Mailing-List: contact couchdb-user-help@incubator.apache.org; run by ezmlm
Precedence: bulk
Reply-To: couchdb-user@incubator.apache.org
Received-SPF: pass (athena.apache.org: domain of dundeemt@gmail.com designates
 209.85.198.249 as permitted sender)
DomainKey-Signature: a=rsa-sha1; c=nofws;
        d=gmail.com; s=gamma;
        h=message-id:date:from:reply-to:to:subject:mime-version:content-type
         :content-transfer-encoding:content-disposition;
        b=eOMikDiGhLBbXw6p8GBNgxjTQnUTXGTVD9p3e90sv+AKGO/5bxH/9+35j/AKTlEXGe
         HdbVThMmqmYNxUrsN0MKj88yI9fKihcjvoAa+fnLNIeMjy5Xc/pL/Wz5bSoaujhm1j6u
         ts4ofnK1sGydd+UBidMqdadY2WVqjnlokH2VY=
Message-ID: <5aaed53f0807030613u78e5345ew14fdca1846ba4b5c@mail.gmail.com>
Date: Thu, 3 Jul 2008 08:13:03 -0500
From: "Jeff Hinrichs - DM&T" <dundeemt@gmail.com>
Reply-To: tech@dundeemt.com
To: couchdb-user@incubator.apache.org
Subject: File Attachments
MIME-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

Can someone point me to the wiki or discussion about how to handle
documents with attached file(s)?  I see references to the subject in
various places but have yet to find the "couchdb" way of doing it.

Do I just base64 encode the file contents and save that along with
filetype and other name attrs in a couchdb document?  Or is it
preferred to mime-encode like cmlenz does with the dump/load scripts?

I see in the wiki under UriTemplates, the method to download a file
attachment, but that would seem to infer that there is a correct way
to attach a file so that the method [WWW]
http://localhost:5984/dbname/docid/_bin/filename does the right thing.


Also, I've seen mention in the list of a 2GB limit to a database,
however I can't seem to find that in the docs -- is this limit
correct?  If it is, are there plans to expand the limit?

Finally, the idea I am toying with is using couchdb to store binary
documents with meta information.  Think of something like a blog, but
instead of the post content being text, it would be a binary document
(PDF, OO, etc), the comments would be other binary documents and their
associated meta data.  Is couchdb a non-optimal solution or would it
be a good solution?  The database would be rather large and constantly
growing -- if couchdb is a good fit, are there limits to database size
that would affect the architecture? partitioning, etc?

This wouldn't be a document revision control system, once an attached
file is stored it won't be changed, although amended documents would
be stored in the "comments" of the original post.

Regards,

Jeff