2010-04-27 10:18:53 utc |
irfn |
Hi guys |
2010-04-27 10:19:03 utc |
tosch_le |
hi! |
2010-04-27 10:19:10 utc |
irfn |
i had a question on Storage Participants |
2010-04-27 10:19:20 utc |
irfn |
hi tosch_le |
2010-04-27 10:19:31 utc |
tosch_le |
what's your question? |
2010-04-27 10:20:10 utc |
irfn |
so it looks like storage participant is meant for catchall scenario where a user intervention is required |
2010-04-27 10:20:28 utc |
tosch_le |
yeah. |
2010-04-27 10:20:58 utc |
tosch_le |
but you don't necessarily need a catchall |
2010-04-27 10:21:05 utc |
irfn |
i understand |
2010-04-27 10:21:30 utc |
irfn |
but i want a participant that only needs user intervention on an error |
2010-04-27 10:21:43 utc |
irfn |
so normally it works unattended |
2010-04-27 10:22:16 utc |
irfn |
but in a particular error scenario |
2010-04-27 10:22:24 utc |
irfn |
i want user intervension |
2010-04-27 10:24:16 utc |
tosch_le |
do you mean an error within the business (unnoticed by ruote) or within the process execution (triggered by an exception)? |
2010-04-27 10:24:50 utc |
tosch_le |
in the latter case, i would suggest to use :on_error: http://ruote.rubyforge.org/common_attributes.html#on_error |
2010-04-27 10:25:57 utc |
irfn |
hmm |
2010-04-27 10:26:08 utc |
tosch_le |
in the first case, you should set a field in the workitem which shows the state (error/no error) and proceed to the on_error subprocess if the field is set |
2010-04-27 10:26:27 utc |
irfn |
ok |
2010-04-27 10:26:52 utc |
irfn |
hmm the only problem there is |
2010-04-27 10:27:16 utc |
irfn |
Generic error handling |
2010-04-27 10:27:37 utc |
irfn |
so on every step of wworkflow i want this feature |
2010-04-27 10:27:37 utc |
tosch_le |
you could even combine both by triggering an error within the process: http://ruote.rubyforge.org/exp/error.html |
2010-04-27 10:28:20 utc |
tosch_le |
wrap a sequence around your workflow and set the :on_error on that sequence |
2010-04-27 10:28:23 utc |
irfn |
what i mean is every step can have user intervension |
2010-04-27 10:28:30 utc |
irfn |
ah |
2010-04-27 10:30:11 utc |
irfn |
that makes sense |
2010-04-27 10:30:39 utc |
irfn |
will this hold for nested flow expressions |
2010-04-27 10:31:09 utc |
irfn |
i mean on_error becomes catch all for that block |
2010-04-27 10:31:10 utc |
irfn |
rt? |
2010-04-27 10:31:14 utc |
tosch_le |
note that :on_error will only be effective if the process is forced into an error -- either when there's an (ruby) exception or the error expression is triggered |
2010-04-27 10:31:29 utc |
irfn |
ok |
2010-04-27 10:31:31 utc |
tosch_le |
sorry, i |
2010-04-27 10:31:57 utc |
tosch_le |
'm no expert on that. can't tell you for sure the on_error will catch for all that block |
2010-04-27 10:32:06 utc |
tosch_le |
but i strongly assume that. |
2010-04-27 10:32:09 utc |
irfn |
oh ok |
2010-04-27 10:32:14 utc |
irfn |
ill test that out |
2010-04-27 10:32:27 utc |
irfn |
thanks for your help |
2010-04-27 10:32:33 utc |
tosch_le |
you should wait for john for a definitive answer or -- even better -- post your question to the ml |
2010-04-27 10:32:40 utc |
irfn |
i will |
2010-04-27 10:32:44 utc |
tosch_le |
never mind, you're welcome! |
2010-04-27 10:32:50 utc |
irfn |
:) |