Class based exception handling

Hi everyone....
I am new to object oriented abap.
I would like to know more about exception handling using classes.
please provide some good material or links on the topic?
Points will be rewarded.......

hi,
Download the PDF from here.
http://www.esnips.com/doc/6d16a298-9227-4d32-acf1-e91164c89daf/3-ABAP-Objects(P283)
Follow this link too for tutorials.
http://www.****************/Tutorials/OOPS/MainPage.htm
Hope this helps, Do reward.
Edited by: Runal Singh on Mar 14, 2008 3:28 PM

Similar Messages

Handling general errors via class based exceptions

Fairly standard scenario: I've got this.
DATA: l_requester TYPE xubname.
CALL FUNCTION 'ENQUEUE_EZMYTAB'
    EXPORTING
      username             = i_username
    EXCEPTIONS
      foreign_lock         = 1
      system_failure       = 2
      OTHERS               = 3.
CASE sy-subrc.
    WHEN 1.
      l_requester = sy-msgv1.
      RAISE EXCEPTION TYPE zcx_my_exception EXPORTING textid   = zcx_my_exception=>lock_error
                                                      user     = i_username
                                                      lockedby = l_requester.
    WHEN 2.
      RAISE EXCEPTION TYPE zcx_my_exception.
ENDCASE.
But what I really want for WHEN 2 is the class based exception of
MESSAGE sy-msgid TYPE sy-msgty NUMBER sy-msgno WITH sy-msgv1...
I.e., in my calling code, I want to catch the exception, and display whatever the original error message was as generated by the ENQUEUE function module.
I suppose I could do this
DATA: ls_error TYPE scx_t100key.
WHEN 2.
ls_error-msgid = sy-msgid.
ls_error-msgty = sy-msgty.
ls_error-msgno = sy-msgno.
RAISE EXCEPTION TYPE zcx_my_exception EXPORTING textid = ls_error
but I can't help feel I'm missing something staggeringly obvious...
matt

Hi Matt,
this may sound too simple:
For me the real advantage of class-based-exception is that they can be raised in an inner block and caught in an outer block at any level.
I never use the exception attributes to communicate any details, but use
IF SY-SUBRC NE 0.
MESSAGE sy-msgid TYPE sy-msgty NUMBER sy-msgno WITH sy-msgv1... INTO lv_string.
Raise Exception type zcx_my_exception.
ENDIF.
Catching the exception I may (roughly) know where it comes from. As the message variables are present, I can check their values or repeat the MESSAGE without INTO. The main function of MESSAGE INTO is for transparency, the ability to see the whole message in debugger in lv_string and the where-used-list if message is created by programmer..
If I really need different reaction for different SY-SUBRC ( = different untyped exception), I'd rather create different class-based exceptions. This helps transparency.
In other oo language you do not have an ABAP message concept with everywhere-present message variables. Why use this crutch of putting the message variable filled by functions exception into exception object attributes only to convert them back to message output?
I prefer [KISS|http://en.wikipedia.org/wiki/KISS_principle].
Regards,
Clemens

How to store a class-based exception as container element?

Hello,
In a BO type method, there are situations where unexpected class-based exceptions can occur, I handle them with TRY ... CATCH ... ENDTRY.
If such an error occurs, my BO type method also returns a BO type method exception using a return code (as usually).
1) Now, my question is how can I store the exception instance in the task container ? (so that to get more information about what happened... it is an unexpected error as I said)
Note that my question may apply to any class instance.
2) To achieve this, I have an idea but not sure if it's realistic (maybe the question is not realistic too ). I'll create a new BO type ZOOINSTANC with string attribute containing the serialized instance, with a method to display it.
Example how it could work, in my BO type method (it returns EXCEPTION parameter of type ZOOINSTANC):
TRY.
CATCH cx_root INTO lo_exc.
    DATA l_exc TYPE string.
    CALL TRANSFORMATION id
        SOURCE oo = lo_exc
        RESULT XML l_exc.
    DATA l_bor_zooinstanc TYPE swc_object.
    swc_create_object l_bor_zooinstanc 'ZOOINSTANC' ''. "no key
    swc_set_property l_bor_zooinstanc l_exc.
    swc_set_element container 'EXCEPTION' l_exc.
    exit_return 1001 '' '' '' ''.
ENDTRY.
One important issue is that I don't know any API to display the instance :-p
Do you have any ideas, advices, ... ?
Thx
sandra

Hi Mike, thank you for your answer, and sorry for the delay.
> I'd rather go with the idea of returning the error details in a structured format that you can view in the container
Yes, but that's the issue as I don't know which exceptions can be triggered, as I'd like to handle all exceptions.
> I'd also ask why it's necessary to use BOR for this?
When a BOR method fails, I want technical support to have all the details of the error, not only a generic error message. It's why I want to store the exception in the workitem container, and be able to display it later.
> It's obviously your own development, so why not use classes. Subclasses of CX_BO_ERROR are recognised by workflow and you can even bring a proper message back to the WF log. There's also some info on this in the second edition of the WF book.
I did a test using demo workflow WS56400159 ("absence"), where I forced CX_BO_ERROR right after SWX_FORMABS_CREATE call in CL_SWF_FORMABSENC / CREATE method, but it seems that only the short text of the exception is kept.
So, I'm a little disappointed to see that we have a great exception "tool" with class-based exceptions, but only the last text is kept in the system! That would have been great to be able to store the whole information (the attributes, and the whole bunch of cascading exceptions (PREVIOUS attribute), instead of only a simple text).
sandra

Class based error handling

Hi evrybody,
i'd like to implemt some own errorclasses, but i cand find anything about the namespace to use. Am i right that evry class based error handling must start with cx? So which namespace is for customoers?
cu
Rainer

Hi Rainer,
if you already have a regulation for class names, just replace CL by CX and apply them accordingly.
For example, if your classes are named ZCL_... or YCL_..., your exception classes will be named ZCX... or YCX_..., respectively.
See T100 message OO 145 for the details. Here is the German version:
+Kurztext
Klassenname &1 ist für den gewählten Klassentyp nicht zulässig
Diagnose
Namenskonvention für Klassen :
Für eine persistente Klasse ist der Präfix 'CL_' bzw. Namensraum+Präfix 'CL_' ( z.B. 'ZCL' oder '/NAMESPACE/CL_' ) vorgeschrieben.
Für eine Exceptionklasse ist der Präfix 'CX_' bzw. NamensraumPräfix 'CX_' ( z.B. 'ZCX' oder '/NAMESPACE/CX_' ) vorgeschrieben.
Best regards,
Thorsten Franz

How do Standard class based exceptions understand an exception?

Dear All,
Can any one explain how does the standard class based exceptions understand and identify the specific exception. For example CX_SY_ZERODIVIDE is an exception class we use to catch the divide by zero exception. How does the class understand it. To clear my point please refer the code below
data: v1 type i,
v2 type i.
try.
v2 = v1 / 0.
catch CX_SY_ZERODIVIDE.
The above code will work and an exception will be handled properly. But how??
suppose if i write..
try.
v1 = Apple
catch CX_SY_ZERODIVIDE.
it obviously doesnt work.
so my point is how does an exception class understand the exception?
I have very poor knowledge about it, i have checked the class CX_SY_ZERODIVIDE and its methods. There is no code in it. .
Any help would be greatly appreciated.
Thanks and Regards,
Abdullah.

1) You can write your own log class. Call the log method from your catch
2) Upgrade to JDK1.4.0 and you can use the Logger function
3) Check jakarta for log4j

Where-used list for class-based exception texts

G'Day,
With traditional exceptions, one can look at a message, go to SE91 and do a where-used for that particular message. This is not possible with ZCX classes, a where-used can be done for the entire class but there doesn't seem to be a way to find where individual exception texts are raised.
Now whilst exception classes do offer the possiblity of a far better structuerd approach that reduces the need for a where-used in first place, this is still a shortcoming of exception classes... or is it? (or am I missing some other way to search for it?)
Cheers,
Mike

D'Oh! That one was so obvious that I was blinded by it
(Believe it or not I've been using ex. classes for some time!)
However it still leaves another issue (which was actually the one that sparked this debate here in the office): In the old message concept it was far easier to navigate from an error to find the offending code. One could click on a message, see the message class/ID, drop into SE91 and (provided the where used is small enough) go straight to the offending code via a where-used.
Unless class-based exceptions are caught and re-raised as an old MESSAGE statement with a message number, this is not so simple anymore. Or is it?
Thanks,
Mike

ENQUEUE and Class-based Exceptions

Hi,
If I use the ENQUEUE Function module in a method which also raises class-based exceptions, I'd like to be able to raise equivalent class-based exceptions which map to the standard ENQUEUE function module exceptions.
Normally I'd simply propogate the FM exception by calling MESSAGE ... RAISING and then raise my own exception with the message that the ENQUEUE FM returned in sy-msg* etc.
But you can't have normal exceptions and class-based ones in the same method interface. Does anyone know if there's a SAP standard CX exception which has all the same messages and parameters in it so that we can raise the relevant class-based exception in the method after an ENQUEUE fails?
Thanks
Paul

Hello Paul,
I am not sure whether I understand your problem correctly, so please excuse me if I am completely off track
I assume you want to transport the information that is normally stored in sy-msg variables via class-based exceptions.
If you define your own exceptions, you can add attributes to the exception class. Any attributes, basically. I did something along this line by adding all relevant fields (message class, type, number, the four variables) as attributes to my exception class. Then, if you raise an exception with RAISE EXCEPTION TYPE <yourExceptionClass> and you use, for example, drag-and-drop from SE80 or the pattern feature for that, in the EXPORTING part of the statement you can fill these fields. As they are public fields, any method catching your exception can look into these and figure out what went wrong with the ENQUEUE.
Hope this helps (if I understood your question right)
Joerg

Working with class based exception and dynamic method calls

Hi Gurus,
we just changed out ERP from EHP6 to EHP7.
Since we did so we are facing an issue with an Z-Report we are using quite often.
This reports looks up Workitems and executes the according methods so that we can go into debugging if we were facing any problems or errors.
since the EHP Upgrade this statement has problems:
data: lt_parmbind type abap_parmbind_tab, lt_excpbind type abap_excpbind_tab, lo_runtime type ref to object. call method lo_runtime->(iv_cls_method) parameter-table lt_parmbind exception-table lt_excpbind.this CALL METHOD Statement has Problem with the Exception Table. We are quite often getting DYN_CALL_METH_EXCP_NOT_FOUND short dumps with Exception "CX_SY_DYN_CALL_EXCP_NOT_FOUND".
The system has problems handling the content of lt_excpbind. if i clear this table the CALL METHOD statement works fine.
AS an example we are trying to call /IDXGC/CL_PD_PROCESS_STEPS-->CREATE_DATA. This method has 2 exceptions
/IDXGC/CX_PROCESS_ERROR
Process Layer Exception
CX_BO_TEMPORARY
Temporary Business Exception
The Content of LT_EXCPBIND is
INDEX
NAME
VALUE
2
/IDXGC/CX_PROCESS_ERROR
1
2
CX_BO_TEMPORARY
2
From my point of view the Problem ist, that they are marked as "class based". I think so because if you looked up the SAP Help for the EXCEPTION-TABLE Statement it is written that is statement only works for none-classbased exception.
I think that restriction is quiet clear. But what i am wondering about is.. that restriction also exists for EHP6. And in EHP6 it work. Does anyone know why? Or how i can change me CALL METHOD Statement that i will work again?
Best Regards
Udo

Class-based exceptions must be caught using try/catch statement.
Calling dynamically a method catchable exceptions are:
CX_SY_DYN_CALL_EXCP_NOT_FOUND
CX_SY_DYN_CALL_ILLEGAL_CLASS
CX_SY_DYN_CALL_ILLEGAL_METHOD
CX_SY_DYN_CALL_PARAM_MISSING
CX_SY_DYN_CALL_PARAM_NOT_FOUND
CX_SY_REF_IS_INITIAL
Anyway catching cx_root (as shown by Matthias) will catch everything is catchable.

Catch class based exceptions

hi
i am calling a outbound proxy class from a inbound proxy class. when i dragged the ourbound proxy the below code is inserted by system directly.
TRY.
CALL METHOD oref1->SEND_OUT
EXPORTING
    OUTPUT = output
CATCH CX_AI_SYSTEM_FAULT .
ENDTRY.
i want to know the difference between writing just
CATCH CX_AI_SYSTEM_FAULT .
and
CATCH CX_AI_SYSTEM_FAULT . into objref. where objref refers to exception class CX_AI_SYSTEM_FAULT .
how does the system behave in either of the case if exception is raised.
thanks

In both cases, no short dump is generated, because the exception is caught. However, no info about the actual error can directly be retrieved like is done in the following example.
DATA: lv_message TYPE string,
      lr_error TYPE REF TO cx_sy_zerodivide,
      y TYPE i VALUE 10,
      x TYPE i.
TRY.
    x = y / 0.
CATCH cx_sy_zerodivide INTO lr_error.
    lv_message = lr_error->get_text( ).
ENDTRY.
But this will only be of interest, when you want to determine the error. If the catch is part of another TRY ENDTRY block, the error is navigated upwards the chain.
Hope this was somehow helpful.

Exception handling across multiple classes

Is it possible to throw the exception from one class to another class for exception handling.............. can anyone tell me the way........

you don't throw it to a specific class. in the method where the exception might be thrown, you can choose to either catch it there, or just re-throw it so that the method before it can choose to catch it or throw it. you know what a stack trace is? it's a chain of methods that were called to get to a certain point in execution, and if an exception is thrown somewhere, either the method in which it is thrown deals with it, or it passes it back to the preceeding method in the stack
why were you trying to throw it to a class?

ABAP Object Exception handling

Hi all,
I would very much like some links to learn how to create Exception objects, and how to use them in my ABAP Object apps.
Thanks in advance!

hi,
Object Services classes work exclusively with Class-Based Exceptions. The exception classes are predefined in the system. They always start with CX_OS_.
Exceptions of the CX_DYNAMIC_CHECK category are passed to the caller using the RAISING clauses of interface methods; class-specific methods of the class actor; attribute access methods of persistent classes; in the transaction object; and the system actors. Such exceptions must be handled here by the user, (in contrast to the general rule that exceptions should be handled within a procedure, not passed along the hierarchy). Thus semantically, the above exceptions of the CX_DYNAMIC_CHECK category belong more to the CX_STATIC_CHECK category. However, they could not be defined as such, due to their incompatibility with existing Object Services applications.
Check out the link.
http://help.sap.com/saphelp_erp2005vp/helpdata/en/66/bc7aec43c211d182b30000e829fbfe/frameset.htm
http://help.sap.com/saphelp_nw04/helpdata/en/fb/35c62bc12711d4b2e80050dadfb92b/content.htm
http://help.sap.com/saphelp_erp2005vp/helpdata/en/17/f9f44295bae2499e8c56ca4f5106fb/frameset.htm
Regards,
Richa.

What is the best way of Exception Handling

There are many ways to handle the exceptions of a program.
1: You could put every exception handler in the most upper abstract classes.
2: You could write a generic exception that handles a lot of things.
3: You could write exceptions handling for every class.
4: you could group the exception handling.
many, many more ways.
Question:
What is the best way to do this?

There are many ways to handle the exceptions of a program.
1: You could put every exception handler in the most upper abstract classes.I woldn't do that always, a class may throw an exception if it's not able to handle it prperly, but it may, so it'll catch it. Throw Exception in abract supper clases only when you are sure the children must always throw the exception, not processing it.
2: You could write a generic exception that handles a lot of things.Bad idea, if you always throw an Exception no informtaion is provaided about whta happened, if you throw a NullPointerException and a NumberFormatException in one of your method I (for example) colud handle properly both problems, one likely to be a programming erros, so check my cod eagain, and the other like to be an user error in introduciong some data, so I wold ask him for the data again.
3: You could write exceptions handling for every class.Write exceptions handling for every class that needs it, throwing the exception if at that moment you don't know what to do with it and catching it if yes. Do not employ too many exception handling: execution of try catch blcks is slower than usuall code.
4: you could group the exception handling.I don't undestand this. What do youmean?
Question:
What is the best way to do this? In general thrwo an exception if the class dosen't know what to do with it, the class which uses yours will handle it. If your class can process the Exception, don't be too lazy and process it.
Employ exceptions always you need them but do not abuse; thay made the code execute slower because the VM machine must check if the exceotions occur.
abraham.

ABAP OO Exception Class based processing

Hi there,
Note: In going forward with SAP WAS 6.20, one can handle exceptions using exception-class based handling using RAISE EXCEPTION TYPE abc and then CATCHing it in TRY/ ENDTRY block. Standard method like GET_TEXT can be used to get the text of the exception raised.
Question: If I know the EXCEPTION CLASS and Exception ID of my exception class, is it possible to get the exception text directly from the repository without creating the exception class object?
E.g. Exception class is CX_MY_SECRET_ID
and Exception IDs for this class are
ID_NOT_FOUND,
ID_EXPIRED,
ID_IS_FOR_SPECIAL_ACCESS
E.g. last two exception IDs are my warning conditions, and if such conditions are encountered, all I want to do is collect the warning messages. Whereas first exception ID condition (i.e. ID_NOT_FOUND) is an error for me, in which I have pass the exception back to the calling program. E.g. the source code is like this:
PERFORM.......
If ID is not found
   RAISE EXCEPTION TYPE ZCX_MY_SECRET_ID
           EXPORTING TEXT_ID = 'ID_NOT_FOUND'.
else if ID has expired
   ...... then do I have to first raise the exception like raised above and CATCH it before I can get its text? or can I get the exception text directly without raising the exception first (as I know i have to retrieve the text of ZCX_MY_SECRET_ID exception class for Exception ID ID_EXPIRED)?
In other words, if a certain condition is warning for my program, can I get the condition for that exception (from exception class based on exception class name and exception ID), or do I have to RAISE it first explicitely and then CATCH it immediately and execute GET_TEXT to get its text?
Thanks very much!

Hello Chetan,
in basic the raise exception type xxx creates an exception object and aborts normal execution continuing in the enclosing try block.
The Abap runtime contains some optimizations regarding the try block has an 'into xxx' clause. But as long you use the same exception you cant take benefit from this.
As far I understand your problem you have two different kind of severities. So I would use 2 differnt exception classes (maybe derived from a common parent type, or one from the other).
So both the code which throws the exception and the one which catches it are aware of the different semantic.
Kind Regards
Klaus

Exception Handling for many bean objects of a container class in a JSP page

Hello,
I have on container bean class. In this container class, there are several others class objects and the getter methods to get these objects out to the JSP pages.
I have one JSP page which will use different objects in the container class object through the getter methods of the container class.
My question is how to implement the exception handler for all the objects in the container so that the JSP page can handle all exceptions if occurrs in any object in the container?
Please give me some suggestions. Thanks
Tu

Thanks for your reply.
Since the container is the accessor class, I have no other super class for this container class, I think I will try the try catch block in the getter methods.

Exception Handling Standards -The exception Exception should never been thrown. Always Subclass Exception and throw the subclassed Classes.

In the current project my exception handling implementation is as follows :
Exception Handling Layer wise :
DL layer :
catch (Exception ex)
bool rethrow = ExceptionPolicy.HandleException(ex, "Ui Policy");
if (rethrow)
throw;
BL Layer
catch (Exception ex)
bool rethrow = ExceptionPolicy.HandleException(ex, "Ui Policy");
if (rethrow)
throw;
UI Layer
catch (Exception ex)
bool rethrow = ExceptionPolicy.HandleException(ex, "Ui Policy");
if (rethrow)
Response.Redirect("ErrorPage.aspx", false);
We have a tool to check the standards. And tool output is as follows :
The exception Exception should never been thrown. Always Subclass Exception and throw the subclassed Classes.
I need suggestions on how to implement the same according to standards.

Your tool is wrong if it says to never throw Exception. This was a common recommendation back in the .NET v1 days but has long since been thrown out. It is perfectly fine to use Exception when you have a general exception that provides no information
that an application can use to make an informed opinion.
The general rules of exception throwing is to throw the most specific exception that makes sense. If there is no specific exception that applies and it would be useful for the caller to handle the exception differently than other exceptions then creating
a custom exception type is warranted. Otherwise throwing Exception is reasonable. As an example you might have an application that pulls back product data given an ID. There is no built in exception that says the ID is invalid. However an invalid ID
is something that an application may want to handle differently than, say, an exception about the product being discontinued. Therefore it might make sense to create an ItemNotFoundException exception that the application can react to.
Conversely there is no benefit in having different exception types for disk full and disk quota met. The application will respond the same in either case.
Michael Taylor
http://blogs.msmvps.com/p3net

Class based exception handling

Similar Messages

Maybe you are looking for