On Fri, Apr 23, 2010 at 7:49 AM, Olle <[email protected]> wrote:
>
> Was "set" intended for setting array elements of process variables?
> Example below tries to set first element of variable 'v1'. It doesn't
> work unlike setting field. Another thing is that setting of process
> variable doesn't parse compound keys.
>
> ---
> require 'rubygems'
> require 'ruote'
> require 'ruote/storage/hash_storage'
>
>
> engine = Ruote::Engine.new(
> Ruote::Worker.new(
> Ruote::HashStorage.new()))
>
> pdef = Ruote.process_definition :name => 'set test' do
> echo 'This works: set f:f1.0'
> set :field => 'f1.0', :value => 'val'
> echo 'f1[0]: ${f:f1.0}'
> echo 'This doesn\'t: set v:v1.0'
> set :var => 'v1.0', :value => 'val'
> echo 'v1[0]: ${v:v1.0}'
> end
>
> wfid = engine.launch pdef, { 'f1' => [] }, { 'v1' => [] }
> engine.wait_for(wfid)
> ------
Hello Oleg,
I will investigate/fix that. I think it's good to have the same behaviour for
process variables and workitem fields.
> Another question is regarding setting process variables inside of
> concurrent expressions. I tried to set variable inside of
> concurrent_iterator, it doesn't work and produce error then run with '-
> d'. If wrap "set" in "sequence" inside of "concurrent_iterator" then
> it just does nothing.
> May be this is a well-known behavior and it's prohibited? I found
> nothing about it in the documentation and tests. So probably it's just
> my inattention.
There are some scoping rules. Mostly subprocesses and concurrent branches get a
new scope.
This works :
---8<---
require 'rubygems'
require 'ruote'
engine = Ruote::Engine.new(Ruote::Worker.new(Ruote::HashStorage.new()))
pdef = Ruote.define do
concurrent_iterator :on => (1..5).to_a do
sequence do
set 'v:v' => '>${v:i}<'
echo '${v:i} nada ${v:v}'
end
end
end
wfid = engine.launch(pdef)
engine.wait_for(wfid)
--->8---
You probably want to do something like
---8<---
pdef = Ruote.define do
concurrent_iterator :on => (1..5).to_a do
sequence do
set 'v:/v' => '>${v:i}<' # <------------------------------ notice the /
echo '${v:i} nada ${v:v}'
end
end
end
--->8---
But then you'll get an output of
1 nada >5<
2 nada >5<
3 nada >5<
4 nada >5<
5 nada >5<
Last 'set' to execute wins.
Try to avoid using process variables in such concurrency cases. Let the merge
do its work.
> And finally, I have a small suggestion regarding "set" syntax.
> Sometimes it needs to append value to array, without care about
> array's index. It could look like this:
> set :field => 'array_field.+', :value => 'some_val'
> Maybe you will find this useful.
Could inc/dec be what you need ?
http://ruote.rubyforge.org/exp/inc.html
Isn't putting too much data manipulation logic in a business process a bit off
? A participant that does all the complex data manipulation in plain ruby is
probably better.
What do you think ?
Best regards,
--
John Mettraux - http://jmettraux.wordpress.com
--
you received this message because you are subscribed to the "ruote users" group.
to post : send email to [email protected]
to unsubscribe : send email to [email protected]
more options : http://groups.google.com/group/openwferu-users?hl=en
