Primary key and unique key not Null
Hello gurus,Asked in an interview about the difference between primary and unique keys.
I talked Unique keys can accept null values, but they are unique through should have the table. So, the next question asked me was "a Unique key NOT NULL can be treated as the primary key?
Tricky question! :)
I said yes! that it meets the requirement to be a primary key for this particular table, but since this isn't really a primary key, that it can't be covered in the foreign keys.
The interviewer wanted just a Yes or no.
Can someone please put some of this?
Thanks in advance!
Two things.
1. unique can also be referenced as a foreign key. If your statement "but because this isn't really a primary key may not be seized key foreign." is not true.
2. primary key and Unique key are different (for example you can have more than one UK in a table but PK's alone) and to know the difference, read some RDBMS concepts.
I'm curious to know what is the outcome of your interview? ;)
Tags: Database
Similar Questions
-
Validation auto-generated Primary Key not Null
Hello
I have a generated Assistant shape, that has a primary key that is generated automatically by a trigger before insert. In my form, primary key is initially empty and after that the user saves the form, then the primary key generates a number and that it fills in the form. I currently have a non-Null on the form validation, after that the user submitted the form and wants to update. How can I keep the validation non-Null on the form so that the user can write again at the start of the database, with the empty primary key field? Auto Wizard generates this out of the constraints of the table, and I would like to know if it is a validation required?
Thank you
MaryWell, you do NOT have NULL validation on the primary key point and that causes the problem that the user create folder? Started to run or remove the 'never '. I don't think it's necessary that you display as 'read only '.
Kind regards
Hari -
where can I use statement box put Null or not primary key or null in the result?
create the table test_target
(
letter_grade varchar2 (2) null,
grade_point number (3,2) not null,
max_grade number (3).
min_grade number 4
);
create the table test_source
(
letter_grade VARCHAR2 (3),
grade_point number (3,2).
max_grade number (3,2).
min_grade number (3).
created_by date,
Modified_By date
);
with the CBC as
(
Select src.table_name src_table_name, src.column_name src_col_name, src.data_type src_data_type, src.data_length src_data_len, src.data_precision src_data_precision, src.data_scale src_data_scale, src.nullable src_nullable
user_tab_columns CBC
where table_name = 'TEST_SOURCE '.
),
As TGT
(
Select tgt.table_name tgt_table_name, tgt.column_name tgt_col_name, tgt.data_type tgt_data_type, tgt.data_length tgt_data_len, tgt.data_precision tgt_data_precision, tgt.data_scale tgt_data_scale, tgt.nullable tgt_nullable
of tgt user_tab_columns
where table_name = 'TEST_TARGET '.
),
col_details as
(
Select src.src_table_name, nvl (tgt.tgt_table_name, first_value (tgt_table_name) more (order of nulls last tgt_table_name)) tgt_table_name;
SRC.src_col_name, src.src_data_type, src.src_data_len, src.src_data_precision, src.src_data_scale, src.src_nullable,
TGT.tgt_col_name, tgt.tgt_data_type, tgt.tgt_data_len, tgt.tgt_data_precision, tgt.tgt_data_scale, tgt.tgt_nullable
the CBC
outer join left tgt
on)
SRC.src_col_name = tgt.tgt_col_name
)
)
Select *.
de)
Select the case sensitive option
When tgt_data_type! = src_data_type or tgt_data_len! = src_data_len or tgt_data_precision! = src_data_precision or tgt_data_scale! = src_data_scale or tgt_nullable! = src_nullable
then 'alter table ' | tgt_table_name | 'Edit ' | tgt_col_name | ' ' || src_data_type | ' (' ||
-case when src_data_type null ('DATE') then
on the other
case
When src_data_type in ('VARCHAR', 'VARCHAR2')
then nvl (to_char (src_data_len), ' ') | ') '
otherwise decode (nvl (src_data_precision-1),-1, null, nvl (to_char (src_data_precision), ' ') |) ', ' || NVL (to_char (src_data_scale), ' ') | ')')
end
end
When tgt_col_name is null
then 'alter table ' | tgt_table_name | 'Add ' | src_col_name | ' ' || src_data_type |
-case when src_data_type null ('DATE') then
on the other
case
When src_data_type in ('VARCHAR', 'VARCHAR2')
then nvl (to_char (src_data_len), ' ') | ') '
otherwise decode (nvl (src_data_precision-1),-1, null, nvl (to_char (src_data_precision), ' ') |) ', ' || NVL (to_char (src_data_scale), ' ') | ')')
end
end
end alter_statement
of col_details
)
where alter_statement is not null;
result:
ALTER table TEST_TARGET change LETTER_GRADE VARCHAR2 (3) / / if it is null or not null primary key or want to see the result as alter table TEST_TARGET change LETTER_GRADE VARCHAR2 (3) null or not null or primary key
ALTER table TEST_TARGET change the NUMBER of MAX_GRADE (3, 2)
ALTER table TEST_TARGET change the MIN_GRADE NUMBER (3, 0)
ALTER table TEST_TARGET add CREATED_BY
ALTER table TEST_TARGET add MODIFIED_BY
Please try:
drop table test_target purge;
drop table test_source purge;
create the table test_target
(
letter_grade varchar2 (2) primary key,.
grade_point number (3,2) not null,
max_grade number (3) unique.
min_grade number 4,
MIN_AGE number 4
);
create the table test_source
(
letter_grade VARCHAR2 (3),
grade_point number (3,2).
max_grade number (3,2).
min_grade number (3).
created_by date,
Modified_By date
);
with the CBC as
(
Select src.table_name src_table_name, src.column_name src_col_name, src.data_type src_data_type, src.data_length src_data_len, src.data_precision src_data_precision, src.data_scale src_data_scale,
CBC. Nullable src_nullable, decode (T.Constraint_Type, 'P', 'Primary Key', 'U', 'Unique', ") as src_cons
user_tab_columns CBC
left join (select Cc.Column_Name, Uc.Constraint_Type
of user_cons_columns cc, uc user_constraints
where Cc.Constraint_Name = Uc.Constraint_Name
and Cc.Table_Name = Uc.Table_Name) t
on T.Column_Name = Src.Column_Name
where table_name = 'TEST_SOURCE '.
),
As TGT
(
Select tgt.table_name tgt_table_name, tgt.column_name tgt_col_name, tgt.data_type tgt_data_type, tgt.data_length tgt_data_len,
TGT.data_precision tgt_data_precision, tgt.data_scale tgt_data_scale, tgt.nullable tgt_nullable,
Decode (T.Constraint_Type, 'P', 'Primary Key', 'U', 'Unique', ") as tgt_cons
of tgt user_tab_columns
left join (select Cc.Column_Name, Uc.Constraint_Type
of user_cons_columns cc, uc user_constraints
where Cc.Constraint_Name = Uc.Constraint_Name
and Cc.Table_Name = Uc.Table_Name) t
on T.Column_Name = TGT. Column_Name
where table_name = 'TEST_TARGET '.
),
col_details as
(
Select src.src_table_name, nvl (tgt.tgt_table_name, first_value (tgt_table_name) more (order of nulls last tgt_table_name)) tgt_table_name;
SRC.src_col_name, src.src_data_type, src.src_data_len, src.src_data_precision, src.src_data_scale, src.src_nullable, src_cons,
TGT.tgt_col_name, tgt.tgt_data_type, tgt.tgt_data_len, tgt.tgt_data_precision, tgt.tgt_data_scale, tgt.tgt_nullable, tgt_cons
the CBC
outer join full tgt
on)
SRC.src_col_name = tgt.tgt_col_name
)
)
Select *.
de)
Select the case sensitive option
When tgt_data_type! = src_data_type or tgt_data_len! = src_data_len or tgt_data_precision! = src_data_precision or tgt_data_scale! = src_data_scale or tgt_nullable! = src_nullable
then 'alter table ' | tgt_table_name | 'Edit ' | tgt_col_name | ' ' || src_data_type | ' (' ||
-case when src_data_type null ('DATE') then
on the other
case
When src_data_type in ('VARCHAR', 'VARCHAR2')
then nvl (to_char (src_data_len), ' ') | ') '
otherwise decode (nvl (src_data_precision-1),-1, null, nvl (to_char (src_data_precision), ' ') |) ', ' || NVL (to_char (src_data_scale), ' ') | ')')
end
end
||
cases where tgt_nullable = 'Y' then 'null '.
of another end 'not null '.
|| tgt_cons
When tgt_col_name is null
then 'alter table ' | tgt_table_name | 'Add ' | src_col_name | ' ' || src_data_type |
-case when src_data_type null ('DATE') then
on the other
case
When src_data_type in ('VARCHAR', 'VARCHAR2')
then nvl (to_char (src_data_len), ' ') | ') '
otherwise decode (nvl (src_data_precision-1),-1, null, nvl (to_char (src_data_precision), ' ') |) ', ' || NVL (to_char (src_data_scale), ' ') | ')')
end
end
|| tgt_cons
When src_col_name is null
then 'alter table' | tgt_table_name: ' drop '. tgt_col_name
end alter_statement
of col_details
)
where alter_statement is not null;
priamry key and unique key you choose user_contraints
Check and change for your condition
Concerning
Mr. Mahir Quluzade
-
Automatic numbering on a primary key (not in order)
Hello
I was wondering how an AutoNumber primary keys without using the sequence when 'on' insert.
Right now I use trigger:
the trigger above works when I'm inserting line-by-line. But now, I tried to insert 300 rows at a time. Here the insertion fails... I think he should there be commit after each insertion (what I think).TRIGGER "scott"."do_numbers" BEFORE INSERT ON "scott"."test" REFERENCING NEW AS NEW FOR EACH ROW declare pragma autonomous_transaction; begin SELECT nvl(MAX(ID),0)+1 INTO :NEW.ID FROM test; commit; END;
Can someone explain to me how can this "automatic" numbering can be made when inserting large number or rows of one table to another at the same time?
Thank you!user13071990 wrote:
HelloI know how to do this with sequence I ask just how it can be done without the help of a.
You would not do otherwise than to use a sequence, otherwise you will have problems in a multiuser environment.
The closest thing would use an update statement in your trigger code...
UPDATE test SET id = NVL(id,0)+1 RETURNING id INTO :new.id;Although this is not ideal and you must use sequences.
And you most certainly not use autonomous transactions within a trigger. It is simply not true, especially when you are trying to manipulate the data on the table of the triggers.
-
No column primary key not defined. error
Hi all
First of all, thanks to DWFAQ for advice a tutorials.
Secondly, I built a contact form the wizard of registration form of update to MEASUREMENT - http://dwfaq.info/home.php?id=4
I hit a problem, when you run the form I get the following error:
Error:
Internal error.
Details of Developer:
tNG_update.prepareSQL:
No primary key column is defined. (UPD_NO_PK_SET)
tNG backtrace - REVIEWS
* tNG_update.executeTransaction
o STARTER. Trigger_Default_Starter
o tNG_update.doTransaction
+ BEFORE. Trigger_Default_saveData
# tNG_update.saveData
+ BEFORE. Trigger_Default_FormValidation
+ tNG_update.prepareSQL*
o tNG_update.getRecordset
o tNG_update.getFakeRsArr
o tNG_update.getLocalRecordset
o tNG_update.getFakeRecordset
Could someone please help - I'm so closely that it made wrong ;-)
Thank you
NJ
Heya,
I received your email and took a look at your script. It seems that the question was in the transaction of PrimaryKey. I sent the file to you with some modifications, where he says:
$upd_contact-> setPrimaryKey ("id", "NUMERIC_TYPE", "GET", "id");
I changed to
$upd_contact-> setPrimaryKey ("id", "NUMERIC_TYPE", "VALUE", "id");
There are other things that I've cleaned up a bit, but I think that's what was causing your error.
Hope that helps!
-
What is the difference between primary key and unique indexes with forced not null?
Primary key is = unique index + not null?
The short answer is Yes.
However, even if the primary key, applying both uniquness and not null, there is a notion of "special".
You can only have one primary key in tables, but you can have multiple unique indexes and constraints not null.
See: https://asktom.oracle.com/pls/asktom/f?p=100:11:0:P11_QUESTION_ID:8743855576462
-
How to refer to the primary key column of newly inserted rows of tabular form
Hello
I use APEX 4.2.0.00.27 with Oracle DB 11.2.0.3.0.
I work with a tabular presentation wizard-created for insert and update a table using the integrated SRM process (sequence 10). I'm trying to use a process of anonymous block of PL/SQL (sequence 30) to make another manipulation of table after the records were inserted or updated. The manual process is associated with my tabular form and I use the variables of name of column binding in my program block.
My (rsn_test) table has 3 columns: test_id (number), test_nbr (number), test_id2 (number). Test_id column is identified as the primary key and the type of the source already exists a sequence rsn_test_seq. Column test_id2 gets its default value 0 to a hidden page element.
I would use my manual process for updating the value of the test_id2 column. If it's 0 then I want to put the value of the column test_id. If it is any other value, then it must remain at this value. My logic works very well for an existing line, but I'm running into a problem with the newly added lines. The new lines get inserted, but the test_id2 column remains the default value 0. I can tell the debugger that the SRM process is triggered first and inserts the line, then my manual dealing with fires. The problem seems to be that the connection variable: TEST_ID for the primary key column remains NULL after insertion. I don't know how to get the value of the column test_id of my newly created line to use in my PL/SQL block to my update.
Process of PL/SQL:
DECLARE BEGIN :P7_SHOW := NULL; :P7_SHOW := NVL(:TEST_ID2,555) || ' and ' || NVL(:TEST_ID,787) || ' and ' || NVL(:TEST_NBR,9999); IF :TEST_ID2 = 0 AND :TEST_ID IS NOT NULL THEN UPDATE rsn_test SET test_id2 = :TEST_ID WHERE test_id = :TEST_ID; ELSE :TEST_ID2 := :TEST_ID2; END IF; END;
Excerpt from the debugger:
0.01625 0.00010 Processes - point: ON_SUBMIT_BEFORE_COMPUTATION 0.01635 0.00008 Branch point: Before Computation 0.01643 0.00003 Process point: AFTER_SUBMIT 0.01646 0.00022 Tabs: Perform Branching for Tab Requests 0.01668 0.00008 Branch point: Before Validation 0.01676 0.00024 Validations: 0.01700 0.00135 Perform basic and predefined validations: 0.01835 0.00020 Perform custom validations: 0.01855 0.00049 ...Validation "TEST_NBR must be numeric" - Type: ITEM_IS_NUMERIC 0.01904 0.00007 ......Skip for row 1 because row hasn't changed 0.01911 0.00016 ......Skip for row 2 because row hasn't changed 0.01927 0.00012 ...Validation "TEST_ID2 must be numeric" - Type: ITEM_IS_NUMERIC 0.01939 0.00007 ......Skip for row 1 because row hasn't changed 0.01945 0.00018 ......Skip for row 2 because row hasn't changed 0.01964 0.00005 Branch point: Before Processing 0.01968 0.00004 Processes - point: AFTER_SUBMIT 0.01972 0.00588 ...Process "ApplyMRU" - Type: MULTI_ROW_UPDATE 0.02560 0.00154 ...Execute Statement: declare function x return varchar2 is begin begin for c1 in ( select "RSN_TEST_SEQ".nextval pk from sys.dual ) loop return c1.pk; end loop; end; return null; end; begin wwv_flow.g_value := x; end; 0.02714 0.00140 ......Row 3: insert into "APPPCSRSN"."RSN_TEST" ( "TEST_ID", "TEST_NBR", "TEST_ID2") values ( :b1, :b2, :b3) 0.02854 0.00011 ...Process "ApplyMRD" - Type: MULTI_ROW_DELETE 0.02865 0.00004 ......Skip because condition or authorization evaluates to FALSE 0.02869 0.00015 ...Process "Process Submit" - Type: PLSQL 0.02884 0.00007 ......Skip for row 1 because row hasn't changed 0.02891 0.00012 ......Skip for row 2 because row hasn't changed 0.02903 0.00012 ......Process row 3 0.02915 0.00429 ...Execute Statement: begin DECLARE BEGIN :P7_SHOW := NULL; :P7_SHOW := NVL(:TEST_ID2,555) || ' and ' || NVL(:TEST_ID,787) || ' and ' || NVL(:TEST_NBR,9999); IF :TEST_ID2 = 0 AND :TEST_ID IS NOT NULL THEN UPDATE rsn_test SET test_id2 = :TEST_NBR WHERE test_id = :TEST_ID; ELSE :TEST_ID2 := :TEST_ID2; END IF; END; end; 0.03344 0.00013 ...Session State: Saved Item "P7_SHOW" New Value="0 and 787 and 1300" 0.03356 0.00004 Branch point: After Processing 0.03360 0.00048 ...Evaluating Branch: "AFTER_PROCESSING" Type: REDIRECT_URL Button: (No Button Pressed) Condition: (Unconditional) 0.03407 0.00013 Redirecting to f?p=290:7:8717971109610:::::&success_msg=0%20row(s)%20updated%2C%201%20row(s)%20inserted.Success%2FEBD244168556408CBA714E3974918C09%2F 0.03420 0.00012 Stop APEX Engine detected 0.03432 0.00007 Stop APEX Engine detected 0.03439 - Final commit
Any suggestions?
I have run tests on
https://apex.Oracle.com/pls/apex/f?p=83488:1 demo/demo
to see your problem.
I have 2 solution for your problem.
I add trial NOT tabular just usual block of PL/SQLBEGIN I'm IN (SELECT TEST_ID FROM RSN_TEST WHERE TEST_ID2 = 0) LOOP UPDATE RSN_TEST SET test_id2 = TEST_ID WHERE test_id = i.TEST_ID; END LOOP; END; and works very well, you can see in the sample.
The other solution is to show new generated TEST_ID
Adding a sequence as a default value for a column in a table field
And to execute your procedure.
I get how is with the good luck of time.
By
-
How to create a validation "not null" unique for all elements in a page?
Hello world
How to create a validation "not null" unique for all elements in a page? I love textfields. Rather than create "not null" validation for each article, I would like to create a single validation control that will be used
Thanks and greetings
UmerNice1 wrote:
Bob, like u said I did the following:(1) less to create a button, there are 9 elements, and for each item, I put in the light of 'yes '.
(2) under the button Delete, there is 1 point and have in light of the 'Yes' for the item
(3) validation of the page defined for the 9 items under 'create a button' and put it to fire when 'create' button clicked
(4) the page validation set for 1 number under 'remove' button and put it on fire when "delete" button clickednow, when I click on the 'create' button it shows same for the item under the "Delete" button that's a required element
Sorry, I do not see this note. The required model will not work, there is no way to attach it to the button.
The best solution is that the answer only a few answers to the top
Create 2 validations page type as a PL/SQL with code
1st validation :P1_ITEM1 IS NOT NULL and :P1_ITEM2 IS NOT NULL ...... and :P1_ITEM9 IS NOT NULL include all 9 items Set the When Button Pressed to the CREATE button 2nd validation :P1_ITEM10 IS NOT NULL Set the When Button Pressed to the DELETE buttonI think that this will be the best way to do it.
Published by: Bob37 on April 27, 2012 12:02
-
Constraint unique but allows null values
Hello
I have a table in Oracle XE (10g) with four matching columns:
PrimaryKey
Book_FKEY
Subject_FKEY
Author_FKEY
The rules for this table are:
A line can contain Book_FKEY and Subject_FKEY (Author_FKEY is Null)
A line can contain Book_FKEY and Author_FKEY (Subject_FKEY is Null)
A line cannot contain only Book_FKEY, but a line must ALWAYS be Book_FKEY
A line cannot contain, Book_FKEY and Subject_FKEY and Author_FKEY
I tried to use the Unique key constraint in Oracle SQL Developer (v2.1.1) but it failed.
Can you help me please?
Thanks a lot for your help.I would use only triggers after exhausting all other possibilities of constraint. You already have the constraints of primary key and foreign key in place, which is good. Technically, I believe that what you need is an extra check constraint to accomplish what you want to do, I think your best bet is to add a Unique combination of Not Null constraints and check on this Board.
First of all, Book_FKEY must always have a value, so make sure not Null column, if not already...
alter tablemodify Book_FKEY not null; Second, I guess that your primary key is a substitute (number generated automatically), then you'll want a unique constraint to enforce a single author of book/object instance...
alter tableadd constraint unique (Book_FKEY, Subject_FKEY, Author_FKEY); And finally, you will need a check constraint to apply other rules...
alter tableadd constraint check ( (Book_FKEY is not null and Subject_FKEY is not null and Author_FKEY is null) or (Book_FKEY is not null and Subject_FKEY is null and Author_FKEY is not null) ) I hope it works for you,
Mark -
Validate the file browse point is not null?
Hello, using APEX 5.0.1. I have a process that needs to run in case a file browse is not null - validation of the PL/SQL Expression. * The file storage type is: BLOB column specified in the attribute of the Source element. for example: P14_STATUS_ID_CURRENT_VALUE = 0 AND: P14_IS_CANCELLED = ' only AND: P14_FILEBROWSE_ITEM IS NOT NULL, even if I download the file, the validation fails! ??? Then I tried this validation, but no result, select FILENAME from wwv_flow_files where name =: p$ _fname and: P14_STATUS_ID_CURRENT_VALUE = 0 AND: P14_IS_CANCELLED = 'n') is this a bug? is there a work around?
Solved: It should be: dbms_lob.getlength(:P14_FILEBROWSE_ITEM) > 0
-
Choose the value is not null in my database using report 6i
Hello world
I have 2 fields in my database and I want to put in my report that the data that is not empty.
I have the following code example, but I don't know where to put in my report and how to use it:
Can someone help me please?SELECT column1 ,column2 ,DECODE (GRADE_Num,NULL, GRADE_Letter,GRADE_NUM) AS grade FROM ....Hello..
Depends on what you want to do...
What is your main query, your column of formulas or perhaps within a function, or a procedure... .so depends on your needs place you only
And the decode line will always search results if this value is null. .because you tell if grade_num is null then show the note by letter... otherwise (way grade_num is not null), you will see the grade_num...
This can be achieved using the NVL function also...
instead of decoding you could write NVL (grade_name, grade_letter)...
NVL function works as follows: If the first argument is null then subititute as the second argument.If your goal is to extract the lines where a certain column is not NULL only, you can then you can also say something like that...
SELECT column1, Column2
From table_name
where (Column1 is not null) AND (column2 is not null); -
How to know the history have not null value
Hi all
I want to find records that have prior LOG_REVIEW_STAGE = "HHH" have (DT_LOG_BEGIN is not null and DT_LOG_END IS NULL and LOG_STATUS = pending)
based on the identification number
Thanks in advance
For Example my Ouput should be ID SORT_ORDER LOG_STATUS LOG_REVIEW_STAGE DT_LOG_BEGIN DT_LOG_END ----------------------------------------------------------------------------------------------- 20 700 Pending FFF 1/26/2004ID SORT_ORDER LOG_STATUS LOG_REVIEW_STAGE DT_LOG_BEGIN DT_LOG_END ----------------------------------------------------------------------------------------------- 10 100 Complete AAA 1/13/2004 1/13/2004 10 200 Complete BBB 1/23/2004 1/23/2004 10 300 Pending CCC 1/23/2004 10 400 DDD 10 601 EEE 10 700 FFF 10 800 GGG 10 900 HHH ---------------------->>>>>>>>> 10 1000 JJJ 10 1100 KKK 20 100 Complete AAA 1/13/2004 1/13/2004 20 200 Complete BBB 1/23/2004 1/23/2004 20 300 Complete CCC 1/23/2004 1/23/2004 20 400 Complete DDD 1/24/2004 1/24/2004 20 601 Complete EEE 1/25/2004 1/25/2004 20 700 Pending FFF 1/26/2004 20 900 HHH ---------------------->>>>>>>>> 20 1000 JJJ 20 1100 KKKcreate table TEMP_TABLE ( id NUMBER(2), sort_order NUMBER(10), log_status VARCHAR2(50), log_review_stage VARCHAR2(50), dt_log_begin DATE, dt_log_end DATE ) ; insert into TEMP_TABLE(id, sort_order, log_status, log_review_stage, dt_log_begin, dt_log_end) values (10, 100, 'Complete', 'AAA', to_date('13-01-2004', 'dd-mm-yyyy'), to_date('13-01-2004', 'dd-mm-yyyy')); insert into TEMP_TABLE(id, sort_order, log_status, log_review_stage, dt_log_begin, dt_log_end) values (10, 200, 'Complete', 'BBB', to_date('23-01-2004', 'dd-mm-yyyy'), to_date('23-01-2004', 'dd-mm-yyyy')); insert into TEMP_TABLE(id, sort_order, log_status, log_review_stage, dt_log_begin, dt_log_end) values (10, 300, 'Pending', 'CCC', to_date('23-01-2004', 'dd-mm-yyyy'), null); insert into TEMP_TABLE(id, sort_order, log_status, log_review_stage, dt_log_begin, dt_log_end) values (10, 400, null, 'DDD', null, null); insert into TEMP_TABLE(id, sort_order, log_status, log_review_stage, dt_log_begin, dt_log_end) values (10, 601, null, 'EEE', null, null); insert into TEMP_TABLE(id, sort_order, log_status, log_review_stage, dt_log_begin, dt_log_end) values (10, 700, null, 'FFF', null, null); insert into TEMP_TABLE(id, sort_order, log_status, log_review_stage, dt_log_begin, dt_log_end) values (10, 800, null, 'GGG', null, null); insert into TEMP_TABLE(id, sort_order, log_status, log_review_stage, dt_log_begin, dt_log_end) values (10, 900, null, 'HHH', null, null); insert into TEMP_TABLE(id, sort_order, log_status, log_review_stage, dt_log_begin, dt_log_end) values (10, 1000, null, 'JJJ', null, null); insert into TEMP_TABLE(id, sort_order, log_status, log_review_stage, dt_log_begin, dt_log_end) values (10, 1100, null, 'KKK', null, null); insert into TEMP_TABLE(id, sort_order, log_status, log_review_stage, dt_log_begin, dt_log_end) values (20, 100, 'Complete', 'AAA', to_date('13-01-2004', 'dd-mm-yyyy'), to_date('13-01-2004', 'dd-mm-yyyy')); insert into TEMP_TABLE(id, sort_order, log_status, log_review_stage, dt_log_begin, dt_log_end) values (20, 200, 'Complete', 'BBB', to_date('23-01-2004', 'dd-mm-yyyy'), to_date('23-01-2004', 'dd-mm-yyyy')); insert into TEMP_TABLE(id, sort_order, log_status, log_review_stage, dt_log_begin, dt_log_end) values (20, 300, 'Complete', 'CCC', to_date('23-01-2004', 'dd-mm-yyyy'), to_date('23-01-2004', 'dd-mm-yyyy')); insert into TEMP_TABLE(id, sort_order, log_status, log_review_stage, dt_log_begin, dt_log_end) values (20, 400, 'Complete', 'DDD', to_date('24-01-2004', 'dd-mm-yyyy'), to_date('24-01-2004', 'dd-mm-yyyy')); insert into TEMP_TABLE(id, sort_order, log_status, log_review_stage, dt_log_begin, dt_log_end) values (20, 601, 'Complete', 'EEE', to_date('25-01-2004', 'dd-mm-yyyy'), to_date('25-01-2004', 'dd-mm-yyyy')); insert into TEMP_TABLE(id, sort_order, log_status, log_review_stage, dt_log_begin, dt_log_end) values (20, 700, 'Complete', 'FFF', to_date('26-01-2004', 'dd-mm-yyyy'), to_date('26-01-2004', 'dd-mm-yyyy')); insert into TEMP_TABLE(id, sort_order, log_status, log_review_stage, dt_log_begin, dt_log_end) values (20, 800, 'Pending', 'GGG', to_date('27-01-2004', 'dd-mm-yyyy'), null); insert into TEMP_TABLE(id, sort_order, log_status, log_review_stage, dt_log_begin, dt_log_end) values (20, 900, null, 'HHH', null, null); insert into TEMP_TABLE(id, sort_order, log_status, log_review_stage, dt_log_begin, dt_log_end) values (20, 1000, null, 'JJJ', null, null); insert into TEMP_TABLE(id, sort_order, log_status, log_review_stage, dt_log_begin, dt_log_end) values (20, 1100, null, 'KKK', null, null); commit;Please provide examples of data, as well as an explanation of your logic. It is very useful!
I think that's what you want:
SELECT id , sort_order , log_status , log_review_stage , dt_log_begin , dt_log_end FROM ( SELECT id , sort_order , log_status , log_review_stage , dt_log_begin , dt_log_end , LEAD(log_review_stage) OVER (PARTITION BY id ORDER BY sort_order) AS next_log_review_stage FROM temp_table ) WHERE log_status = 'Pending' AND dt_log_begin IS NOT NULL AND dt_log_end IS NULL AND next_log_review_stage = 'HHH' ;When run on your dataset, it returns:
ID SORT_ORDER LOG_STATUS LOG_REVIEW_STAGE DT_LOG_BEGIN DT_LOG_END -------------------- -------------------- --------------- -------------------------------------------------- ------------------- ------------------- 20 800 Pending GGG 01/27/2004 00:00:00If this is incorrect, please explain why.
Thank you!
-
Difference of path between primary key and a Unique Index
Hi all
Is there a specific way the oracle optimizer to treat differently the Primary key and Unique index?
Oracle Version
Sample data test for Index NormalSQL> select * from v$version; BANNER -------------------------------------------------------------------------------- Oracle Database 11g Enterprise Edition Release 11.2.0.3.0 - 64bit Production PL/SQL Release 11.2.0.3.0 - Production CORE 11.2.0.3.0 Production TNS for IBM/AIX RISC System/6000: Version 11.2.0.3.0 - Production NLSRTL Version 11.2.0.3.0 - Production SQL>
Examples of test when using primary key dataSQL> create table t_test_tab(col1 number, col2 number, col3 varchar2(12)); Table created. SQL> create sequence seq_t_test_tab start with 1 increment by 1 ; Sequence created. SQL> insert into t_test_tab select seq_t_test_tab.nextval, round(dbms_random.value(1,999)) , 'B'||round(dbms_random.value(1,50))||'A' from dual connect by level < 100000; 99999 rows created. SQL> commit; Commit complete. SQL> exec dbms_stats.gather_table_stats(USER_OWNER','T_TEST_TAB',cascade => true); PL/SQL procedure successfully completed. SQL> select col1 from t_test_tab; 99999 rows selected. Execution Plan ---------------------------------------------------------- Plan hash value: 1565504962 -------------------------------------------------------------------------------- | Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time | -------------------------------------------------------------------------------- | 0 | SELECT STATEMENT | | 99999 | 488K| 74 (3)| 00:00:01 | | 1 | TABLE ACCESS FULL| T_TEST_TAB | 99999 | 488K| 74 (3)| 00:00:01 | -------------------------------------------------------------------------------- Statistics ---------------------------------------------------------- 1 recursive calls 0 db block gets 6915 consistent gets 259 physical reads 0 redo size 1829388 bytes sent via SQL*Net to client 73850 bytes received via SQL*Net from client 6668 SQL*Net roundtrips to/from client 0 sorts (memory) 0 sorts (disk) 99999 rows processed SQL> create index idx_t_test_tab on t_test_tab(col1); Index created. SQL> exec dbms_stats.gather_table_stats('USER_OWNER','T_TEST_TAB',cascade => true); PL/SQL procedure successfully completed. SQL> select col1 from t_test_tab; 99999 rows selected. Execution Plan ---------------------------------------------------------- Plan hash value: 1565504962 -------------------------------------------------------------------------------- | Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time | -------------------------------------------------------------------------------- | 0 | SELECT STATEMENT | | 99999 | 488K| 74 (3)| 00:00:01 | | 1 | TABLE ACCESS FULL| T_TEST_TAB | 99999 | 488K| 74 (3)| 00:00:01 | -------------------------------------------------------------------------------- Statistics ---------------------------------------------------------- 1 recursive calls 0 db block gets 6915 consistent gets 0 physical reads 0 redo size 1829388 bytes sent via SQL*Net to client 73850 bytes received via SQL*Net from client 6668 SQL*Net roundtrips to/from client 0 sorts (memory) 0 sorts (disk) 99999 rows processed SQL>
If you see here the same even as the statistics were gathered,SQL> create table t_test_tab1(col1 number, col2 number, col3 varchar2(12)); Table created. SQL> create sequence seq_t_test_tab1 start with 1 increment by 1 ; Sequence created. SQL> insert into t_test_tab1 select seq_t_test_tab1.nextval, round(dbms_random.value(1,999)) , 'B'||round(dbms_random.value(1,50))||'A' from dual connect by level < 100000; 99999 rows created. SQL> commit; Commit complete. SQL> exec dbms_stats.gather_table_stats('USER_OWNER','T_TEST_TAB1',cascade => true); PL/SQL procedure successfully completed. SQL> select col1 from t_test_tab1; 99999 rows selected. Execution Plan ---------------------------------------------------------- Plan hash value: 1727568366 --------------------------------------------------------------------------------- | Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time | --------------------------------------------------------------------------------- | 0 | SELECT STATEMENT | | 99999 | 488K| 74 (3)| 00:00:01 | | 1 | TABLE ACCESS FULL| T_TEST_TAB1 | 99999 | 488K| 74 (3)| 00:00:01 | --------------------------------------------------------------------------------- Statistics ---------------------------------------------------------- 1 recursive calls 0 db block gets 6915 consistent gets 0 physical reads 0 redo size 1829388 bytes sent via SQL*Net to client 73850 bytes received via SQL*Net from client 6668 SQL*Net roundtrips to/from client 0 sorts (memory) 0 sorts (disk) 99999 rows processed SQL> alter table t_test_tab1 add constraint pk_t_test_tab1 primary key (col1); Table altered. SQL> exec dbms_stats.gather_table_stats('USER_OWNER','T_TEST_TAB1',cascade => true); PL/SQL procedure successfully completed. SQL> select col1 from t_test_tab1; 99999 rows selected. Execution Plan ---------------------------------------------------------- Plan hash value: 2995826579 --------------------------------------------------------------------------------------- | Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time | --------------------------------------------------------------------------------------- | 0 | SELECT STATEMENT | | 99999 | 488K| 59 (2)| 00:00:01 | | 1 | INDEX FAST FULL SCAN| PK_T_TEST_TAB1 | 99999 | 488K| 59 (2)| 00:00:01 | --------------------------------------------------------------------------------------- Statistics ---------------------------------------------------------- 1 recursive calls 0 db block gets 6867 consistent gets 0 physical reads 0 redo size 1829388 bytes sent via SQL*Net to client 73850 bytes received via SQL*Net from client 6668 SQL*Net roundtrips to/from client 0 sorts (memory) 0 sorts (disk) 99999 rows processed SQL>
* In the 1st table T_TEST_TAB, table always use FULL table access after creating indexes.
* And in the 2nd table T_TEST_TAB1, table uses PRIMARY KEY as expected.
Any comments?
Kind regards
BPat>
* In the 1st table T_TEST_TAB, table always use FULL table access after creating indexes.
* And in the 2nd table T_TEST_TAB1, table uses PRIMARY KEY as expected.
>
Yes - for the first table a full table scan will be used as the currently selected column is nullable and indexes do not include null values.The index can be used for the second query, since all the data (first column) is available between the index and there may be no NULL values because of the primary key. If you check constraints, you find that the there is now a CHECK constraint to ensure that the first column cannot be null.
For a full and interesting discussion see the explanation of this and a related issue on the question I ask in this thread
What SYS tables (not seen) contain the value NULL spec /not/ column definition? and my response he posted: 23 April 2012 09:02I ask the question is based on a question here which is similar to yours
Columns becoming nullable after a fall of primary key? -
Potential problems for tables without primary keys and unique keys
GoldenGate 11.2.1.0.3/Solaris 10
DB: Oracle for Oracle (Source and target is 11.2.0.3)
Topology: unidirectional
In our one-way configuration GG, little of the tables being replicated is not a primary key or a Unique key.
Last week when we have implemented GG for the test, we received warnings for these table below.
Replication seems to work very well for these tables.GGSCI > add trandata WMHS_UD.crtn_dtl 2013-01-12 11:34:33 WARNING OGG-00869 No unique key is defined for table 'CRTN_DTL'. All viable columns will be used to represent the key, but may not guarantee uniqueness. KEYCOLS may be used to define the key. Logging of supplemental redo data enabled for table WMHS_UD.crtn_dtl.
Googling, I think that there may be performance degradation when you replicate tables without PK or the United Kingdom.
But are there other potential problems such as data of a certain kind not replicated to the lack of PK/UK?It really depends on the data.
By default, GG is combining all columns as a virtual primary key but don't no conflict control by default. So when you can be sure that the record you insert into the table is unique, then it will work.
BUT as soon as you insert the same record, which is already inserted, then you will encounter problems.Let me show what happens when you use an initial charge because it makes it easier to describe:
We start at 10:00 the capture for a table. Now, you insert a record at 10:00 in the tables. When you now start an initial charge to 10.02, then check you have inserted in the database to 10.01 will be repeated two times. During the IPL as the initial charge is made to 10.02 and it includes data of 10.01 AND it will be replicated again through the process of capture/replicate. -
A composite primary key needs of the Unique ID column or not?
Hello dear,
I have a table that's the primary key is a key that is composed of two columns and there are like 5 columns of information. You think I'm going to use two columns as the primary key of this table or create a unique ID column.
Example:
Employees table:
Services of table:
The table of the problem is the Department managers:
Now, I'll add employee id and service id in this table and Manager info etc... (info does not exist in the employee table), should I use as primary or create the id column in this table?
Thank you very much.You must use the new ID as primary (alter table add constraint primary key,...) you cannot use foreign key as primary :)
Maybe you are looking for
-
Game will get songs off the auxiliary power of the hard drives connected to my computer?
ITunes game will get songs on my auxiliary hard drive?
-
How can I set the window size by default when Firefox opens (on a Mac)?
Firefox opens random sizes, sometimes full screen, sometimes not
-
I got a call from a man speaking Indian, two days in a row telling me that he needed me to go to my computer right now and give him my ID so that it can resolve the error messages that Microsoft had received when I used my computer. The first time th
-
How to download and install outlook express for free
How to download and install outlook express for free
-
WRT1900ACS - is there a way to reset/refresh/clear/initialize the DHCP client table?
I have a WRT1900ACS with the 1.0.1.174588 firmware. There are several things that are a little weird. I noticed thre were some features missing in the network map, so I looked at the table of the DHCP client to see if they were there. The customer ta