Development

Compare the results of the following query with the clause “fetch first 2 rows only”

with 
 t1(a) as (select * from table(odcinumberlist(1,3)))
,t2(a,b) as (select * from table(ku$_objnumpairlist(
                                 sys.ku$_objnumpair(1,1),
                                 sys.ku$_objnumpair(1,2),
                                 sys.ku$_objnumpair(1,3),
                                 sys.ku$_objnumpair(3,1),
                                 sys.ku$_objnumpair(3,2),
                                 sys.ku$_objnumpair(3,3)
                                 )))
,t(id) as (select * from table(odcinumberlist(1,2,3,4,5,6,7)))
select
  *
from t,
     lateral(select t1.a,t2.b
             from t1,t2 
             where t1.a = t2.a 
               and t1.a = t.id
             order by t2.b
             fetch first 2 rows only
             )(+)
order by id;

        ID          A          B
---------- ---------- ----------
         1          1          1
         1          3          1
         2          1          1
         2          3          1
         3          1          1
         3          3          1
         4          1          1
         4          3          1
         5          1          1
         5          3          1
         6          1          1
         6          3          1
         7          1          1
         7          3          1

14 rows selected.

with this one (i’ve just commented out the line with “fetch-first-rows-only”:

with 
 t1(a) as (select * from table(odcinumberlist(1,3)))
,t2(a,b) as (select * from table(ku$_objnumpairlist(
                                 sys.ku$_objnumpair(1,1),
                                 sys.ku$_objnumpair(1,2),
                                 sys.ku$_objnumpair(1,3),
                                 sys.ku$_objnumpair(3,1),
                                 sys.ku$_objnumpair(3,2),
                                 sys.ku$_objnumpair(3,3)
                                 )))
,t(id) as (select * from table(odcinumberlist(1,2,3,4,5,6,7)))
select
  *
from t,
     lateral(select t1.a,t2.b
             from t1,t2 
             where t1.a = t2.a 
               and t1.a = t.id
             order by t2.b
--             fetch first 2 rows only
             )(+)
order by id;

        ID          A          B
---------- ---------- ----------
         1          1          2
         1          1          3
         1          1          1
         2
         3          3          2
         3          3          1
         3          3          3
         4
         5
         6
         7

11 rows selected.

Obviously, the first query should return less rows than second one, but we can see that it returned more rows and join predicate “and t1.a = t.id” was ignored, because A and B are not empty and “A” is not equal to t.ID.

Everyone knows that rownum in inline views blocks many query transformations, for example pushing/pulling predicates, scalar subquery unnesting, etc, and many people use it for such purposes as a workaround to avoid unwanted transformations(or even CBO bugs).

select * 
from (select * from table(odcinumberlist(1,1,1,2,2,2,3,3,3)) order by 1)
where rownum <= 2
  and column_value = 3;


COLUMN_VALUE
------------
           3
           3

select * 
from (select *
      from (select * from table(odcinumberlist(1,1,1,2,2,2,3,3,3)) order by 1)
      where rownum <= 2
     )
where column_value = 3;

no rows selected

with 
 t1(a) as (select * from table(odcinumberlist(1,3)))
,t2(b) as (select * from table(odcinumberlist(1,1,3,3)))
,t(id) as (select * from table(odcinumberlist(1,2,3)))
select
  *
from t,
     lateral(select/*+ no_decorrelate */ rownum rn 
             from t1,t2 
             where t1.a=t2.b and t1.a = t.id
            )(+)
order by 1,2;

        ID         RN
---------- ----------
         1          1
         1          2
         2
         3          1
         3          2

with 
 t1(a) as (select * from table(odcinumberlist(1,3)))
,t2(b) as (select * from table(odcinumberlist(1,1,3,3)))
,t(id) as (select * from table(odcinumberlist(1,2,3)))
select
  *
from t,
     lateral(select rownum rn 
             from t1,t2 
             where t1.a=t2.b and t1.a = t.id
            )(+)
order by 1,2;

        ID         RN
---------- ----------
         1          1
         1          2
         2
         3          3
         3          4

Of course, we can draw conclusions even from these results: we can see that in case of decorrelation(query with hint) rownum was calculated before the join. But to be sure we can check optimizer’s trace 10053:

******* UNPARSED QUERY IS *******
SELECT VALUE(KOKBF$2) "ID", "VW_DCL_76980902"."RN" "RN"
  FROM TABLE("ODCINUMBERLIST"(1, 2, 3)) "KOKBF$2",
       (SELECT ROWNUM "RN_0", VALUE(KOKBF$0) "ITEM_3"
          FROM TABLE("ODCINUMBERLIST"(1, 3)) "KOKBF$0",
               TABLE("ODCINUMBERLIST"(1, 1, 3, 3)) "KOKBF$1"
         WHERE VALUE(KOKBF$0) = VALUE(KOKBF$1)
        ) "VW_DCL_76980902"
 WHERE "VW_DCL_76980902"."ITEM_3"(+) = VALUE(KOKBF$2)
 ORDER BY VALUE(KOKBF$2), "VW_DCL_76980902"."RN"

*************************

select
  *
from t,
     lateral(select rownum rn 
             from t1,t2 
             where t1.a=t2.b and t1.a = t.id)(+)
order by 1,2;

was transformed to

select
  t.id, dcl.rn
from t,
     (select rownum rn 
      from t1,t2 
      where t1.a=t2.b) dcl
where dcl.a(+) = t.id
order by 1,2;

And it confirms that rownum was calculated on the different dataset (t1-t2 join) without join filter by table t.
I created SR with Severity 1 (SR #3-19117219271) more than a month ago, but unfortunately Oracle development do not want to fix this bug and moreover they say that is not a bug. So I think this is a dangerous precedent and probably soon we will not be able to be sure in the calculation of rownum and old fixes…

Back in 2015 I wrote a blog entry to help our customers get a clear picture about the Oracle ADF vs Oracle JET "debate", over the years I kept pointing people to this writeup as a simple summary of what we are aiming to do and as a guideline for picking the technology you should use. However, the technology world is an ever changing place, and I believe that it is time to do a bit of an update of that entry to reflect the current status.

While the basic distinctions between the two technologies have not changed, specifically: 

• ADF for Java developer, JET for JavaScript developer
• ADF is server side framework, JET is client side focused
• ADF is both front and back-end framework, JET is UI layer only

Some of the surrounding eco-system has changed and evolved.

Over the past several years we saw an increase in the adoption of UI development that is focused around JavaScript-based client layer. This is something you can see for example in the results from the Stackoverflow yearly survey - back in 2016 JavaScript was used by 54% of developers and in 2018 the number stood at ~70%. 

Today, most of the new projects that we see being developed will pick up a JavaScript based UI layer as their solution for front end development. The adoption of this JS/HTML UI is not limited to web apps, many mobile apps also leverage the architecture to create mobile optimized UIs deployed on devices using either a hybrid or a Progressive Web Application (PWA) architecture. Development organizations like the fact that a client-centric JavaScript/HTML architecture allows them to use a single set of skills to support multi-channel apps. 

Back in 2015 many of the Oracle PaaS services were already leveraging this architecture along with Oracle JET - Oracle's Toolkit for building JavaScript based UIs. Today this trend started to extend to Oracle SaaS - Oracle's cloud apps offering. Several of the modules in Oracle SaaS have introduced UIs built with Oracle JET - leveraging its capabilities to create engaging and dynamic user interfaces. In the coming years you are going to see more and more modules offering new UIs based on Orale JET. This is a transition for Oracle's SaaS org from server-side generation UI to a client centric architecture. 

In parallel Oracle JET also added a bunch of capabilities targeting mobile specific scenarios including the ability to render the components to look like native controls on both iOS and Android, the ability to package apps as hybrid mobile apps, and support for offline usage patterns. All those capabilities are further pushing the usage of Oracle JET at Oracle as the UI layer for the various mobile apps we deliver.

One thing to keep in mind, since JET only provides the UI layer, there is still a need to build the backend of the application. This backend will handle data access and business logic - and it will then expose REST services that will be consumed by the JET UI. The existing backend layer for the Oracle SaaS apps was developed with Oracle ADF Business Components, and now they are able to leverage the investment they already did in that layer and expose these components through REST services. This way they get to keep the complex logic already implemented in that layer and just create a new modernized UI layer. 

If you track down the "what's new" document for the versions of ADF that we released since 2015 you'll see that enhancing the out-of-the-box functionality we provide around REST enablement has been a focus area for us. As a result of this effort, when you are exposing ADF BC as REST you get capabilities such as pagination, query by example, dynamic sorting, relationship management, and more built-into the services you expose.

This approach is something that ADF customers should consider if they decide to adopt the trend of new UIs being built with JavaScript based toolkit. Regardless of the UI framework that you'll choose, remember that the investment you made in the ADF backend can still be leveraged easily.

One of the other difference I quoted between ADF and JET was the development experience. ADF was focused on providing a more visual and declarative approach to Java developers, while JET had you use code editors to manually write your code.

This is another thing that changed since 2015. Oracle now offers a development platform for JET applications that is focused on a more visual way of building the app - this is Oracle Visual Builder Cloud Service. If you haven't seen a demo before - take 10 minutes and watch this video as an example. As you can see, Visual Builder provides a WYSIWYG UI design, action flow diagrams for business logic, page flow diagrams, property inspectors and more. All this while still allowing you to directly access the code when needed for direct manipulation. 

Image 1 - Visual Oracle JET Development in Oracle Visual Builder (Components, Structure Pane, Visual Layout Editor, Property Inspector)

With this environment you can become much more productive building UIs that leverage the power of Oracle JET. Note that Visual Builder is not just about simplifying the layout of your pages, it is also simplifying how you work with variables, binding and business logic - reducing the amount of coding you need to do.

Visual Builder also simplifies the learning curve for people who are new to JavaScript development - making the transition to this new technology smoother. Our SaaS teams have also decided to leverage VB to help them deliver new generations of UI faster.

For customers who already have ADF based applications, another appealing aspect of VB is its ability to understand the structure of ADF BC based REST services and the functionality they provide. Check out this demo to see how you can quickly use Visual Builder to build a JET based mobile app on top of ADF BC REST backend in a declarative way.

A little side note - beyond the ability of VB to create UIs, you can also use it to quickly create tables in a database and business objects (including business logic) on top of them that are exposed as REST services. The technology that drives that layer of VB is based on the same ADF Business Components - another thing that will be of course familiar to ADF developers if they decide to pick up Visual Builder.

So what should you do if you are an Oracle ADF customer today? As always the answer is "it depends" :-)

The first thing to think about is - "Does Oracle ADF answers my needs and does the application I built with it work as expected?" - assuming that the answer to both is positive, then you can continue as you were.

If, however, there are technical or architectural needs where your existing ADF app doesn't provide the answers you need (for example on the UI side or the mobile device support needs), looking at Oracle JET and Visual Builder is definitely something you should do. Also if you are about to embark on a new project, you might want to consider if a new client-centric architecture for the UI layer is something you can benefit from, and if you think it is - again try out JET and VB and see how you like them. In both cases remember that you can leverage your existing investments in the business objects layer and expose it as REST services to be consumed by new UIs. In fact you can have the two apps running in parallel.

Image 2- JET based architecture (MVVM) compared to ADF Architecture (MVC) - sharing business services

One more note, many developers like to keep their skills fresh and learn new things, if you are an Oracle ADF developer, I think that picking up Visual Builder and playing with it can get you on the road to learning new JavaScript skills quite quickly while still keeping you in a fun friendly visual cloud IDE - give it a try.

Just short note: If your imported Oracle Linux image hangs on boot in the Oracle cloud, just set GRUB_DISABLE_UUID=”true” in /etc/default/grub

End-users are strange. Sometime they need the system we are developing to remind them when they do silly things. For example some users want us to remind them if they are trying to navigate away from a page where they did changes to data, but didn't click "save". Below I'll show you an approach for implementing such a behavior in Oracle Visual Builder Cloud Service.

What we are going to do is cancel navigation actions until users acknowledge that they are ok to leave the page.

I will leave it up to you to decide when this should go into effect. While some people might claim that this should be the default behavior across the app, this is debatable. For example, if I go into a "create record" type of page, and then without doing any changes decide to leave that page - should I be prompted for a "you have unsaved changes" message? Isn't me leaving the page the equivalent of saying - I don't need to do changes? As you can see the decision is not clear cut - so in the demo below I let you control when we start this "changes were made" status simply by pressing a button. In real applications these can be done for example when a value in a field changes. At the end of the day, all you need to do is set a boolean variable that tracks whether we are now in "changes were made" status.

In the demo I added a simple dialog to the shell page (the page that acts as the containing template to the rest of the pages) - this dialog has a "you have unsaved changes, are you sure you want to leave?" type of warning and two buttons "Yes" and "No". (Quick tip/reminder - you can see how to add a dialog to a page here, and don't forget to include the popup import in the page's json file).

I add an action chain to the shell page that will be invoked on the vbBeforeExit event - in there I check the value of the "changes made" variable and if changes were made - I show the dialog. Then I use a return action to return an object type variable that has a boolean variable called "cancelled" set to true.  Returning such an object tells the flow to stop the navigation.

Now all I needed to add were action chains to the buttons for "yes" and "no" to close the dialog, and for the "yes" scenario to also set the changes made boolean variable to no - so the next time we click to navigate away we don't show the dialog.

Check out the video to see the runtime behavior and the various parts that make up the solution.

Three interesting myths about rowlimiting clause vs rownum have recently been posted on our Russian forum:

1. TopN query with rownum<=N is always faster than "fetch first N rows only" (ie. row_number()over(order by ...)<=N)
2. “fetch first N rows only” is always faster than rownum<=N
3. “SORT ORDER BY STOPKEY” stores just N top records during sorting, while “WINDOW SORT PUSHED RANK” sorts all input and stores all records sorted in memory.

Interestingly that after Vyacheslav posted first statement as an axiom and someone posted old tests(from 2009) and few people made own tests which showed that “fetch first N rows” is about 2-3 times faster than the query with rownum, the final decision was that “fetch first” is always faster.

First of all I want to show that statement #3 is wrong and “WINDOW SORT PUSHED RANK” with row_number works similarly as “SORT ORDER BY STOPKEY”:
It’s pretty easy to show using sort trace:
Let’s create simple small table Tests1 with 1000 rows where A is in range 1-1000 (just 1 block):

create table test1(a not null, b) as
  select level, level from dual connect by level<=1000;

alter session set max_dump_file_size=unlimited;
ALTER SESSION SET EVENTS '10032 trace name context forever, level 10';

ALTER SESSION SET tracefile_identifier = 'rownum';
select * from (select * from test1 order by a) where rownum<=10;

ALTER SESSION SET tracefile_identifier = 'rownumber';
select * from test1 order by a fetch first 10 rows only;

And we can see from the trace files that both queries did the same number of comparisons:

----- Current SQL Statement for this session (sql_id=bbg66rcbt76zt) -----
select * from (select * from test1 order by a) where rownum<=10

---- Sort Statistics ------------------------------
Input records                             1000
Output records                            10
Total number of comparisons performed     999
  Comparisons performed by in-memory sort 999
Total amount of memory used               2048
Uses version 1 sort
---- End of Sort Statistics -----------------------

----- Current SQL Statement for this session (sql_id=duuy4bvaz3d0q) -----
select * from test1 order by a fetch first 10 rows only

---- Sort Statistics ------------------------------
Input records                             1000
Output records                            10
Total number of comparisons performed     999
  Comparisons performed by in-memory sort 999
Total amount of memory used               2048
Uses version 1 sort
---- End of Sort Statistics -----------------------

And if we change the order of rows in the table both of these queries will do the same number of comparisons again:

create table test1(a not null, b) as
  select 1000-level, level from dual connect by level<=1000;

alter session set max_dump_file_size=unlimited;
ALTER SESSION SET EVENTS '10032 trace name context forever, level 10';

ALTER SESSION SET tracefile_identifier = 'rownum';
select * from (select * from test1 order by a) where rownum<=10;


ALTER SESSION SET tracefile_identifier = 'rownumber';
select * from test1 order by a fetch first 10 rows only;

----- Current SQL Statement for this session (sql_id=bbg66rcbt76zt) -----
select * from (select * from test1 order by a) where rownum<=10

---- Sort Statistics ------------------------------
Input records                             1000
Output records                            1000
Total number of comparisons performed     4976
  Comparisons performed by in-memory sort 4976
Total amount of memory used               2048
Uses version 1 sort
---- End of Sort Statistics -----------------------

----- Current SQL Statement for this session (sql_id=duuy4bvaz3d0q) -----
select * from test1 order by a fetch first 10 rows only

---- Sort Statistics ------------------------------
Input records                             1000
Output records                            1000
Total number of comparisons performed     4976
  Comparisons performed by in-memory sort 4976
Total amount of memory used               2048
Uses version 1 sort
---- End of Sort Statistics -----------------------

Now let’s talk about statements #1 and #2:
We know that rownum forces optimizer_mode to switch to “first K rows”, because of the parameter “_optimizer_rownum_pred_based_fkr”

SQL> @param_ rownum

NAME                               VALUE  DEFLT  TYPE      DESCRIPTION
---------------------------------- ------ ------ --------- ------------------------------------------------------
_optimizer_rownum_bind_default     10     TRUE   number    Default value to use for rownum bind
_optimizer_rownum_pred_based_fkr   TRUE   TRUE   boolean   enable the use of first K rows due to rownum predicate
_px_rownum_pd                      TRUE   TRUE   boolean   turn off/on parallel rownum pushdown optimization

while fetch first/row_number doesn’t (it will be changed after the patch #22174392) and it leads to the following consequences:
1. first_rows disables serial direct reads optimization(or smartscan on Exadata), that’s why the tests with big tables showed that “fetch first” were much faster than the query with rownum.
So if we set “_serial_direct_read”=always, we get the same performance in both tests (within the margin of error).

2. In cases when index access (index full scan/index range scan) is better, CBO differently calculates the cardinality of underlying INDEX FULL(range) SCAN:
the query with rownum is optimized for first_k_rows and the cardinality of index access is equal to K rows, but CBO doesn’t reduce cardinality for “fetch first”, so the cost of index access is much higher, compare them:

SQL> explain plan for
  2  select *
  3  from (select * from test order by a,b)
  4  where rownum<=10;

--------------------------------------------------------------------------------------------
| Id  | Operation                     | Name       | Rows  | Bytes | Cost (%CPU)| Time     |
--------------------------------------------------------------------------------------------
|   0 | SELECT STATEMENT              |            |    10 |   390 |     4   (0)| 00:00:01 |
|*  1 |  COUNT STOPKEY                |            |       |       |            |          |
|   2 |   VIEW                        |            |    10 |   390 |     4   (0)| 00:00:01 |
|   3 |    TABLE ACCESS BY INDEX ROWID| TEST       |  1000K|    12M|     4   (0)| 00:00:01 |
|   4 |     INDEX FULL SCAN           | IX_TEST_AB |    10 |       |     3   (0)| 00:00:01 |
--------------------------------------------------------------------------------------------

Predicate Information (identified by operation id):
---------------------------------------------------

   1 - filter(ROWNUM<=10)

SQL> explain plan for
  2  select *
  3  from test
  4  order by a,b
  5  fetch first 10 rows only;

-----------------------------------------------------------------------------------------
| Id  | Operation                | Name | Rows  | Bytes |TempSpc| Cost (%CPU)| Time     |
-----------------------------------------------------------------------------------------
|   0 | SELECT STATEMENT         |      |    10 |   780 |       |  5438   (1)| 00:00:01 |
|*  1 |  VIEW                    |      |    10 |   780 |       |  5438   (1)| 00:00:01 |
|*  2 |   WINDOW SORT PUSHED RANK|      |  1000K|    12M|    22M|  5438   (1)| 00:00:01 |
|   3 |    TABLE ACCESS FULL     | TEST |  1000K|    12M|       |   690   (1)| 00:00:01 |
-----------------------------------------------------------------------------------------

Predicate Information (identified by operation id):
---------------------------------------------------

   1 - filter("from$_subquery$_002"."rowlimit_$$_rownumber"<=10)
   2 - filter(ROW_NUMBER() OVER ( ORDER BY "TEST"."A","TEST"."B")<=10)

SQL> explain plan for
  2  select/*+ first_rows */ *
  3  from test
  4  order by a,b
  5  fetch first 10 rows only;

--------------------------------------------------------------------------------------------
| Id  | Operation                     | Name       | Rows  | Bytes | Cost (%CPU)| Time     |
--------------------------------------------------------------------------------------------
|   0 | SELECT STATEMENT              |            |    10 |   780 | 27376   (1)| 00:00:02 |
|*  1 |  VIEW                         |            |    10 |   780 | 27376   (1)| 00:00:02 |
|*  2 |   WINDOW NOSORT STOPKEY       |            |  1000K|    12M| 27376   (1)| 00:00:02 |
|   3 |    TABLE ACCESS BY INDEX ROWID| TEST       |  1000K|    12M| 27376   (1)| 00:00:02 |
|   4 |     INDEX FULL SCAN           | IX_TEST_AB |  1000K|       |  2637   (1)| 00:00:01 |
--------------------------------------------------------------------------------------------

Predicate Information (identified by operation id):
---------------------------------------------------

   1 - filter("from$_subquery$_002"."rowlimit_$$_rownumber"<=10)
   2 - filter(ROW_NUMBER() OVER ( ORDER BY "TEST"."A","TEST"."B")<=10)

SQL> explain plan for
  2  select/*+ index(test (a,b)) */ *
  3  from test
  4  order by a,b
  5  fetch first 10 rows only;

--------------------------------------------------------------------------------------------
| Id  | Operation                     | Name       | Rows  | Bytes | Cost (%CPU)| Time     |
--------------------------------------------------------------------------------------------
|   0 | SELECT STATEMENT              |            |    10 |   780 | 27376   (1)| 00:00:02 |
|*  1 |  VIEW                         |            |    10 |   780 | 27376   (1)| 00:00:02 |
|*  2 |   WINDOW NOSORT STOPKEY       |            |  1000K|    12M| 27376   (1)| 00:00:02 |
|   3 |    TABLE ACCESS BY INDEX ROWID| TEST       |  1000K|    12M| 27376   (1)| 00:00:02 |
|   4 |     INDEX FULL SCAN           | IX_TEST_AB |  1000K|       |  2637   (1)| 00:00:01 |
--------------------------------------------------------------------------------------------

Predicate Information (identified by operation id):
---------------------------------------------------

   1 - filter("from$_subquery$_002"."rowlimit_$$_rownumber"<=10)
   2 - filter(ROW_NUMBER() OVER ( ORDER BY "TEST"."A","TEST"."B")<=10)

ps. I thought to post this note later, since I hadn’t time enough to add other interesting details about the different TopN variants, including “with tie”, rank(), etc, so I’ll post another note with more details later.

The latest and greatest flavor of the Oracle Database in the cloud is Oracle Autonomous Transaction Processing (ATP). One of the Autonomous DB flavors that, it is optimized for OLTP (On-Line Transaction Processing) Applications - the type that you and I usually work on.

One new feature in the world of ATP is the way that you connect to the DB, connection is done leveraging wallets to make sure that your data is secured even though you are connecting over public internet. Here are instructions on how to get such a wallet file for your instance of ATP.

We introduced an enhancement to the latest version of Developer Cloud Service that allows you to connect to ATP from your CI/CD automation jobs. This can help you automate CI/CD for SQL scripts that you need to run against that DB.

As I mentioned in past blogs, DevCS has built in support for the SQLcl utility, allowing you to run SQL scripts against an Oracle database as part of your CI/CD chain. If you want to connect the SQLcl utility in DevCS to ATP, it will need to have access to your wallet.zip file. You can achieve this by uploading the file into your git repository.

Then in your SQLcl configuration you'll specify the user/pass like before, and then point the field titled Credentials File to the wallet.zip file location. (in the screenshot below, the zip file is at the top of the git repo connected to the build - so there is no need to add a path). In the next field, titled connection string, you specify the name used in the wallet's tnsnames.ora file to connect to the DB.

Now you can continue as usual and provide inline SQL or point to SQL files from your git repository.

Recently I had to install the patch for fixing cross-platform PDB transport bug onto the docker images with Oracle, so these are easy way how to do it:

1. create directory “patches” and create “install_patches.sh”:

#!/bin/bash

unzip  -u ./*.zip
CURDIR=`pwd`

for D in *; do
    if [ -d "${D}" ]; then
        echo =================================================
        echo " *** Processing patch # ${D}... "   # your processing here
        cd "${D}"
        opatch apply -silent
    fi
    cd $CURDIR
done

2. add the following commands into the dockerfile:

USER root

# Copy DB install file
COPY --chown=oracle:dba patches $INSTALL_DIR/patches

# Install DB software binaries
USER oracle
RUN chmod ug+x $INSTALL_DIR/patches/*.sh && \
    sync && \
    cd $INSTALL_DIR/patches && \
    ./install_patches.sh

3. put downloaded patches into the “patches” directory and build image.

For example, dockerfile for 18.3:

FROM oracle/database:18.3.0-ee

MAINTAINER Sayan Malakshinov <sayan@orasql.org>

USER root

# Copy patches:
COPY --chown=oracle:dba patches $INSTALL_DIR/patches

# Install patches:
USER oracle
RUN chmod ug+x $INSTALL_DIR/patches/*.sh && \
    sync && \
    cd $INSTALL_DIR/patches && \
    ./install_patches.sh

ps. I’ve also create the request for that in the official Docker github: https://github.com/oracle/docker-images/issues/1070

This is a quick blog to show two techniques for adding calculated fields to an Oracle Visual Builder application.

Both techniques do the calculation on the client side (in the browser). Keep in mind that you might want to consider doing the calculation on the back-end of your application and get the calculated value delivered directly to your client - in some cases this results in better performance. But sometimes you don't have access to modify the backend, or you can't do calculations there, so here we go:

1. For simple calculation you can just use the value property of a field to do the calculation for you.

For example if you need to know the yearly salary you can take the value in a field and just add *12 to it.

You can also use this to calculate values from multiple fields for example [[$current.data.firstName + " " +$current.data.lastName]] - will get you a field with the full name.

2. For more complex calculation you might need to write some logic to arrive at your calculated value, for example if you have multiple if/then conditions. To do that you can create a client side JavaScript function in your page's JS section. Then you refer to the function from your UI component's value attribute using something like {{$functions.myFunc($current.data.salary)}}

As you'll see in the demo, if you switch to the code view of your application the code editor in Oracle VB will give you code insight into the functions you have for your page, helping you eliminate coding errors.

I posted in the past a basic introduction to filtering lists in Visual Builder - showing how to use the filterCriterion to filter the records shown in a ServiceDataProvider. Since I recorded this video, a few things changed, and I also saw several customers asking how can they use more complex conditions that involve more than one filter.

In the video below I show how to define a basic filter with the latest versions (note that in VB 18.4.1 you no longer need to surround value with quotes ""), and then I show how to create a more complex condition that involves two filter criteria and set them to work with either an or or an and operator.

When you are using business components in Visual Builder, the filterCriterion is translated into a "q" parameter that is passed to the GET method (more about this q query parameter here). If you find that you are not getting the records you are expecting, check out the browser's Network tab to see what query parameter was passed in your call to the REST service (intro to this debugging technique here).

As you'll see the filterCriterion contains an array of "criteria" so you can specify several of them. In the video I'm using an approach that Brian Fry showed me that gives you a more declarative way to populate the array dragging and dropping multiple "criteria type" variables into the same array.

Note however that the important thing is what is actually being populated in the json file that defines the action. You should go into this view and verify that you have the right structure there. You can also directly manipulate that source to achieve the filter you need.

As you'll see in the video there are some cases where the design time for this filterCriterion adds an entry into the JSON that might not match what you want (we are tracking this issue). So as mentioned - if things don't work as expected direct manipulation of the JSON might be required. 

At Oracle OpenWorld this year I did a session about the future of Oracle ADF, and one of the demos I did there was showing the powerful combination of Oracle ADF backend with a new Oracle JET UI layer and how Oracle Visual Builder makes this integration very simple.

While we have many happy Oracle ADF customers, we do hear from some of them about new UI requirements that might justify thinking about adopting a new UI architecture for some modules. These type of requirements align with an industry trend towards adopting a more client centric UI architecture that leverages the power of JavaScript on the client. While ADF (which is more of a server centric architecture) does let you leverage JavaScript on the client and provides hook points for that in ADF Faces, some customers prefer a more "puristic" approach for new user interfaces that they are planning to build. Oracle's solution for such a UI architecture is based on Oracle JET - an open source set of libraries we developed and share with the community at http://oraclejet.org.

Oracle Visual Builder provides developers with a simpler approach to building Oracle JET based UIs - for both web and on-device mobile applications. Focusing on a visual UI design approach it drastically reduce the amount of manual coding you need to do to create JET based UIs. 

UIs that you build in Visual Builder connect at the back to REST services, and this is where you can leverage Oracle ADF. In version 12 of JDeveloper we introduced the ability to publish ADF Business Components as REST services through a simple wizard. Note that out-of-the-box you get a very powerful set of services that support things like query by example, pagination, sorting and more. If you haven't explored this functionality already, check out the videos showing how to do it here, and this video covering cloud hosting these services.

Once you have this ADF based REST services layer - you'll be glad to hear that in Visual Builder we have specific support to simplify consuming these REST services. Specifically - we understand the meta-data descriptions that these REST services provide and then are able to create services and endpoints mapping for you.

You leverage our "Service from specification" dialog to add your ADF services to your Visual Builder app - and from that point on, it's quite simple to build new JET UIs accessing the data.

In the video below I show how simple it is to build a JET-based on-device mobile app that leverage a set of REST services that were created from Oracle JDeveloper 12. Check it out: