Skip to Content.
Sympa Menu

grouper-dev - Re: [grouper-dev] RE: xml export/import v1.6

Subject: Grouper Developers Forum

List archive

Re: [grouper-dev] RE: xml export/import v1.6

Chronological Thread 
  • From: Tom Zeller <>
  • To: Tom Barton <>
  • Cc: "GW Brown, Information Systems and Computing" <>, Chris Hyzer <>,
  • Subject: Re: [grouper-dev] RE: xml export/import v1.6
  • Date: Thu, 21 Jan 2010 10:07:49 -0600
  • Domainkey-signature: a=rsa-sha1; c=nofws;; s=gamma; h=mime-version:sender:in-reply-to:references:from:date :x-google-sender-auth:message-id:subject:to:cc:content-type :content-transfer-encoding; b=HELwSTO7AJMS59xQDnUdWEv7bcwTTHyytYDhAsgF20v9LsDRq+s4p7FPa3SiYWgTkD I7mTixCZ2ybXmh3XLw7FttTDTVN0qSMKGkvy2bgelD20JTGT/UMUUbWWWvSVQobG4Asp vdSsyhQmOk7IY0EMV0D92yyPIo/jwr7T3NAJ0=

> GW Brown, Information Systems and Computing wrote:
>> --On 21 January 2010 09:28 -0600 Tom Barton
>> <>
>> wrote:
>>> Chris Hyzer wrote:
>>>>> Alternatively, would it make sense to include an attribute of
>>>>> uuid-bearing child elements that contain the name of the object being
>>>>> referenced? Eg,
>>>> Well, if it were easy, I might say yes.  But I didn't want to mess
>>>> with the default Xstream serialization.  Plus, I also didn't want
>>>> people to rely on this either i.e. start using it for something,
>>>> since the xml export isn't for provisioning, right?  :)
>>> Understood. It's just that my sanity-checker fires up when I see us
>>> mapping structured data into comments.
>>> Gary, who were you anticipating would use those comments, humans or
>>> automatons?
>> My first concern is being able to understand/troubleshoot the data without
>> having to lookup all those uuids. Comments would not be very friendly for
>> machine parsing but we are only expecting Grouper to parse the file and it
>> doesn't need the data here.
>> I would suggest that we go with comments for now and wait and see if
>> anyone requests additional attributes / tags.
> I have no problem with this.
> Tom

Could the comments be written as valid xml nodes, rather than <!--
-->, so that the same parser used to read the machine data could read
the comments as well ? I'm thinking that's easy enough to do in code
and may be easier for folks to read using tools other than their eyes.
I do realize we don't want to encourage using those comments for
"real" things, though.


Archive powered by MHonArc 2.6.16.

Top of Page