Andrew Borley wrote:
Hi everyone,
I have posted candidates for the C++ SCA and SDO M2 release here:
http://people.apache.org/~ajborley/cpp-1.0-incubator-M2-RC1
Please take a look, read the docs, build & run the libraries and
samples & let us know how you find it so that we can either re-spin
the release or vote on it asap.
The website documentation is currently based on the M1 release and
will be re-written to sync with what is in the release.
About Tuscany SCA C++
===================
Tuscany SCA C++ provides a runtime implementation for the for the Service
Component Architecture 0.96 Assembly specification and the 0.95 C++
Client
& Implementation Model specification, written in C++ and will currently
support C++, Python and Ruby component implementation types.
The specifications can be found at
http://www.osoa.org/display/Main/Service+Component+Architecture+Specifications
It is possible to interoperate with Tuscany SCA Java via the Web
Service bindings
for services and references but restrictions apply. This is not yet a
complete
implementation and known restrictions are described below.
Supported SCA Assembly Model features
* All features are supported unless listed under the known restrictions
below. See SCA Assembly Model specification.
Supported language bindings
* Component implementations written in C++. See the SCA C++ Client and
Implementation Model specification.
* Component implementations written in Python. See the
doc/PythonExtension.html
documentation.
* Component implementations written in Python. See the
doc/RubyExtension.html
documentation.
* Component interfaces described by C++ classes. See SCA Client and
Implementation Model specification.
Supported service and reference bindings
* The web service binding is supported. This implementation will support
web services which using document literal SOAP bindings conforming
to the
WS-I basic profile (rpc/encoded is not yet supported).
Known restrictions
* Local service interfaces cannot use overloaded operations (the SCA
specification limits remote service interfaces to not using overloaded
operations).
* Each WSDL definition for a web service binding must be in a single
WSDL
document.
* No load time validation of the deployed SCA application (run time
validation only).
* No metadata API.
About Tuscany SDO for C++
=========================
Tuscany SDO is an implementation of the Service Data Objects 2.01
specification for C++ developers found at
http://www.osoa.org/display/Main/Service+Data+Objects+Specifications
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
Looks good! I tested the RC1 distribution on a Redhat Enterprise Linux 4
and everything worked smoothly.
Also reviewed all the docs and structure of the distributions and only
found a few issues that should be easy to fix. A few HTML files are
missing Apache license headers, a few files should be excluded from the
distros. I found minor issues in the various readmes and docs, but most
are really just suggestions.
Everything I found is in the following JIRAs:
http://issues.apache.org/jira/browse/TUSCANY-842
http://issues.apache.org/jira/browse/TUSCANY-843
http://issues.apache.org/jira/browse/TUSCANY-844
http://issues.apache.org/jira/browse/TUSCANY-845
http://issues.apache.org/jira/browse/TUSCANY-847
http://issues.apache.org/jira/browse/TUSCANY-848
http://issues.apache.org/jira/browse/TUSCANY-849
http://issues.apache.org/jira/browse/TUSCANY-850
http://issues.apache.org/jira/browse/TUSCANY-851
http://issues.apache.org/jira/browse/TUSCANY-852
http://issues.apache.org/jira/browse/TUSCANY-853
http://issues.apache.org/jira/browse/TUSCANY-854
Overall I was really impressed with the quality and the flow of the
Readmes, GettingStarted and the User guide type docs that you added for
the new Ruby and Python extensions.
The only thing I have not tested yet is the PHP client included the
BigBank sample, I'll try to do this by tomorrow.
Last comment, maybe the high level description you have in this email
could be added to the README? What do you think?
--
Jean-Sebastien
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
