Development

Sometimes the Oracle APEX documentation announces some packages will become deprecated in a release. It's not that those packages are suddenly gone, but you should not use them anymore. Your code will run fine still, but in the future, APEX might take it out completely, so it's best to replace them with the new package.

One of those packages announced in Oracle APEX 5.1 that are deprecated, and which I used a lot, was apex_util.string_to_table.

For example, in APEX Office Print (AOP) we read the session state of page items and we have some code like this:

declare
  l_string          varchar2(4000) := 'P1_X:P1_Y';
  l_page_items_arr  apex_application_global.vc_arr2;
begin  
  l_page_items_arr := apex_util.string_to_table(p_string => l_string, p_separator => ':');
  for i in 1..l_page_items_arr.count
  loop
    sys.htp.p(l_page_items_arr(i)||':'||apex_util.get_session_state(l_page_items_arr(i)));
  end loop;
end;    

As the function is deprecated and APEX is already on release 18.1, it's good to start replacing those calls. The new function you can use is apex_string.split.

The above code becomes then:

declare
  l_string          varchar2(4000) := 'P1_X:P1_Y';
  l_page_items_arr  apex_t_varchar2;
begin  
  l_page_items_arr := apex_string.split(p_str => l_string, p_sep => ':');
  for i in 1..l_page_items_arr.count
  loop
    sys.htp.p(l_page_items_arr(i)||':'||apex_util.get_session_state(l_page_items_arr(i)));
  end loop;
end;

Depending on your application, you might need to be careful. For example with AOP, we support customers with versions of Oracle APEX 5.0, 5.1 and 18.1. We can't really force customers to move to higher APEX versions, so the way we solve it is by using conditional compilation of our code. If we see you are on APEX 5.1 or above we will use apex_string.split if not, and you are still on an earlier version, we will use apex_util.string_to_table.

Here's an example of what the code with conditional compilation looks like:

  $if wwv_flow_api.c_current >= 20160824
  $then
    l_page_items_arr := apex_string.split(p_str=>l_string, p_sep=>':');
  $else
    l_page_items_arr := apex_util.string_to_table(p_string=>l_string, p_separator=>':');
  $end

Note the conditional compilation you also need to do on the variable if they are different, or you can choose to conditional compile on the entire function.

To conclude, I recommend with every new release of Oracle APEX to look for deprecated components and search for those and make notes to change those when needed.

Downtime during upgrades can cause a lot of headaches to E-Business Suite customers. No more!

Attend the free webinar by Oracle ACE Associate and Chicago OUG President Alfredo Abate and learn how you can patch your EBS installations without any downtime.

Register at this link.

In Visual Builder Cloud Service (VBCS) it is very easy to create a CRUD application where one page shows you a list of records, and then clicking on one of them you are directed to another page where you edit the record. But what if instead you want to edit the record on the same page - or in a popup window when you click the record in the table?

This is what this blog is all about - the results looks like this:

The video combines several techniques and tips, some of which I covered in details in other blog entires. Since the video is a bit on the long side (14 min) - here is a break down of what it shows and a way for you to skip to the parts that interest you:

• Reflecting selected table record in separate fields on the page (0:48)
• Hooking up a POST REST save endpoint to a button (3:35)
• Refreshing a data provider/table from an external event (4:45)
• Adding the missing dialog JET component to a page (5:48)
• Adding page level JavaScript methods and hook it to UI components (8:36)
• Adding an edit hyperlink to a table (12:50)

 

While working on some customers' applications, I ran into a few performance and functionality tips related to list of values in Visual Builder Cloud Service (VBCS). While it is very tempting to use the built in quick start that binds a list to results from a service, in some cases you might want to take a different approach.

One reason is performance - some lists don't change very often - and it makes more sense to fetch them only once instead of at every entry into a page. VBCS offers additional scopes for variables - and in the demo below I show how to use an application scope to only fetch a list of countries once (starting at 1:15). I also show how to define an array that will store the values of the list in a way that will allow you to access them from other locations in your app, and not just the specific UI component.

The other scenario that the demo shows relates to situations where you need to get additional information on the record you selected in the list. For example, your list might have the code and label but might contain additional meaningful fields. What if you need access to those values for the selected record?

In the demo below (6:40), I use a little JavaSciprt utility method that I add to the page to get the full details of the selected record from the list. The code used is (replace the two bold names with the id field and the value you want to return):

PageModule.prototype.FindThreeLetter = function(list,value) {
 return  list.find(record => record.alpha2_code === value).alpha3_code;
}

In the past, any array used for an LOV had to have "label" and "code" fields, but Oracle JET now allows you to set other fields to act in those roles this is shown at 5:54 using the options-keys property of the list component - a combobox in my case.

Check it out:

  select
    c.cust_first_name,
    c.cust_last_name,
    c.cust_city,
    cursor(select o.order_total, order_name,
              cursor(select p.product_name, i.quantity, i.unit_price
                       from demo_order_items i, demo_product_info p
                      where o.order_id = i.order_id
                        and i.product_id = p.product_id
                    ) product                
             from demo_orders o
            where c.customer_id = o.customer_id
          ) orders
  from demo_customers c



In the latest version of ORDS (18.2) you get by default the 500 error without the error number:


Fix is the same, set pagination to 0 and you are good to go.

A few days ago I got a question on how to hide the title row from the Interactive Report Pivot view.
So the person didn't want to show the red area:


The solution to this problem is to add the following CSS to your page:


table.a-IRR-table--pivot tr:nth-child(3) {
    display:none;
}

The result is this - the title is gone:


Doing this blog post is not about giving the solution to the above problem. I find it more important to show you the process to come to your answer. It comes down to find the right elements on the page which you can manipulate with CSS or JavaScript. To hide something, you can either use CSS with display: none or a JavaScript function (or JQuery hide()). The first thing you do is a search for the element. You want to use the Developer Tools of your browser for that. Most of the time you can right click on your page and do Inspect Element. The browser will show the HTML that is behind what you see on the page.


In the above screenshot, I see that row is a TR in a Table.
So the next step is to find a way to select that element. Typically you would use the id or class tag and look that up. The TR in our case doesn't have any of those, so I went up a line in the hierarchy until I find a good selector. The table has a class a-IRR-table--pivot which we can use.
Once we have the selector, we want to go to the real element, so we navigate back down. Now you need to know a bit of JavaScript or CSS or search on the internet how to do that. You can add elements after each other and it will drill down in the hierarchy again.
In our case, the TR is the third TR in the table, and there's a function to select that, which I used in CSS (nth-child).

If this is all new to you, learning about JavaScript and CSS selectors is a great start. For example, W3School is a nice site to get started learning more about HTML, CSS, JavaScript, and general web.

Just a quick note I'll do two presentations at the APEX Meetup in Munich on Thursday, July 19th 2018.

In the first presentation I'll bring you to a virtual and augmented world, entirely build in Oracle Application Express (APEX). There are 30 Google Cardboards available to make the experience complete. Fun guaranteed! :)


At KScope I was also interviewed by Bob Rhubart on my talks over there, which the AR/VR presentation was one of them.

In my second presentation at Munich I'll show the upcoming version of APEX Office Print (AOP).
I'll show some features nobody has seen before :) With every major release of AOP I feel like this:

If you are in the Munich area I would love to meet you at the meetup.

With every new version of Oracle Application Express (APEX) new features are added and the life of a developer is made even easier. If the feature set is not enough or you see you need to build the same functionality more often, you can always extend APEX with plug-ins.

There are six different types of plugins: dynamic action, item,  region, process, authentication scheme and authorization scheme.

Plug-ins are absolutely fantastic to extend the native functionalities of APEX in a declarative way. The APEX plugin becomes a declarative option in the APEX Builder and has the [Plug-in] text next to it. In the next screenshot, you see the dynamic actions being extended by two Plug-ins.


If you are searching for an APEX plug-in, I typically go to APEX World > Plug-ins. The nice thing about that site is that the plug-ins seem to be maintained, so if a plug-in is not supported anymore it gets the status deprecated.

!! And here lays the catch with using plug-ins. When you decide to use a plug-in in your project, you become responsible for this and need to make sure it's compatible with every release of Oracle APEX. Many plug-ins are open source and many plug-in developers maintain their plug-ins, but it's really important you understand that at the end you are responsible for things you put in your application. If the plug-in is not secure or it breaks in the next release of APEX, you need to find a solution. So use plug-ins with care and see for example how many likes the plug-in has or what the comments are about the plug-in or author. Oracle is not reviewing or supporting the plug-ins !!

When I saw the tweet of Travis, I thought to do a blog post on my top 5 plugins I use in almost every project.


Here we go:

1. Built with love using Oracle APEX

I'm proud to built applications with Oracle Application Express, and this plug-in makes it very clear :) At the bottom of the app, you will see this text:


Note that in Oracle APEX 18.1 this text in included by default and you don't even need to add the plugin. Nevertheless, I wanted to include it in this list as it should be there in every app, even the ones built before APEX 18.1 :)

2. Select2

When a select list (or drop-down) has many values, it takes too long to find the right value. Select2 makes it easy to search for values, it also supports lazy loading and multiple select.


3. APEX Office Print

APEX Office Print extends APEX so it becomes possible to export to native Excel files and generate documents in Word, Powerpoint, PDF, HTML and Text, all based on your own template. It has many more features, I blogged about some before.


4. Dropzone

APEX 18.1 has declarative multi-file upload, but still, I love the Dropzone plugin developed by Daniel Hochleitner. You can drag multiple files from your desktop straight in your APEX app. Daniel is one of my favorite plug-in developers. When he releases something, you know it will be good.



5. Modal LOV

This is a newer plugin and I haven't used it that much yet, but I'm sure I will do. The nice thing with this item type plugin is that it also supports Interactive Grid. Where Select2 stays within the page, this Modal LOV comes with a modal list of values (pop-up) which is great if you want to show multiple columns or need more context for the record you need to select.


There are many more plug-ins out there, most of them work on APEX 5.x and upwards. For example, Pretius has some cool plug-ins too, the one to create nested reports I recently used in a project. Another site you can find plug-ins is APEX-Plugin.com.

Let me start this post with a conversation between an end-user (Sarah) and a developer (Harry):

End-user: "Hey there, I'm receiving an error in the app."
Developer: "Oh, sorry to hear that. What is the message saying?"
End-user: "Unable to process row of table EBA_PROJ_STATUS_CATS.  ORA-02292: integrity constraint (XXX.SYS_C0090660) violated - child record found"
Developer: "Oh, what are you trying to do?"
End-user: "I'm trying to delete a category."
Developer: "Oh, most likely this category is in use, so you can't delete the category, you first need ..."
End-user: "Ehh?!"

When defining this error handling function the error the user gets is more like a notification message and embedded in your app. You can also define a custom message, in the above package there's a lookup in an error_lookup table, but as it can't find the constraint name, it falls back to the normal message.


The real power comes when you start to combine the error handling function with a call to also log session state information. Then you know exactly which record this error was produced for.

There are a couple of ways to include the session state:

Team Development

I typically include a feedback page in my apps. When the user logs feedback by clicking on the feedback link, this is saved in Team Development. The really cool thing is that whenever feedback is logged, automatically the session state of items and some other info like the browser that was being used at the moment of the logging is included. But you can also log feedback through an APEX API:

apex_util.submit_feedback (
    p_comment         => 'Unexpected Error',
    p_type            => 3,
    p_application_id  => v('APP_ID'),
    p_page_id         => v('APP_PAGE_ID'),
    p_email           => v('APP_USER'),
    p_label_01        => 'Session',
    p_attribute_01    => v('APP_SESSION'),
    p_label_02        => 'Language',
    p_attribute_02    => v('AI_LANGUAGE'),
    p_label_03        => 'Error orq_sqlcode',
    p_attribute_03    => p_error.ora_sqlcode,
    p_label_04        => 'Error message',
    p_attribute_04    => p_error.message,
    p_label_05        => 'UI Error message',
    p_attribute_05    => l_result.message
);

Logger 

Logger is a PL/SQL logging and debugging framework. If you don't know it yet, you should definitely check it out. In my opinion, Logger is the best way to instrument your PL/SQL code. Logger has many cool features, one of them is the ability to log your APEX items:

logger.log_apex_items('Debug Items from Error log');
With the above methods, you know which record the end-user was looking at and what the context was. Note that you might find this information too if you look at their session, but it would take more time to figure things out.

Be pro-active

Now, to prevent the conversation from happening again, you can take it one step further and start logging and monitoring those errors. Whenever errors happen you can, for example, log it in your own error table, or in your support ticket system and send yourself an email or notification.
Then instead of the end-user calling you, you call them and say "Hey, I saw you had some issues...".

By monitoring errors in your application, you can pro-actively take actions :)

Note that APEX itself also stores Application Errors. You find under Monitor Activity > Application Errors:


The report gives the error and the session, so you look further into what happened:


So, even when you didn't have an error handling function in place, you can still start monitoring errors that happen in your app. I know the readers of this blog are really smart so you might not see any errors, but still, it might be worthwhile to check it once and a while :)

You find another example of the error handling function in my Git account. I included an example of logging in your own error table and sending an email.

.gist .blob-wrapper.data { max-height:200px; overflow:auto; }

Two questions that pop-up a lot when I'm at a conference or when doing consulting are:

• How can I export my data from APEX to Excel?
• How can I print to PDF? Or how can I get a document/report with my data?

How can I export my data from APEX to Excel?

People typically want to export data from a Classic Report, Interactive Report, Interactive Grid or a combination of those to Excel.

What APEX provides out-of-the-box is the export to CSV format, which can be opened in Excel.

The biggest issue with CSV is that it's not native Excel format. Depending on the settings of Excel (or better your OS globalization settings) the CSV will open incorrectly. Instead of different columns, you see one big line. You also get an annoying message that some functions will be lost as it's not a native Excel format.


You can customize the CSV separator, so the columns are recognized. But with a global application (users with different settings), it's still a pain. Maybe the biggest issue people have with CSV export is that it's just plain text, so the markup or customizations (sum, group by, ...) are lost.

You can enable the CSV export in the attributes section of the respective components:


When you have BI Publisher (BIP) setup and in APEX specified as Print Server, you have a few more options. In the Classic Report, you find it in the Printing section - there's an option for Excel. In the Interactive Report, there's an option for XLS, the Interactive Grid doesn't have an option.

BI Publisher is expensive and comes with a big infrastructure and maintenance overhead, so this is not an option for many APEX people. But even the companies who have it, are looking at other solutions because although you get a native Excel file, it's cumbersome to use and BIP doesn't export your Interactive Report exactly as you see it on the screen with the customizations you did.

So how to get around those issues then? There are some APEX plugins to export an Interactive Report and Grid as you see it on the screen. The plugin of Pavel is probably the most popular one.
If you need to export one IR/IG at a time to Excel in a pre-defined Excel file, this might be an option for you. If you want to use your own Excel template, the ability to export multiple IR/IG at the same time or want more flexibility all around, you want to read on...

The solution

APEX Office Print (AOP). The AOP plugin extends APEX so you can specify the Excel file you want to start from, your template, in combination with the different APEX reports (Classic Report, Interactive Report, Interactive Grid) and get the output in Excel (or other formats). AOP is really easy to use, yet flexible and full of features no other solution provides. I'll touch on three different aspects customers love.

Interactive Report/Grid to Excel with AOP - WYSIWYG (!)

This feature is what customers love about AOP and something you won't find anywhere else. You can print one or more Interactive Reports and Grids directly to Excel, exactly as you see it on the screen. So if the end-user made a break, added some highlights or did some computations, it's all known by AOP. Even the Group by and Pivot are no problem. The implementation is super simple; in Excel, you can define your template; a title, a logo etc. Where you want to see the Interactive Report or Grid you specify {&interactive_1}, {&interactive_2} and for the Interactive Grid you specify {&static_id&}. In the AOP APEX plugin, you specify the template, and the static ids of the Interactive Report / Grid regions and that is it! AOP is doing the merge... if in the template the special tags are seen, AOP will generate the IR/IG. Not a screenshot - REAL table data! Here's an example with one Interactive Report:


In your Excel you can add multiple tags, on the same sheet and on different sheets... and this doesn't only work in Excel, but also in Word and PDF!

But there is even more... what if you look at the Interactive Report as a chart?
You got it... AOP even understands this. You can plot the table data with {&interactive} and by using {$interactive} it will generate the chart ... and that is a native Office chart, you can still change it in Excel!

Here's an example of the output generated by AOP with three interactive reports, one as a chart:


All the above goodies you can do through the AOP PL/SQL API too. Some people use this to schedule their reports and email them out on a daily basis, so they don't even have to go into APEX.

For me, the Interactive Report and Grid feature are one of the killer features of AOP.

Advanced templates in Excel with AOP

AOP is really flexible in how you build your template. The templating engine supports hierarchical data, angular expressions, conditions, blocks of data so you can view data next to each other and it supports HTML expressions too.

Here's an example of a template which loops over the orders and shows the product of that order. It contains a condition to show an "X" when the quantity is higher than 2 and it also has an expression to calculate the price of the line (unit price * quantity).


The data source specified in the plugin is of type SQL. AOP supports the cursor technique in SQL to create hierarchical data:


And (a part of) the output looks like this:


Multiple sheets in one Excel file with AOP

We have one customer who basically dumps their entire database in Excel. Every table has its own sheet in Excel. You just need to put the right tags in the different sheets and you are done.

AOP also supports the dynamic generation of sheets in Excel, so you get for example one sheet per customer and on that sheet the orders of that customer. The template looks like this (the magic tag is {!customers}):


The output is this:

We built this feature a while back based on some customers feedback.

Dynamic column generation in Excel with AOP

This is a new feature we have been working for in AOP 4.0. By using the {:tag} we can generate columns dynamically now too:


This might be useful if you want to pivot the data or want to see it in a different format. This feature is also available for Word tables. Another way of pivoting is doing it in Oracle or in an Interactive Report. This feature took us a long time to develop, but we think it's worth it.

I hope by the above I demonstrated why I believe APEX Office Print (AOP) is "THE" solution if you want to export your data from APEX (or the Oracle Database) into Excel.


Let's move on to the second question...

How can I print to PDF? Or how can I get a document/report with my data?

Oracle Application Express (APEX) has two integrated ways to print to PDF: either you use XSL-FO or you use BI Publisher. But the reason people still ask the question of how to print to PDF is that the one is too hard to implement (XSL-FO) and the other (BI Publisher) is too expensive, too hard to maintain and not user-friendly enough.

Again APEX Office Print (AOP) is the way to go. AOP is so easy to use, so well integrated with APEX, that most developers love to work with it. Based on a template you create in Word, Excel, Powerpoint, HTML or Text you can output to PDF. In combination with the AOP plugin or PL/SQL API, it's easy to define where your data and template is, and AOP does the merge for you.

Building the template

It begins the same as with any print engine... You don't want to learn a new tool to build your template in. You want to have a fast result. So the way you get there with AOP is, use the AOP plugin, define your data source and let AOP generate the template for you. AOP will look at your data and create a starter template for you (in Word, Excel, HTML or Text) with the tags you can use based on your data and some explanation how to use the tags.

Here's an example where AOP generates a Word template based on the SQL Query specified in the Data Source:



So now you have a template you can start from. Next, you customize the template to your needs... or you can even let the business user customize the template. The only thing to know is how to use the specific {tags}. As a developer, I always thought my time would be better spent than changing the logo on a template or changing some sentences over and over again. With AOP my dream comes true; as a developer, I can concentrate on my query (data), the business user can create the template themselves and send the new version or upload it straight into the app whenever changes are required.

When customers show me what they did with AOP; from creating templates for invoices, bills of materials, certificates to full-blown books, I'm really impressed by their creativity. If you imagine it, you can probably do it :)

Here's the AOP plugin, where we specify where the customized Word template can be found (in Static Application Files) and set the output to PDF:


Features in AOP that people love

When you download APEX Office Print, it comes with a Sample app, which shows the features of AOP in action. Here's a screenshot of some of the Examples you find in the AOP Sample App:


As this blog post is getting long, I won't highlight all the features of AOP and why they rock so much, but I do want to take two features you probably won't find anywhere else.

Native Office Charts and JET Charts in PDF

AOP supports the creation of native Office Charts, so you can even customize the charts further in Word. But sometimes people want to see exactly the chart they have on the screen, it is a JET chart, a Fusion chart, Highchart or any other library... With AOP you can get those charts straight into your PDF! The only thing you have to do is specifying the static id of the region and in your template, you put {%region} ... AOP will screenshot what the user sees and replace the tag with a sharp image. So even when the customer removed a series from the legend, it's exactly like that in the PDF.



HTML content in PDF

At the APEX World conference, a customer showed their use case of APEX together with AOP. Before they had to manage different Word documents and PDFs, but it was so hard as they had to update different documents every time again, it got out of sync and it was just a pain overall to deal with. So they replaced all this by Oracle APEX and Rich Text Editors. They created a structured database, so the information was in there once, but by using APEX Office Print (AOP) they generate all the different documents (Word/PDF) they need.

AOP will interpret the HTML when it sees an underscore in the tag e.g. {_tag}, then it will translate that HTML into native Word styling. If a PDF is requested, the Word is converted to PDF, so the PDF contains real bold text, or real colors etc.

Here's an example of how Rich Text is rendered to PDF.


AOP also understands when you use for example HTML expressions in your Classic or Interactive Report, or you do some inline styling. It took us a very long time to develop this feature, but the feedback we get from our customer base made it worthwhile :)

So far I showed Word as starting template for your PDF, but sometimes Powerpoint is a great start too, and not many people know about that. In Powerpoint you can make pixel perfect templates too and go to PDF is as easy as coming from Word.

In our upcoming release of AOP 4.0, we spend a lot of time improving our PDF feature set. We will introduce PDF split and merge and the ability to prepend and append files to any of your documents.


Some last words

If you are interested in what APEX Office Print (AOP) is all about, I recommend to sit down and watch this 45 minutes video I did at the APEX Connect conference. In that presentation, I go from downloading, installing to using and show many features of AOP live.



We at APEX R&D are committed to bringing the best possible print engine to APEX, which makes your life easier. We find it important to listen to you and support you however we can. We really want you to be successful. So if you have feedback for us in ways we can help you, even more, let us know, we care about you. We won't rest before we let everybody know about our mission and want to stay "the" printing solution for APEX.

Sometimes I get emails from developers who tell me they have to do a comparison between the print engines for Oracle APEX, but they love AOP. If you include some of the above features (IR/IG to PDF or Excel, JET Charts, and HTML to PDF) in your requirements, you are guaranteed to work with APEX Office Print, there's nothing else that comes even close to those features :)

AOP's philosophy has been to be as integrated as possible in APEX, as easy as building APEX applications, yet flexible enough to build really advanced reports. We make printing and exporting of data in APEX easy.

If you read until here, you are amazing, now I rest my case :)

This is a quick demo that combines two techniques I showed in previous blogs - filtering lists, and accessing the value of a  selected row in a table. Leveraging these two together it's quite easy to crate a page that has two tables on it - one is the parent and the other is the child, once you select a record in the parent the child table will update to see only the related child records.

Here is a quick demo:

The two steps we are doing are:

• Create an action flow on the change of first-selected-row attribute of the table
• In the flow use the assign variable function to set the filterCriterion of the child table to check for the value selected in the master

As you can see - quite simple.

 

 From there go to Active Sessions:



You will see all active sessions at that moment. Looking at the Session Id or Owner (User) you can identify the session easily:


Clicking on the session id shows the details: which page views have been done, which calls, the session state information and the browser they are using.

But even more interesting, you can set the Debug Level for that session :)

When the user requests a new page or action, you see a Debug ID of that request.


Clicking on the Debug ID, you see straight away all the debug info and hopefully it gives you more insight why something is not behaving as expected.



A real use case: custom APEX app

I had a real strange issue which I couldn't explain at first... an app that was running for several years suddenly didn't show info in a classic report, it got "no data found". When logging out and back in, it would show the data in the report just fine. The user said it was not consistent, sometimes it works, sometimes not... even worse, I couldn't reproduce the issue. So I told her to call me whenever it happened again.
One day she calls, so I followed the above to set debug on for her session and then I saw it... the issue was due to pagination. In a previous record she had paginated to the "second page", but for the current record there was no "second page". With the debug information I could see exactly why it was behaving like that... APEX rewrote the query rows > :first_row, which was set to 16, but for that specific record there were not more than 16 records, so it would show no data found.
Once I figured that out, I could quickly fix the issue by Resetting Pagination on opening of the page.

Debug Levels

You can set different Debug Levels. Level 9 (= APEX Trace) gives you most info whereas debug level 1, only shows the errors, but not much other info. I typically go with APEX Trace (level 9).

The different debug levels with the description:


Trace Mode

In case you want to go a step futher you can also set Trace Mode to SQL Trace.


This will do behind the scenes: alter session set events '10046 trace name context forever, level 12’;
To find out where the trace file is stored, go to SQL Workshop > SQL Scripts and run

SELECT VALUE FROM V$DIAG_INFO WHERE NAME = 'Diag Trace';
It will return the path of the trace file. When looking into that directory you want to search for the filename which contains the APEX session id (2644211946422) and time you ran the trace.


In Oracle SQL Developer you can then look at those trace files a bit more easily. You can also use TKPROF or other tools.


When I really have performance issues and I need to investigate further, I like to use Method R Workbench. The Profiler interpretes the trace file(s) and gives an explanation what's going on.

And with the different tools on the left, you can drill down in the files.

I'm definitely not a specialist in reading those trace files, but the above tools really help me understanding them. When I'm really stuck I contact Cary Millsap - or I call him Mr Trace - he's the father of those tools and knows trace files inside out :)

A second use case: APEX Builder

I was testing our APEX Office Print plugin in APEX 18.1 and for some reason APEX was behaving differently than earlier versions, but I didn't understand why. I followed the above method again to turn debug and trace on for my own session - so even when you are in the APEX Builder you can see what APEX is doing behind the scenes.
I hope by this post you see the light when you are in the dark. Let the force be with you :)

Oracle Database Cloud is increasingly becoming an attractive option to run databases. However, moving all our data to the cloud still represents an interesting problem. Attend the free webinar by Oracle ACE Director and OCM Kamran Aghayev and learn the eight ways you can migrate your databases to the Oracle Cloud. Register at this link.

Oracle APEX 18.1 has many new features, one of them is called Social Login.

On the World Cup 2018 Challenge, you can see the implementation of this new feature. The site allows you to sign-up or login with Facebook, Google, and your own email address.


It's even nicer that if you register with your email, but later decide to sign-up with Google or Facebook, it will recognize you as the same user if the email address is the same.

To get the Social Login to work I had to do the following...

Facebook

To enable Facebook login in your own app, you first have to create an app on Facebook. Creating an application is straightforward by following the wizards, just make sure you create a website app.


Google

To enable Google login in your own app, you first have to create a project on Google. Adrian did a really nice blog post which walks you through creating your project and setup Google authentication in your APEX application.




To hook-up Google and Facebook to our own APEX app, we have to let APEX know which credentials it should use, namely the info you find in the previous screenshots.
Web Credentials 

Go to App Builder > Workspace Utilities > All Workspace Utilities and click on the Web Credentials link

I added the Web Credentials for Facebook and Google. Web Credentials store the necessary info (Client ID = App ID and Client Secret = App Secret) of the OAuth2 authentication. OAuth2 is a standard these days which most sites are using to authenticate you as a user. Web Credentials are stored on Workspace Level so you can reuse those credentials in all the APEX apps in the same workspace.


Authentication Scheme 

We need to create the different authentication schemes. The Custom Authentication is to authenticate with email, next we have FACEBOOK, and GOOGLE (and Application Express Authentication which is there by default, but not used in this app).

Custom Authentication Scheme

I blogged before about Create a Custom Authentication and Authorization Scheme in Oracle APEX. The package I use in that blog post is pretty similar to the one of the World Cup app. In the Authentication Scheme, you define the authentication function. I also have a post-authentication procedure that sets some application items.



Facebook Authentication Scheme

Normally the authentication scheme of Facebook would look a bit different as Oracle APEX has built-in Facebook authentication, but for that to work, you need to load the SSL certificate in the Oracle wallet. On the platform the World Cup is running, the database is 12.1 and unfortunately, there's a bug in the database with multi-site or wildcard certificates (which Facebook has). So I had to workaround the issue, but I still used a new feature of APEX 18.1, instead of Facebook Authentication I used Generic OAuth2 Provider.

This is how it looks like:


As we are using the Generic OAuth2 Provider, we have to define the different OAuth URLs manually. When you look at my URLs they look a bit strange...

To get around the SSL issue I set up a reverse proxy in Apache which handles the SSL, so anytime the database does a call to http://apexrnd.localdomain it goes through the reverse proxy.
Note that in Oracle DB 12.2 and above the SSL bug is not there, so you don't need to do the above. I've been using the technique many times before if I don't want to deal with the SSL certificates and configuring the Oracle wallet. Adrian did a post about APEX Social Sign-In without a wallet, which might be of interest if you are on Oracle XE for example.

So what else is happening in the authentication scheme? You have to give the scope of what you want to get back from Facebook. In our case, we use the email as username and for additional attributes, we also want to know the first name, last name and the picture. It's really important you set those additional attributes, otherwise, APEX won't pass the full JSON through and takes a shortcut as it just needs the email.

The User info Endpoint URL is special:
http://apexrnd.localdomain/graph.facebook.com/v2.10/me?fields=#USER_ATTRIBUTES#&access_token=#ACCESS_TOKEN#

Special thanks to Christian of the APEX Dev team, without his help, I wouldn't have figured that one out. Thanks again, Christian!

The next big bit is the post_authenticate procedure which contains the logic to map the Facebook user to the World Cup app user. If it finds the user, it will set some application items again, just like in the custom authentication, but if it doesn't find the user (the first time somebody connects through Facebook), it will create a World Cup user. The most important part of that logic is the part to get the name and picture. Here we parse the JSON the authentication scheme holds in memory.

apex_json.get_varchar2('first_name')
apex_json.get_varchar2('last_name')
apex_json.get_varchar2('picture.data.url')

And then the final bit you have to be careful with, that in the authentication scheme "Switch in Session" is set to "Enabled". This setting is the magic bit to have your APEX application multiple authentication schemes and be able to use one or the other.


Google Authentication Scheme

The Google authentication is simpler than the Facebook one, as we don't have to do the workaround for the certificate as Oracle understands the Google certificate. So here I use the standard APEX 18.1 feature to authenticate against Google. The username attribute is again the email, and the "additional user attribute" is "profile" as that holds the name and picture of the person.


The rest of the authentication scheme is very similar to the one of Facebook. Again don't forget to switch in session to enable.

Login buttons

To call the different authentication schemes on our login page we included different buttons:


The Login button is a normal Submit and will do the Custom Authentication as that is the default authentication (see - Current in Shared Components > Authentication Schemes).

The Facebook button has a Request defined in the link: APEX_AUTHENTICATION=FACEBOOK, this is the way that APEX let you switch authentication schemes on the fly. Very cool! :)


The Google button is similar, but then the request is APEX_AUTHENTICATION=GOOGLE
(note the name after the equal sign needs to be the same as your authentication scheme)


I hope by me showing how the Social Authentication of Oracle APEX 18.1 was implemented in the World Cup 2018 Challenge, it will help you to do the same in your own APEX application.

I really love this new feature of APEX 18.1. The implementation is very elegant, user-friendly and flexible enough to handle most of the OAuth2 authentications out there. Note that Facebook and Google upgrade their APIs to get user info, so depending on when you read this, things might have changed. Facebook is typically backward compatible for a long time, but know that the current implementation in APEX is for API v2.10 and the default Facebook authentication is v3.0. As far as I experienced, the user info didn't change between the API versions. I'll do another blog post how you can debug your authentication as it might help you get other info than the one I got for the World Cup app. Feel free to add a comment if you have any question.

Since 2006 it's a tradition... every two years we launch a site where you can bet on the games of the World Cup (or Euro Cup). This year you find the app at https://www.wc2018challenge.com

You can read more about the history and see how things look like over time, or you can look on this blog at other posts in the different years.

The initial goal of the app was to showcase what you can do with Oracle Application Express (APEX). Many companies have Excel sheets where they keep the scores of the games and keep some kind of ranking for their employees. When I saw in 2006 that Excel sheet, I thought, oh well, I can do this in APEX, and it would give us way more benefits... results straight away, no sending of Excel sheets or merging data, much more attractive design with APEX etc. and from then on this app lives its own life.

Every two years I updated the app with the latest and greatest of Oracle APEX at that time.

Today the site is built in Oracle APEX 18.1 and it showcases some of the new features.
The look and feel is completely upgraded. Instead of a custom theme, the site is now using Universal Theme. You might think, it doesn't look like a typical APEX app, but it is! Just some minimal changes in CSS and a background image makes the difference.

The other big change is the Social Authentication, which is now using the built-in capabilities of APEX 18.1 instead of a custom authentication scheme I used the previous years. You can authenticate with Google, Facebook and with your own email (custom).

Some other changes came with JET charts and some smaller enhancements that came with APEX 5.1 and 18.1.

Some people asked me how certain features were done, so I'll do some separate blog posts about how Universal Theme was adapted on the landing page and how Social Authentication was included and what issues we had along the line. If you wonder how anything else was done, I'm happy to do some more posts to explain.

Finally, I would like to thank a few people who helped to make the site ready for this year: Erik, Eduardo, Miguel, Diego, Galan, and Theo, thanks so much!

A common requirement in applications is to have dependent lists (also known as cascading lists) - meaning have the value selected in one place influence the values that could be select in another place. For example when you select a state, we'll only show you cities in that state in the city list.

In the short demo video below, I'm showing you how to implement this cascading lists solution with the new Visual Builder Cloud Service.

The solution is quite simple

You catch the event of a value change in the first list, and in the action chain that is invoked you set a filterCriterion on the second list. (See this entry for a quick introduction to filterCriterion).

Since the list is connected to a ServiceDataProvider, there is no further action you need to take - the change to the SDP will be reflected in the UI component automatically.

Quick tips - make sure you reference the id of the column and that your operators are properly defined and enclosed in double quotes.

 

Oracle Application Express (APEX) 18.1 has been out now for a couple of days.

I typically don't wait long before doing the upgrade, as with every new release you get many new features I want to use. Also if you want to stay on top of the game, you just want to move as fast as you can. I typically start testing the Early Adopter releases and then when apex.oracle.com gets updated, I do more testing, but having it on your own system with applications that are used day-in-day-out is a different level.

So I thought to share how we update our environment in a safe way.

The first thing we do is put our maintenance pages on. We use an Apache Reverse Proxy in front of Apache Tomcat with ORDS which is connected to the Database. By specifying some ErrorDocuments the maintenance pages are being used the moment there's an error.

For example, you can add this to your httpd.conf:

ErrorDocument 404 https://s3.amazonaws.com/apexRnD/website/maintenance.html
ErrorDocument 500 https://s3.amazonaws.com/apexRnD/website/maintenance.html
ErrorDocument 503 https://s3.amazonaws.com/apexRnD/website/maintenance.html

A dataset with all sessions of the upcoming Oracle OpenWorld 2017 conference is nice to have – for experiments and demonstrations with many technologies. The session catalog is exposed at a website here.

With searching, filtering and scrolling, all available sessions can be inspected. If data is available in a browser, it can be retrieved programmatically and persisted locally in for example a JSON document. A typical approach for this is web scraping: having a server side program act like a browser, retrieve the HTML from the web site and query the data from the response. This process is described for example in this article – https://codeburst.io/an-introduction-to-web-scraping-with-node-js-1045b55c63f7 – for Node and the Cheerio library.

However, server side screen scraping of HTML will only be successful when the HTML is static. Dynamic HTML is constructed in the browser by executing JavaScript code that manipulates the browser DOM. If that is the mechanism behind a web site, server side scraping is at the very least considerably more complex (as it requires the server to emulate a modern web browser to a large degree). Selenium has been used in such cases – to provide a server side, programmatically accessible browser engine. Alternatively, screen scraping can also be performed inside the browser itself – as is supported for example by the Getsy library.

As you will find in this article – when server side scraping fails, client side scraping may be a much to complex solution. It is very well possible that the rich client web application is using a REST API that provides the data as a JSON document. An API that our server side program can also easily leverage. That turned out the case for the OOW 2017 website – so instead of complex HTML parsing and server side or even client side scraping, the challenge at hand resolves to nothing more than a little bit of REST calling. Read the complete article here.

PaaS Partner Community

For regular information on business process management and integration become a member in the SOA & BPM Partner Community for registration please visit www.oracle.com/goto/emea/soa (OPN account required) If you need support with your account please contact the Oracle Partner Business Center.

Blog Twitter LinkedIn Facebook Wiki

Technorati Tags: SOA Community,Oracle SOA,Oracle BPM,OPN,Jürgen Kress