openldap/doc/man/man3/lber-decode.3

.TH LBER_DECODE 3 "22 September 1998" "OpenLDAP LDVERSION"
.SH NAME
ber_get_next, ber_skiptag, ber_peek_tag, ber_scanf, ber_get_int, ber_get_stringb, ber_get_stringa, ber_get_null, ber_get_boolean, ber_get_bitstring, ber_first_element, ber_next_element \- LBER simplified Basic Encoding Rules library routines for decoding
.SH SYNOPSIS
.nf
.ft B
#include <lber.h>
.ft
.fi
.LP
.nf
.ft B
typedef struct berelement BerElement;
.ft
.fi
.LP
.nf
.ft B
typedef struct sockbuf Sockbuf;
.ft
.fi
.LP
.nf
.ft B
typedef struct berval {
    unsigned long bv_len;
    char *bv_val;
};
.ft
.fi
.LP
.nf
.ft B
ber_get_next(sb, len, ber)
Sockbuf *sb;
unsigned long \(**len;
BerElement \(**ber;
.ft
.fi
.LP
.nf
.ft B
ber_skip_tag(ber, len)
BerElement \(**ber;
unsigned long \(**len;
.ft
.fi
.LP
.nf
.ft B
ber_peek_tag(ber, len)
BerElement \(**ber;
unsigned long \(**len;
.ft
.fi
.LP
.nf
.ft B
ber_get_int(ber, num)
BerElement \(**ber;
long \(**num;
.ft
.fi
.LP
.nf
.ft B
ber_get_stringb(ber, buf, len)
BerElement \(**ber;
char \(**buf;
unsigned long \(**len;
.ft
.fi
.LP
.nf
.ft B
ber_get_stringa(ber, buf)
BerElement \(**ber;
char \(***buf;
.ft
.fi
.LP
.nf
.ft B
ber_get_stringal(ber, bv)
BerElement \(**ber;
struct berval \(***bv;
.ft
.fi
.LP
.nf
.ft B
ber_get_null(ber)
BerElement \(**ber;
.ft
.fi
.LP
.nf
.ft B
ber_get_boolean(ber, bool)
BerElement \(**ber;
int \(**bool;
.ft
.fi
.LP
.nf
.ft B
ber_get_bitstringa(ber, buf, blen)
BerElement \(**ber;
char \(***buf;
unsigned long \(**blen;
.ft
.fi
.LP
.nf
.ft B
ber_first_element(ber, len, cookie)
BerElement \(**ber;
unsigned long \(**len;
char \(***cookie;
.ft
.fi
.LP
.nf
.ft B
ber_next_element(ber, len, cookie)
BerElement \(**ber;
unsigned long \(**len;
char \(**cookie;
.ft
.fi
.LP
.nf
.ft B
ber_scanf(ber, fmt [, arg...] )
BerElement \(**ber;
char \(**fmt;
.ft
.fi
.LP
.nf
.ft B
ber_bvfree(bv)
struct berval \(**bv;
.ft
.fi
.LP
.nf
.ft B
ber_bvecfree(bvec)
struct berval \(***bvec;
.SH DESCRIPTION
.LP
These routines provide a subroutine interface to a simplified
implementation of the Basic Encoding Rules of ASN.1.  The version
of BER these routines support is the one defined for the LDAP
protocol.  The encoding rules are the same as BER, except that 
only definite form lengths are used, and bitstrings and octet strings
are always encoded in primitive form.  In addition, these lightweight
BER routines restrict tags and class to fit in a single octet (this
means the actual tag must be less than 31).  When a "tag" is specified
in the descriptions below, it refers to the tag, class, and primitive
or constructed bit in the first octet of the encoding.  This man page
describes the decoding routines in the lber library.  See lber-encode(3)
for details on the corresponding encoding routines.
.LP
Normally, the only routines that need be called by an application
are ber_get_next() to get the next BER element and ber_scanf()
to do the actual decoding.  In some cases, ber_peek_tag() may also
need to be called in normal usage.  The other routines are provided for those
applications that need more control than ber_scanf() provides.  In
general, these routines return the tag of the element decoded, or
-1 if an error occurred.
.LP
The ber_get_next() routine is used to read the next BER element from
the given Sockbuf, \fIsb\fP.  A Sockbuf consists of the descriptor
(usually socket, but a file descriptor works just as well) from which
to read, and a BerElement structure used
to maintain a buffer.  On the first call, the \fIsb_ber\fP struct should
be zeroed.  It strips off and returns the
leading tag byte, strips off and returns the length of the
entire element in \fIlen\fP,
and sets up \fIber\fP for subsequent calls to ber_scanf() et al to decode
the element.
.LP
The ber_scanf() routine is used to decode a BER element in much the
same way that scanf(3) works.  It reads from \fIber\fP, a pointer to a
BerElement such as returned by ber_get_next(), interprets the
bytes according to the format string \fIfmt\fP, and stores the
results in its additional arguments.  The format string contains
conversion specifications which are used to direct the interpretation
of the BER element.  The format string can contain the following
characters.
.RS
.LP
.TP 3
.SM a
Octet string.  A char ** should be supplied.  Memory is allocated,
filled with the contents of the octet string, null-terminated, and
returned in the parameter.
.TP
.SM s
Octet string.  A char * buffer should be supplied, followed by a pointer
to an integer initialized to the size of the buffer.  Upon return, the
null-terminated octet string is put into the buffer, and the integer is
set to the actual size of the octet string.
.TP
.SM O
Octet string.  A struct ber_val ** should be supplied, which upon return
points to a malloced struct berval containing the octet string and its
length.  ber_bvfree() can be called to free the malloced memory.
.TP
.SM b
Boolean.  A pointer to an integer should be supplied.
.TP
.SM i
Integer.  A pointer to an integer should be supplied.
.TP
.SM B
Bitstring.  A char ** should be supplied which will point to the malloced
bits, followed by an unsigned long *, which will point to the length
(in bits) of the bitstring returned.
.TP
.SM n
Null.  No parameter is required.  The element is simply skipped if
it is recognized.
.TP
.SM v
Sequence of octet strings.  A char *** should be supplied, which upon
return points to a malloced null-terminated array of char *'s
containing the octet strings.  NULL is returned if the sequence is empty.
.TP
.SM V
Sequence of octet strings with lengths.
A struct berval *** should be supplied, which upon
return points to a malloced null-terminated array of struct berval *'s
containing the octet strings and their lengths.
NULL is returned if the sequence is empty.  ber_bvecfree() can be called
to free the malloced memory.
.TP
.SM x
Skip element.  The next element is skipped.
.TP
.SM {
Begin sequence.  No parameter is required.  The initial sequence tag
and length are skipped.
.TP
.SM }
End sequence.  No parameter is required and no action is taken.
.TP
.SM [
Begin set.  No parameter is required.  The initial set tag
and length are skipped.
.TP
.SM ]
End set.  No parameter is required and no action is taken.
.RE
.LP
The ber_get_int() routine tries to interpret the next element as an integer,
returning the result in \fInum\fP.  The tag of whatever it finds is returned
on success, -1 on failure.
.LP
The ber_get_stringb() routine is used to read an octet string into a
preallocated buffer.  The \fIlen\fP parameter should be initialized to
the size of the buffer, and will contain the length of the octet string
read upon return.  The buffer should be big enough to take the octet
string value plus a terminating NULL byte.
.LP
The ber_get_stringa() routine is used to malloc space into which an octet
string is read.
.LP
The ber_get_stringal() routine is used to malloc space into which an octet
string and its length are read.  It takes a struct berval **, and returns
the result in this parameter.
.LP
The ber_get_null() routine is used to read a NULL element.  It returns
the tag of the element it skips over.
.LP
The ber_get_boolean() routine is used to read a boolean value.  It is called
the same way that ber_get_int() is called.
.LP
The ber_get_bitstringa() routine is used to read a bitstring value.  It
takes a char ** which will hold the malloced bits, followed by an
unsigned long *, which will point to the length (in bits) of the
bitstring returned.
.LP
The ber_first_element() routine is used to return the tag and length
of the first element in a set or sequence.  It also returns in \fIcookie\fP
a magic cookie parameter that should be passed to subsequent calls to
ber_next_element(), which returns similar information.
.SH EXAMPLES
Assume the variable \fIber\fP contains a lightweight BER encoding of
the following ASN.1 object:
.LP
.nf
      AlmostASearchRequest := SEQUENCE {
          baseObject      DistinguishedName,
          scope           ENUMERATED {
              baseObject    (0),
              singleLevel   (1),
              wholeSubtree  (2)
          },
          derefAliases    ENUMERATED {
              neverDerefaliases   (0),
              derefInSearching    (1),
              derefFindingBaseObj (2),
              alwaysDerefAliases  (3)
          },
          sizelimit       INTEGER (0 .. 65535),
          timelimit       INTEGER (0 .. 65535),
          attrsOnly       BOOLEAN,
          attributes      SEQUENCE OF AttributeType
      }
.fi
.LP
The element can be decoded using ber_scanf() as follows.
.LP
.nf
      int    scope, ali, size, time, attrsonly;
      char   *dn, **attrs;

      if ( ber_scanf( ber, "{aiiiib{v}}", &dn, &scope, &ali,
          &size, &time, &attrsonly, &attrs ) == -1 )
              /* error */
      else
              /* success */
.fi
.SH ERRORS
If an error occurs during decoding, generally these routines return -1.
.LP
.SH NOTES
.LP
The return values for all of these functions are declared in the
<lber.h> header file.  Some routines may malloc memory.
.SH SEE ALSO
.BR lber-encode (3)
.BR ldap-parse (3)
.BR ldap-sync (3)
.BR ldap-async (3)
.LP
Yeong, W., Howes, T., and Hardcastle-Kille, S., "Lightweight Directory Access
Protocol", OSI-DS-26, April 1992.
.LP
Information Processing - Open Systems Interconnection - Model and Notation -
Service Definition - Specification of Basic Encoding Rules for Abstract
Syntax Notation One, International Organization for Standardization,
International Standard 8825.
.SH AUTHOR
Tim Howes, University of Michigan
.SH ACKNOWLEDGEMENTS
.B	OpenLDAP
is developed and maintained by The OpenLDAP Project (http://www.openldap.org/).
.B	OpenLDAP
is derived from University of Michigan LDAP 3.3 Release.