1 dept. of cis, temple univ. cis616/661 – principles of data management v. megalooikonomou...

1

Dept. of CIS, Temple Univ.CIS616/661 – Principles of Data Management

V. MegalooikonomouIntegrity Constraints

(based on slides by C. Faloutsos at CMU)

Constraints:

Integrity constraints in the E-R Model:

Key cardinalities of a Relationship

Overview Domain; Ref. Integrity constraints Assertions and Triggers Functional dependencies

Domain Constraints Domain Types, eg, SQL

Fixed Length characters Int; float; (date)

null values eg., create table student( ssn char(9) not

null, ...)

Referential Integrity constraints

‘foreign keys’ - eg:create table takes(

ssn char(9) not null, c-id char(5) not null, grade integer, primary key(ssn, c-id), foreign key ssn references

student, foreign key c-id references class)


…foreign key ssn references student, foreign key c-id references class)

Effect: expects that ssn exists in ‘student’ table blocks ops that violate that - how??

insertion? deletion/update?


…foreign key ssn references student on delete cascade on update cascade,...

-> eliminate all student enrollments other options (set to null, to default

etc)

Weapons for IC: assertions

create assertion <assertion-name> check <predicate>

triggers (~ assertions with ‘teeth’) on operation, if condition, then

action

Triggers - example

define trigger zerograde on update takes

(if new takes.grade < 0 then takes.grade = 0)

Triggers - discussion more complicated: “managers

have higher salaries than their subordinates” - a trigger can automatically boost mgrs salaries

triggers: tricky (infinite loops…)

Overview Domain; Ref. Integrity constraints Assertions and Triggers Functional dependencies

why definition Armstrong’s “axioms” closure and cover

Functional dependencies motivation: ‘good’ tables

takes1 (ssn, c-id, grade, name, address)

‘good’ or ‘bad’?

Functional dependencies

takes1 (ssn, c-id, grade, name, address)

Ssn c-id Grade Name Address

123 413 A smith Main

123 415 B smith Main



‘Bad’ - why?





Functional Dependencies Redundancy

space inconsistencies insertion/deletion anomalies (later…)

What caused the problem?

Functional dependencies ‘name’ depends on the ‘ssn’ define ‘depends’






Definition: ‘a’ functionally determines ‘b’





ba


Informally: ‘if you know ‘a’, there is only one ‘b’ to match’





Functional dependenciesformally:

if two tuples agree on the ‘X’ attribute,they *must* agree on the ‘Y’ attribute, too(eg., if ssn is the same, so should address)

… a functional dependency is a generalization of the notion of a key

])[2][1][2][1( ytytxtxtYX

Functional dependencies ‘X’, ‘Y’ can be sets of attributes other examples??





Functional dependencies ssn -> name, address ssn, c-id -> grade






K is a superkey for relation R iff K -> R

K is a candidate key for relation R iff:K -> Rfor no a K, a -> R


Closure of a set of FD: all implied FDs - eg.:ssn -> name, addressssn, c-id -> grade

implyssn, c-id -> grade, name, addressssn, c-id -> ssn

FDs - Armstrong’s axioms

Closure of a set of FD: all implied FDs - eg.:ssn -> name, addressssn, c-id -> grade

how to find all the implied ones, systematically?


“Armstrong’s axioms” guarantee soundness and completeness:

Reflexivity: eg., ssn, name -> ssn Augmentation

eg., ssn->name then ssn,grade-> ssn,grade

YXXY

YWXWYX

FDs - Armstrong’s axioms Transitivity

ssn->address address-> county-tax-rateTHEN:

ssn-> county-tax-rate

ZXZY

YX


Reflexivity: Augmentation: Transitivity:

ZX

ZY

YX

YXXY

YWXWYX

‘sound’ and ‘complete’

FDs – finding the closure F+

F+ = Frepeat

for each functional dependency f in F+

apply reflexivity and augmentation rules on f add the resulting functional dependencies to F+

for each pair of functional dependencies f1and f2 in F+

if f1 and f2 can be combined using transitivity then add the resulting functional dependency to F+

until F+ does not change any further

We can further simplify manual computation of F+ by using the following additional rules.


Additional rules: Union

Decomposition Pseudo-transitivity

ZXW

ZYW

YX

ZX

YXYZX

YZXZX

YX


Prove ‘Union’ from three axioms:

YZXZX

YX

?


Prove ‘Union’ from three axioms:

YZXtytransitiviand

thusXisXXbut

XZXXXwaugm

YZXZZwaugm

ZX

YX

)4()3(

;

)4(/.)2(

)3(/.)1(

)2(

)1(


Prove Pseudo-transitivity:

ZXW

ZYW

YX

?

ZXZY

YX

YXXY

YWXWYX


Prove Decomposition

ZXZY

YX

YXXY

YWXWYX

ZX

YXYZX

?

FDs - Closure F+

Given a set F of FD (on a schema)F+ is the set of all implied FD. Eg.,takes(ssn, c-id, grade, name,

address)ssn, c-id -> grade

ssn-> name, address}F

FDs - Closure F+

ssn, c-id -> grade ssn-> name, address ssn-> ssn ssn, c-id-> address c-id, address-> c-id ...

F+

FDs - Closure F+ R=(A,B,C,G,H,I)

F= { A->BA->CCG->HCG->IB->H}

Some members of F+:A->HAG->ICG->HI

FDs - Closure A+

Given a set F of FD (on a schema)A+ is the set of all attributes determined

by A:takes(ssn, c-id, grade, name, address)

ssn, c-id -> grade ssn-> name, address

{ssn}+ =??

}F

FDs - Closure A+

takes(ssn, c-id, grade, name, address)ssn, c-id -> grade

ssn-> name, address

{ssn}+ ={ssn, name, address }

}F

FDs - Closure A+


ssn-> name, address

{c-id}+ = ??

}F

FDs - Closure A+


ssn-> name, address

{c-id, ssn}+ = ??

}F

FDs - Closure A+

if A+ = {all attributes of table}then ‘A’ is a candidate key

FDs - Closure A+ Algorithm to compute +, the closure of

under Fresult := ;while (changes to result) do

for each in F do begin

if result then result := result

end

FDs - Closure A+ (example)

R = (A, B, C, G, H, I) F = {A B, A C, CG H, CG I, B H} (AG)+

1.result = AG2.result = ABCG (A C and A B)3.result = ABCGH (CG H and CG AGBC)4.result = ABCGHI (CG I and CG AGBCH)

Is AG a candidate key? 1. Is AG a super key?

1. Does AG R? 2. Is any subset of AG a superkey?

1. Does A+ R?2. Does G+ R?

FDs - A+ closureDiagrams

AB->C (1)A->BC (2)B->C (3)A->B (4)

CA

B

FDs - ‘canonical cover’ Fc

Given a set F of FD (on a schema)Fc is a minimal set of equivalent FD.

Eg.,takes(ssn, c-id, grade, name, address)

ssn, c-id -> grade ssn-> name, address ssn,name-> name, address ssn, c-id-> grade, name

F



F

Fc

FDs - ‘canonical cover’ Fc why do we need it? define it properly compute it efficiently

FDs - ‘canonical cover’ Fc why do we need it?

easier to compute candidate keys define it properly compute it efficiently

FDs - ‘canonical cover’ Fc define it properly - three properties

every FD a->b has no extraneous attributes on the RHS

ditto for the LHS all LHS parts are unique

‘extraneous’ attribute: if the closure is the same, before and

after its elimination or if F-before implies F-after and vice-

versa




F


Algorithm: examine each FD; drop extraneous

LHS or RHS attributes merge FDs with same LHS repeat until no change


Trace algo forAB->C (1)A->BC (2)B->C (3)A->B (4)


Trace algo forAB->C (1)A->BC (2)B->C (3)A->B (4) (4) and (2)

merge:

AB->C (1)A->BC (2)B->C (3)


AB->C (1)A->BC (2)B->C (3)

in (2): ‘C’ is extr.

AB->C (1)A->B (2’)B->C (3)

FDs - ‘canonical cover’ FcAB->C (1)A->B (2’)B->C (3)

in (1): ‘A’ is extr.

B->C (1’)A->B (2’)B->C (3)

FDs - ‘canonical cover’ FcB->C (1’)A->B (2’)B->C (3)

(1’) and (3) merge

A->B (2’)B->C (3)

nothing is extraneous: ‘canonical cover’

FDs - ‘canonical cover’ FcAFTER

A->B (2’)B->C (3)

BEFOREAB->C (1)A->BC (2)B->C (3)A->B (4)

Overview - conclusions Domain; Ref. Integrity constraints Assertions and Triggers Functional dependencies

why definition Armstrong’s “axioms” closure and cover

1 dept. of cis, temple univ. cis616/661 – principles of data management v. megalooikonomou...

Documents