subversion-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Kylo Ginsberg <k...@kylo.net>
Subject Re: svnsync "207 Multi-Status" failure with http access method and property "svn:"
Date Tue, 05 Jan 2010 16:28:46 GMT
So one bug/enhancement here is to improve client reporting when it
encounters a 207.

But back to the root of my original problem: I can create a "svn:"
property via the svn access method without any errors which then
results in errors when accessed via the http access method.  That
seems like a bug, yes?  But is the problem that the svn access method
*allows* this bogus property, or that http chokes on it?

I'd be willing to look into a patch if someone more knowledgeable
agrees on what the bug is and can provide some pointers on where to
start.

Kylo

On Tue, Jan 5, 2010 at 5:53 AM, C. Michael Pilato <cmpilato@collab.net> wrote:
> Sure.  Ideally, the client would be able to say, "I had trouble parsing the
> XML in the response, namely this bit here:  <S:>asdf</S:>"
>
> Wanna dive in and look into a patch?
>
> Engebakken Geir wrote:
>> Right, but why doesn't the client handle this more "graciously", by
>> examining the result and maybe point the user in a direction to find the
>> cause of the "failure", instead of just spitting out "207 Multi-status"?
>>
>> In our case we get this message diffing two revisions of a file, but we
>> don’t know why it fails (although some time ago someone said something
>> about cvs2svn not converting things correctly in all circumstances,
>> leading to this error?).
>>
>> Anyhow, the bottom line for me is that diffing these revisions still
>> fails, and I don’t know why!
>>
>>
>> Geir
>>
>> -----Original Message----- From: C. Michael Pilato
>> [mailto:cmpilato@collab.net] Sent: 5. januar 2010 13:54 To: Engebakken
>> Geir Cc: Kylo Ginsberg; users@subversion.apache.org;
>> dev@subversion.apache.org Subject: Re: svnsync "207 Multi-Status" failure
>> with http access method and property "svn:"
>>
>> Engebakken Geir wrote:
>>> It seems to me that the 207 message is not an error message seen from
>>> Apache, rather an informative message,  but that SVN interprets it as
>>> an error somehow. I would be glad to find out more about this ........
>>
>> "207 Multi-Status" isn't an error at all.  It's a code from Apache that
>> means, "I've got multiple different pieces of status information to
>> provide to you -- see the response body for details."
>>
>> In Subversion, those multiple bits of status generally correspond to
>> individual statii (heh) of property requests.  Subversion says, "Apache,
>> please send me properties 'foo', 'bar', and 'baz' on file.txt."  But what
>> if there's no 'bar' property on that file?  Apache can't reply "200 OK",
>> because all is not well.  It can't reply "404 Item Not Found", because
>> some of the items *were* found.  So it replies "207 Multi-Status", and
>> the body of the response contains the property values and 200 statuses
>> for the requested properties that actually exist, plus 404 statuses for
>> the properties that *don't* exist.
>>
>> -- C. Michael Pilato <cmpilato@collab.net> CollabNet   <>
>> www.collab.net   <>   Distributed Development On Demand
>>
>
>
> --
> C. Michael Pilato <cmpilato@collab.net>
> CollabNet   <>   www.collab.net   <>   Distributed Development On Demand
>
>

Mime
View raw message