accumulo-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Marc Parisi <m...@accumulo.net>
Subject Re: Setting Charset in getBytes() call.
Date Tue, 30 Oct 2012 22:28:54 GMT
Instead of taking the platform encoding ( which can be changed as von cloud
suggested ), we're creating a class specific private definition? This isn't
particularly dynamic and an override for those who want to use a platform
specific encoding. This is a maintenance nightmare. I think before you move
forward with it, think of someone who doesn't want this change and how
difficult it would be to utilize their own or a different encoding scheme.

At the very least you can create a singleton that creates a dynamic
encoding for that, which can be modified via the accumulo configuration.

On Tue, Oct 30, 2012 at 4:02 PM, Eric Newton <eric.newton@gmail.com> wrote:

> > The current reliance on Text is in my opinion the greatest deficit of the
> > client API-
>
>
>
> Heh... if only that were true!
>
> But it is a big wart: requires users' to include the hadoop-core library in
> client code, which is unfortunate.  But it's probably going to be required
> for the importDirectory() call anyhow.
>
> -Eric
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message