Re: Safe to use the colon character in the path segment of URL identifiers?


I only have experience with the functionality of encoders which
totally encode all reserved characters in the strings you give them.
Have you found any encoders which actually split the URL up before
encoding it and then reassembling it?

It would be nice to see what you find out about different
implementations of the split encode and combine following the
contextual reserved character rules in the RFC.



2009/5/4 Peter Krantz <>:
> Hi!
> We are in the process of identifying legal information resources in
> RDF. Each resource has a property that identifies it uniquely in this
> domain of the form YYYY:NN (e.g. "2006:45"). We'd like to reuse this
> property in the url-path segment when constructing URI:s for these
> resources since the process and organization behind it guarantees its
> uniqueness.
> According to RFC 1738 [1] the colon character is OK in the url-path,
> but since it is one of the reserved characters for other segments I
> guess there could be some libraries that erronously would escape it.
> Has anyone on this list had any trouble with using the colon character
> in URL identifiers in semweb (or other) toolsets?
> Regards,
> Peter
> [1]:

Received on Monday, 4 May 2009 22:08:06 UTC