i know we can use SHOW TABLE DATABASENAME.TABLENAME to get the DDL of a certain table in teradata,
like,
show table customerservice.employee;
result:
CREATE SET TABLE customerservice.employee ,FALLBACK ,
NO BEFORE JOURNAL,
NO AFTER JOURNAL,
CHECKSUM = DEFAULT,
DEFAULT MERGEBLOCKRATIO
(
employee_number INTEGER,
manager_employee_number INTEGER,
department_number INTEGER,
job_code INTEGER,
last_name CHAR(20) CHARACTER SET LATIN NOT CASESPECIFIC NOT NULL,
first_name VARCHAR(30) CHARACTER SET LATIN NOT CASESPECIFIC NOT NULL,
hire_date DATE FORMAT 'YY/MM/DD' NOT NULL,
birthdate DATE FORMAT 'YY/MM/DD' NOT NULL,
salary_amount DECIMAL(10,2) NOT NULL)
UNIQUE PRIMARY INDEX ( employee_number );
If i want to show DDL of a certain database, is it possible ?
Related
According to the documentation, PRAGMA schema.index_list(table-name); should list all of the indexes including primary keys.
However, when I use it I only get the secondary indexes.
https://www.sqlite.org/pragma.html#pragma_index_info
What would cause the primary keys to not be listed?
CREATE TABLE Employee
(
EmployeeKey INTEGER PRIMARY KEY,
FirstName nvarChar(25) NOT NULL,
MiddleName nvarChar(25) NULL,
LastName nVarChar(25) NOT NULL,
Title nVarChar(100) null,
EmployeeId nvarChar(50) NOT NULL,
ManagerKey INT NULL REferences Employee(EmployeeKey),
OfficePhone VARCHAR(15) NULL ,
CellPhone VARCHAR(15) NULL ,
CreatedDate DateTime NOT NULL DEFAULT CURRENT_TIME,
UpdatedDate DateTime NULL
)
CREATE UNIQUE INDEX index_name ON Employee(EmployeeId);
EmployeeKey won't be shown as an index as it's an alias of the rowid column.
The rowid column being inherent for all tables that are not defined as WITHOUT ROWID tables.
The rowid column could be considered the MASTER index.
If you were to not specify INTEGER PRIMARY KEY as the PRIMARY KEY, e.g. you specified EmployeeKey TEXT PRIMARY KEY, then an index would be listed as SQLite is quite specific about what constitutes an alias of the rowid (e.g. EmployeeKey INT PRIMARY KEY is not an alias of the rowid, so would have an index).
You may wish to have a look at CREATE TABLE - Rowid.
I am currently getting the following error when trying to add to one of my tables:
Error adding record: FOREIGN KEY constraint failed: (INSERT INTO Stock(StockID,ItemName,MinAmountRequired,AmountInStock,Order? (Yes/No),DataLastUpdated,OrderNumber,SupplierRefrence,PurchaseID`) VALUES (1,",0,0,",",0,0,0);)
Currently, This is how I have my tables set up:
Stock
CREATE TABLE Stock (
StockID INTEGER NOT NULL PRIMARY KEY AUTOINCREMENT UNIQUE,
ItemName TEXT NOT NULL,
MinAmountRequired INTEGER NOT NULL,
AmountInStock INTEGER NOT NULL,
Order? (Yes/No) TEXT NOT NULL,
DataLastUpdated TEXT NOT NULL,
OrderNumber INTEGER NOT NULL UNIQUE,
SupplierReference INTEGER NOT NULL,
PurchaseID INTEGER NOT NULL UNIQUE,
FOREIGN KEY(PurchaseID) REFERENCES Purchase(PurchaseID),
FOREIGN KEY(OrderNumber) REFERENCES Orders(OrderNumber),
FOREIGN KEY(SupplierReference) REFERENCES Supplier(SupplierReference));
Orders
CREATE TABLE Orders (
OrderNumber INTEGER NOT NULL PRIMARY KEY AUTOINCREMENT UNIQUE,
CustomerReferenceNumber INTEGER NOT NULL UNIQUE,
OrderDate TEXT NOT NULL,
ItemName TEXT NOT NULL UNIQUE);
Suppliers
CREATE TABLE Supplier (
SupplierReference INTEGER NOT NULL PRIMARY KEY AUTOINCREMENT UNIQUE,
Name TEXT NOT NULL UNIQUE,
Address TEXT NOT NULL UNIQUE,
ContactNumber INTEGER NOT NULL UNIQUE);
Purchases
CREATE TABLE Purchase (
PurchaseID INTEGER NOT NULL PRIMARY KEY AUTOINCREMENT UNIQUE,
Date TEXT NOT NULL,
AmountSpent REAL NOT NULL);
You have defined these foreign keys in table Stock:
FOREIGN KEY(PurchaseID) REFERENCES Purchase(PurchaseID),
FOREIGN KEY(OrderNumber) REFERENCES Orders(OrderNumber),
FOREIGN KEY(SupplierReference) REFERENCES Supplier(SupplierReference)
meaning that the values in columns PurchaseID, OrderNumber, SupplierReference need to reference values in columns of the tables Purchase, Orders and Supplier.
But you want to store 0 to all of these columns which I'm sure that is not the value of any of the referenced columns, since these referenced columns are defined as
PRIMARY KEY AUTOINCREMENT
and so their values are > 0.
Pass valid values that do exist in these 3 tables and the statement will execute succesfully.
I need to copy one table to another and both tables contain column with AUTOINCREMENT. Is it possible to insert a defined value into AUTOINCREMENT column.
Tables:
CREATE TABLE tmptimetables (
_id INTEGER NOT NULL PRIMARY KEY AUTOINCREMENT,
_title NVARCHAR(256) NOT NULL,
_weeks INTEGER NOT NULL,
_first_week_date INTEGER NOT NULL,
_auto_complete INTEGER NOT NULL,
_first_lesson_time INTEGER NOT NULL,
_lesson_duration INTEGER NOT NULL,
_break_duration INTEGER NOT NULL,
_color INTEGER NOT NULL,
_symbol NCHAR(1) NOT NULL
);
CREATE TABLE timetables (
id INTEGER NOT NULL PRIMARY KEY AUTOINCREMENT,
title NVARCHAR(256) NOT NULL,
weeks INTEGER NOT NULL,
first_week_date INTEGER NOT NULL,
auto_complete INTEGER NOT NULL,
first_lesson_time INTEGER NOT NULL,
lesson_duration INTEGER NOT NULL,
break_duration INTEGER NOT NULL,
color INTEGER NOT NULL,
symbol NCHAR(1) NOT NULL
);
My SQL request:
INSERT INTO timetables (
auto_complete,
break_duration,
color,
first_lesson_time,
first_week_date, id,
lesson_duration,
symbol,
title,
weeks
)
SELECT
_auto_complete,
_break_duration,
_color,
_first_lesson_time,
_first_week_date,
_id,
_lesson_duration,
_symbol,
_title,
_weeks
FROM tmptimetables
AUTOINCREMENT can only be used for a column that is INTEGER PRIMARY KEY, it is INTEGER PRIMARY KEY that is the factor that makes the column a special column whereby if the value is not provided when inserting a row that a unique integer will be assigned.
So be the column INTEGER PRIMARY KEY or INTEGER PRIMARY KEY AUTOINCREMENT you can specify an integer value and a row may be inserted with the given value.
A row will not be inserted with a given value if that value is not unique.
For example if the table timetables is currently :-
Then
INSERT INTO timetables (id,title,weeks,first_week_date,auto_complete,first_lesson_time,lesson_duration,break_duration,color,symbol) VALUES (null,'mytitle',78,86000,23,1800,900,200,16,'E');
Would insert a new row with the id as determined by SQLite's algorithm for providing a unique id (probably 5).
If the id were changed to be provided (i.e. not null) say to 10 as per :-
INSERT INTO timetables (id,title,weeks,first_week_date,auto_complete,first_lesson_time,lesson_duration,break_duration,color,symbol) VALUES (10,'mytitle',78,86000,23,1800,900,200,16,'E');
Then the id for the new row would be 10.
However if then using (the same SQL but with the last column value changed) :-
INSERT INTO timetables (id,title,weeks,first_week_date,auto_complete,first_lesson_time,lesson_duration,break_duration,color,symbol) VALUES (10,'mytitle',78,86000,23,1800,900,200,16,'Z');
A new row would not be inserted as a row with an id of 10 already exists.
Finally if the id is not given (null is used) but the SQL is otherwise the same a new row is inserted with a unique id being provided by SQLite e.g.
INSERT INTO timetables (id,title,weeks,first_week_date,auto_complete,first_lesson_time,lesson_duration,break_duration,color,symbol) VALUES (null,'mytitle',78,86000,23,1800,900,200,16,'Z');
So the end result of following the above is :-
AUTOINCREMENT
The AUTOINCREMENT keyword, only usable for an INTEGER PRIMARY COLUMN, invokes a different algorithm for determining the next sequence to ensure that the next sequence/id is always greater, whilst without AUTOINCREMENT a lower sequence/id can be applied.
The AUTOINCREMENT keyword does not specify that if a value for the column is not provided then a sequence/id is applied it is INTEGER PRIMARY KEY that specifies that. Well actually, by default, i.e. unless WITHOUT ROWID is specified, this happens for all tables. It's just the the special rowid column is hidden. Specifying <column_name> INTEGER PRIMARY KEY (where is a valid column name) creates an alias of the rowid.
For example using SELECT rowid,* FROM timetables produces :-
SQLite Autoincrement
Rowid Tables
I have created one table in oracle data base my table script is
CREATE TABLE wsc_widget_bundle (
id VARCHAR (50),
widgetBundle BLOB NULL,
devicesoftwareclass VARCHAR(30) NOT NULL,
widgetProfileId VARCHAR (50) NOT NULL,
bundleHash BLOB NULL,
widgetLocale VARCHAR (6) NOT NULL ,
status INT,
primary key(id),
unique(widgetProfileId, devicesoftwareclass,status),
foreign key(widgetProfileId) references wsc_widget_profile(id)
);
When i create ddl for that is looks like
create table "DEV1"."WSC_WIDGET_BUNDLE"(
"ID" VARCHAR2(50) not null,
"WIDGETBUNDLE" BLOB,
"DEVICESOFTWARECLASS" VARCHAR2(30) not null,
"WIDGETPROFILEID" VARCHAR2(50) not null,
"BUNDLEHASH" BLOB,
"WIDGETLOCALE" VARCHAR2(6) not null,
"STATUS" NUMBER,
constraint "SYS_C00323290" primary key ("ID")
);
alter table "DEV1"."WSC_WIDGET_BUNDLE"
add constraint "SYS_C00323292"
foreign key ("WIDGETPROFILEID")
references "MTP440_DEV1"."WSC_WIDGET_PROFILE"("ID");
create unique index "MTP440_DEV1"."SYS_C00323290" on "MTP440_DEV1"."WSC_WIDGET_BUNDLE"("ID");
create unique index "MTP440_DEV1"."SYS_C00323291" on "MTP440_DEV1"."WSC_WIDGET_BUNDLE"("WIDGETPROFILEID","DEVICESOFTWARECLASS","STATUS");
create index "MTP440_DEV1"."TEST" on "MTP440_DEV1"."WSC_WIDGET_BUNDLE"("DEVICESOFTWARECLASS","STATUS","WIDGETLOCALE","WIDGETPROFILEID");
Now i want to write alter script to alter unique key constrain of my table but as at creation of table I didn't mention the name of my unique key name it is given by system like SYS_C00323291
So how can I write alter script to drop that unique key whose name is not known to me and generation new one
You can find the name of the constraint by querying the user_constraints and user_cons_columns views.
Alter table x
drop constraint pk;
Alter table x
add constraint New_constraint_name PRIMARY KEY (colname);
I need to convert a table from MySQL to SQLite, but I can't figure out how to convert an enum field, because I can't find ENUM type in SQLite.
The aforementioned field is pType in the following table:
CREATE TABLE `prices` (
`id` INT(11) NOT NULL AUTO_INCREMENT,
`pName` VARCHAR(100) NOT NULL DEFAULT '',
`pType` ENUM('M','R','H') NOT NULL DEFAULT 'M',
`pField` VARCHAR(50) NULL DEFAULT NULL,
`pFieldExt` VARCHAR(50) NULL DEFAULT NULL,
`cmp_id` INT(11) NOT NULL DEFAULT '0',
PRIMARY KEY (`id`)
)
ENGINE=MyISAM
ROW_FORMAT=DEFAULT
I need a field with only three values for the user to chose, and I would like to enforce that in the DB, not just in my application.
SQLite way is to use a CHECK constraint.
Some examples:
CREATE TABLE prices (
id INTEGER PRIMARY KEY,
pName TEXT CHECK( LENGTH(pName) <= 100 ) NOT NULL DEFAULT '',
pType TEXT CHECK( pType IN ('M','R','H') ) NOT NULL DEFAULT 'M',
pField TEXT CHECK( LENGTH(pField) <= 50 ) NULL DEFAULT NULL,
pFieldExt TEXT CHECK( LENGTH(pFieldExt) <= 50 ) NULL DEFAULT NULL,
cmp_id INTEGER NOT NULL DEFAULT '0'
)
This will limit the pType column to just the values M, R, and H, just
like enum("M", "R", "H") would do in some other SQL engines.
There is no enum type in SQLite, only the following:
NULL
INTEGER
REAL
TEXT
BLOB
Source: http://www.sqlite.org/datatype3.html
I'm afraid a small, custom enum table will be required in your case.
To expand on MPelletier’s answer, you can create the tables like so:
CREATE TABLE Price (
PriceId INTEGER PRIMARY KEY AUTOINCREMENT NOT NULL,
Name VARCHAR(100) NOT NULL,
Type CHAR(1) NOT NULL DEFAULT ('M') REFERENCES PriceType(Type)
);
CREATE TABLE PriceType (
Type CHAR(1) PRIMARY KEY NOT NULL,
Seq INTEGER
);
INSERT INTO PriceType(Type, Seq) VALUES ('M',1);
INSERT INTO PriceType(Type, Seq) VALUES ('R',2);
INSERT INTO PriceType(Type, Seq) VALUES ('H',3);
Now the enumeration values are available directly in the Price table as they would be using an ENUM: you don’t need to join to the PriceType table to get the Type values, you only need to use it if you want to determine the sequence of the ENUMs.
Foreign key constraints were introduced in SQLite version 3.6.19.