incubator-couchdb-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Roger Binns <>
Subject Re: Silent corruption of large numbers
Date Tue, 10 Nov 2009 08:11:14 GMT
Hash: SHA1

Chris Anderson wrote:
> If your concern is this strong, then you should be encoding your
> numbers as strings and writing your own number handling code.

The point is that they aren't my numbers :-)  They are the numbers of the
users of my component.  That may be another developer or more likely yet
another component.  Additionally my component is about interoperability so I
can't just go ahead and encode numbers differently.

> number handling) but we can't, as a pragmatic matter, be responsible
> for the behavior of JavaScript or other query-server engines.

Well you can take a little "blame" for the one that comes with the CouchDB
distribution :-)

> If you need application level support for precise numbers, 

All I expected was to get out what I put in.

> or just rely on CouchDB documents (not the view
> servers) to handle large numbers accurately.

I am changing my code anyway to get the complete documents (include_docs) as
opposed to getting just the keys I am interested in via the view.  In theory
this may result in more data being transferred but I can live with that.

> I think it's fine that our JS engine is lossy on large numbers. If it
> turns out to be trivial to patch for better math, I'd be in support.

I understand that it is just the way JS works and it is impractical to fix.
 I do think it would be valuable for there to be a test/compliance suite in
CouchDB that at least detects and documents the issue for any view server.
That is also valuable information for anyone choosing a view server.

Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Mozilla -


View raw message