Regex in the SHibboleth2.xml

[Rick Hoodenpyle]

Thank you Christopher. I will try that.



Appreciate it.



Rick



*From:* users <users-bounces at shibboleth.net> *On Behalf Of *Christopher
Bongaarts via users
*Sent:* 29 April 2022 20:12
*To:* users at shibboleth.net
*Cc:* Christopher Bongaarts <cab at umn.edu>
*Subject:* Re: Regex in the SHibboleth2.xml



On 4/29/2022 2:05 PM, Rick Hoodenpyle wrote:

I often have to separate username from the email address using the
following:



<!--<AttributeResolver type="Transform" source="upn">

              <Regex match="^(.+)@(.+)$" <%5e(.+)@(.+)$>
dest="LoginID">$1</Regex>

              <Regex match="^(.+)@(.+)$" <%5e(.+)@(.+)$>
dest="domain">$2</Regex>

</AttributeResolver>-->



I have a client though that passes an ID through the UID attribute. It
begins with a letter K and ends with the numerical ID, example K98456.



The value that this must equate to in my application is a numerical only
value, so I am needing to find out how to remove the K and only pass the
numbers. I have thought about maybe using the following, but have not had a
chance to test it and was wondering if someone out there might already know
how to do this.



<!--<AttributeResolver type="Transform" source="uid">

              <Regex match="^(.+)[0-9](.+)$" dest="kuid">$1</Regex>

</AttributeResolver>-->

Given K followed by numbers, and extracting just the numbers, you would use
something like:

<Regex match="^K([0-9]+)$" dest="kuid">$1</Regex>

-- 

%%  Christopher A. Bongaarts   %%  cab at umn.edu          %%

%%  OIT - Identity Management  %%  http://umn.edu/~cab  %%

%%  University of Minnesota    %%  +1 (612) 625-1809    %%
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://shibboleth.net/pipermail/users/attachments/20220429/862ac236/attachment.htm>