So, the question is: should I start developing the <trycatch> task and
also leave the onFail part for targets?

Have anybody really analyzed the way in which I've done onfail -- do
you have some claims on its implementation?

Also, how should I address the problem of getting log only for the
part really included in the <try> block? (I'm still hadn't enough time
to dive that deep into nant design, probably it's easy, I'm just not
sure in that)

MA> I like onFail attribute slightly better then current nant.onfailure. Setting
MA> some special property looks little weird for me. But it is enough for most
MA> projects, I think.

MA> Martin

>> Hello Gert,
>>
>> such a task does really makes sense, still, I feel we have two
>> (a bit) different approaches to solve the problem.
>>
>> There is one "pro" for my approach -- it is possible to add "onfail"
>> attribute to a target element, thus, if target B depends on A, the B
>> gets executed and A fails during dependant targets execution, we can
>> do something on A failure:
>>
>> <project default="B">
>>   <target name="B" depends="A">
>>     ...
>>   </target>
>>   <target name="A" onfail="A.failed">
>>     ...
>>   </target>
>>   ...
>> </project>
>>
>> Probably, there is some sense in combining both of the approaches:
>> implementing <trycatch> task and leaving "onfail" attribute for target
>> only (I'd added it for task first and the next day found out, that I
>> need to handle the case similar to the one I've described above, so I
>> added it for target also).
>>
>> Anyway, whatever the final idea of the solution will be, I'm ready to
>> participate in its development.
>>

-- 
Best regards,
 Ivan                            mailto:[EMAIL PROTECTED]



-------------------------------------------------------
This SF.net email is sponsored by: SF.net Giveback Program.
SourceForge.net hosts over 70,000 Open Source Projects.
See the people who have HELPED US provide better services:
Click here: http://sourceforge.net/supporters.php
_______________________________________________
nant-developers mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/nant-developers

Reply via email to