Mailing-List: contact user-help@couchdb.apache.org; run by ezmlm
Precedence: bulk
Reply-To: user@couchdb.apache.org
Received-SPF: pass (athena.apache.org: local policy)
DomainKey-Signature: a=rsa-sha1; c=nofws; d=ecstatic.com; h=content-type
	:mime-version:subject:from:in-reply-to:date
	:content-transfer-encoding:message-id:references:to; q=dns; s=
	ecstatic.com; b=VLFVt9DckSTEA2AMy5/jEaWc94ly909eoMhh74J9CvYnAkL0
	HbpbA5dT4/QAdyyMpKqeMAr2e8GjbyEzT0kAfzCXG5cPi7Qq4BMVZZXxN3vBw+C/
	qVUGuzl7anqKj9ICZ0FZNWlfHoFRi/6WHpWsufhY2K9yHIINUe5zsU5C8Gg=
Content-Type: text/plain; charset=us-ascii
Mime-Version: 1.0 (Apple Message framework v1084)
Subject: Re: _changes?include_docs=true fails if any documents deleted?
From: Roger Rohrbach <roger@ecstatic.com>
In-Reply-To: <BANLkTikqLmkmBUe_EgACHczuU7Fw2q=u=w@mail.gmail.com>
Date: Tue, 14 Jun 2011 18:40:35 +0200
Content-Transfer-Encoding: quoted-printable
Message-Id: <DDB6F846-F441-457E-A360-871E7100A70E@ecstatic.com>
References: <275045A2-B106-48C1-B4A5-DD9D9A1316B1@ecstatic.com>
 <BANLkTikqLmkmBUe_EgACHczuU7Fw2q=u=w@mail.gmail.com>
To: user@couchdb.apache.org

Yes, indeed; it works now.  Thanks.  I've since installed Couchbase on =
my laptop, so will isolate any future platform-specific bugs and report =
them to Cloudant...

> Roger,
>=20
> This was a recent regression on Cloudant itself, it's now fixed,
> please let us know if you still encounter the error.
>=20
> B.
>=20
> On 14 June 2011 16:40, Roger Rohrbach <roger@ecstatic.com> wrote:
>> Hi, all.  I'm new to CouchDB (am actually using Cloudant's hosted =
service, via Heroku), and have (possibly) discovered something =
troubling: it is not possible to make a request to =
_changes?include_docs=3Dtrue if there are any deleted documents.  =
CouchDB craps out at the deleted document, rendering malformed JSON.  If =
I remove the include_docs option, I get the correct results, but of =
course then I must request each new, updated or deleted record =
individually.  Surely one should be able to request the documents and =
process them in one request (in fact, one can, as long as none were =
deleted).
>>=20
>> This seems like a glaring defect.  I just wanted to run it by this =
list before reporting it as a bug.  Is there something I'm =
misapprehending?
>>=20
>> Thanks,
>>=20
>> Roger
>>=20
>>=20