Open Menu

Synergex Blog


Developing in Visual Studio

By Steve Ives, Posted on June 30, 2017 at 7:56 pm

Most Synergy developers would love to use the very latest and greatest development tools to develop and maintain their Synergy applications, but how do you get started? At the recent DevPartner conference in Atlanta product manager Marty Lewis not only discussed the concepts of how to get started, but actually demonstrated the entire process with a real Synergy application. Check out his presentation entitled Developing Synergy Code in Visual Studio:

By the way, this video is just one of many from the 2017 DevPartner Conference.


How “easy” can easy be – Google it!

By Richard Morris, Posted on May 24, 2017 at 4:45 am

The post conference workshop developed from the ground up a vinyl record collection catalogue system that allowed the management of your favourite records. The workshop stepped through the various aspects of the system including a desktop WPF maintenance program and remote management of your cloud based catalogue using the Synergy DBMS manager.

Having all your vinyl discs catalogued is great – but when I’m in the record shop browsing through all the latest available titles I often wonder – have I got this one? I’ve got a lot of vinyl’s and many more CD’s so remembering all the one’s I have can be difficult, and there is nothing worse than buying a duplicate.

So our OnVinyl app we built at the conference allows us to scan the barcode of the album in the store and check it against our catalogue – if we have it then the details are displayed and we don’t go and buy it again.

Getting all the album information into our catalogue would have been a long and rather boring job – especially having to trawl the web looking for the artwork to associate with each album we have. We need an easier solution – and so to Google we turn.

We all know how good Google is at searching for basically anything you can think of really, but did you know you can use this powerful searching capability from within your Synergy programs? Google provide a search API that allows you to simply and very easily perform any search you require. This is the facility we added into our OnVinyl Album Maintenance program – enter (or scan if you have a barcode scanner to hand) the barcode on the album and pass this through to the Google search API. Our implementation retrieved the album name, artist, cost and artwork so we are able to populate all the fields in the maintenance form – without any typing!

The Google search API provides a REST API that you can call directly from within your Synergy program. For full details visit https://developers.google.com/custom-search/json-api/v1/using_rest. To use the API you need a Google account. The first step is to create an “API key” which allows you access to various Google API’s and identifies you to those API’s. Visit the Google API Manager console to create an API Key. Once you have an API key then you can set up a custom “search engine” that allow you to customize how your searching is performed. You can limit the results to certain web sites for example. This is all configured using the Google control panel (https://cse.google.com/all). To begin you create a new “search engine” and name it. Once created it will be assigned a unique search engine ID – you’ll need this!

In your Synergy code define some constants;

The API is a REST implementation that has a defined URI;

And then make the required REST call to retrieve the search results;

And that’s the call to your custom search engine complete. The response from Google will be in the form of a JSON string which you need to parse and out and extract the individual elements you require. The code for that is available in the OnVinyl project.

And the results in your application;

Easy really is easy with Synergy and Google searching!

On a side note – I’ve been wondering where Jodah had got to, and then this appeared in my inbox;

I’m sure I recognise those eyes…. Have you seen or got a picture of Jodah’s identifying name tag?

 

 

 


Code Gremlins?

By Richard Morris, Posted on May 17, 2017 at 8:27 am

This year’s DevPartner conference was held in Atlanta, Georgia and was a blast. I guess I say “the best ever”” each year, so I won’t this time – but it was. The whole conference ran flawlessly, although throughout the week I did see a few strange things in the corner of my eye – but you know how it is. You think you are seeing things… But maybe not.

We had lots of new faces in the audience seeing what Synergy has to offer which is always great, and they got to see a few new faces presenting at the DevPartner conference for the first time.

Although Phil could be considered a veteran now – his third conference – he was joined on stage by first-timer Tate, a fellow support engineer. They presented a cool session on BOTS – not those things you sit on, the automated robots that answer whatever questions you have. There are loads of bots around and the duo showed us what it’s all about and how to build you own!

Another new face was Jacklin who presented all the ways to get fully up to speed with Synergy today including all the various media feeds and the cool YouTube channel we have that’s being loaded with great training videos – and of course you can always re-watch all the conference sessions.

At the beginning of March I blogged about “Physio My Way” which is an app to monitor a patients compliance to perform their prescribed upper limb exercises. Ashley presented the full Physio My Way story and received great feedback – if you get the time check out the Synergex YouTube channel in the coming days to see the full video! It’s a complete mobile solution written end to end in Synergy.

And so to my post-conference workshop. The goal was to write, from scratch, a complete application called “On Vinyl”. “On Vinyl” is a vinyl collection management and inquiry system. The system is written entirely in Synergy .Net and uses the Symphony Framework.

We started off by code generating the base data objects – the classes that expose the Synergy Repository structures as classes with full get/set properties for the fields. We also code generated classes to provide complete data management using Symphony Harmony.

Next we built a simple maintenance program to allow us to manage our collection of vinyl albums. To save typing we also implemented the ability to perform a Google search passing in the album barcode and returning/displaying all the album information, including the cover artwork. I’ll blog about just how easy it was to implement Google searching soon.

Next we hosted all the server logic on a remote cloud server behind Symphony Bridge and showed how to manage the data remotely using the Synergy DBMS Manager (available on the download page at www.symphonyframework.net).

Next we moved to the client portion of the project – to build and deploy an app on the Android and iOS platforms to enable us to check if we have a particular vinyl in our collection. We built up the technology stack. First layer was the client data layer – code generated data objects built into a portable class library. Next we created the connection library which used commands to search for vinyl details using the barcode:

If we didn’t have the album in our collection we could call our stored procedure to perform the required Google search:

And if we chose to buy the album we could insert it into our collection directly from the device:

Our final portable library was the common UI code – using Xamarin Forms. This provides the ability to write the UI once and deploy to different target devices such as Android and iOS.

Up to this point, although there were a couple of “what was that…” moment’s things all built and ran just fine.

The final piece of the puzzle was to write the device specific code – and our first target was Android. We created the project, added the required resource (images, etc.) and after a couple of minor typo’s the project built. But something was not quite right and we could not get the program to successful deploy and run.

We’d hit the end of the day so aborted the attempt and regrouped to try to find out what had gone wrong.  Once we have the solution I’ll post the entire project to GitHub and blog about it!

Now I don’t really believe in code gremlins, ghosts or phantoms, but throughout the week there were some strange goings-on, and to be honest I was having flash-backs to conference time 2009! Surely he was not back to try to mess up my workshop again?

And where did Jodah get to……?


I didn’t know you could do that!

By Richard Morris, Posted on March 23, 2017 at 2:34 pm

Working with the many customers we have is a great opportunity to advance the software we write. Some time ago I blogged about Symphony Bridge – a server based utility that allows you to expose Synergy data and logic through Symphony Harmony using an SQL-like syntax. In essence you can write queries on the client like “select * from part” and the bridge server will return a collection of part objects that are based on a structure you have created in your repository. You can then bind these to grids or input controls on you windows or mobile applications or display them in web pages. In fact you can use the data in any Synergy .Net or non-Synergy application.

Another aspect of Symphony Harmony/Symphony Bridge is the ability to call a remote procedure using the syntax “exec @MyNameSpace.MyClass.MyMethod”. You can pass in arguments and again a data object or a collection of data objects are returned.

During a recent visit with one of our customers who is currently utilizing symphony Harmony and Symphony Bridge to provide third-party access to their systems to allow enquiries on orders, etc. I was asked about “providing the ability to return a standard object that contains the call status AND the response data”. Simple, no problem: Create an RPS structure that defines your response details – call status, error codes, and error message for example. Code generate the required data object and then customize it by exposing a collection of <DataObjectBase> (the base class of all data objects!) As I said – simple.

But the customer was not overly impressed – and they had an idea! Oh dear me thinks – these are seasoned C# developers and I’m not convinced I’ll be able to implement their ideas in Synergy .Net! But how wrong I was…. and without any changes to the Symphony Framework. The basic requirement is to have a “response” class that they can extend with any serializable “type” – be that a single instance or a collection of “type” – without any restrictions.

The starting point was the same – create the required response structure in the repository and code-generate the “response” class. This is the resulting class definition;

public partial class Server_response_Data extends Symphony.Conductor.Model.DataObjectBase

The code generation process creates all the required properties and the logic to serialize the data into JSON or BSON.

The next step was to create a second class with the same name as the response class but this time it will accept a generic – generics are quite cool and allow you to defer the specification of a type until that type is instantiated. So our generic response class definition looks like this;

public class Server_response_Data<T(new)> extends Server_response_Data

By extending the base code-generated class we get all the required Symphony capabilities. The “T” denotes the generic “type”. The (new) ensures that any type you provide must have a parameter-less constructor which ensures that this class can creates an instance of it. We can expose the property so it’s serializable;

{JsonProperty}
public property Result, T
       method get
       endmethod
       method set
       endmethod
endproperty

And in the constructor we can ensure there is an instance (to prevent null object exceptions);

public method Server_response_Data
       endparams
proc
       Result = new T()
endmethod

And that is about it. You can then write your server methods to create an instance of your response class, populate the generic “response.Result” with the required “type” and set the response properties accordingly. For example;

data response = new Server_response_Data<Part_Data>()

Will create a response class with the “Result” as a type of Part_Data data object. If you need to return a collection of Part_Data objects;

data response = new Server_response_Data<List<Part_Data>>()

It’s really as easy as that. On the client you simply make the remote EXEC call through Symphony Harmony;

result = DataExecute.RunDataExecute(connector,
       &      "exec @MyNameSpace.MyClass.MyMethod",
       &      new Server_response_Data<List<Part_Data>>()).Result

You can also await the call on a separate thread.

Big shout-out to Gareth for the idea and his persistence to make me get it working!

I’ll be using these techniques in my DevPartner 2017 post conference workshop as we build a mobile app from start to finish using Symphony Harmony/Symphony Bridge to access remote data and logic.


Let’s get Physical – at DevPartner 2017!

By Richard Morris, Posted on March 8, 2017 at 8:38 am

The DevPartner 2017 conference is rapidly approaching and so I thought I’d follow on from Steve’s recent blog about the conference and what a content packed agenda we have this year. As usual we have the hugely successful customer demonstrations – always great to see how developer are making the most from Synergy today. There are guest speakers taking about testing your software – and I thought if it compiled it was tested, so I’ll be glued to that one. We also have a student Physiotherapist guest speaker – not your usual topic for a Synergy software conference I must agree!

Way back in 1981 a certain blond bombshell by the name of Olivia Newton-John hit our screens in tight pink leggings telling us all to “get physical”. For many around my age she was the pin-up of our times. I’m sure I can recite all the lyrics today! Although we won’t be seeing Miss Newton-John at the conference (sorry) we will be “getting physical”!

As part of an undergraduate degree I’ve developed an interactive mobile app:

Exercise compliance is one of the most important parts in patient rehabilitation, and yet it is given minimal consideration. The Chartered Society of Physiotherapists (CSP) detail that the first step in rehabilitating a patient back to their pre-injury function lies with their Physiotherapist’s ability to educate and motivate their patient through a clear and concise home exercise program. Patients can regularly expect to receive a printed sheet of generic exercises laden with hand-written amendments which look to tailor the exercises to the patient’s needs. Given that more than 70% of adults in the U.K. own a smartphone, there has never been a better time to take the technological leap and focus of connecting with and engaging patients much more remotely. That’s where Physio My Way comes in.”

And so Physio My Way was born. Written entirely in Synergy .Net it’s a true mobile application that is designed to educate patients to perform the correct exercises and monitor their compliance with the exercise schedules defines by their therapist. In-app options include the ability to review the various assigned exercises and stream videos that show just how they should be performed. Guidance is provided through spoken instructions and monitoring patient compliance is recorded using the devices accelerometer.

Movement patterns and screen activity can provide a clear and detailed account of every single repetition that a patient completes. Their Physiotherapist will have secure access to all of their patients’ data through their own personal log in. Specifically, this means that all exercise progressions or regressions will now be based on objective outcome markers, rather than a patient’s subjective recollection of their exercise completion. This gives superior reliability of patient feedback and is based on key Physiotherapy findings such as severity, intensity and frequency of pain and rate of perceived exertion at every interval.”

Our guest student Physiotherapist , Ashley, will be presenting an interesting session on his design and the theory behind the Physio My Way app and the results from his studies – did it improve people’s compliance with their prescribed exercise plan? And I’ll be dissecting the technologies used to develop and deploy data in the cloud and the app to Google Play Store and the Apple App Store:

At the post conference workshop on 12th May I’ll be building, from the ground up, a cross-device mobile application through lecture and demonstration. The session will target both Android and iPhone devices. Using the latest Synergy tools, Xamarin forms and Visual Studio.  We will initially build and deploy a simple cloud based server. From here we will build a client connection module to communicate to our server and also implement the required code to manage Synergy device licensing. Next we will layer on the UI – XAML based cross device compliant forms. We’ll include some cool Xamarin components and finally build and deploy to both iPhone and Android. It’s a not-to-be-missed day if you are interested in expanding your Synergy applications capabilities onto the mobile device platforms.

Make sure to sign up and bring your exercise gear as we “get physical” at DevPartner 2017!


CodeGen 5.1.7 Released

By Steve Ives, Posted on February 7, 2017 at 10:25 am

We are pleased to announce that Professional Services has just released CodeGen 5.1.7. The main feature of the release is the addition of experimental support for generating code for the MySQL and PostgreSQL relational databases. Developers can use a new command line option -database to specify their database of choice. This causes the SQL-compatible data types that are injected by the field loop expansion token <FIELD_SQLTYPE> to be customized based on the chosen database. The default database continues to be Microsoft SQL Server.

Before we consider support for these new databases to be final we would appreciate any feedback from developers working with MySQL or PostgreSQL to confirm whether we have chosen appropriate data type mappings. Additional information can be found in the CodeGen documentation.

 


Wheel, Scroll, Oops.

By Richard Morris, Posted on October 14, 2016 at 6:45 am

If you answer “yes” to the following questions, then please read on: Do you have a Synergy UI Toolkit application? Do you use standard (not ActiveX) list processing with a load method? Do you run your software on Microsoft Windows 10?

Windows 10 offers a new feature that allows you to mouse over a list and use the mouse wheel to scroll the list, without the list actually getting focus. It’s a great feature, but if you have a standard list displayed in your UI Toolkit application which uses a load method – then that mouse-over scroll operation will attempt to “process” the list and cause the list load method to execute. Does not sound too bad – but if you have method data being passed through from the l_select() or l_input() routines then this data will not be passed to your load method, because you are not actually in l_select() or l_input(). Also, because the list has not gained focus you have potentially not been through your “my list is gaining focus so set up the load parameters” logic, which again means when your load method executes it’s in an unknown state.

When your load method executes in this “unknown” state and you try to access method data or your uninitialized load data then a segmentation fault may occur. The user uses the Wheel, the list attempts to Scroll and Oops your application crashes.

Thankfully, the Synergex team have found the issue and resolved it – and the fix will be in the upcoming 10.3.3b patch. If you are experiencing this issue today and need a resolution now, you can contact support who can provide you with a hotfix.


Replicating Data to SQL Server – Made Easy

By Steve Ives, Posted on July 28, 2016 at 5:29 pm

For some time now we have published various examples of how to replicate ISAM data to a relational database such as SQL Server in near to real time. Until now however, all of these examples have required that the ISAM files that were to be replicated needed be modified by the addition of a new “replication key” field and the addition of a corresponding key in the file. Generally this new field and key would be populated with a timestamp value that was unique to each record in the file. While this technique guarantees that every ISAM file can be replicated, it also made it hard work to do so because the record layout and key configuration of each ISAM file needed to be changed.

However, almost all ISAM files already have at least one unique key, and when that is the case one of those existing those keys could be used to achieve replication without requiring changes to the original record layouts or files. When this technique is combined with the capabilities of I/O hooks it is now possible to achieve data replication with only minimal effort, often with no changes to the ISAM files being replicated, and with only minimal modification of the original application code.

I am pleased to announce that I have just published a new example of doing exactly that. You can find the example code on GitHub at https://github.com/SteveIves/SqlReplicationIoHooks. Of course if you are interested in implementing data replication to a relational database but need some assistance in doing so, then we’re here to help; just contact your Synergex account manager for further information.


Merging Data and Forms with the PSG PDF API

By Steve Ives, Posted on July 6, 2016 at 9:53 pm

When I introduced the PSG PDF API during the recent DevPartner Conference in Washington DC I received several questions about whether it was possible to define the layout of something like a standard form using one PDF file, and then simply merge in data in order to create another PDF file. I also received some suggestions about how this might be done, and I am pleased to report that one of those suggestions panned out into a workable solution, at least on the Windows platform.

The solution involves the use of a third-party product named PDFtk Pro. The bad news is that this one isn’t open source and neither is it free. But the good news is it only costs US$ 3.99, which I figured wouldn’t be a problem if you need the functionality that it provides.

Once you have PDFtk Pro installed and in your PATH you can then call the new SetBackgroundFile method on your PdfFile object, specifying the name of the existing PDF file to use as the page background for the pages in the PDF file that you are currently creating. All that actually happens is when you subsequently save your PDF file, by calling one of the Print, Preview or Save methods, the code executes a PDFtk Pro command that merges your PDF file with the background file that you specified earlier. Here’s an example of what the code looks like:

;;Create an instance of the PdfFile class
pdf = new PdfFile()

;;Name the other PDF file that defines page background content
if (!pdf.SetBackgroundFile(“FORMS:DeliveryTicketForm.pdf”,errorMessage)
    throw new Exception(errorMessage)

;;Other code to define the content of the PDF file

 

;;Show the results
pdf.Preview()

There are several possible benefits of using this approach, not least of which is the potential for a significant reduction in processing overhead when creating complex forms. Another tangible benefit will be the ability to create background forms and other documents using any Windows application that can create print output; Microsoft Word or Excel for example. Remember that in Windows 10 Microsoft has included the “Print to PDF” option, so now any Windows application that can create print output can be used to create PDF background documents.

I have re-worked the existing Delivery Ticket example that is distributed with the PDF API so that it first creates a “form” in one PDF file, then creates a second PDF file containing an actual delivery ticket with data, using the form created earlier as a page background.

I have just checked the code changes into the GitHub repository so this new feature is available for use right away, and I am looking forward to receiving any feedback that you may have. I will of course continue to research possible ways of doing this on the other platforms (Unix, Linux and OpenVMS) but for now at least we have a solution for the Windows platform that most of us are using.


Symphony takes a REST

By Richard Morris, Posted on at 3:08 am

The Symphony Harmony namespace allows access to data and logic through an SQL like syntax. For example you can select records from a file using a query such as “SELECT ID, DESCRIPTION FROM PART WHERE QUANTITY > 12 ORDER BY QUANTITY”. All matching records are returned from the query in the form of Symphony Data Objects. The data can be local or accessed via Synergy xfServer. The Symphony Bridge utility allows you to expose your query-able database via a standard Windows Communication Foundation (WCF) web service. So far so good.

Steve Ives and I recently had the opportunity to spend a week working together in the UK to “bang” heads together. Steve has always been an exponent of providing RESTful services to access logic and data which can be consumed by just about anything. So we set about using CodeGen to build a standard Restful service that will utilize the Symphony Framework to enable dynamic access to data and ultimately logic.

We soon had the basic service up and running. Out first implementation handled the standard GET verb – and returned all the records in the file. No filtering, no selection, just all the records retuned as a JSON collection. This is the standard API;

rest_1

Now remember that Symphony Harmony allows you to filter the data you are requesting, so we next implemented the ability to provide the “where” clause to the query. So for example;

rest_2

And using ARC (Advanced Rest Client which is a Google Chrome plug-in) we can test and query the service;

rest_3

And we get back just the selected customer details – all those customers where CUSTST has value of CA.

As well as being able to filter the data we can also limit the results returned by Harmony to just the fields we need; this has the benefit of reducing the data being brought across the wire. But how can our REST server build the required data objects to just include the fields we select? By doing runtime code generation! Within our code generated data objects we added the ability to dynamically build the response data object to only include those fields requested by the client. The calling syntax, as provided by the API, is;

rest_4

And again using ARC to test our server we can issue a command like;

rest_5

This is requesting all records from CUSMAS where the CUSNM2 field contains the word “LAWN” and limiting the response data object to just three fields. The response JSON looks like;

rest_6

Two perfectly formed data object that are limited by the fields in the selection list. If your Symphony Harmony connection to your data is via xfServer then only those selected fields will have been loaded from the file, again improving performance.

We also added the ability to limit the amount of data retuned by adding a “maxrows” option;

rest_7

We have already added the ability to the Symphony Harmony namespace to perform inserts, updates and deletes using standard SQL syntax and we’ll be adding these capabilities to the appropriate rest verbs POST, PUT and DELETE. Watch this blog feed for more information.


CodeGen 5.1.3 Released

By Steve Ives, Posted on June 30, 2016 at 1:11 pm

Tomorrow morning I’m heading back home to California having spent the last two weeks in the United Kingdom. The second week was totally chill time; I spent time with family and caught up with some old friends. But the first week was all about work; I spent a few days working with Richard Morris (it’s been WAY too long since that happened) and I can tell you that we worked on some pretty cool stuff. I’m not going to tell you what that is right now, but It’s something that many of you may be able to leverage in the not too distant future, and you’ll be able to read all about it in the coming weeks. For now I wanted to let you know that we found that we needed to add some new features to CodeGen to achieve what we were trying to do, so I am happy to announce that CodeGen 5.1.3 is now available for download.


PSG PDF API Moves to GitHub

By Steve Ives, Posted on May 28, 2016 at 11:05 am

This is just a brief update on the current status of the PDF API that I have mentioned previously on this forum. During the recent DevPartner Conference in Washington DC I received some really great feedback from several developers already using the API, and some very positive reactions from several others who hope to start working with it in the near future.

During my conference presentation about the API I mentioned that I was considering making the code a little easier to access by moving it out of the Code Exchange and on to GitHub. Well it turns out that was a popular idea too, so I am pleased to announce that I have done just that; you can now obtain the code from its new home at https://github.com/Synergex/SynPSG_PDF. And if any of you DBL developers out there want to get involved in improving and extending the API, we will be happy to consider any pull requests that you send to us.


Drilling for Data

By Richard Morris, Posted on April 20, 2016 at 5:44 am

One cool aspect of a Synergy UI Toolkit program has been list processing. When we introduced the concept of a list load method to populate the list as items were required was a huge step towards decoupling the program logic from the user interface. Because of this separation of concerns it’s actually very easy using the Symphony Framework to load a modern WPF data grid using an existing UI Toolkit list load method.

Our starting point has to be the UI, and that’s being rewritten in XAML. XAML allows data binding between class properties and the column elements of a data grid. We exposes our Synergy data by use of a Symphony Data Object. These data object classes represent your repository based data structures. The fields within your structures are exposed as the properties that the data grid will data-bind to.

Once we have the data grid defined we need to define the hooks between the collection of data we are going to expose to the data grid and the host programs list load method. First we call a generic method that will allow us to signal the required list loading requirements back to the host program. This snippet is going to call a generic method that will then raise the event back to the host:

d1

We are passing the load method name and then various parameters that define the actual data we are going to load and any additional list load method data. Now we can raise the load method request to the host DBL program:

d2

In the host DBL program we need to handle the event so we register an event handler:

d3

The following code snippet dispatches the load request to the existing UI Toolkit load method. There are two formats to the request depending if the list has associated list method data:

d4

So now we need to change the load method. If you have coded your list load method according to the standards laid out in the UI Toolkit manual there should only be a single line of code to change:

d5

The DoingTK test is a flag we have set to indicate if the program is running with the traditional UI Toolkit or our new WPF UI.

We shall be drilling down into the ability to handle list loading during the DevPartner 2016 pre-conference workshop as we all migrate an existing UI Toolkit program to a modern WPF user interface.


STOP! Validation Alert

By Richard Morris, Posted on April 12, 2016 at 12:00 pm

It’s the tried and trusted way to get the users attention. At the slighting hint of an issue with their data entry you put up a big dialog box complete with warning icons and meaningful information.

c1

We’ve all done it, and many programs we write today may still do it. Writing code using the Synergy UI Toolkit its common practice to write a change method to perform field level validation. When the user has changed the data and left the field – either by tabbing to the next or clicking the “save my soul” button – the change method executes and validates the entry. It is here were we stop the user in their tracks. How dare they give us invalid data – don’t they know what they should have entered? It’s an ever so regimented approach. The user must acknowledge their mistake by politely pressing the “OK” button before we allow them to continue. Users are usually not OK with this interruption to their daily schedule so there must be a nicer way to say “hey there, this data is not quite as I need it, fancy taking a look before we try to commit it to the database and get firm with you and tell you how bad you are doing at data entry?”

When migrating to a new Windows Presentation Foundation UI we can do things a little different and guide the user through the process of entering the correct data we need to complete a form or window. We will still use the same change method validation logic however as there is no reason to change what we know works.

When using the Symphony Framework you create Symphony Data Objects – these classes represent your repository based data structures. The fields within your structures are exposed as properties that the UI will data-bind to. These data objects are a little bit cleverer that just a collection of properties. Based on the attributes in the repository it knows what fields have change methods associated with them. Because of this the data object can raise an event that we can listen for – an event that says “this field needs validation by means of this named change method”. Here is a snippet of code registering the event handler:

c2

The event handler simply raises the required “change method” event back to the host DBL program;

c3

Back in the DBL program we can now listen for the change method events. Here is the event handler being registered:

c4

Remember, we are now back in the host DBL code so we can now dispatch to the actual change methods registered against the field. This is a code snippet and not the complete event handler code:

c5

We are calling into the original change method and passing through the required structure data and method data. Inside the change method we will have code that validate the entry and then, as this snippet shows, we can perform the error reporting:

c6

If the code is running as a UI Toolkit program the normal message box dialog is used to display the message. However, when running with the new WPF UI the code provides the required error information against the field. No message boxes are displayed. To the user they will see:

c7

The edit control background is coloured to indicate an issue with the data and the tooltip gives full details of the problem. When the user has entered valid data, the field reverts back to the standard renditions:

c8

We shall be exploring the ability to handle field change method processing during the DevPartner 2016 pre-conference workshop as we all migrate an existing UI Toolkit program to a modern WPF user interface.

 


What’s on the menu today?

By Richard Morris, Posted on April 6, 2016 at 5:51 am

Menu processing is a fundamental requirement in any UI Toolkit program. Without menu columns and entries your program would not function. You also signal menu events from toolbars, and buttons placed on windows and lists. Once a menu entry of any sort is fired, even if it’s signalled from within the DBL code, there is usually a using statement that dispatches to the required segment of logic. When configuring a new Windows Presentation Foundation UI this technique is still valid.

The Symphony Framework UI Toolkit assists the developer to bridge the connection between the DBL host program and the commands associated with menu bars, toolbar, and buttons – basically any WPF control that can bind to a command, and that includes selecting an item within a data grid!

Before we head into how the Symphony Framework UI Toolkit helps let’s first look at some simple UI Toolkit script and code snippets. Firstly we can create menu entries in scripts:

m1

Here we have a menu column called “mfile” with an entry called “f_new”. The important element is the entry name “f_new” as this is the menu signal name that is used within the DBL code. You can create toolbars in code that will signal a menu event, for example:

m2

Here we have the same menu event name of “f_new” being signalled when the toolbar button is selected. We can also add buttons to tab containers;

m3

Windows and lists can also have buttons added to them both in scrip and in code;

m4 m5

If we need to enable or disable the “f_new” capability within the UI Toolkit application we need to do this for all instances, for example;

m6

The Symphony Framework UI Toolkit removes the verbosity of this structure of programming while retaining the same logical concept. Using the Symphony.UIToolkit.Menu.MenuController class allows you to create a command;

m7

Now we have the command we can bind it to any WPF control that accepts a command binding. So for a menu;

m8

We can utilise the same command on, for example, an Infragistics ribbon control;

m9

Using a regular tool bar;

m10

And of course as a regular button on an input form or data grid view;

m11

In the host DBL code we can now simplify the control of the commands. By enabling the single command it changes the executable status of the all UI controls that it is bound to;

m12

When the command is executed the toolkit menu entry is signalled to the host program. We need to define an event handler;

m13

This is a code snippet and not the complete event handler code:

m14

This code is setting up the required menu entry name and then calling into the host DBL code to process the menu select. In the host program we need to ensure that we dispatch to the menu processing code segment. In this example the original toolkit code was processing an input form. We continue to process the input window using the i_input() routine if the toolkit is still the active environment. If we are running using the new WPF UI then we simple process the menu event;

m15

We shall be covering menu, toolbar and command processing during the DevPartner 2016 pre-conference workshop as we all migrate an existing UI Toolkit program to a modern WPF user interface.

 


Page 1 of 512345
RSS

Subscribe to the RSS Feed!

Recent Posts Categories Tag Cloud Archives