<html>
<head>
<meta content="text/html; charset=windows-1252"
http-equiv="Content-Type">
</head>
<body bgcolor="#FFFFFF" text="#000000">
<p><br>
</p>
<br>
<div class="moz-cite-prefix">On 1/11/17 8:07 PM, Cantor, Scott
wrote:<br>
</div>
<blockquote cite="mid:9111C59E-BC8E-4144-9ED3-8D8D25F4257C@osu.edu"
type="cite">
<pre wrap="">On 1/11/17, 7:56 PM, "users on behalf of Brent Putman" <a class="moz-txt-link-rfc2396E" href="mailto:users-bounces@shibboleth.netonbehalfofputmanb@georgetown.edu"><users-bounces@shibboleth.net on behalf of putmanb@georgetown.edu></a> wrote:
</pre>
<blockquote type="cite">
<pre wrap="">But I'm not clear how we'd support user-configuration of the "trigger" for any given instance of logging.
</pre>
</blockquote>
<pre wrap="">
Probably the same way I did the timers, with a scriptable trigger using a context to store the rules. </pre>
</blockquote>
<br>
I never looked at that in detail. I guess context data can drive
it, but then the question is: what data...<br>
<br>
Within the action, one does have access to all the SWF
RequestContext stuff. So I guess theoretically many things are
possible from that, like using the current flow ID (from the active
FlowDefinition), and current state ID (from the current
StateDefinition), as well as of course all the various -Scope data.
<br>
<br>
So maybe outside-in could be triggers based on (flow ID + state ID),
injected either into context tree or as flow/conversation scope
data.<br>
<br>
Haven't tried it, but that might be generally useful for other
action trigger-y things as well.<br>
</body>
</html>