|
Hi all,
i am trying to expose C# object to COM and in this process i got a query and not sure how to progress further. so here i am seeking for help.
i want to define a interface in C# in such a way the consumer of the COM object should be able to use the method in the following way.
comObject.methodName(123)
the above statement will return a value, assume where 123 is the parameter which could id or some thing like that
comObject.methodName(123) = 100
the above statement should set the value 100 something internal member having id 123
in both the above statement the methodName is the same.
so the question is how can we define an interface in C# that can be exposed to COM using regasm.exe.
i know it is possible to define such a way in C++ using ATL but not sure how can i achieve the same in C#.
[Comvisible(true)]
interface ExposeToCom
{
// don't know what all to add to make the method appear to COM to be get and set
int methodName(int);
}
Thanks for your help,
|
|
|
|
|
|
You are use a method as if it was a property. You cannot assign a value to a function!
Rather think of comObject.methodName(123, 100), i.e. a function with 2 parameters (key and value).
|
|
|
|
|
Thanks for your response.
Actually, i am trying re-write a COM component developed in ATL which is production. i want to give the same interface so that consumer has minimal impact.
code snippet from my existence code.
STDMETHOD(get_methodA)( VARIANT storeID, VARIANT *pVal);
STDMETHOD(put_methodA)( VARIANT storeID, VARIANT newVal);
STDMETHOD(get_methodB)( VARIANT storeID, VARIANT index, VARIANT *pVal);
STDMETHOD(put_methodB)( VARIANT storeID, VARIANT index, VARIANT newVal);
when i extract the typelib for this
[id(0x00000001), propget, helpstring("property methodA")]
HRESULT methodA([in] VARIANT storeID, [out, retval] VARIANT* pVal);
[id(0x00000001), propput, helpstring("property methodA")]
HRESULT methodA([in] VARIANT storeID, [in] VARIANT pVal);
[id(0x00000004), propget, helpstring("property methodB")]
HRESULT methodB([in] VARIANT storeID, [in] VARIANT index, [out, retval] VARIANT* pVal);
[id(0x00000004), propput, helpstring("property methodB")]
HRESULT methodB([in] VARIANT storeID, [in] VARIANT index, [in] VARIANT pVal);
so the consumer can use as
methodA(123) -- return a value
methodA(123) = 100 -- this will set a value
methodB(123, 3) -- return a value
methodB(123, 3) = 100 -- this will set a value.
i am trying to mimic the same methods defined in my ATL in a C# interface and use .net com interoperability using a regasm.
i am not able to find out how achieve this in C#. if this is done in ATL i was thinking there could be a way to achieve this.
thanks for your time
modified 27-Jul-17 13:36pm.
|
|
|
|
|
If you have a type library, you can convert it into a .NET assembly using tlbimp.
This space for rent
|
|
|
|
|
Not sure, but I believe you have to turn on this feature in the AssemblyInfo page. Then, you have to put it on the Type you want to make visible.
Ben Scharbach
Temporalwars.Com
YouTube:Ben Scharbach
|
|
|
|
|
Hi,
I was looking for some guidance on best practices regarding error handling.
public class DemoController : Controller
{
private readonly ILogger _logger;
public DemoController(ILogger logger)
{
_logger = logger;
}
public ActionResult Index()
{
try
{
}
catch (Exception ex)
{
_logger.Error(ex.ToString());
}
return View();
}
}
I was therefore wondering if anyone would have any suggestions to improve this code?
Thanks
The Dark Knight
modified 26-Jul-17 16:34pm.
|
|
|
|
|
Ask the person who told you it was sub-standard what their issues are. They could be referring to the fact that catching just "Exception" is too generic, that you should aim for specific exceptions (that rarely happens in the real world though). Or it could be the fact that you just log the exception and do nothing with it, you go on to return the View as normal so the user has no real knowledge that the code failed.
|
|
|
|
|
Thanks, makes more sense to me now.
Paul McGann
|
|
|
|
|
It's usually good practice to inform the user that an error occurred, and therefore the output might not be trustworthy or the input was not accepted by the system.
"There are three kinds of lies: lies, damned lies and statistics."
- Benjamin Disraeli
|
|
|
|
|
Thanks Benjamin, i'll keep this in mind for future.
Paul McGann
|
|
|
|
|
The "best wisdom" I have found re error-handling is in the writings of Eric Lippert, a key .NET architect: his "classic" piece often cited by other language gurus: [^].
Also: [^], [^].
A summary based on Eric's writings: [^].
There's a 3 part series on "Dr. Spock Engineer" blog I think is worth studying that starts here: [^].
I favor always throwing an Exception except in very special cases like within an enumeration of data where partial results are acceptable and usable.
«Differences between Big-Endians, who broke eggs at the larger end, and Little-Endians gave rise to six rebellions: one Emperor lost his life, another his crown. The Lilliputian religion says an egg should be broken on the convenient end, which is now interpreted by the Lilliputians as the smaller end. Big-Endians gained favor in Blefuscu.» J. Swift, 'Gulliver's Travels,' 1726CE
|
|
|
|
|
Thanks Bill, great articles and advice appreciate everyone's feedback.
Paul McGann
|
|
|
|
|
Logging the exception is useful, but I would recommend also throwing (not rethrowing) it. What I mean is:
try {
DoSomething();
}
catch (Exception ex) {
this._logger.Log(ex);
throw;
}
/ravi
|
|
|
|
|
Yes, instead of writing the same old boiler-code (as they call it), I would use PostSharp to create code-injection attributes which would automatically wrap the method with your try-catch{} exception-logger. This reduces the redundant code and improves the visibility of your class code.
Ben Scharbach
Temporalwars.Com
YouTube:Ben Scharbach
|
|
|
|
|
Greetings,
I feel like stupid when asking this question: Is it possible to create an API (or REST API) from a C# winform so other application can consume it? However, I have seen lots of posts about how to consume API from a winform.
What I have done is to create an command-line in winform so other app calls, it passes the values in the parameter.
For example:
myapp.exe -LastName:[name value] -DOB:[value]
I have run Google search and could not find anything related to this question.
Your guidance/suggestions/recommendations are truly appreciated.
JPL
|
|
|
|
|
It rather depends on what myapp is doing. You need to explain more about exactly what problem you are trying to solve.
|
|
|
|
|
Though it may be possible, it is generally not recommended: some one might run the program in a non-interactive session (e.g. by calling it from a windows service, or IIS). Then any dependency on a UI will cause trouble. Note that there is no user on the server where IIS is running, so nobody will be able to click OK on a message box showing an error. And even when you got such things right: even the link to System.Windows.Forms anywhere in a dll/exe causes trouble in such circumstances, you might have to copy System.Windows.Forms.dll into the folder of your application.
Better separate the logic completely from any UI. The UI of your "normal" app will then reference that logic and use it. And a command line app will reference the same logic. Or others may directly reference the dll(s) with the logic and use it directly. That's the way to go.
|
|
|
|
|
Consider three scenarios:
1. you want to provide a library of methods and data ... without visual UI ... to external clients: you create a Class Library which compiles to a 'dll.
2. you want to provide a re-usable Control which integrates methods and data with a visual UI to external clients: you create a WinForm Control (Component) which compiles to a 'dll.
3. you want to provide a Win application which integrates methods and data with a visual UI to external clients, which can be started with the System.Diagnostics resources (Process.Start), to which you can pass command line arguments etc. : you create an 'exe.
In all three cases, I'd follow Bernard's advice, and separate out logic, algorithms, data constant values, etc., into a Class Library that had no dependencies on the UI.
However, you can add a reference to a WinForm '.exe to another WinForm Project, and then access methods and data in that '.exe without starting it with Process.Start; without running it by invoking its Main Method. If you have static Classes defined: their contents can be used immediately; other non-static Classes will, of course, need to be instantiated with 'new to use their contents.
But, while .NET let's you "get away with this," you're, imho, creating a "monster" 
«Differences between Big-Endians, who broke eggs at the larger end, and Little-Endians gave rise to six rebellions: one Emperor lost his life, another his crown. The Lilliputian religion says an egg should be broken on the convenient end, which is now interpreted by the Lilliputians as the smaller end. Big-Endians gained favor in Blefuscu.» J. Swift, 'Gulliver's Travels,' 1726CE
|
|
|
|
|
You can use WCF to host an API in your app. This link is for hosting in a Console app but the principal is the same
Self Hosted WCF REST Service or Hosting WCF REST Service in Console Application[^]
Obviously your winform app needs to be running for it to be listening for requests though. If you don't want to use WCF you can create your own lightweight protocol just by listening on a given port and having the client connect to that port and send basic commands in a format that you devise.
|
|
|
|
|
Hi,
I am working on ASP.NET MVC5.
I have a two class that has many fields. In a different view, I want to insert two fields from on table and a list of one field from the other class.
Table A
CentreId
AdminId (UserId)
CentreName
Address
Phone
Table B
UserId
FirstName
LastName
Email
Password
The view will have fields like below
First Name
Last Name
Email
Centres
-------List the centres based on the UserId-------
Button Save
When the save button is clicked, I wish to add the details to another table like below
Table C
TeacherId
AdminId
CenterId
How can I accomplish this? Please help
Thanks
|
|
|
|
|
And? What have you tried? Where are you stuck? What help do you need?
And where does TeacherId come from?
Bad command or file name. Bad, bad command! Sit! Stay! Staaaay...
AntiTwitter: @DalekDave is now a follower!
|
|
|
|
|
Teacher is a new table and hence TeacherId.
I added code in the Controller as below
Public ActionResult Educator()
{
int curId = (int)Session["UserId"];
var centres = from c in db.Centres where c.AdminId == curId select c;
return View();
}
And in View, I have
@html.Label("First Name")
@Html.EditorFor()
@html.Label("Last Name")
@Html.EditorFor()
@html.Label("Email")
@Html.EditorFor()
Next, I want a collection of Checkboxes from the Centres result.
When I click submit, this record goes to the Users table and also the Teachers table.
Thanks
|
|
|
|
|
sunsher wrote: Next, I want a collection of Checkboxes from the Centres result.
I'd recommend reviewing the ng-loop directive. Most books will show how to do this in one of the first chapters.
Bastard Programmer from Hell 
If you can't read my code, try converting it here[^]
|
|
|
|
|
i have a server which send packet in the following format continuously after 60 second
u2u 325 16.6 6.5 330 329 | u2u 15 17 20 22 19 30 | u2u n s e w r s ||
i want to show this packet in the given format
label1 : 325 label2: 16.6 label3: 6.5 and so on in c sharp window form application
plz help me i am begginner in c sharp
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
