Opened 10 years ago
Last modified 10 years ago
#164 new
sliver_id should not use #
Reported by: | ahelsing@bbn.com | Owned by: | somebody |
---|---|---|---|
Priority: | minor | Milestone: | |
Component: | AM | Version: | SPIRAL5 |
Keywords: | Cc: | ||
Dependencies: |
Description
The sliver_id
field is not supposed to include a '#'. The current spec says only alphanumeric plus hyphen. (http://groups.geni.net/geni/wiki/GeniApiIdentifiers#Name)
The adopted AM API draft changeset allows underscore or '.' as well. Use one of those if you need another separator.
See http://groups.geni.net/geni/wiki/GAPI_AM_API_DRAFT#ChangeSetO1:Allowothercharactersinslivernames
Change History (2)
comment:1 Changed 10 years ago by
comment:2 Changed 10 years ago by
We chose not to include it because that is 1 of the characters that must be converted when translating from a publicid to a URN (RFC 3151). It isn't impossible to allow it - it just gets harder for tools & aggregates that want to convert.
I don't feel strongly about this one (which is why I left it at 'minor' priority, since 'trivial' always sounds to me like I'm presuming that it is easy to do).
For now, can you email me & Jonathon Duerig <duerig@flux.utah.edu> explaining why allowing '#' in the sliver URN would really really make your life easier? I suspect we can just do this as a change for APIv4 - that you will have already done.
Doing this will threaten the ability to do proper ndl-rspec conversions. I don't believe '#' is disallowed in urns standards?