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 includes SPF record at
 spf.trusted-forwarder.org)
MIME-Version: 1.0
In-Reply-To: 
 <CAPNuaCh8AvRZ_oeT7SFSxKtXeyvx859pDoR1V6rU6uNcBiz9Ww@mail.gmail.com>
References: <1380218246.19728.YahooMailNeo@web121006.mail.ne1.yahoo.com>
 <CAMGHjASHsPXC+5tXt1OhAvZY0txyi0MVP9jj-MicxEALef3N0w@mail.gmail.com>
 <CAPNuaCh8AvRZ_oeT7SFSxKtXeyvx859pDoR1V6rU6uNcBiz9Ww@mail.gmail.com>
From: Mark Hahn <mark@reevuit.com>
Date: Thu, 26 Sep 2013 11:16:05 -0700
Message-ID: 
 <CAK0S_BBu7EmJCB4p1eFKfNjGKjM-sQh6y0Nf3PhLgu_ztnuOtQ@mail.gmail.com>
Subject: Re: Put method
To: user <user@couchdb.apache.org>
Content-Type: multipart/alternative; boundary=bcaec52bea1965cf7404e74d5fb9

--bcaec52bea1965cf7404e74d5fb9
Content-Type: text/plain; charset=ISO-8859-1

I use an update handler for every single update.  I put in a couchdb wiki
entry about how to update any part of a document.  An update looks like
{"/date/year" : "2013"} to just change the year in a nested property.  It
also creates missing property paths.  My code is much simpler using this.


On Thu, Sep 26, 2013 at 11:11 AM, Stanley Iriele <siriele2x3@gmail.com>wrote:

> @mark I have actually found that when my implicit schema is scattered
> across application code...like during a GET , change random stuff, PUT
> sequence it gets hairy and the code is more contrived... But when I use
> update handlers...I know everything in there is atomic and I can handle all
> of the weird cases there....and should it fail... I just have to
> reconstruct the http request.... Plus 1 call should be faster than a fetch
> and put.... But the actual put is slower because you're executing a
> function.... But I see what you mean...I just adore that guys because most
> other nosql datastores force you to deal with conflicts via referring data
> and what not... But now couch :-)
> On Sep 26, 2013 12:05 PM, "Mark Deibert" <mark.deibert@gmail.com> wrote:
>
> > @Stanley: You're definitely right, the update functions are powerful,
> but I
> > don't use them to keep logic out of my database. I like the database to
> be
> > as close to pure data as possible. I suppose I actually could use them
> just
> > for doc "patching". I should think about this... :-)
> >
> >
> > On Thu, Sep 26, 2013 at 1:57 PM, Ashraf Janan <ashrafjanan@yahoo.com>
> > wrote:
> >
> > > Hello sir!
> > > Thank you for answering and helping.
> > > Please sir , i will do the following:
> > >
> > > type:student
> > > name: Tom
> > > age:37
> > > _id:6789
> > >
> > > -----
> > > type: message
> > > sender: Pablo
> > > title: "important information"
> > > receiverid:6789
> > > text : "---------"
> > >
> > > in this case must i know the _id for a student to can send a message to
> > > him.
> > > and in the Url to get all the messages for that person i will do the
> > > following:
> > > http:// url/name of database/ _design/name of
> > > view/_view/student?key="6789"
> > >
> > > is that correct ? please
> > >
> > > Best regards
> > > Ashraf
> >
>

--bcaec52bea1965cf7404e74d5fb9--