Database Reference
In-Depth Information
• Step 2: for each attribute in the attributes of tuple t, break its
security level into parts.
• Step 3: for each security level part in parts in step 2, if this
part of the security level is equal to the security level of the
user, it will be asserted by user beliefs.
• Step 4: pack the new attribute class level.
• Step 5: return the attribute class level.
Figure  3.19 illustrates the flow chart for the verify operation in the
belief-consistent model.
3.6.5 Update Operation Procedure
The SQL statement for the update operation is described as follows:
UPDATE R
SET [A 1 = a 1 ,A 2 = a 2 ,...,A n = a n ]
WHERE P
where R is an MLS relation, A 1 ,A 2 ,...,A n are attributes from  R,
a 1 ,a 2 ,...,a n are values from domains of A 1 ,A 2 ,...,A n , and
P is an update condition that defines the tuples that are to be
updated. If a subject with the security class level L runs a command
to update an MLS relation R for all t ∈ R, the update operation is
implemented as follows [35]:
• Step 1: get the security class level of the subject that runs the
update operation.
• Step 2: if Pl (t [tuple class]) is equal to the security class level
of the subject L, tuple t will be updated.
• Step 3: if Pl (t [tuple class]) is lower than the security level of
the user, ib(security level of user, t [tuple class]) = security level
of the user. A new tuple t′ based on tuple t will be inserted on
the level equal to the security level of the user. The attribute
values of tuple t will not be changed.
• Step 4: if Pl (t [tuple class]) is lower than the security level of
the user, Ib (security level of user, t [tuple class]) = 0. If existing
tuples ti i from R have ti i [primary key] = t[primary key] and Pl
(t i [primary key]) = Pl(t[primary key]) and Pl (ti i [tuple class])
< the security level of the user, the user will choose a tuple,
t′, and a new tuple, t″, based on t′ will be inserted on the user
Search WWH ::




Custom Search