Does anyone have an example of how to use regexsplit

[Leonard Kroll]

Hi all,

I am having problems getting regexsplit to work.
The problem is with the regular expression specified.
I have a title attribute that looks like Faculty/pt and I want to pull off just the Faculty portion.
I have tried a regex of ^(Staff|Student|Faculty) and it only works if the title is "Faculty". (exact match)
If the title is "Faculty/pt" the regular expression does not match.

The attribute values for title can be Student, Student/pt, Staff, Faculty and Faculty/pt, Faculty/do

Also tried regex="^(\w*)"

Not sure what to try next. 
Here is what my attribure defs look like:

    <resolver:AttributeDefinition xsi:type="ad:Scoped" id="eduPersonScopedAffiliation" scope="umb.edu" 
						sourceAttributeID="SPLIT_TITLE">
        <resolver:Dependency ref="SPLIT_TITLE" />
        <resolver:AttributeEncoder xsi:type="enc:SAML1ScopedString" 
				name="urn:mace:dir:attribute-def:eduPersonScopedAffiliation" />
        <resolver:AttributeEncoder xsi:type="enc:SAML2ScopedString" name="urn:oid:1.3.6.1.4.1.5923.1.1.1.9" 
				friendlyName="eduPersonScopedAffiliation" />
    </resolver:AttributeDefinition>

<!-- -->

<resolver:AttributeDefinition xsi:type="RegexSplit" xmlns="urn:mace:shibboleth:2.0:resolver:ad"
                              id="SPLIT_TITLE"
                              sourceAttributeID="title"  
		      	regex="^(Student|faculty|Faculty|Staff)" >

     <resolver:Dependency ref="myLDAP" />



Christopher, thank you for getting me this far.
I could not have done it without you.

~lennyK

-----Original Message-----
From: users-bounces at shibboleth.net [mailto:users-bounces at shibboleth.net] On Behalf Of Christopher Bongaarts
Sent: Tuesday, September 25, 2012 4:12 PM
To: users at shibboleth.net
Subject: Re: Does anyone have an example of how to use regexsplit

On 9/25/2012 12:52 PM, Leonard Kroll wrote:
>
> Hi, I need to split the title attribute. I want to pull "Student", "Staff" and "Faculty" off the beginning of the attribute.
> i.e. "Faculty/pt", "Staff/ne", "Student/li"
>
> Does anyone have an example  of regexsplit showing how to use it?

No, but might be able to give some suggestions...

> <resolver:AttributeDefinition xsi:type="ad:Scoped" id="eduPersonScopedAffiliation" scope="umb.edu"
>                                                            
> sourceAttributeID="title">

I think you want sourceAttributeID="SPLIT_TITLE" if you want your ePSA to be based on the result of the splitting rather than the raw title attribute.

>          <resolver:Dependency ref="myLDAP" />

I think you have your dependencies backward if you want ePSA based on the first part of the title.  This one would be ref="SPLIT_TITLE".

> <resolver:AttributeDefinition xsi:type="RegexSplit" xmlns="urn:mace:shibboleth:2.0:resolver:ad"
>                                id="SPLIT_TITLE"
>                                sourceAttributeID="title"
>                                regex="[Student][Faculty][Staff]">

You need a Java 5 regex, with the first match group assigned as the value... something like:

   regex="^(Student|Faculty|Staff)"

>       <resolver:Dependency ref="eduPersonScopedAffiliation" />

This one would ref="myLDAP".

-- 
%%  Christopher A. Bongaarts   %%  cab at umn.edu          %%
%%  OIT - Identity Management  %%  http://umn.edu/~cab  %%
%%  University of Minnesota    %%  +1 (612) 625-1809    %%
--
To unsubscribe from this list send an email to users-unsubscribe at shibboleth.net