Columns last_ddl_time and timestamp in dba_objects

Hi,
i just ran into strange thing.
i created a procedure called aaa. then i ran create or replace procedure aaa with the same code of the original procedure.
i was surprised to see that the fields of last_ddl_time and timestamp haven't change.
Of course, when i did this experiment with changing one line in the original package - the fields of last_ddl_time and timestamp did change.
The question is if it's suppose to be this way and if so, is there any way to make this fields change even if the "new" code is exectly the same.
thanks

Hi,
i demonstrated the thing here.
i forgot to mention that the DB version i'm using is 11.2.0.3.
SQL> select last_ddl_time,timestamp from dba_objects where object_name = 'AAAAAAA';
no rows selected
SQL> create or replace procedure AAAAAAA is
begin
null;
end; 2 3 4
5 /
Procedure created.
SQL> select last_ddl_time,timestamp from dba_objects where object_name = 'AAAAAAA';
LAST_DDL_TIME TIMESTAMP
17-APR-2013 17:37:09 2013-04-17:17:37:09
SQL> create or replace procedure AAAAAAA is
begin
null;
end; 2 3 4
5 /
Procedure created.
SQL> select last_ddl_time,timestamp from dba_objects where object_name = 'AAAAAAA';
LAST_DDL_TIME TIMESTAMP
17-APR-2013 17:37:09 2013-04-17:17:37:09
SQL> create or replace procedure AAAAAAA is
begin
null;
null;
end; 2 3 4 5
6 /
Procedure created.
SQL> select last_ddl_time,timestamp from dba_objects where object_name = 'AAAAAAA';
LAST_DDL_TIME TIMESTAMP
17-APR-2013 17:38:02 2013-04-17:17:38:02

Similar Messages

  • LAST_DDL_TIME, and TIMESTAMP in the USER_OBJECTS

    Hi,
    I need to know the difference between the LAST_DDL_TIME, and TIMESTAMP in the USER_OBJECTS table concerning with the packages.I know both will give the latest compile time of the package , but i need to know why there is two seperate column. I also could find the date and time of LAST_DDL_TIME, TIMESTAMP to be different some time for a particular pacakge. Please help me in understanding the difference in detail.
    Regards,
    Vijayanand.C

    LAST_DDL_TIME is the timestamp for the last DDL operation on the object (including GRANTS and revokes).
    TIMESTAMP is the timestamp for the specification of the object.
    In more detail,
    TIMESTAMP relates only to changes to the specification of the object. DDL such as GRANTS for the object would alter the LAST_DDL_TIME, but not the TIMESTAMP.
    DDL which modifies the specification of the object, such as in the case of adding a column to a table or modifying a column attribute, would update both LAST_DDL_TIME and TIMESTAMP.
    Hope this helps.
    Janet.

  • LAST_DDL_TIME, and TIMESTAMP in the USER_OBJECTS table

    Hi,
    I need to know the difference between the LAST_DDL_TIME, and TIMESTAMP in the USER_OBJECTS table concerning with the packages.I know both will give the latest compile time of the package , but i need to know why there is two seperate column. I also could find the date and time of LAST_DDL_TIME, TIMESTAMP to be different some time for a particular pacakge. Please help me in understanding the difference in detail.
    Regards,
    Vijayanand.C

    Here is an example.
    SQL> create or replace package mypackage as
    2 procedure abc(def in varchar2);
    3 end;
    4 /
    Package created.
    SQL> select last_ddl_time,timestamp from user_objects
    2 where object_name = 'MYPACKAGE';
    LAST_DDL_TIME TIMESTAMP
    19-MAY-2004 11:21:31 2004-05-19:11:21:31
    SQL> alter package mypackage compile;
    Package altered.
    SQL> select last_ddl_time,timestamp from user_objects
    2 where object_name = 'MYPACKAGE';
    LAST_DDL_TIME TIMESTAMP
    19-MAY-2004 11:23:26 2004-05-19:11:21:31
    I hope this helps.

  • When LAST_DDL_TIME and TIMESTAMP fields are uptaded?

    Hi,
    I need know when the fields LAST_DDL_TIME and TIMESTAMP are updated.
    - Which updated objects (tables, procedures, etc) can change them?
    - Which references are updated too? All?
    Thanks by any help
    Moises Rodrigues
    System Analist
    CPM Braxis - http://www.cpmbraxis.com

    Hi,
    Please refrain from posting duplicate threads.
    Already answered Re: When LAST_DDL_TIME and TIMESTAMP fields are uptaded?
    Aman....

  • AQ tables have DATE columns in 9i and TIMESTAMP in 10g

    Hi,
    I created a queue table on 9i and 10g database's using
    eg:
    dbms_aqadm.create_queue_table(queue_table => 'TEST',
    queue_payload_type => 'RAW',
    multiple_consumers => TRUE,
    compatible => '8.1.3');
    But in 9i
    ENQ_TIME DATE
    and 10g
    ENQ_TIME TIMESTAMP(6)
    Why does date column change to timestamp?
    Thanks
    Reynold.

    My strong suggestion would be to convert all dates to char and then go from there. Doing anything other than that will force Applications to convert the dates (and I can only assume the timestamps). It is the nature of the application/datatype design. Converting everything to a CONSISTANT format, even if that is a varchar goes a long way to removing many unknowns.
    My two cents,
    Scott

  • [iOS4] CMDeviceMotion and timestamp problem

    Hi,
    Excuse my English as it isn't my primary language.
    I hava a problem using an instance of CMDeviceMotion. I need data from both the gyro and accelerometer and the frequency of sample is set at the maximum, 100.
    I also need the timestamp at which the sample was taken. However, at the beginning the timestamp doesn't change even if the acceleration and gyro change. I got something like that :
    0.003779 -0.050910 0.001185 933.604339
    0.035029 -0.042457 0.019893 933.604339
    0.062266 -0.009147 0.084819 933.604339
    0.069300 0.012292 0.092631 933.604339
    0.047862 0.018517 0.091456 933.604339
    0.035304 0.029885 0.084575 933.604339
    0.032496 0.039498 0.083583 933.604339
    We have acceleration.x, .y, .z and the timestamp. As you can see the timestamp doesn't change for about 60 upgrades ~6 seconds.
    Besides, when it finally starts upgrading, it sometimes sample twice the accelreation/gyro at the same timestamp :
    0.115215 0.005342 -0.051157 1385.964002
    *0.120561 0.019797 -0.054112 1385.977830*
    *0.122637 0.031470 -0.051182 1385.977830*
    0.113840 0.042271 -0.045493 1385.992829
    Here is my code :
    In the viewDidLoad method :
    motionManager = [[CMMotionManager alloc] init];
    motionManager.deviceMotionUpdateInterval = 0.01;
    frequence = 100;
    Then I have a method invoked by a button :
    [motionManager startDeviceMotionUpdatesToQueue:[NSOperationQueue currentQueue]
    withHandler: ^(CMDeviceMotion *motionData, NSError *error)
    CMRotationRate rotation = motionData.rotationRate;
    CMAcceleration acceleration = motionData.userAcceleration;
    NSLog(@"rotation rate = [%f, %f, %f, %f]", rotation.x, rotation.y, rotation.z, motionData.timestamp);
    NSLog(@"acceleration rate = [%f, %f, %f, %f]", acceleration.x, acceleration.y, acceleration.z, motionData.timestamp);
    Any help appreciated,
    Thanks for your time
    Loïs

    Oracle FAQ page: http://www.oracle.com/technetwork/database/enterprise-edition/jdbc-faq-090281.html#08_01
    Well obviously that was really stupid.
    I'm adding some queries to an older part of our code base and while reading through the documentation for one of the classes I found that we weren't using our normal Hibernate queries because we needed to ensure that we weren't sending Timestamps to Oracle. Apparently the Timestamp would be converted to a Date because the column in the database was of type DATE. This was causing problems because the DATE column's index was being ignored and the table contains millions of records.
    Not sure I follow that logic.
    If the value was going into the database then it would update the index.
    From that only one of the following could be true.
    1. It wasn't going into the database.
    2. It was being truncated to a date.
    The Oracle FAQ seems to indicate that this has been fixed in the 11.1 JDBC drivers. I know that I can use 11.1 JDBC drivers with a 10.2.0 database, but will it use the new 11.1 mappings for DATE and TIMESTAMP or the old 10.2.0 mappings? I would agree with your interpretation of the FAQ.
    And I would then follow it up by testing both with the old driver and the new.

  • Oracle ODBC driver and TIMESTAMP with timezone

    Does anyone know if it is possible to return data from a "TIMESTAMP WITH TIMEZONE" column using the Oracle ODBC driver and an ADO Recordset?
    I am using the Oracle driver version 10.2.0.2 and TIMESTAMP fields work fine.
    I can call Recordset->Open() with a query like "SELECT * FROM TABLE" when the table contains a column of type timestamp with timezone but when I execute a statement to see if there are results as in
    if (!(srcRecsetPtr->BOF && srcRecsetPtr->EndOfFile))
    my application throws an unhandled exception and exits. The exception is not a COM exception and I'm not sure how to get back additional information if that's possible.
    The only information I've been able to find in searching TechNet and MetaLink is that a workaround is to wrap the columns in a TO_CHAR or TO_DATE conversion first but that's not a good solution for my problem since I am executing user specified SQL which may join multiple tables.
    I've found one other note that says the documentation should be corrected and that these fields are NOT supported period (Bug #4011640).
    I've experimented with the Bind Timestamp as Date option in the ODBC connection and with various ALTER SESSION settings to attempt to change the NLS_TIMESTAMP_TZ_FORMAT but I have been unable to get past the problem.
    Any ideas are greatly appreciated.
    Thanks,
    Troy

    Hi Justin
    Thanks for your help.
    I tried what you mentioned and I could connect myself via SQL*Plus without passing a password and a login, Extern authentification seems to work and my user seems to be right configurated.
    But he problem goes on via ODBC. When I test connection in the ODBC Data Source Administrator of Windows XP, test fails and seems to forbid connection without userID and password. When I try to connect via ODBC in my program the same problem appears : "Unable to connect SQLSTATE=28000 [Oracle][ODBC][ORA]Ora-01017: Invalid username/password;logon denied" I could not connect in this way.. What's wrong ?

  • Update trigger fails with value too large for column error on timestamp

    Hello there,
    I've got a problem with several update triggers. I've several triggers monitoring a set of tables.
    Upon each update the updated data is compared with the current values in the table columns.
    If different values are detected the update timestamp is set with the current_timestamp. That
    way we have a timestamp that reflects real changes in relevant data. I attached an example for
    that kind of trigger below. The triggers on each monitored table only differ in the columns that
    are compared.
    CREATE OR REPLACE TRIGGER T_ava01_obj_cont
    BEFORE UPDATE on ava01_obj_cont
    FOR EACH ROW
    DECLARE
      v_changed  boolean := false;
    BEGIN
      IF NOT v_changed THEN
        v_changed := (:old.cr_adv_id IS NULL AND :new.cr_adv_id IS NOT NULL) OR
                     (:old.cr_adv_id IS NOT NULL AND :new.cr_adv_id IS NULL)OR
                     (:old.cr_adv_id IS NOT NULL AND :new.cr_adv_id IS NOT NULL AND :old.cr_adv_id != :new.cr_adv_id);
      END IF;
      IF NOT v_changed THEN
        v_changed := (:old.is_euzins_relevant IS NULL AND :new.is_euzins_relevant IS NOT NULL) OR
                     (:old.is_euzins_relevant IS NOT NULL AND :new.is_euzins_relevant IS NULL)OR
                     (:old.is_euzins_relevant IS NOT NULL AND :new.is_euzins_relevant IS NOT NULL AND :old.is_euzins_relevant != :new.is_euzins_relevant);
      END IF;
    [.. more values being compared ..]
        IF v_changed THEN
        :new.update_ts := current_timestamp;
      END IF;
    END T_ava01_obj_cont;Really relevant is the statement
    :new.update_ts := current_timestamp;So far so good. The problem is, it works the most of time. Only sometimes it fails with the following error:
    SQL state [72000]; error code [12899]; ORA-12899: value too large for column "LGT_CLASS_AVALOQ"."AVA01_OBJ_CONT"."UPDATE_TS"
    (actual: 28, maximum: 11)
    I can't see how the value systimestamp or current_timestamp (I tried both) should be too large for
    a column defined as TIMESTAMP(6). We've got tables where more updates occur then elsewhere.
    Thats where the most of the errors pop up. Other tables with fewer updates show errors only
    sporadicly or even never. I can't see a kind of error pattern. It's like that every 10.000th update
    or less failes.
    I was desperate enough to try some language dependend transformation like
    IF v_changed THEN
        l_update_date := systimestamp || '';
        select value into l_timestamp_format from nls_database_parameters where parameter = 'NLS_TIMESTAMP_TZ_FORMAT';
        :new.update_ts := to_timestamp_tz(l_update_date, l_timestamp_format);
    END IF;to be sure the format is right. It didn't change a thing.
    We are using Oracle Version 10.2.0.4.0 Production.
    Did anyone encounter that kind of behaviour and solve it? I'm now pretty certain that it has to
    be an oracle bug. What is the forum's opinion on that? Would you suggest to file a bug report?
    Thanks in advance for your help.
    Kind regards
    Jan

    Could you please edit your post and use formatting and tags.  This is pretty much unreadable and the forum boogered up some of your code.
    Instructions are here: http://forums.oracle.com/forums/help.jspa                                                                                                                                                                                                                                                                                                                                                                                                                                       

  • How to view both real data and timestamp informatio​n in TDM data viewer?

    Attached, I have a VI that uses NI-Scope (NI-5122) to digitize and store data.  I have added some TDMS functions in order to save waveform data in binary format ( our acquisition time is around 3 hours and the files become very large)  As you can see, I am using TDM file viewer to view the data, the problem I am having is that I can only see the voltages and not time stamps.  I am interested in viewing and analyzing timestamped data.  I do see that the graph contains both voltages and timestamps and idealy I would like to see both in the value tab.
    Please let me know what is the best way to see both sets of data. 
    Thanks!
    Attachments:
    niScopeFetchWDT_TDM1.vi ‏55 KB

    The TDMS file does preserve the timing information for a waveform, you can see it if you go to the 'Properties' tab.  The attributes are there to calculate the waveform timestamp data, however, it does not look like the code behind the TDMS file viewer supports displaying the timestamp values in a column on the 'Values' tab.  This seems like a good case for filing a product suggestion.  As for a work-around, if you are set on using a TDMS file and viewing the timing information you would require a custom version of a TDMS file viewer.
    -Bob
    -Bob

  • Mysql  And TimeStamp

    I am using Mysql.
    I want to get the TimeStamp column from database.
    I used rs.getString(6) where 6 is the column index.
    But it throws Exception like this:
    java.sql.SQLException: Cannot convert value '0000-00-00 00:00:00' from column 6 to TIMESTAMP.
         at com.mysql.jdbc.ResultSet.getTimestampFromString(ResultSet.java:6243)
         at com.mysql.jdbc.ResultSet.getTimestampInternal(ResultSet.java:6273)
         at com.mysql.jdbc.ResultSet.getTimestamp(ResultSet.java:5884)
    Help me

    The exception you posted does not match what you said you did.
    The exception specifically says that the getTimestamp() method is being called.
    The error suggests to me that the database field, not your code, is a varchar. If so then you must get it as a String. And then convert it.
    However I would suggest that if you want to keep a timestamp in the database then you use a field type that is appropriate for that rather than a varchar.

  • Date and Timestamp not updating

    Following on from my previous posts, the date and timestamp of updated files are not being updated in RoboSource Control, i.e. the Modified column in the RoboSource Control Explorer. However, it looks like changes are being stored on the server when topics are checked in, as I can right-click a file in source control that I have changed, choose View, and see the amendments in the HTML.
    Basically, this means that source control isn't recognising that files have changed and therefore when someone tries to get the latest version it doesn’t think there is one because the date and time are the same as previous version.
    Has anyone experienced this behaviour before or know why the Modified column has suddenly stopped being updated?
    Thanks
    Jonathan

    After some digging and random Googling the following article seems to have fixed my issue .
    It seems that if you are running RoboSource Control on 64-bit machines, there are problems with checking in files.
    http://helpx.adobe.com/robohelp/kb/cant-check-files-robosource-control.html.

  • DATE and TIMESTAMP problem - fixable with 11.1 JDBC Driver?

    I'm adding some queries to an older part of our code base and while reading through the documentation for one of the classes I found that we weren't using our normal Hibernate queries because we needed to ensure that we weren't sending Timestamps to Oracle. Apparently the Timestamp would be converted to a Date because the column in the database was of type DATE. This was causing problems because the DATE column's index was being ignored and the table contains millions of records.
    The Oracle FAQ seems to indicate that this has been fixed in the 11.1 JDBC drivers. I know that I can use 11.1 JDBC drivers with a 10.2.0 database, but will it use the new 11.1 mappings for DATE and TIMESTAMP or the old 10.2.0 mappings? Oracle FAQ page: http://www.oracle.com/technetwork/database/enterprise-edition/jdbc-faq-090281.html#08_01

    Oracle FAQ page: http://www.oracle.com/technetwork/database/enterprise-edition/jdbc-faq-090281.html#08_01
    Well obviously that was really stupid.
    I'm adding some queries to an older part of our code base and while reading through the documentation for one of the classes I found that we weren't using our normal Hibernate queries because we needed to ensure that we weren't sending Timestamps to Oracle. Apparently the Timestamp would be converted to a Date because the column in the database was of type DATE. This was causing problems because the DATE column's index was being ignored and the table contains millions of records.
    Not sure I follow that logic.
    If the value was going into the database then it would update the index.
    From that only one of the following could be true.
    1. It wasn't going into the database.
    2. It was being truncated to a date.
    The Oracle FAQ seems to indicate that this has been fixed in the 11.1 JDBC drivers. I know that I can use 11.1 JDBC drivers with a 10.2.0 database, but will it use the new 11.1 mappings for DATE and TIMESTAMP or the old 10.2.0 mappings? I would agree with your interpretation of the FAQ.
    And I would then follow it up by testing both with the old driver and the new.

  • Reading MS Project column names and data on the fly from a selected View

    Hi guys,
    I have several views on my project file (MSPROJECT 2010) and I want to build a macro so that;
    1. User can select any view ( Views can have diffrent columns and the user may add new columns as well)
    2. User runs the Macro and all the coulmns along with the tasks displayed in the view will be written to a excel file. ( I don't want to build several macro's for each view, I'm thinking of a common method which would work for any selected view)
    The problem I'm facing is that how will i read the column names and data for a particular view on the fly without hard coding them inside the vba code ?
    The solution needs to work on a master schedule as well.
    Appreciate your feedback.

    Just to get you started the following code writes the field name and data for the active task to the Immediate window.
    Sub CopyData()
    Dim fld As TableField
    For Each fld In ActiveProject.TaskTables(ActiveProject.CurrentTable).TableFields
    If fld.Field >= 0 Then
    Debug.Print Application.FieldConstantToFieldName(fld.Field), ActiveCell.Task.GetField(fld.Field)
    End If
    Next fld
    End Sub
    Rod Gill
    Author of the one and only Project VBA Book
    www.project-systems.co.nz

  • Interactive Report Column Headings and Row Height

    I am using the div style="width:350px;" method to control the width of columns in various interactive reports. That works fine except for the following:
    1. When creating filters, the <div...> stuff shows up along with the actual column heading, thus confusing some end users
    2. Even when I uncheck the "Use same text for single row view" checkbox and then provide a simple single row view label, the <div> stuff still shows up on the single row view
    Does anyone have a better solution?
    Also, does anyone know of a way to limit the row height within an interactive report row? I have some columns of data that contain a large amount of HTML data and I'd like to be able to limit the number of rows that show on the report.
    Is Oracle planning to provide some better control over the Interactive Report columns in another version? The Interactive Report is such a huge improvement in usability in APEX - it would be great to take it to another level by providing some better control over column width and row height.
    Edited by: DaleB on Jun 18, 2009 8:54 AM
    Edited by: DaleB on Jun 18, 2009 8:54 AM

    Dale,
    Unfortunately we don't have much we can use to do what you would like. I would have said it's impossible until version 4 but you could actually do something similar to what Roel has done. His trick is in the edit button. He changed the edit button to use an "onload" call to a JavaScript process. You could do the same but call a process that goes across the rows and styles each column. Now because you don't have a way to identify the column (can't use the order because the end user could change it) you'll have to write the code to look at the top row first and then style the appropriate column. As far as I can tell, this would be quite difficult and inefficient. Having said that if you need it that bad and would like some help with it, put up an example application on apex.oracle.com and provide the workspace/username/password and I'll take a look.
    Regards,
    Dan
    http://danielmcghan.us
    http://sourceforge.net/projects/tapigen
    You can reward this reply by marking it as either Helpful or Correct ;-)

  • HT5824 I've been using numbers for iPad and I accidentally deleted a column. And the auto-save feature saved that and now I can't undo it. Does iCloud storage have backups of saved files from like 10 minutes ago? I have number

    I've been using numbers for iPad and I accidentally deleted a column. And the auto-save feature saved that and now I can't undo it. Does iCloud storage have backups of saved files from like 10 minutes ago? I have numbers synced to iCloud.
    I tried to undo it but the app crashed.
    I am hoping there's like a previously saved version of my file in iCloud somewhere. 

    No, iOS does not do short term incremental back ups to iCloud such as you get with Time Capsule.It backs up when you do soo manually, or if set properly, when the device is plugged in and connected to WiFi. It will not have saved your changes as of 10 minutes prior.

Maybe you are looking for