Fujitsu UDS/SQL V2.5 用户手册
General information
Relational model
U929-J-Z125-9-76
33
D
ok
us
c
h
abl
on
en
1
9x
24 V
e
rs
ion
7
.3
us
f
ü
r F
ram
eM
ak
er
V
7
.x
v
o
m
1
4.
02
.2
00
7
©
c
o
gni
tas
Gm
b
H
20
01-
20
07
18
. Ok
tobe
r 20
07
S
ta
nd
1
1:
29
.4
8
P
fa
d:
G
:\
v
og
t\
fs
c
\ud
s
\M
anu
al
e
\e
n\
ud
s
e
nt
_e\
u
ds
ent
.k
0
2
Relationships between tables
In the case of relational data organization, record types are linked by means of certain
record elements with matching contents. A table in a relational schema thus corresponds
to a record type in a CODASYL schema. For example, to link a table with another table by
means of record element contents, the identifying primary key of the first table must also be
included as a (redundant) foreign key in the second table (see
record elements with matching contents. A table in a relational schema thus corresponds
to a record type in a CODASYL schema. For example, to link a table with another table by
means of record element contents, the identifying primary key of the first table must also be
included as a (redundant) foreign key in the second table (see
, DEPARTMENT and
EMPLOYEE tables).
A foreign key must match the values of another table’s primary key. Foreign keys create
logical links between tables. Foreign keys are defined by the users themselves.
logical links between tables. Foreign keys are defined by the users themselves.
Tables are the only fixed data structures in the relational model. Most relational database
systems neither monitor nor maintain the logical relationships between tables. Users create
these relationships by means of values when processing the database, i.e. the tables are
linked logically on the basis of their contents. Users are thus not restricted by structures
prescribed by the system.
systems neither monitor nor maintain the logical relationships between tables. Users create
these relationships by means of values when processing the database, i.e. the tables are
linked logically on the basis of their contents. Users are thus not restricted by structures
prescribed by the system.
Data manipulation and retrieval in relational databases
The relational model defines basic database functions which are analogous to set opera-
tions. Thus, in addition to the principle of processing data in the form of tables, another
major feature characterizing the relational model is a defined set of basic operations which
can be used to retrieve information from relational databases.
The relational model does not define a language for data manipulation and retrieval.
tions. Thus, in addition to the principle of processing data in the form of tables, another
major feature characterizing the relational model is a defined set of basic operations which
can be used to retrieve information from relational databases.
The relational model does not define a language for data manipulation and retrieval.
The main features of data manipulation languages (DML) in relational systems are the set-
oriented operations projection, selection and join and the absence of any predefined struc-
tures. In other words, relational data manipulation languages describe the desired result
itself, instead of describing how that result is to be obtained, i.e. users no longer need to
specify how the desired result is to be ascertained from the database - they need merely
state what they want to see.
oriented operations projection, selection and join and the absence of any predefined struc-
tures. In other words, relational data manipulation languages describe the desired result
itself, instead of describing how that result is to be obtained, i.e. users no longer need to
specify how the desired result is to be ascertained from the database - they need merely
state what they want to see.
The relational model is supported in UDS/SQL through the inclusion of SQL (Structured
Query Language), a language for relational database systems.
Query Language), a language for relational database systems.