Can we call the option "SupportBuggyDotNetArrays" :-)
+0
-Dug
Glen Daniels <[EMAIL PROTECTED]> on 02/20/2003 02:21:02 PM
Please respond to [EMAIL PROTECTED]
To: "'[EMAIL PROTECTED]'" <[EMAIL PROTECTED]>
cc:
Subject: VOTE : Should we do this patch? (was RE: Request that zero
length array interop patch be applied in ti me f or 1.1 final)
OK committers, time to step up and decide this thing.
Please vote +1 if you think we should reopen the bug and fix it by applying
the patch, -1 if you think we shouldn't, and +/-0 if you're ambivalent.
I myself am for including the patch, since it doesn't change our default
behavior, and does allow people who are having problems with buggy .NET
installations to get up and running, which at the end of the day is what
we're really hoping for anyway. We should make sure there are comments
describing why it's there, but I see no harm in including it.
So: +1
--Glen
> -----Original Message-----
> From: David Marquard [mailto:[EMAIL PROTECTED]]
> Sent: Thursday, February 20, 2003 1:47 PM
> To: '[EMAIL PROTECTED]'; '[EMAIL PROTECTED]'
> Subject: RE: Request that zero length array interop patch be
> applied in
> ti me f or 1.1 final
>
>
> It is fixed in .Net 1.1 beta, but the only shipping,
> supported version of
> .Net is 1.0 at this point. .Net 1.1 final won't be released
> for quite some
> time yet.
>
> The problem still exists because the only shipping, supported
> version of
> .Net can't communicate with an axis server at this point, and
> still won't be
> able to in the near- to mid-future.
>
> Please reconsider the decision to mark the bug invalid. Even
> though the bug
> is in a 3rd party client, the patch fixes a problem that
> users (like myself)
> will encounter in "real world" usage. The patch is still
> extremely small and
> well contained (the core of which is about a 4 line change to
> one method),
> and provides a big interop win without sacraficing axis correctness.
>
> Dave
>
> -----Original Message-----
> From: James M Snell [mailto:[EMAIL PROTECTED]]
> Sent: Thursday, February 20, 2003 10:21 AM
> To: [EMAIL PROTECTED]
> Subject: RE: Request that zero length array interop patch be
> applied in
> ti me f or 1.1 final
>
>
> Hmm.. ok, I was reluctant to do this anyway ... and since
> it's been fixed
> in .NET 1.1, then I retract my original "this sounds
> reasonable" and give
> it a -1.
>
> - James Snell
> IBM Emerging Technologies
> [EMAIL PROTECTED]
> (559) 587-1233 (office)
> (700) 544-9035 (t/l)
> Programming Web Services With SOAP
> O'Reilly & Associates, ISBN 0596000952
>
> Have I not commanded you? Be strong and courageous.
> Do not be terrified, do not be discouraged, for the Lord your
> God will be with you whereever you go. - Joshua 1:9
>
>
>
> [EMAIL PROTECTED]
> 02/20/2003 07:29 AM
> Please respond to axis-dev
>
>
> To
> [EMAIL PROTECTED]
> cc
>
> bcc
>
> Subject
> RE: Request that zero length array interop patch be applied
> in ti me
> f or 1.1 final
>
>
>
> I would be in favour of this.
> Note that this is fixed in .NET 1.1, when this patch is
> applied I would
> propose to add a note to bugzilla for removal from the next
> Axis release
> after .NET 1.1 is out. This to avoid accumulating bloat.
>
> Thanks,
>
> - ArentJan
>
> -----Original Message-----
> From: David Marquard [mailto:[EMAIL PROTECTED]]
> Sent: woensdag 19 februari 2003 23:46
> To: '[EMAIL PROTECTED]'
> Subject: Request that zero length array interop patch be
> applied in time
> f or 1.1 final
>
>
> I'd like to request that the patch in bug #17021
> (http://nagoya.apache.org/bugzilla/show_bug.cgi?id=17021) be
> applied in
> time
> for the 1.1 final release.
>
> Currently, .Net 1.0 clients cannot deserialize objects from axis that
> contain zero length arrays due to a bug in Microsoft's deserializer.
> Basically, .Net 1.0's deserializer goes belly up if a zero
> length array
> doesn't have an explicit closing tag (i.e., <myArray/> fails, but
> <myArray></myArray> works).
>
> The patch adds a new configuration option to axis called
> "sendMinimizedElements", which when set to "false" causes
> axis to always
> send explicit closing tags for empty elements. This cleanly
> works around
> the
> Microsoft bug, but does not sacrafice axis's correctness. The default
> value
> of the new option is "true", which does not change axis's current
> behavior.
>
> The patch is extremely small (< 20 lines changed) and simple.
> It does not
> change axis's behavior unless a user explicitely sets the new
> configuration
> option to "false". All in all, it's a very safe patch that
> provides big
> bang
> for the buck on the interop side of things.
>
> Dave
> _________________________________________
> Dave Marquard ([EMAIL PROTECTED])
>
