[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
[ecf-dev] file-based discovery


perhaps we might deprecate the old discovery mechanism in ECF 3.5 and
remove it in ECF 3.6. This would give consumers more time to migrate
their code.

Anyway, if this way would be to extensive, +1 for the removal in ECF 3.5.



On Friday, February 4, 2011, Wim Jongman <wim.jongman@xxxxxxxxx> wrote:
> +1
> On Wed, Feb 2, 2011 at 5:12 AM, Markus Alexander Kuppe <ecf-dev_eclipse.org@xxxxxxxxxxx> wrote:
> On 02/01/2011 06:29 PM, Scott Lewis wrote:
>> Hi Folks,
>> Part of the Remote Service Admin specification is what is called the
>> 'Endpoint Description Extender Format'.    This is essentially an xml
>> file format that allows Endpoint Descriptions (meta-data about a remote
>> service endpoint) to be read from an xml file.  See [1] section 122.8.
>> This endpoint description extender format essentially replaces (and
>> obsoletes) the ECF file-based discovery [2].   This old format came from
>> the work on RFC 119 in the OSGi enterprise experts group, and the
>> implementation that ECF has been using was originally contributed by
>> some great folks at Siemens...along with some integration work by Markus
>> and myself.
>> In moving over to use/support RSA, I would like to *remove* the current
>> file-based discovery mechanism from ECF 3.5...in favor of the standard
>> Endpoint Description Extender Format (which is supported by the new ECF
>> RSA implementation code).  This will mean that existing ECF remote
>> service clients that use the old file-based discovery would need to move
>> to using the new EDEF format, and that the old version of those ECF
>> implementation plugins will not work with ECF 3.5 remote services (which
>> is now based upon RSA [3]).
>> I think it will be better to eliminate the support for the old format in
>> favor of the OSGi RSA standard format, rather than keeping both
>> active/available.  Although this will imply some work by everyone
>> involved (i.e. me/us/ECF contributors as well as consumers), it doesn't
>> seem feasible to me to continue supporting the old (non-standardized)
>> format when a standard exists and is now fully supported by ECF.  Also,
>> I don't think it's going to be possible to continue for me or other ECF
>> committers to support the code that reads/manipulates the old format.
>> So, first...any comments about this decision (i.e. to remove support for
>> the old format from ECF remote services...in favor of the EDEF that's in
>> the RSA implementation)?
>> Second...at a certain point it will make sense to remove this project
>> from the ECF hudson build:
>> https://build.ecf-project.org/hudson/job/C-HEAD-discovery.local.feature/
>> because this feature (discovery local) is no longer necessary...and in
>> fact some of the dependencies that it has (on the old ECF remote
>> services implementation) are now gone...meaning that this code won't
>> continue to compile against master (that's why it's failing now).  So in
>> essence what I would like to do is remove this project from the Hudson
>> build.
>> Thanks,
>> Scott
>> [1] http://www.osgi.org/download/r4v42/r4.enterprise.pdf
>> [2] http://wiki.eclipse.org/File-based_Discovery
>> [3] ECF RSA support/plan bug:
>> https://bugs.eclipse.org/bugs/show_bug.cgi?id=324215
> +1 unless somebody from the community steps up and a) fix the current
> compile errors b) agrees to maintaine the code base in the future.
> Markus
> _______________________________________________
> ecf-dev mailing list
> ecf-dev@xxxxxxxxxxx
> https://dev.eclipse.org/mailman/listinfo/ecf-dev