The caller of kg_unseal_v1 passes a gss_qop_t * for the qop_state
parameter, so make it use that type instead of an int *.  Noted by
David Benjamin <davidben@mit.edu>.

Add a template-based array constructor for convenient marshalling of
structured values as JSON array values.  Use it to simplify
export_cred.c.

Return error codes (0, ENOMEM, or EINVAL) from JSON support functions
instead of returning results directly.  This makes error handling
simpler for functions which assemble JSON objects and then return a
krb5_error_code values.  Adjust all callers.  Use shims in
export_cred.c to minimize changes there; it will be redesigned
internally in a subsequent commit.

Add k5-int.h static functions to duplicate byte ranges, optionally
with a trailing zero byte, and set an error code like k5alloc does.
Use them where they would shorten existing code.

Mostly this gets rid of the trailing space on line 2 after
bb76891f5386526bdf91bc790c614fc9296cb5fa.

KRB5_CONF_ prefix should be used for the krb5/kdc.conf parameters.
Use KRB5_CC_CONF_ prefix for cache configuration variables.

Since there is no overlap between the clpreauth and kdcpreauth
interface declarations, there's no particular reason to combine them
into one header.  For backward compatibility and convenience, leave
behind a preauth_plugin.h which includes both.

RFC 6680 requires that gss_export_name_composite begin the output
token with 04 02.  So we must produce a composite token even if the
name has no authdata, and be able to consume a composite token with no
authdata attributes.

[ghudson@mit.edu: expanded commit message]

ticket: 7400 (new)

It might have been safe to access the krb5 verifier cred without a
lock before constrained delegation, but it is less likely to be safe
now that we might access both the initiator and acceptor parts of the
cred.  Hold a lock on the cred for the full accept_sec_context
operation.

ticket: 7366 (new)