| Current Path : /usr/share/doc/krb5-devel-1.10.3/ccapi/ |
| Current File : //usr/share/doc/krb5-devel-1.10.3/ccapi/ccache-api-v2.html |
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN"
"http://www.w3.org/TR/REC-html40/loose.dtd">
<HTML>
<HEAD>
<TITLE>Credentials Cache API v2 Specification</TITLE>
</HEAD>
<BODY>
<H1>Credentials Cache API v2 Specification</H1>
<H3>This version of the API is deprecated. <BR>
Please refer to CCAPI version 3 or later for the current API.</H3>
<UL>
<LI><A HREF="#TypeDefinitions">Type Definitions</A>
<UL CLASS=small>
<LI><A HREF="#cc_int32">cc_int32 & cc_uint32</A>
<LI><A HREF="#cc_time_t">cc_time_t</A>
<LI><A HREF="#cc_nc_flags">cc_nc_flags</A>
<LI><A HREF="#typedefCC_Cred">cc_cred</A>
<LI><A HREF="#typedefCC_Data">cc_data</A>
<LI><A HREF="#typedefV4Cred_type">V4Cred_type</A>
<LI><A HREF="#enumStringToKey_Type">StringToKey_Type</A>
<LI><A HREF="#enumCc_cred_vers">cc_cred_vers</A>
<LI><A HREF="#unionCred_ptr_union">cred_ptr_union</A>
<LI><A HREF="#unionCred_union">cred_union</A>
<LI><A HREF="#typdefInfoNC">infoNC</A>
<LI><A HREF="#opaque_types">opaque types</A>
</UL>
<LI><A HREF="#ErrorCodeDefinitions">Error Code Definitions</A>
<LI><A HREF="#ImplementationNotes">Implementation Notes</A>
<LI><A HREF="#FunctionDefinitions">Function Definitions</A>
<UL>
<LI><A HREF="#MainCacheFunctions">Main Cache Functions</A>
<UL CLASS=small>
<LI><A HREF="#cc_initialize">cc_initialize</A>
<LI><A HREF="#cc_create">cc_create</A>
<LI><A HREF="#cc_open">cc_open</A>
<LI><A HREF="#cc_close">cc_close</A>
<LI><A HREF="#cc_destroy">cc_destroy</A>
<LI><A HREF="#cc_shutdown">cc_shutdown</A>
<LI><A HREF="#cc_initialize">cc_initialize</A>
<LI><A HREF="#cc_get_change_time">cc_get_change_time</A>
<LI><A HREF="#cc_get_NC_info">cc_get_NC_info</A>
<LI><A HREF="#cc_seq_fetch_NCs_begin">cc_seq_fetch_NCs_begin</A>
<LI><A HREF="#cc_seq_fetch_NCs_next">cc_seq_fetch_NCs_next</A>
<LI><A HREF="#cc_seq_fetch_NCs_end">cc_seq_fetch_NCs_end</A>
</UL>
<LI><A HREF="#NCFunctions">NC Functions</A>
<UL CLASS=small>
<LI><A HREF="#cc_get_name">cc_get_name</A>
<LI><A HREF="#cc_get_cred_version">cc_get_cred_version</A>
<LI><A HREF="#cc_get_principal">cc_get_principal</A>
<LI><A HREF="#cc_get_instance">cc_get_instance</A>
<LI><A HREF="#cc_remove_cred">cc_remove_cred</A>
<LI><A HREF="#cc_store">cc_store</A>
<LI><A HREF="#cc_get_instance">cc_get_instance</A>
<LI><A HREF="#cc_seq_fetch_creds_begin">cc_seq_fetch_creds_begin</A>
<LI><A HREF="#cc_seq_fetch_creds_next">cc_seq_fetch_creds_next</A>
<LI><A HREF="#cc_seq_fetch_creds_end">cc_seq_fetch_creds_end</A>
<LI><A HREF="#cc_lock_request">cc_lock_request</A>
</UL>
<LI><A HREF="#LiberationFunctions">Liberation Functions</A>
<UL CLASS=small>
<LI><A HREF="#cc_free_principal">cc_free_principal</A>
<LI><A HREF="#cc_free_name">cc_free_name</A>
<LI><A HREF="#cc_free_creds">cc_free_creds</A>
<LI><A HREF="#cc_free_NC_info">cc_free_NC_info</A>
</UL>
</UL>
<LI><A HREF="#RevisionNotes">Revision History/Notes</A>
<UL>
<LI><A HREF="#ChangesRev3">Changes made during revsion 3</A>
</UL>
<LI><A HREF="#FutureIdeas">Ideas for Future Versions</A>
<LI><A HREF="#Abstract">Abstract</A>
</UL>
<P>
<HR>
<A NAME="Abstract"></A>
<H2>Abstract</H2>
<P>This is the specification for an API which provides Credentials
Cache services for both
<A HREF="http://web.mit.edu/kerberos/www/">Kerberos V5</A> and V4.
The idea behind this API is that multiple Kerberos implementations
can share a single Credentials Cache, mediated by this API
specification. On the Microsoft Windows platform this will allow
single-signon, even when more than one Kerberos DLL is in use on a
particular system. Ideally, this problem could be solved by
standardizing the Kerberos V5 API library interface. However, the
Kerberos API is complicated enough that this would be hard to
accomplish. Standardizing the interface for credentials cache access
is much simpler. This API has also been adopted in the MIT Kerberos
for the Macintosh implementation.
<P>This specification has been revised to allow storage and
manipulation of both V4 and V5 tickets. A cache contains one or more
"Named Cache"s. It is assumed that V4 and V5 credentials would each
be stored in separate "Named Cache"s and not mixed in a single "Named
Cache".
<P>Below, "NC" refers to "Named Cache".<BR>
<P>
<HR>
<A NAME="RevisionNotes"></A>
<H2><B>Revision History/Notes</B></H2>
<H4>Original version (Draft Version 1)</H4>
<P>1/27/96 by
<A HREF="http://web.mit.edu/tytso/www/home.html">Theodore Ts'o</A>
<H4>Revision 2 (Draft Version 1)</H4>
<P>970628 by <A HREF="http://www.umich.edu/~sgr">Steve Rothwell</A>
for the V4Cache Team (Paul Hill, Jenny Khuon, Jean Luker, Dave
Detlefs, Allan Bjorklund, & Steve Rothwell)
<P><A NAME="ChangesRev3"></A>
<H4>Revision 3 (Draft Version 1)</H4>
<P>970725 by Steve Rothwell after initial implementation and alpha
release. The term "credentials cache" was previously used to mean
both "the main cache" and individual "named cache"s within the main
cache. I have started using the term "NC" for "named cache" to make
the distinction clearer and to reduce the overloading of the word
"cache".
<H5>Changes made for revision 3 of this API:</H5>
<UL>
<LI>Added cred version type to cc_create() & cc_open()
<LI>New functions
<UL>
<LI>cc_get_NC_info(), returns NC_info list for all NCs
<LI>cc_free_NC_info(), frees NC_info list
<LI>cc_get_cred_version(), returns version type of NC
<LI>cc_get_name(), returns name of NC
<LI>cc_free_name(), frees name aquired via cc_get_name()
<LI>cc_seq_fetch_NCs(), iterate over all NCs
</UL>
<LI>New return codes
<UL>
<LI>CC_BAD_PARM
<LI>CC_ERR_CACHE_ATTACH
<LI>CC_ERR_CACHE_RELEASE
<LI>CC_ERR_CACHE_FULL
<LI>CC_ERR_CRED_VERSION
</UL>
<LI>Modified functions
<UL>
<LI>cc_create(), cc_open(), pass version type of NC
<LI>cc_store(), cc_remove(), cc_
</UL>
<LI>New & Modified typedefs & data structures
<UL>
<LI>cc_cred_vers { CC_CRED_VUNKNOWN, CC_CRED_V4, CC_CRED_V5 }
<LI>cred_ptr_union : contains pointer to credentials (either V4
or V5)
<LI>cred_union : contains version type and cred_ptr_union
<LI>modified V4Cred_type
<LI>enum StringToKey_Type { STK_AFS or STK_DES }
<LI>copies of the maximum V4 string size indicators
KRB_PRINCIPAL_SZ, KRB_SERVICE_SZ, KRB_INSTANCE_SZ,
KRB_REALM_SZ, ADDR_SZ
</UL>
</UL>
<H4>Revision 4 (Draft Version 1)</H4>
<P>970908 by Steve Rothwell to incorporate changes initiated by Ted
Tso. Further changes are expected in the comments for cc_create() and
cc_get_change_time().
<H4>Revision 4a (Final Version 1)</H4>
<P>980603 by <A HREF="mailto:smcguire@mit.edu">Scott McGuire</A> to
correct typographical errors, HTML errors, and minor clarifications.
Final API Version 1 spec.
<H4>Revision 5 (Draft Version 2)</H4>
<P>990201 by <A HREF="mailto:smcguire@mit.edu">Scott McGuire</A>.
<UL>
<LI>Increased API version number to 2.
<LI>Added enum's defining version numbers.
<LI>Changes to cc_initialize() to specify how to deal with
different API version numbers.
<LI>Added description of cc_int32 and cc_uint32 types.
<LI>Change some cc_int32's to cc_uint32's.
<LI>Changed way cc_create() will behave when called on an existing
cache.
<LI>Replaced cc_seq_fetch_NCs() with cc_seq_fetch_NCs_begin(),
cc_seq_fetch_NCs_next(), and cc_seq_fetch_NCs_end();
<LI>Replaced cc_seq_fetch_creds() with cc_seq_fetch_creds_begin(),
cc_seq_fetch_creds_next(), and cc_seq_fetch_creds_end();
<LI>Replaced enum type references in structs and function
paramenters with cc_int32 references;
<LI>Replaced int type references in function parameters with
cc_int32;
<LI>Added return type of cc_int32 to all functions;
<LI>Removed #ifdef from cred_union structure;
<LI>Constant definitions and changes to V4Cred_type structure;
<LI>Removed incorrect const ccache_p * parameters from cc_store()
and cc_remove_cred();
<LI>Added CC_NOERROR and CC_BAD_PARM as possible return codes from
all functions (except no CC_BAD_PARM from cc_shutdown() );
<LI>Added CC_ERR_CRED_VERSION as possible return code from
cc_open() and cc_create();
<LI>Moved infoNC structure definition up to be with rest of
structure definitions;
<LI>Changed "struct _infoNC" to "infoNC" in parameter type
references.
<LI>cc_free_principal() and cc_free_name() now take char **
instead of char * for final parameter. (This change was made
between rev 4a and rev 5, but I'm re-emphasizing it here.)
<LI>Added Implementation Notes section with requirement that all
functions must be atomic and name requirements for Windows DLL's.
<LI>Renamed "the proposed changes to this API are" section to
"Ideas for Future Versions" -- but removed all items but one
because they'd all been done.
<LI>Removed most of the notes about differences with the Win NT/95
implementation of the API -- the differences have been reconciled.
<LI>Removed unnecessary and inconsistent italicizing.
</UL>
<H4>Revsion 5a (Final Version 2)</H4>
<P>990723 by <A HREF="mailto:smcguire@mit.edu">Scott McGuire</A>.
<UL>
<LI>cc_create(): Removed text about "expected" form of name.
Removed note about "the alpha version does not do this."
<LI>cc_destroy(): Clarified that you do not need to call
cc_close() on the cache_pointer after calling this function.
<LI>Removed note about Windows cc_get_instance() and
cc_set_instance() functions, they are no longer part of the
Windows code!
</UL>
<H2><A NAME="FutureIdeas"></A>Ideas for Future Versions</H2>
<UL>
<LI>Define Get/Set functions for all components of _cc_creds?
(This will allow future changes to the data structure to be
transparent to the caller. This also makes backward compatibility
much easier to maintain.)
</UL>
<P><!-- -----------------------------------------------------------------------------------><!-- type definitions --><!-- -----------------------------------------------------------------------------------><BR>
<HR>
<A NAME="TypeDefinitions"></A>
<H2>Type definitions</H2>
<PRE>// enums for API versions used in cc_initialize()
enum {
CC_API_VER_1 = 1,
CC_API_VER_2 = 2
};
<A NAME="cc_int32"></A>
// cc_int32 and cc_uint32 are not exactly defined in this API due
// to a lack of standard 32-bit integer size between platforms
// (although there is the C9X standard).
// However, we will place the following constraints:
//
// cc_int32 is a signed integer that is at least 32 bits wide.
// cc_uint32 is an unsigned integer that is at least 32 bits wide
<A NAME="cc_time_t"></A>
typedef cc_int32 cc_time_t; //see <A HREF="#cc_time_t_Note">notes below</A>
<A NAME="cc_nc_flags"></A>
typedef cc_uint32 cc_nc_flags;
<A NAME="opaque_types"></A>
typedef struct opaque_dll_control_block_type* apiCB;
typedef struct opaque_ccache_pointer_type* ccache_p;
typedef struct opaque_credential_iterator_type* ccache_cit;
// These really are intended to be opaque. All implementations of the cache API must have
// them but what they are is implementation specific. In the case of SGR's implementation,
// the cc_ctx returned available after a call to cc_initialize, is a CCache_ctx class object. The
// code that normally calls the cc_initialize function is straight C, which means the calling
// application doesn't have a chance in hell of manipulating this directly. The API is designed
// so that it does not have to. It does have to pass the pointer to the class around, one reason
// being so that the destructor can eventually be called.
<A NAME="typedefCC_Data"></A>
typedef struct _cc_data {
cc_uint32 type;
cc_uint32 length;
unsigned char* data;
} cc_data;
<A NAME="typedefCC_Cred"></A>
typedef struct _cc_creds {
char* client; /* client's principal identifier */
char* server; /* server's principal identifier */
cc_data keyblock; /* session encryption key info */
cc_time_t authtime;
cc_time_t starttime;
cc_time_t endtime;
cc_time_t renew_till;
cc_uint32 is_skey; /* true if ticket is encrypted in
another ticket's skey */
cc_uint32 ticket_flags; /* flags in ticket */
cc_data** addresses; /* addrs in ticket */
cc_data ticket; /* ticket string itself */
cc_data second_ticket; /* second ticket, if related to
ticket (via DUPLICATE-SKEY or
ENC-TKT-IN-SKEY) */
cc_data** authdata; /* authorization data */
} cc_creds;
// use an enumerated type so all callers infer the same meaning
// these values are what krbv4win uses internally.
<A NAME="enumStringToKey_Type"></A>
enum StringToKey_Type { STK_AFS = 0, STK_DES = 1 };
enum { MAX_V4_CRED_LEN = 1250 };
// V4 Credentials
<A NAME="typedefV4Cred_type"></A>
enum {
KRB_NAME_SZ = 40,
KRB_INSTANCE_SZ = 40,
KRB_REALM_SZ = 40
};
typedef struct _V4credential {
unsigned char kversion;
char principal[KRB_NAME_SZ+1];
char principal_instance[KRB_INSTANCE_SZ+1];
char service[KRB_NAME_SZ+1];
char service_instance[KRB_INSTANCE_SZ+1];
char realm[KRB_REALM_SZ+1];
unsigned char session_key[8];
cc_int32 kvno; // k95 used BYTE skvno
cc_int32 str_to_key; // k4 infers dynamically, k95 stores; of type enum StringToKey_Type
long issue_date; // k95 called this issue_time
cc_int32 lifetime; // k95 used LONG expiration_time
cc_uint32 address; // IP Address of local host as an unsigned 32-bit integer
cc_int32 ticket_sz; // k95 used BYTE, k4 ktext uses int to hold up to 1250
unsigned char ticket[MAX_V4_CRED_LEN];
unsigned long oops; // zero to catch runaways
} V4Cred_type;
<A NAME="enumCc_cred_vers"></A>
enum cc_cred_vers {
CC_CRED_VUNKNOWN = 0, // For validation
CC_CRED_V4 = 1,
CC_CRED_V5 = 2,
CC_CRED_VMAX = 3 // For validation
};
<A NAME="unionCred_ptr_union"></A>
typedef union cred_ptr_union_type {
V4Cred_type* pV4Cred;
cc_creds* pV5Cred;
} cred_ptr_union;
<A NAME="unionCred_union"></A>
typedef struct cred_union_type {
cc_int32 cred_type; // cc_cred_vers
<A HREF="#unionCred_ptr_union">cred_ptr_union</A> cred;
} cred_union;
<A NAME="typdefInfoNC"></A>
typedef struct _infoNC {
char* name;
char* principal;
cc_int32 vers; // cc_cred_vers
} infoNC;</PRE>
<H4>The cc_data structure</H4>
<P>The cc_data structure is used to store the following elements:
<UL>
<LI>keyblock
<LI>addresses
<LI>ticket (and second_ticket)
<LI>authorization data
</UL>
<P>For cc_creds.ticket and cc_creds.second_ticket, the cc_data.type
field MUST be zero. For the cc_creds.addresses, cc_creds.authdata,
and cc_data.keyblock, the cc_data.type field should be the address
type, authorization data type, and encryption type, as defined by the
Kerberos V5 protocol definition.
<H4><A NAME="cc_time_t_Note"></A>cc_time_t</H4>
<P>The cc_time_t fields are used to represent time. The time must be
stored as the number of seconds since midnight GMT on January 1,
1970.
<H4>Principal names</H4>
<P>Principal names are stored as C strings in this API. The C strings
may contain UTF-8 encoded strings for internationalization
purposes.<BR>
<HR>
<A NAME="ErrorCodeDefinitions"></A>
<H2>Error Codes Definition</H2>
<P><TABLE BORDER=1>
<TR>
<TD>
<P>0
</TD><TD>
<P>CC_NOERROR
</TD><TD>
<P>"Successful return"
</TD></TR>
<TR>
<TD>
<P>1
</TD><TD>
<P>CC_BADNAME
</TD><TD>
<P>"Bad credential cache name format"
</TD></TR>
<TR>
<TD>
<P>2
</TD><TD>
<P>CC_NOTFOUD
</TD><TD>
<P>"Matching credential not found"
</TD></TR>
<TR>
<TD>
<P>3
</TD><TD>
<P>CC_END
</TD><TD>
<P>"End of credential cache reached"
</TD></TR>
<TR>
<TD>
<P>4
</TD><TD>
<P>CC_IO
</TD><TD>
<P>"Credentials cache I/O operation failed"
</TD></TR>
<TR>
<TD>
<P>5
</TD><TD>
<P>CC_WRITE
</TD><TD>
<P>"Error writing to credentials cache file"
</TD></TR>
<TR>
<TD>
<P>6
</TD><TD>
<P>CC_NOMEM
</TD><TD>
<P>"No memory"
</TD></TR>
<TR>
<TD>
<P>7
</TD><TD>
<P>CC_FORMAT
</TD><TD>
<P>"Corrupted credentials cache"
</TD></TR>
<TR>
<TD>
<P>8
</TD><TD>
<P>CC_LOCKED
</TD><TD>
<P>"The credentials cache or NC is locked"
</TD></TR>
<TR>
<TD>
<P>9
</TD><TD>
<P>CC_BAD_API_VERSION
</TD><TD>
<P>"Unsupported API version"
</TD></TR>
<TR>
<TD>
<P>10
</TD><TD>
<P>CC_NO_EXIST
</TD><TD>
<P>"Credentials cache or NC does not exist"
</TD></TR>
<TR>
<TD>
<P>11
</TD><TD>
<P>CC_NOT_SUPP
</TD><TD>
<P>"Function not supported"
</TD></TR>
<TR>
<TD>
<P>12
</TD><TD>
<P>CC_BAD_PARM
</TD><TD>
<P>"Bad Paramter Passed"
</TD></TR>
<TR>
<TD>
<P>13
</TD><TD>
<P>CC_ERR_CACHE_ATTACH
</TD><TD>
<P>"Failed to attach cache"
</TD></TR>
<TR>
<TD>
<P>14
</TD><TD>
<P>CC_ERR_CACHE_RELEASE
</TD><TD>
<P>"Failed to release cache"
</TD></TR>
<TR>
<TD>
<P>15
</TD><TD>
<P>CC_ERR_CACHE_FULL
</TD><TD>
<P>"Cache FULL"
</TD></TR>
<TR>
<TD>
<P>16
</TD><TD>
<P>CC_ERR_CRED_VERSION
</TD><TD>
<P>"Wrong Cred Version"
</TD></TR>
</TABLE>
<P>
<HR>
<A NAME="ImplementationNotes"></A>
<H2>Implementation Notes</H2>
<H4>All functions are atomic</H4>
<P>All Credentials Cache API functions must be atomic.
<P><B>Windows</B>
<P>DLLs should be named KrbCC16.dll and KrbCC32.dll.
<P>
<HR>
<A NAME="FunctionDefinitions"></A>
<H2>Function definitions</H2>
<P><A NAME="MainCacheFunctions"></A>
<H3>Main Cache Functions</H3>
<P><!-- -----------------------------------------------------------------------------------><!-- cc_initialize --><!-- ----------------------------------------------------------------------------------->
<P><A NAME="cc_initialize"></A>
<H4>cc_initialize</H4>
<PRE>cc_int32 cc_initialize(apiCB** cc_ctx, cc_int32 api_version, cc_int32* api_supported, char** vendor)</PRE>
<P>This function performs any initialization required by the
<I>API</I>. It must be called before any other function in the
<I>API</I> is called. The cc_ctx returned by this function must be
passed to all other <I>API</I> functions as the first argument.
<P>The application must pass in the maximum version number of the API
it supports in the <B>api_version</B> parameter.
<P>If <B>api_supported</B> non-NULL, then cc_initialize will store
the maximum API version number supported by the library implementing
the API there.
<P>If the version requested by <B>api_version</B> is not equal to the
version supported by the library, CC_BAD_API_VERSION will be returned
as the error code (along with the version the library does support in
<B>api_supported</B>) and cc_initialize should not allocate any
memory.
<P>If the <B>vendor</B> is non-NULL, then cc_initialize will store a
pointer to a read/only C string which contains a string describing
the vendor which implemented the credentials cache <I>API</I>.
<P>Possible error codes: CC_NOERROR, CC_NOMEM, CC_BAD_API_VERSION,
CC_BAD_PARM
<P><!-- -----------------------------------------------------------------------------------><!-- cc_shutdown --><!-- -----------------------------------------------------------------------------------><BR>
<A NAME="cc_shutdown"></A>
<H4>cc_shutdown</H4>
<PRE>cc_int32 cc_shutdown(apiCB** cc_ctx)</PRE>
<P>This function performs any cleanup required by the <I>API</I>.
cc_ctx will be NULL on return. The application program must call
cc_initialize() again before making any credentials cache <I>API</I>
calls.
<P>Possible error codes: CC_NOERROR, CC_NO_EXIST, CC_BAD_PARM
<P><!-- -----------------------------------------------------------------------------------><!-- cc_get_change_time --><!-- -----------------------------------------------------------------------------------><BR>
<A NAME="cc_get_change_time"></A>
<H4>cc_get_change_time</H4>
<PRE>cc_int32 cc_get_change_time(apiCB* cc_ctx, cc_time_t* time)</PRE>
<P>This function returns the time of the most recent change for the
entire cache. There is ONE timestamp maintained for the entire cache.
By maintaining a local copy the caller can deduce whether "something
changed" or not.
<P>Possible error codes: CC_NOERROR, CC_NO_EXIST, CC_NOMEM,
CC_BAD_PARM
<P><!-- -----------------------------------------------------------------------------------><!-- cc_get_NC_info --><!-- -----------------------------------------------------------------------------------><BR>
<A NAME="cc_get_NC_info"></A>
<H4>cc_get_NC_info</H4>
<PRE>cc_int32 cc_get_NC_info(apiCB* cc_ctx, infoNC*** ppNCi)</PRE>
<P>cc_get_NC_info() is a wrapper for cc_seq_fetch_NCs(),
cc_get_name() cc_get_cred_version(), and cc_get_principal(). It
returns all the information needed to uniquely identify each NC in
the cache (name and cred_version) and the associated principal.
Specifically it returns a null terminated list of pointers to infoNC
structs. Each infoNC struct contain a pointer to the NC's name, a
pointer to the the principal associated with the NC, and the version
number (as an enumerated type) of the credentials stored in this NC.
<P>The ppNCi (the entire data structure) aquired by this routine
should be freed with cc_free_NC_info().
<P>Possible error codes: CC_NOERROR, CC_NO_EXIST, CC_NOMEM,
CC_BAD_PARM
<P><!-- -----------------------------------------------------------------------------------><!-- cc_open --><!-- -----------------------------------------------------------------------------------><BR>
<A NAME="cc_open"></A>
<H4>cc_open</H4>
<PRE>cc_int32 cc_open(apiCB* cc_ctx, const char* name, cc_int32 cred_vers, cc_uint32 cc_flags,
ccache_p** ccache_pointer)</PRE>
<P>Opens an already exising NC identified by both <B>name</B>, and
<B>cred_vers</B>. It fills in the parameter **ccache_pointer with a
pointer to the NC.
<P>The list of cache names, principals, and credentials versions may
be retrieved via cc_seq_fetch_NCs(), cc_get_name(),
cc_get_cred_version(), & cc_get_principal() OR via
cc_get_NC_info().
<P>Possible error codes: CC_NOERROR, CC_BADNAME, CC_NO_EXIST,
CC_NOMEM, CC_ERR_CRED_VERSION, CC_BAD_PARM
<P><!-- -----------------------------------------------------------------------------------><!-- cc_create --><!-- -----------------------------------------------------------------------------------><BR>
<A NAME="cc_create"></A>
<H4>cc_create</H4>
<PRE>cc_int32 cc_create(apiCB* cc_ctx, const char* name, const char* principal,
cc_int32 cred_vers, cc_uint32 cc_flags, ccache_p** ccache_pointer)</PRE>
<P>Create a new NC. The NC is uniquely identified by the combination
of it's name and the "cc_creds_vers" (i.e. which credentials version
it holds). The principal given is also associated with the NC. A NULL
<B>name</B> is not allowed (and CC_BADNAME should be returned if one
is passed in). If <B>name</B> is non-null and there is already a NC
named <B>name</B>, all credentials in the cache are removed, and
handle for the existing cache is returned. If there is already a NC
named name, all existing handles for this cache remain valid. The NC
is created with a primary principal specified by <B>principal</B>.
<P><I>(Removed text about the "expected" form of the NC name.)</I>
<P>An NC is intended to hold credentials for a single principal in a
single realm, and for a single credentials version (i.e. V4 or V5).
The cache can contain credentials for other credential versions,
other realms, and even other principals, but each in a separate NC.
This rule will allow callers that can only handle a single principal
in a single realm to continue to work by dealing with only one NC.
Callers that can deal with multiple principals, multiple realms,
and/or multiple credentials versions can do so by dealing with
multiple NCs. By doing it this way, the callers that are able to
handle multiple principals, realms, and/or versions can do so without
interfering with "differently abled" code.
<P>The list of cache names, principals, & cred_versions may be
retrieved via cc_get_NC_info().
<P>Possible error codes: CC_NOERROR, CC_BADNAME, CC_BAD_PARM,
CC_NO_EXIST, CC_NOMEM, CC_ERR_CRED_VERSION
<P><!-- -----------------------------------------------------------------------------------><!-- cc_close --><!-- -----------------------------------------------------------------------------------><BR>
<A NAME="cc_close"></A>
<H4>cc_close</H4>
<PRE>cc_int32 cc_close(apiCB* cc_ctx, ccache_p** ccache_pointer)</PRE>
<P>Close the NC. The <B>ccache_pointer</B> related memory is
deallocated, and ccache_pointer is set to NULL before being returned
to caller.
<P>Possible error codes: CC_NOERROR, CC_NO_EXIST, CC_BAD_PARM
<P><!-- -----------------------------------------------------------------------------------><!-- cc_destroy --><!-- -----------------------------------------------------------------------------------><BR>
<A NAME="cc_destroy"></A>
<H4>cc_destroy</H4>
<PRE>cc_int32 cc_destroy(apiCB* cc_ctx, ccache_p** ccache_pointer)</PRE>
<P>Destroy the NC pointed to by <B>ccache_pointer</B>. The
<B>ccache_pointer</B> related memory is deallocated, and
ccache_pointer is set to NULL before being returned to caller. The
caller does not need to call cc_close() on the <B>cache_pointer</B>
afterwards.
<P>Possible error codes: CC_NOERROR, CC_NO_EXIST, CC_BAD_PARM
<P><!-- -----------------------------------------------------------------------------------><!-- cc_seq_fetch_NCs --><!-- -----------------------------------------------------------------------------------><BR>
<P><A NAME="cc_seq_fetch_NCs_begin"></A>
<H4>cc_seq_fetch_NCs_begin</H4>
<PRE>cc_int32 cc_seq_fetch_NCs_begin(apiCB* cc_ctx, ccache_cit** itNCs)</PRE>
<P>Used to allocate memory and initialize the iterator *itNCs. Use
cc_seq_fetch_NCs_end() to deallocate the memory used by *itNCs.
<P>Possible error codes: CC_NOERROR, CC_NO_EXIST, CC_BAD_PARM,
CC_NOMEM
<H4><A NAME="cc_seq_fetch_NCs_next"></A></H4>
<H4>cc_seq_fetch_NCs_next</H4>
<PRE>cc_int32 cc_seq_fetch_NCs_next(apiCB* cc_ctx, ccache_p** ccache_pointer, ccache_cit* itNCs)</PRE>
<P>Used to sequentially open every NC in the cache.
<P>Ccache_pointer must be a pointer to a ccache_p*. The
ccache_pointer returned may be used to get information about the NC
by calling cc_get_name(), cc_get_cred_version(), and
cc_get_principal(). Ccache_pointer's returned must be freed via
cc_close() between calls to cc_seq_fetch_NCs_next().
<P>itNCs must be a pointer to a ccache_cit* variable provided by the
calling application and which is used by cc_seq_fetch_NCs_next() to
determine the next NC to return. It must have been initialized by
cc_seq_fetch_NCs_begin().
<P>If changes are made to the credentials cache while it iterator is
being used, it must return at least the intersection, and at most the
union, of the set of NC's that were in the cache when the iteration
began and the set of NC's that are in the cache when it ends.
<P>When the last NC in the sequence is returned, the return code from
cc_seq_fetch_NCs_next() will be CC_END.
<P>Possible error codes: CC_NOERROR, CC_END, CC_NO_EXIST.
CC_BAD_PARM, CC_NOMEM
<H4> </H4>
<H4><A NAME="cc_seq_fetch_NCs_end"></A></H4>
<H4>cc_seq_fetch_NCs_end</H4>
<PRE>cc_int32 cc_seq_fetch_NCs_end(apiCB* cc_ctx, ccache_cit** itNCs)</PRE>
<P>Deallocates the memory used by *itNCs, and sets *itNCs to NULL.
<P>Possible error codes: CC_NOERROR, CC_NO_EXIST, CC_BAD_PARM
<P>
<P><A NAME="NCFunctions"></A>
<H3>NC Functions</H3>
<P><!-- -----------------------------------------------------------------------------------><!-- cc_get_name --><!-- -----------------------------------------------------------------------------------><A NAME="cc_get_name"></A>
<H4>cc_get_name</H4>
<PRE>cc_int32 cc_get_name(apiCB* cc_ctx, const ccache_p* ccache_pointer, char** name)</PRE>
<P>cc_get_name() returns the name of the NC indicated by
ccache_pointer. The name can be used in cc_open() or cc_create(). The
combination of the name and the credentials version uniqeuly identify
an NC. The returned name should be freed via cc_free_name().
<P>Possible error codes: CC_NOERROR, CC_NOMEM, CC_NO_EXIST,
CC_BAD_PARM
<P><!-- -----------------------------------------------------------------------------------><!-- cc_get_cred_version --><!-- -----------------------------------------------------------------------------------><BR>
<A NAME="cc_get_cred_version"></A>
<H4>cc_get_cred_version</H4>
<PRE>cc_int32 cc_get_cred_version(apiCB* cc_ctx, const ccache_p* ccache_pointer, cc_int32* cred_vers)</PRE>
<P>cc_get_cred_version() returns one of the enumerated type
cc_cred_vers in cred_vers. The expected values are CC_CRED_V4, or
CC_CRED_V5. The combination of the name and the credentials version
uniquely identify an NC.
<P>Possible error codes: CC_NOERROR, CC_NO_EXIST, CC_BAD_PARM
<P><!-- -----------------------------------------------------------------------------------><!-- cc_set_principal --><!-- -----------------------------------------------------------------------------------><BR>
<A NAME="cc_set_principal"></A>
<H4>cc_set_principal</H4>
<PRE>cc_int32 cc_set_principal(apiCB* cc_ctx, const ccache_p* ccache_pointer, const cc_int32 cred_vers,
const char* principal)</PRE>
<P>Set the primary principal for the NC indicated by ccache_pointer.
This is the complement to cc_get_principal().
<P>cred_vers is used as a double check.
<P>principal points to a null terminated string that will be copied
into the NC. This new principal will be returned if you call
cc_get_principal() for this NC.
<P>Possible error codes: CC_NOERROR, CC_NOMEM, CC_NO_EXIST,
CC_ERR_CRED_VERSION, CC_BAD_PARM<BR>
<P><!-- -----------------------------------------------------------------------------------><!-- cc_get_principal --><!-- -----------------------------------------------------------------------------------><BR>
<A NAME="cc_get_principal"></A>
<H4>cc_get_principal</H4>
<PRE>cc_int32 cc_get_principal(apiCB* cc_ctx, const ccache_p* ccache_pointer, char** principal)</PRE>
<P>Return the primary principal for the NC that was set via
cc_create() or cc_set_principal(). The returned principal should be
freed via cc_free_principal() .
<P>Possible error codes: CC_NOERROR, CC_NOMEM, CC_NO_EXIST,
CC_BAD_PARM<BR>
<P><!-- -----------------------------------------------------------------------------------><!-- cc_store --><!-- -----------------------------------------------------------------------------------><BR>
<A NAME="cc_store"></A>
<H4>cc_store</H4>
<PRE>cc_int32 cc_store(apiCB* cc_ctx, ccache_p* ccache_pointer, const <A HREF="#unionCred_union">cred_union</A> cred)</PRE>
<P>Store (make a copy of) <B>cred</B> in the NC indicated by
<B>ccache_pointer</B>.
<P>A cred_union contains a cred_type indicator and a cred_ptr_union.
A cred_ptr_union can contain either a V4Cred_type pointer or a
cc_creds (V5 creds) pointer. Cred_type indicates which type of
pointer is in the cred_ptr_union. This also allows the <I>API</I> to
enforce the credentials version declared in cc_create() or cc_open().
<P>Possible error codes: CC_NOERROR, CC_NO_EXIST, CC_ERR_CACHE_FULL,
CC_ERR_CRED_VERSION, CC_BAD_PARM
<P><!-- -----------------------------------------------------------------------------------><!-- cc_remove_cred --><!-- -----------------------------------------------------------------------------------><BR>
<A NAME="cc_remove_cred"></A>
<H4>cc_remove_cred</H4>
<PRE>cc_int32 cc_remove_cred(apiCB* cc_ctx, ccache_p* ccache_pointer, const <A HREF="#unionCred_union">cred_union</A> cred)</PRE>
<P>Removes the credential <B>cred</B> from <B>ccache_pointer</B>. The
credentials in the NC indicated by ccache_pointer are searched to
find a matching credential. If found, that credential is removed from
the NC. The cred parameter is not modified and should be freed via
cc_free_creds(). It is legitimate to call this function during a
sequential fetch, and the deletion of a credential already returned
by cc_seq_fetch_creds() should not disturb sequence of credentials
returned by cc_seq_fetch_creds().
<P>Use of cred_union is the same as is explained in cc_store().
<P>Possible error codes: CC_NOERROR, CC_NO_EXIST, CC_NOTFOUND,
CC_ERR_CRED_VERSION, CC_BAD_PARM
<P><!-- -----------------------------------------------------------------------------------><!-- cc_seq_fetch_creds --><!-- -----------------------------------------------------------------------------------><BR>
<A NAME="cc_seq_fetch_creds_begin"></A>
<H4>cc_seq_fetch_creds_begin</H4>
<PRE>cc_int32 cc_seq_fetch_creds_begin(apiCB* cc_ctx, const ccache_p* ccache_pointer, ccache_cit** itCreds)</PRE>
<P>Allocates memory for and initializes *itCreds. This memory must be
deallocated using cc_seq_fetch_creds_end().
<P>Ccache_pointer must be a valid pointer to the NC containing the
creds to be returned by the iterator.
<P>Possible error codes: CC_NOERROR, CC_NO_EXIST, CC_BAD_PARM,
CC_NOMEM
<P>
<P><A NAME="cc_seq_fetch_creds_next"></A>
<H4>cc_seq_fetch_creds_next</H4>
<PRE>cc_int32 cc_seq_fetch_creds_next(apiCB* cc_ctx, cred_union** cred, ccache_cit* itCreds)</PRE>
<P>cc_seq_fetch_creds_next() is used to sequentially read every set
of credentials in an NC. The NC has been indicated in the call to
cc_seq_fetch_creds_begin().
<P>itCreds must be a pointer to a ccache_cit* variable provided by
the calling application and which is used by
cc_seq_fetch_creds_next() to determine the next cached credential to
return. The ccache_cit* variable must be initialized by calling
cc_seq_fetch_creds_begin().
<P>The credentials are filled into the cred_union pointed to by
creds. Note that the cred_union contains elements which are
dynamically allocated, so must be freed using cc_free_creds() between
calls to cc_seq_fetch_creds_next().
<P>If changes are made to the NC while it iterator is being used, it
must return at least the intersection, and at most the union, of the
set of credentials that were in the NC when the iteration began and
the set of credentials that are in the NC when it ends.
<P>When the last credential in the sequence is returned, the return
code from cc_seq_fetch_creds_next() will be CC_END.
<P>Possible error codes: CC_NOERROR, CC_END, CC_NO_EXIST,
CC_BAD_PARM, CC_NOMEM
<P>
<P><A NAME="cc_seq_fetch_creds_end"></A>
<H4>cc_seq_fetch_creds_end</H4>
<PRE>cc_int32 cc_seq_fetch_creds_end(apiCB* cc_ctx, ccache_cit** itCreds)</PRE>
<P>Deallocates memory used by *itCreds and sets *itCreds to NULL.
<P>Possible error codes: CC_NOERROR, CC_NO_EXIST, CC_BAD_PARM
<P><!-- -----------------------------------------------------------------------------------><!-- cc_lock_request --><!-- -----------------------------------------------------------------------------------><BR>
<A NAME="cc_lock_request"></A>
<H4>cc_lock_request</H4>
<PRE>cc_int32 cc_lock_request(apiCB* cc_ctx, const ccache_p* ccache_pointer, cc_int32 lock_type)</PRE>
<DL>
<DD>99/02/11 - smcguire
<DD><B>As of this date there is no locking in the Win NT/95
or Machintosh implementations. The description below may not be
completely accurate as to how this function should be
implemented.</B>
</DL>
<P>This function is currently NOT IMPLEMENTED. All functions attach
to the cache, take action, and detach from the cache before returning
to the caller.
<P>This function will lock or unlock the NC based on the argument
value of lock_type:
<PRE> CC_LOCK_UNLOCK 1 Unlock the NC
CC_LOCK_READER 2 Lock the NC for reading
CC_LOCK_WRITER 3 Lock the NC for writing
CC_LOCK_NOBLOCK 16 Don't block, but return an error code if
the request cannot be satisfied.
</PRE>
<P>Locking is done on a per-thread basis. At most one thread may have
the credentials locked for writing; if so, there must not be any
threads that have the credentials locked for reading.
<P>Multiple threads may have the cache locked for reading, as long as
there is not a writer lock asserted on the cache.
<P>If a thread has a cache locked for reading, that lock may be
upgraded to a writer lock by calling cc_lock_request() with a
lock_type of CC_LOCK_WRITER. If a thread has the cache locked for
reading or writing, a request to cc_lock_request() for a reader or
writer lock, respectively, is a no-op. If a thread does not have the
cache locked, and calls cc_lock_request() with a lock_type of
CC_LOCK_UNLOCK, this is also a no-op.
<P>A request for CC_LOCK_READER and CC_LOCK_WRITER may be made
non-blocking by logical or'ing the value CC_LOCK_NOBLOCK. In that
case, if it is not possible to satisfy the lock request, the error
CC_LOCKED will be returned.
<P>
<P><A NAME="LiberationFunctions"></A>
<H3>Liberation Functions</H3>
<P><!-- -----------------------------------------------------------------------------------><!-- cc_free_principal --><!-- -----------------------------------------------------------------------------------><A NAME="cc_free_principal"></A>
<H4>cc_free_principal</H4>
<PRE>cc_int32 cc_free_principal(apiCB* cc_ctx, char** principal)</PRE>
<P>This function frees the <B>principal</B> returned by
cc_get_principal() and sets *principal to NULL.
<P>Possible error codes: CC_NOERROR, CC_NO_EXIST, CC_BAD_PARM
<P><!-- -----------------------------------------------------------------------------------><!-- cc_free_name --><!-- -----------------------------------------------------------------------------------><BR>
<A NAME="cc_free_name"></A>
<H4>cc_free_name</H4>
<PRE>cc_int32 cc_free_name(apiCB* cc_ctx, char** name)</PRE>
<P>This function frees the <B>name</B> returned by cc_get_name() and
sets *name to NULL.
<P>Possible error codes: CC_NOERROR, CC_NO_EXIST, CC_BAD_PARM
<P><!-- -----------------------------------------------------------------------------------><!-- cc_free_creds --><!-- -----------------------------------------------------------------------------------><BR>
<A NAME="cc_free_creds"></A>
<H4>cc_free_creds</H4>
<PRE>cc_int32 cc_free_creds(apiCB* cc_ctx, cred_union** creds)</PRE>
<P>This function frees all storage associated with creds returned by
cc_seq_fetch_creds() and sets the creds pointer to NULL.
<P>Possible error codes: CC_NOERROR, CC_NO_EXIST, CC_BAD_PARM
<P><!-- -----------------------------------------------------------------------------------><!-- cc_free_NC_info --><!-- -----------------------------------------------------------------------------------><BR>
<A NAME="cc_free_NC_info"></A>
<H4>cc_free_NC_info</H4>
<PRE>cc_int32 cc_free_NC_info(apiCB* cc_ctx, infoNC*** ppNCi)</PRE>
<P>This routine frees all storage aquired by cc_get_NC_info() and
sets ppNCi to NULL.
<P>Possible error codes: CC_NOERROR, CC_NO_EXIST, CC_BAD_PARM
<!-- Begin MIT-use only web reporting counter -->
<IMG SRC="http://counter.mit.edu/tally" WIDTH=1 HEIGHT=1 ALT="">
<!-- End MIT-use only web reporting counter -->
</BODY></HTML>
Copyright 2K16 - 2K18 Indonesian Hacker Rulez
;?>)