Mailing-List: contact felix-dev-help@incubator.apache.org; run by ezmlm
Precedence: bulk
Reply-To: felix-dev@incubator.apache.org
Received-SPF: pass (herse.apache.org: domain of chris.custine@gmail.com
 designates 64.233.162.232 as permitted sender)
DomainKey-Signature: a=rsa-sha1; c=nofws;
        d=gmail.com; s=beta;
        h=received:message-id:date:from:to:subject:in-reply-to:mime-version:content-type:references;
        b=YAE2V9MBp7QWlCGQeH9TjnVxJoiByamdRlNBzb8gDyTMnukKLOpbD9ssn3QNT3+TG463zfnAuArTW8QDQhUmmeaS2HkzYFR0lobJ10nJpx8UsF6wohEVOziwPUbkxLzvcAdvXYgS96kZAhdRmN77UykCg/aIzdEcoTiK//u3BII=
Message-ID: <43b026c70701290540o570309a4v4cb2bae0009282a6@mail.gmail.com>
Date: Mon, 29 Jan 2007 08:40:57 -0500
From: "Chris Custine" <chris.custine@gmail.com>
To: felix-dev@incubator.apache.org
Subject: Re: bundleContext.installBundle()
In-Reply-To: <ae03024e0701290113g73c829e1n9aede01e41ae3e2f@mail.gmail.com>
MIME-Version: 1.0
Content-Type: multipart/alternative;
	boundary="----=_Part_106617_4011889.1170078057712"
References: <43b026c70701282048k3fad0243geb3db38d4f40531e@mail.gmail.com>
	 <ae03024e0701282321t30521809s39d003f09a9a81bc@mail.gmail.com>
	 <487a994c0701290039r5208e4edi73ed1551bd88858d@mail.gmail.com>
	 <ae03024e0701290041m2c85fc84sbb13b822e47f1ea8@mail.gmail.com>
	 <487a994c0701290101y304447e2wb35b68c7f2f2b907@mail.gmail.com>
	 <ae03024e0701290113g73c829e1n9aede01e41ae3e2f@mail.gmail.com>

------=_Part_106617_4011889.1170078057712
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

Outstanding guys!  I love it when something gets fixed while I sleep  ;-)

Thanks,
Chris

On 1/29/07, Felix Meschberger <Felix.Meschberger@day.com> wrote:
>
> Thanks alot.
>
> Regards
> Felix
>
> On 1/29/07, Karl Pauls <karlpauls@gmail.com> wrote:
> > > The problem is, that no matter the URLStreamHandler, if there is no
> > > scheme - either from the context or the url string - the URL
> > > constructor fails. And this - at least in my case - is the situation
> > > here ...
> >
> > Interesting. I was missing the case where we have an invalid scheme.
> > Should be fixed now.
> >
> > regards,
> >
> > Karl
> >
> > > Regards
> > > Felix
> > >
> > > On 1/29/07, Karl Pauls <karlpauls@gmail.com> wrote:
> > > > Hello,
> > > >
> > > > this is really strange since we do pass in a fake URLHandler but
> with
> > > > an empty context. Maybe that is what causing the trouble. Anyways,
> I'm
> > > > going to remove this asap (it was a temporary placeholder anyways).
> > > >
> > > > regards,
> > > >
> > > > Karl
> > > >
> > > > On 1/29/07, Felix Meschberger <Felix.Meschberger@day.com> wrote:
> > > > > Hi,
> > > > >
> > > > > This seems to have only been introduced lately into  Felix. I use
> the
> > > > > same API call and just put a name there. To fix, I just prepend
> the
> > > > > string "inputstream:" (not a good name after all :-) ), which
> works.
> > > > >
> > > > > The API spec just says, the location is "used as the identity of
> the
> > > > > bundle". That is each bundle has  its unique location name. The
> > > > > BundleContext.installBundle(String location) spec says, that the
> > > > > bundle is retrieved from the named location in a Framework
> dependent
> > > > > manner and that the location string is "typically in the form of a
> > > > > URL".
> > > > >
> > > > > All in all, I can live with the situation of the location string
> to be
> > > > > an URL. On the other hand as the spec does not require it, Felix
> > > > > should probably handle the case of a missing scheme name properly
> when
> > > > > creating the code source URL. Perhaps with something like this :
> > > > >
> > > > >       URL dummy = new URL(null, "location:", fakeHandler); //
> might by Constant
> > > > >       URL codeSource = new URL(dummy, locationString,
> fakeHandler);
> > > > >
> > > > > Regards
> > > > > Felix
> > > > >
> > > > > On 1/29/07, Chris Custine <chris.custine@gmail.com> wrote:
> > > > > > Need a little insight from others here...  When installing a
> bundle from an
> > > > > > InputStream using the form:
> > > > > >
> > > > > > bundleContext.installBundle(String location, InputStream is)
> > > > > >
> > > > > > does the location String have to be a valid URL string?  I have
> a couple of
> > > > > > scenarios where existing code running on Equinox and
> Knopflerfish handles
> > > > > > this fine, but in Felix we are using that String to create a
> valid URL
> > > > > > object for the CodeSource constructor at Felix.java:2921 so we
> get a
> > > > > > MalformedURLException at SecureAction.java:161
> > > > > >
> > > > > > I am just trying to figure out if it is pure luck that Equinox
> and KF work
> > > > > > with bogus location Strings (only when installing from
> InputStream) or if it
> > > > > > is something that we need to look at in Felix.  I have run into
> this twice
> > > > > > this week, once with a client's code and once in Spring-OSGi
> test harness.
> > > > > >
> > > > > > Chris
> > > > > >
> > > > > >
> > > > >
> > > >
> > > >
> > > > --
> > > > Karl Pauls
> > > > karlpauls@gmail.com
> > > >
> > >
> >
> >
> > --
> > Karl Pauls
> > karlpauls@gmail.com
> >
>

------=_Part_106617_4011889.1170078057712--