Table creation/Postal addresses
You are encouraged to solve this task according to the task description, using any language you may know.
In this task, the goal is to create a table to store addresses. You may assume that all the addresses to be stored will be located in the USA. As such, you will need (in addition to a field holding a unique identifier) a field holding the street address, a field holding the city, a field holding the state code, and a field holding the zipcode. Choose appropriate types for each field.
For non-database languages, show how you would open a connection to a database (your choice of which) and create an address table in it. You should follow the existing models here for how you would structure the table.
ALGOL 68
<lang algol68>MODE ADDRESS = STRUCT( INT page, FLEX[50]CHAR street, FLEX[25]CHAR city, FLEX[2]CHAR state, FLEX[10]CHAR zip ); FORMAT address repr = $"Page: "gl"Street: "gl"City: "gl"State: "gl"Zip: "gll$;
INT errno; FILE sequence; errno := open(sequence, "sequence.txt", stand back channel); SEMA sequence sema := LEVEL 1;
OP NEXTVAL = ([]CHAR table name)INT: (
INT out;
# INT table page = 0; # # only one sequence implemented #
# DOWN sequence sema; # # NO interprocess concurrency protection #
on open error(sequence,
(REF FILE f)BOOL: (
reset(sequence); #set(table page,1,1);#
put(sequence, 0);
try again;
FALSE
)
);
try again:
reset(sequence); #set(table page,1,1);# get(sequence,out);
out +:=1;
reset(sequence); #set(table page,1,1);# put(sequence,out);
# UP sequence sema; #
out
);
OP INIT = (REF ADDRESS self)REF ADDRESS: ( page OF self := NEXTVAL "address"; self);
REF ADDRESS john brown = INIT LOC ADDRESS;
john brown := (page OF john brown, "10 Downing Street","London","England","SW1A 2AA");
printf((address repr, john brown));
FILE address table; errno := open(address table,"address.txt",stand back channel);
- set(address table, page OF john brown,1,1); - standard set page not available in a68g #
put bin(address table, john brown); close(address table)</lang> Output:
Page: +1 Street: 10 Downing Strreet City: London State: England Zip: SW1A 2AA
C
<lang c>#include <stdio.h>
- include <stdlib.h>
- include <sqlite3.h>
const char *code = "CREATE TABLE address (\n" " addrID INTEGER PRIMARY KEY AUTOINCREMENT,\n" " addrStreet TEXT NOT NULL,\n" " addrCity TEXT NOT NULL,\n" " addrState TEXT NOT NULL,\n" " addrZIP TEXT NOT NULL)\n" ;
int main() {
sqlite3 *db = NULL;
char *errmsg;
if ( sqlite3_open("address.db", &db) == SQLITE_OK ) {
if ( sqlite3_exec(db, code, NULL, NULL, &errmsg) != SQLITE_OK ) {
fprintf(stderr, errmsg);
sqlite3_free(errmsg);
sqlite3_close(db);
exit(EXIT_FAILURE);
}
sqlite3_close(db);
} else {
fprintf(stderr, "cannot open db...\n");
sqlite3_close(db);
exit(EXIT_FAILURE);
}
return EXIT_SUCCESS;
}</lang>
DB2 UDB
<lang sql>CREATE TABLE Address ( addrID Integer generated by default as identity, addrStreet Varchar(50) not null, addrCity Varchar(25) not null, addrState Char(2) not null, addrZIP Char(10) not null )</lang>
J
J is a programming language, not a database, but it ships with a database built in the programming language called JDB. Using that, assuming hd is your database, then:
<lang j> Create__hd 'Address';noun define addrID autoid; addrStreet varchar addrCity varchar addrState char addrZip char )</lang>
Of course J can connect external databases too, using e.g. ODBC. See the list of J database topics.
MySQL
<lang mysql>CREATE TABLE `Address` (
`addrID` int(11) NOT NULL auto_increment, `addrStreet` varchar(50) NOT NULL default , `addrCity` varchar(25) NOT NULL default , `addrState` char(2) NOT NULL default , `addrZIP` char(10) NOT NULL default , PRIMARY KEY (`addrID`)
);</lang>
Oracle
<lang sql>CREATE SEQUENCE seq_address_pk START BY 100 INCREMENT BY 1 / CREATE TABLE address (
addrID NUMBER DEFAULT seq_address_pk.nextval, street VARCHAR2( 50 ) NOT NULL, city VARCHAR2( 25 ) NOT NULL, state VARCHAR2( 2 ) NOT NULL, zip VARCHAR2( 20 ) NOT NULL, CONSTRAINT address_pk1 PRIMARY KEY ( addrID )
) /</lang>
Oz
The SQLite version that comes with Ozsqlite does not understand "AUTOINCREMENT". <lang oz>declare
[Sqlite] = {Module.link ['x-ozlib:/sqlite/Sqlite.ozf']}
DB = {Sqlite.open 'test.db'}
in
try
{Sqlite.exec DB
"CREATE TABLE address ("
#"addrID INTEGER PRIMARY KEY,"
#"addrStreet TEXT NOT NULL,"
#"addrCity TEXT NOT NULL,"
#"addrState TEXT NOT NULL,"
#"addrZIP TEXT NOT NULL"
#")" _}
catch E then
{Inspector.configure widgetShowStrings true}
{Inspect E}
finally
{Sqlite.close DB}
end</lang>
Perl
<lang perl>use DBI;
my $db = DBI->connect('DBI:mysql:database:server','login','password');
my $statment = <<EOF; CREATE TABLE `Address` (
`addrID` int(11) NOT NULL auto_increment, `addrStreet` varchar(50) NOT NULL default , `addrCity` varchar(25) NOT NULL default , `addrState` char(2) NOT NULL default , `addrZIP` char(10) NOT NULL default , PRIMARY KEY (`addrID`)
); EOF
my $exec = $db->prepare($statment); $exec->execute;</lang>
This example uses mysql, but DBI supports a extensive list of database drivers. See dbi.perl.org for more info.
PHP+SQLite
not tested <lang php><?php $db = new SQLite3(':memory:'); $db->exec("
CREATE TABLE address (
addrID INTEGER PRIMARY KEY AUTOINCREMENT,
addrStreet TEXT NOT NULL,
addrCity TEXT NOT NULL,
addrState TEXT NOT NULL,
addrZIP TEXT NOT NULL
)
"); ?></lang>
PicoLisp
PicoLisp has built-in database functionality, in the form of (non-relational) entity/relations, built on top of persistent objects (so-called external symbols)
Define an "address" entity, and create the database: <lang PicoLisp>(class +Adr +Entity) (rel nm (+Sn +Idx +String)) # Name [Soundex index] (rel str (+String)) # Street (rel zip (+Ref +String)) # ZIP [Non-unique index] (rel cit (+Fold +Idx +String)) # City [Folded substring index] (rel st (+String)) # State (rel tel (+Fold +Ref +String)) # Phone [Folded non-unique index] (rel em (+Ref +String)) # EMail [Non-unique index] (rel txt (+Blob)) # Memo (rel jpg (+Blob)) # Photo
(pool "address.db") # Create database</lang> Create a first entry, and show it: <lang PicoLisp>(show
(new! '(+Adr) # Create a record
'nm "FSF Inc."
'str "51 Franklin St"
'st "Boston, MA"
'zip "02110-1301" ) )</lang>
Output:
{2} (+Adr)
zip "02110-1301"
st "Boston, MA"
str "51 Franklin St"
nm "FSF Inc."
Interactive "select": <lang PicoLisp>(select nm zip +Adr nm "FSF") # Select name, zip from Adr where name = FSF*</lang> Output:
"FSF Inc." "02110-1301" {2}
PostgreSQL
<lang sql>CREATE SEQUENCE address_seq start 100; CREATE TABLE address (
addrID int4 PRIMARY KEY DEFAULT nextval('address_seq'),
street varchar(50) not null,
city varchar(25) not null,
state varchar(2) not null,
zip varchar(20) not null
);</lang>
PureBasic+SQLite
Easiest approach with sqlite. Further possible: PostgresQL or each other over ODBC. <lang Purebasic> UseSQLiteDatabase() Procedure CheckDatabaseUpdate(Database, Query$)
Result = DatabaseUpdate(Database, Query$)
If Result = 0
Print(DatabaseError())
EndIf
ProcedureReturn Result
EndProcedure openconsole() DatabaseFile$ = GetCurrentDirectory()+"/rosettadb.sdb" If CreateFile(0, DatabaseFile$)
CloseFile(0)
If OpenDatabase(0, DatabaseFile$, "", "")
CheckDatabaseUpdate(0,"CREATE TABLE address ( addrID INTEGER PRIMARY KEY AUTOINCREMENT, addrStreet TEXT Not NULL, addrCity TEXT Not NULL, addrState TEXT Not NULL, addrZIP TEXT Not NULL)")
CloseDatabase(0)
Else
print("Can't open database !")
EndIf
Else
print("Can't create the database file !")
EndIf closeconsole() </lang>
Python+SQLite
<lang python>>>> import sqlite3 >>> conn = sqlite3.connect(':memory:') >>> conn.execute(CREATE TABLE address ( addrID INTEGER PRIMARY KEY AUTOINCREMENT, addrStreet TEXT NOT NULL, addrCity TEXT NOT NULL, addrState TEXT NOT NULL, addrZIP TEXT NOT NULL
))
<sqlite3.Cursor object at 0x013265C0> >>> </lang>
Ruby
With PStore
PStore implements a permanent key store with transactions. This is a NoSQL database. Each transaction reads the entire database into memory, and then writes it again, so PStore is not good for large databases.
<lang ruby>require 'pstore' require 'set'
Address = Struct.new :id, :street, :city, :state, :zip
db = PStore.new("addresses.pstore") db.transaction do
db[:next] ||= 0 # Next available Address#id db[:ids] ||= Set[] # Set of all ids in db
end</lang>
To put an Address inside this PStore:
<lang ruby>db.transaction do
id = (db[:next] += 1)
db[id] = Address.new(id,
"1600 Pennsylvania Avenue NW",
"Washington", "DC", 20500)
db[:ids].add id
end</lang>
With SQLite
<lang ruby>require 'sqlite3'
db = SQLite3::Database.new(':memory:') db.execute("
CREATE TABLE address (
addrID INTEGER PRIMARY KEY AUTOINCREMENT,
addrStreet TEXT NOT NULL,
addrCity TEXT NOT NULL,
addrState TEXT NOT NULL,
addrZIP TEXT NOT NULL
)
")</lang>
SAS
<lang sql>DATA address;
LENGTH addrID 8. street 50$ city 25$ state 2$ zip 20$; STOP;
RUN;</lang>
Scheme
This example works with Chicken Scheme, using its sql-de-lite library:
<lang scheme> (use sql-de-lite)
(define *db* (open-database "addresses"))
(exec ; create and run the SQL statement
(sql *db*
"CREATE TABLE address (
addrID INTEGER PRIMARY KEY AUTOINCREMENT,
addrStreet TEXT NOT NULL,
addrCity TEXT NOT NULL,
addrState TEXT NOT NULL,
addrZIP TEXT NOT NULL
)"
))
(close-database *db*) ; finally, close database </lang>
Tcl+SQLite
<lang tcl>package require sqlite3
sqlite3 db address.db db eval {
CREATE TABLE address (
addrID INTEGER PRIMARY KEY AUTOINCREMENT, addrStreet TEXT NOT NULL, addrCity TEXT NOT NULL, addrState TEXT NOT NULL, addrZIP TEXT NOT NULL
)
}</lang>
Transact-SQL (MSSQL)
<lang sql>CREATE TABLE #Address (
addrID int NOT NULL Identity(1,1) PRIMARY KEY, addrStreet varchar(50) NOT NULL , addrCity varchar(25) NOT NULL , addrState char(2) NOT NULL , addrZIP char(10) NOT NULL
) drop table #Address</lang>