  size_t objidlen)   /* number of subidentifiers */
{
    fprint_description(stdout, objid, objidlen);
}
 
void
fprint_description(FILE *f,
                   oid *objid,
                   size_t objidlen)   /* number of subidentifiers */
{
    struct tree *tp = get_tree(objid, objidlen, tree_head);
    struct tree *subtree = tree_head;
    fprintf(f, "%s OBJECT-TYPE\n", tp->label);
    print_tree_node(f, tp);
    fprintf(f, "::= {");
    while (objidlen > 1) {
        for(; subtree; subtree = subtree->next_peer){
            if (*objid == subtree->subid){
                if (strncmp( subtree->label, ANON, ANON_LEN))
                    fprintf(f, " %s(%lu)", subtree->label, subtree->subid);
                else
                    fprintf(f, " %lu", subtree->subid);
                break;
            }
        }
        if (subtree == 0) break;
        objid++; objidlen--; subtree = subtree->child_list;
        if (subtree == 0) break;
    }
    fprintf(f, " %lu }\n", *objid);
}
 
void
print_tree_node(FILE *f,
                struct tree *tp)
{
    const char *cp;
    char str[MAXTOKEN];
    int i, prevmod;
    if (tp) {
        module_name(tp->modid, str);
        fprintf(f, "  -- FROM\t%s", str);
        for (i = 1, prevmod = tp->modid; i < tp->number_modules; i++) {
            if (prevmod != tp->module_list[i]) {
              module_name(tp->module_list[i], str);
              fprintf(f, ", %s", str);
            }
            prevmod = tp->module_list[i];
        }
        fprintf(f, "\n");
        if (tp->tc_index != -1) {
            fprintf(f, "  -- TEXTUAL CONVENTION %s\n", get_tc_descriptor(tp->tc_index));
        }
        switch (tp->type) {
        case TYPE_OBJID:        cp = "OBJECT IDENTIFIER"; break;
        case TYPE_OCTETSTR:     cp = "OCTET STRING"; break;
        case TYPE_INTEGER:      cp = "INTEGER"; break;
        case TYPE_NETADDR:      cp = "NetworkAddress"; break;
        case TYPE_IPADDR:       cp = "IpAddress"; break;
        case TYPE_COUNTER:      cp = "Counter"; break;
        case TYPE_GAUGE:        cp = "Gauge"; break;
        case TYPE_TIMETICKS:    cp = "TimeTicks"; break;
        case TYPE_OPAQUE:       cp = "Opaque"; break;
        case TYPE_NULL:         cp = "NULL"; break;
        case TYPE_COUNTER64:    cp = "Counter64"; break;
        case TYPE_BITSTRING:    cp = "BIT STRING"; break;
        case TYPE_NSAPADDRESS:  cp = "NsapAddress"; break;
        case TYPE_UINTEGER:     cp = "UInteger32"; break;
        case 0:                  cp = NULL; break;
        default:                sprintf(str,"type_%d", tp->type); cp = str;
        }
#if SNMP_TESTING_CODE
        if (!cp && (tp->ranges || tp->enums)) { /* ranges without type ? */
            sprintf(str,"?0 with %s %s ?",
            tp->ranges ? "Range" : "",
            tp->enums ? "Enum" : "");
            cp = str;
        }
#endif /* SNMP_TESTING_CODE */
        if (cp) fprintf(f, "  SYNTAX\t%s", cp);
        if (tp->ranges) {
            struct range_list *rp = tp->ranges;
            int first = 1;
            fprintf(f, " (");
            while (rp) {
                if (first) first = 0;
                else fprintf(f, " | ");
                if (rp->low == rp->high) fprintf(f, "%d", rp->low);
                else fprintf(f, "%d..%d", rp->low, rp->high);
                rp = rp->next;
            }
            fprintf(f, ") ");
        }
        if (tp->enums) {
            struct enum_list *ep = tp->enums;
            int first = 1;
            fprintf(f," { ");
            while (ep) {
                if (first) first = 0;
                else fprintf(f, ", ");
                fprintf(f, "%s(%d)", ep->label, ep->value);
                ep = ep->next;
            }
            fprintf(f," } ");
        }
        if (cp) fprintf(f, "\n");
        if (tp->hint) fprintf(f, "  DISPLAY-HINT\t\"%s\"\n", tp->hint);
        if (tp->units) fprintf(f, "  UNITS\t\"%s\"\n", tp->units);
        switch (tp->access) {
        case MIB_ACCESS_READONLY:       cp = "read-only"; break;
        case MIB_ACCESS_READWRITE:      cp = "read-write"; break;
        case MIB_ACCESS_WRITEONLY:      cp = "write-only"; break;
        case MIB_ACCESS_NOACCESS:       cp = "not-accessible"; break;
        case MIB_ACCESS_NOTIFY:         cp = "accessible-for-notify"; break;
        case MIB_ACCESS_CREATE:         cp = "read-create"; break;
        case 0:                          cp = NULL; break;
        default:                        sprintf(str,"access_%d", tp->access); cp = str;
        }
        if (cp) fprintf(f, "  MAX-ACCESS\t%s\n", cp);
        switch (tp->status) {
        case MIB_STATUS_MANDATORY:      cp = "mandatory"; break;
        case MIB_STATUS_OPTIONAL:       cp = "optional"; break;
        case MIB_STATUS_OBSOLETE:       cp = "obsolete"; break;
        case MIB_STATUS_DEPRECATED:     cp = "deprecated"; break;
        case MIB_STATUS_CURRENT:        cp = "current"; break;
        case 0:                          cp = NULL; break;
        default:                        sprintf(str,"status_%d", tp->status); cp = str;
        }
#if SNMP_TESTING_CODE
        if (!cp && (tp->indexes)) { /* index without status ? */
            sprintf(str,"?0 with %s ?",
            tp->indexes ? "Index" : "");
            cp = str;
        }
#endif /* SNMP_TESTING_CODE */
        if (cp) fprintf(f, "  STATUS\t%s\n", cp);
        if (tp->indexes) {
            struct index_list *ip = tp->indexes;
            int first=1;
            fprintf(f, "  INDEXES\t");
            fprintf(f," { ");
            while (ip) {
                if (first) first = 0;
                else fprintf(f, ", ");
                if (ip->isimplied)
                    fprintf(f, "IMPLIED ");
                fprintf(f, "%s", ip->ilabel);
                ip = ip->next;
            }
            fprintf(f," }\n");
        }
        if (tp->description) fprintf(f, "  DESCRIPTION\t\"%s\"\n", tp->description);
    }
    else
        fprintf(f, "No description\n");
}
 
int
get_module_node(const char *fname,
                const char *module,
                oid *objid,
                size_t *objidlen)
{
    int modid, rc = 0;
    struct tree *tp;
    char *name, *cp;
 
    if ( !strcmp(module, "ANY") )
        modid = -1;
    else {
        read_module(module);
        modid = which_module( module );
        if (modid == -1) return 0;
    }
 
                /* Isolate the first component of the name ... */
    name = strdup(fname);
    cp = strchr( name, '.' );
    if ( cp != NULL ) {
        *cp = '\0';
        cp++;
    }
                /* ... and locate it in the tree. */
    tp = find_tree_node(name, modid);
    if (tp){
        size_t maxlen = *objidlen;
 
                /* Set the first element of the object ID */
        if (node_to_oid(tp, objid, objidlen)) {
            rc = 1;
 
                /* If the name requested was more than one element,
                   tag on the rest of the components */
            if (cp != NULL)
                rc = _add_strings_to_oid(tp, cp, objid, objidlen, maxlen);
        }
    }
 
    free(name);
    return (rc);
}
 
 
/*
 * Populate object identifier from a node in the MIB hierarchy.
 * Build up the object ID, working backwards,
 * starting from the end of the objid buffer.
 * When the top of the MIB tree is reached, adjust the buffer.
 *
 * The buffer length is set to the number of subidentifiers
 * for the object identifier associated with the MIB node.
 * Returns the number of subidentifiers copied.
 *
 * If 0 is returned, the objid buffer is too small,
 * and the buffer contents are indeterminate.
 * The buffer length can be used to create a larger buffer.
 */
int
node_to_oid(struct tree *tp, oid *objid, size_t *objidlen)
{
    int numids, lenids;
    oid *op;
 
    if (!tp || !objid || !objidlen)
        return 0;
 
    lenids = (int)*objidlen;
    op = objid + lenids;  /* points after the last element */
 
    for(numids = 0; tp; tp = tp->parent, numids++)
    {
        if (numids >= lenids) continue;
        --op;
        *op = tp->subid;
    }
 
    *objidlen = (size_t)numids;
    if (numids > lenids) {
        return 0;
    }
 
    if (numids < lenids)
        memmove(objid, op, numids * sizeof(oid));
 
    return (numids);
}
 
static int
_add_strings_to_oid(struct tree *tp, char *cp,
             oid *objid, size_t *objidlen,
             size_t maxlen)
{
    int subid;
    struct tree *tp2 = NULL;
    char *cp2 = NULL;
    char doingquote = 0;
 
        while ( cp != NULL ) {
            cp2 = strchr( cp, '.' );    /* Isolate the next entry */
            if ( cp2 != NULL ) {
                *cp2 = '\0';
                cp2++;
            }
 
            if ( *cp == '"' || *cp == '\'') { /* Is it the beggining
                                                 of a quoted string */
              doingquote = *cp++;
              /* insert length if requested */
              if (doingquote == '"') {
                if (*objidlen >= maxlen)
                    return 0;
                objid[ *objidlen ] = (strchr(cp,doingquote) - cp);
                (*objidlen)++;
              }
 
              while(*cp != doingquote) {
                if (*objidlen >= maxlen)
                    return 0;
                objid[ *objidlen ] = *cp++;
                (*objidlen)++;
              }
 
              tp = NULL; /* must be pure numeric from here, right? */
              cp = cp2;
              continue;
            }
 
                                        /* Is it numeric ? */
            if ( isdigit( *cp ) )
                subid=(strtol(cp,0,0));
            else
                subid = -1;
 
                                        /* Search for the appropriate child */
            if ( tp != NULL )
                tp2 = tp->child_list;
            while ( tp2 != NULL ) {
                if (( (int)tp2->subid == subid ) ||
                    ( !strcasecmp( tp2->label, cp ))) {
                        if (*objidlen >= maxlen)
                            return 0;
                        objid[ *objidlen ] = tp2->subid;
                        (*objidlen)++;
                        tp = tp2;
                        break;
                }
                tp2 = tp2->next_peer;
            }
            if ( tp2 == NULL ) {
                if ( subid == -1 ) {
                    return 0;
                }
                                /* pure numeric from now on */
                if (*objidlen >= maxlen)
                    return 0;
                objid[ *objidlen ] = subid;
                (*objidlen)++;
                tp = NULL;
            }
            cp = cp2;
        }
 
        return 1;
}
 
 
/*
 * see comments on find_best_tree_node for usage after first time.
 */
int
get_wild_node(const char *name,
              oid *objid,
              size_t *objidlen)
{
    struct tree *tp = find_best_tree_node(name, tree_head, NULL);
    if (!tp)
        return 0;
    return get_node(tp->label, objid, objidlen);
}
 
int
get_node(const char *name,
         oid *objid,
         size_t *objidlen)
{
    char *cp;
    int res;
 
    if (( cp=strchr(name, ':')) == NULL )
        res = get_module_node( name, "ANY", objid, objidlen );
    else {
        char *module;
                /*
                 *  requested name is of the form
                 *      "module:subidentifier"
                 */
        module = (char *)malloc((size_t)(cp-name+1));
        memcpy(module,name,(size_t)(cp-name));
        module[cp-name] = 0;
        cp++;           /* cp now point to the subidentifier */
        if (*cp == ':') cp++;
 
                        /* 'cp' and 'name' *do* go that way round! */
        res = get_module_node( cp, module, objid, objidlen );
        free(module);
    }
    if (res == 0) {
        SET_SNMP_ERROR(SNMPERR_UNKNOWN_OBJID);
    }
 
    return res;
}
 
#ifdef testing
 
main(int argc, char* argv[])
{
    oid objid[MAX_OID_LEN];
    int objidlen = MAX_OID_LEN;
    int count;
    struct variable_list variable;
 
    init_mib();
    if (argc < 2)
        print_subtree(stdout, tree_head, 0);
    variable.type = ASN_INTEGER;
    variable.val.integer = 3;
    variable.val_len = 4;
    for (argc--; argc; argc--, argv++) {
        objidlen = MAX_OID_LEN;
        printf("read_objid(%s) = %d\n",
               argv[1], read_objid(argv[1], objid, &objidlen));
        for(count = 0; count < objidlen; count++)
            printf("%d.", objid[count]);
        printf("\n");
        print_variable(objid, objidlen, &variable);
    }
}
 
#endif /* testing */
 
/* initialize: no peers included in the report. */
void clear_tree_flags(register struct tree *tp)
{
    for (; tp; tp = tp->next_peer)
    {
        tp->reported = 0;
        if (tp->child_list)
            clear_tree_flags(tp->child_list); /*RECURSE*/
    }
}
 
/*
 * Update: 1998-07-17 <jhy@gsu.edu>
 * Added print_oid_report* functions.
 */
static int print_subtree_oid_report_labeledoid = 0;
static int print_subtree_oid_report_oid = 0;
static int print_subtree_oid_report_symbolic = 0;
static int print_subtree_oid_report_suffix = 0;
 
/* These methods recurse. */
static void print_parent_labeledoid(FILE *, struct tree *);
static void print_parent_oid(FILE *, struct tree *);
static void print_parent_label(FILE *, struct tree *);
static void print_subtree_oid_report(FILE *, struct tree *, int);
 
 
void
print_oid_report (FILE *fp)
{
    struct tree *tp;
    clear_tree_flags(tree_head);
    for (tp = tree_head ; tp ; tp=tp->next_peer)
        print_subtree_oid_report (fp, tp, 0);
}
 
void
print_oid_report_enable_labeledoid (void)
{
    print_subtree_oid_report_labeledoid = 1;
}
 
void
print_oid_report_enable_oid (void)
{
    print_subtree_oid_report_oid = 1;
}
 
void
print_oid_report_enable_suffix (void)
{
    print_subtree_oid_report_suffix = 1;
}
 
void
print_oid_report_enable_symbolic (void)
{
    print_subtree_oid_report_symbolic = 1;
}
 
/*
 * helper methods for print_subtree_oid_report()
 * each one traverses back up the node tree
 * until there is no parent.  Then, the label combination
 * is output, such that the parent is displayed first.
 *
 * Warning: these methods are all recursive.
 */
 
static void
print_parent_labeledoid(FILE *f,
                        struct tree *tp)
{
    if(tp)
    {
        if(tp->parent)
        {
            print_parent_labeledoid(f, tp->parent); /*RECURSE*/
        }
        fprintf(f, ".%s(%lu)", tp->label, tp->subid);
    }
}
 
static void
print_parent_oid(FILE *f,
                 struct tree *tp)
{
    if(tp)
    {
        if(tp->parent)
        {
            print_parent_oid(f, tp->parent); /*RECURSE*/
        }
        fprintf(f, ".%lu", tp->subid);
    }
}
 
static void
print_parent_label(FILE *f,
                   struct tree *tp)
{
    if(tp)
    {
        if(tp->parent)
        {
            print_parent_label(f, tp->parent); /*RECURSE*/
        }
        fprintf(f, ".%s", tp->label);
    }
}
 
/*
 * print_subtree_oid_report():
 *
 * This methods generates variations on the original print_subtree() report.
 * Traverse the tree depth first, from least to greatest sub-identifier.
 * Warning: this methods recurses and calls methods that recurse.
 */
 
static void
print_subtree_oid_report(FILE *f,
                         struct tree *tree,
                         int count)
{
    struct tree *tp;
 
    count++;
 
    /* sanity check */
    if(!tree)
    {
        return;
    }
 
    /*
     * find the not reported peer with the lowest sub-identifier.
     * if no more, break the loop and cleanup.
     * set "reported" flag, and create report for this peer.
     * recurse using the children of this peer, if any.
     */
    while (1)
    {
        register struct tree *ntp;
 
        tp = 0;
        for (ntp = tree->child_list; ntp; ntp = ntp->next_peer)
        {
            if (ntp->reported) continue;
 
            if (!tp || (tp->subid > ntp->subid))
                tp = ntp;
        }
        if (!tp) break;
 
        tp->reported = 1;
 
        if(print_subtree_oid_report_labeledoid)
        {
            print_parent_labeledoid(f, tp);
            fprintf(f, "\n");
        }
        if(print_subtree_oid_report_oid)
        {
            print_parent_oid(f, tp);
            fprintf(f, "\n");
        }
        if(print_subtree_oid_report_symbolic)
        {
            print_parent_label(f, tp);
            fprintf(f, "\n");
        }
        if(print_subtree_oid_report_suffix)
        {
            int i;
            for(i = 0; i < count; i++)
                fprintf(f, "  ");
            fprintf(f, "%s(%ld) type=%d", tp->label, tp->subid, tp->type);
            if (tp->tc_index != -1) fprintf(f, " tc=%d", tp->tc_index);
            if (tp->hint) fprintf(f, " hint=%s", tp->hint);
            if (tp->units) fprintf(f, " units=%s", tp->units);
 
            fprintf(f, "\n");
        }
        print_subtree_oid_report(f, tp, count); /*RECURSE*/
    }
}
 
 
/*
 * Convert timeticks to hours, minutes, seconds string.
 * CMU compatible does not show centiseconds.
 */
char *uptime_string(u_long timeticks, char *buf)
{
    char tbuf[64];
    char * cp;
    uptimeString(timeticks, tbuf);
    cp = strrchr(tbuf, '.');
#ifdef CMU_COMPATIBLE
        if (cp) *cp = '\0';
#endif
    strcpy(buf, tbuf);
    return buf;
}
 
#ifdef CMU_COMPATIBLE
 
int mib_TxtToOid(char *Buf, oid **OidP, size_t *LenP)
{
    return read_objid(Buf, *OidP, LenP);
}
 
int mib_OidToTxt(oid *O, size_t OidLen, char *Buf, size_t BufLen)
{
    _sprint_objid(Buf, O, OidLen);
    return 1;
}
 
#endif /* CMU_COMPATIBLE */
Browse

Tools

Subversion Repositories openrisc

[/] [openrisc/] [trunk/] [rtos/] [ecos-2.0/] [packages/] [net/] [snmp/] [lib/] [v2_0/] [src/] [mib.c] - Blame information for rev 174

Line No.	Rev	Author	Line
1	27	unneback	`//==========================================================================`
2			`//`
3			`// ./lib/current/src/mib.c`
4			`//`
5			`//`
6			`//==========================================================================`
7			`//####ECOSGPLCOPYRIGHTBEGIN####`
8			`// -------------------------------------------`
9			`// This file is part of eCos, the Embedded Configurable Operating System.`
10			`// Copyright (C) 1998, 1999, 2000, 2001, 2002 Red Hat, Inc.`
11			`//`
12			`// eCos is free software; you can redistribute it and/or modify it under`
13			`// the terms of the GNU General Public License as published by the Free`
14			`// Software Foundation; either version 2 or (at your option) any later version.`
15			`//`
16			`// eCos is distributed in the hope that it will be useful, but WITHOUT ANY`
17			`// WARRANTY; without even the implied warranty of MERCHANTABILITY or`
18			`// FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License`
19			`// for more details.`
20			`//`
21			`// You should have received a copy of the GNU General Public License along`
22			`// with eCos; if not, write to the Free Software Foundation, Inc.,`
23			`// 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA.`
24			`//`
25			`// As a special exception, if other files instantiate templates or use macros`
26			`// or inline functions from this file, or you compile this file and link it`
27			`// with other works to produce a work based on this file, this file does not`
28			`// by itself cause the resulting work to be covered by the GNU General Public`
29			`// License. However the source code for this file must still be made available`
30			`// in accordance with section (3) of the GNU General Public License.`
31			`//`
32			`// This exception does not invalidate any other reasons why a work based on`
33			`// this file might be covered by the GNU General Public License.`
34			`//`
35			`// Alternative licenses for eCos may be arranged by contacting Red Hat, Inc.`
36			`// at http://sources.redhat.com/ecos/ecos-license/`
37			`// -------------------------------------------`
38			`//####ECOSGPLCOPYRIGHTEND####`
39			`//####UCDSNMPCOPYRIGHTBEGIN####`
40			`//`