[Nfd-dev] How to treat ".." in an NDN URI?

[Thompson, Jeff]

Hi Junxiao,

Using your example of the URL for RFC3986, the following link works:
http://tools.ietf.org/html/rfcblahblahblah/../rfc3986#section-3.3

So ".." is illegal in a URI, but legal in a URL?  Maybe the support for .. In a URL is non-standard?

- Jeff T


From: Junxiao Shi <shijunxiao at email.arizona.edu<mailto:shijunxiao at email.arizona.edu>>
Date: Tuesday, August 5, 2014 9:27 AM
To: Jeff Thompson <jefft0 at remap.ucla.edu<mailto:jefft0 at remap.ucla.edu>>
Cc: nfd-dev <nfd-dev at lists.cs.ucla.edu<mailto:nfd-dev at lists.cs.ucla.edu>>
Subject: Re: [Nfd-dev] How to treat ".." in an NDN URI?

Hi JeffT

TLV spec cites RFC3986 for URI syntax. The processing of ".." doesn't need to be mentioned in TLV spec because it's inherited from RFC3986.

RFC3986<http://tools.ietf.org/html/rfc3986#section-3.3> says:
The path segments "." and "..", also known as dot-segments, are defined for relative reference within the path name hierarchy.  They are intended for use at the beginning of a relative-path reference to indicate relative position within the hierarchical tree of names.
Therefore, if ".." appears within an absolute ndn URI, the entire URI is invalid and should raise an error.

Yours, Junxiao

On Tue, Aug 5, 2014 at 6:22 AM, Thompson, Jeff <jefft0 at remap.ucla.edu<mailto:jefft0 at remap.ucla.edu>> wrote:
… Right now, both ndn-cxx and ndn-cpp treat ".." as a illegal encoding for a component and drop it. So, "/a/b/../c" simply becomes "/a/b/c". But ndnx (ndnd-tlv) treat ".." as "up one level" and it becomes "/a/c".

The question is whether the TLV specification should spell out the correct behavior.

- Jeff T

From: <Thompson>, Jeff Thompson <jefft0 at remap.ucla.edu<mailto:jefft0 at remap.ucla.edu>>
Date: Tuesday, August 5, 2014 2:51 AM
To: nfd-dev <nfd-dev at lists.cs.ucla.edu<mailto:nfd-dev at lists.cs.ucla.edu>>
Subject: How to treat ".." in an NDN URI?

The TLV specification for the NDN URI scheme says "To unambiguously represent name components that would collide with the use of . and .. for relative URIs, any component that consists solely of one or more periods is encoded using three additional periods.".
http://named-data.net/doc/ndn-tlv/name.html#ndn-uri-scheme

If an NDN URI uses the "relative" value of "..", how should the URI be decoded.  Specifically, should it be treated as "up one level" like in a Unix path?  For example, should the URI "/a/b/../c" be decoded as the name "/a/c"?

(This question comes from an issue on the ndn-js Redmine: http://redmine.named-data.net/issues/1818 ).

Thanks,
- Jeff T

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.lists.cs.ucla.edu/pipermail/nfd-dev/attachments/20140806/3b1ad30d/attachment.html>