[Date Prev][Date Next] [Chronological] [Thread] [Top]

RE: EBCDIC

> -----Original Message-----
> From: Kurt D. Zeilenga [mailto:Kurt@OpenLDAP.org]

> At 09:11 AM 2002-07-12, Howard Chu wrote:
> >Are you running OpenLDAP on mainframe Linux?
>
> Yes.
>
> >What's your approach to EBCDIC translation?
>
> Run Linux, avoid EBCDIC.

Ah, it runs everything in ASCII huh. Handy...

> >It seems like this is a major pain; can't conveniently leave everything in
> >ASCII/UTF8 unless we use hex constants instead of character constants all
> >over the source code. It would also make debug msgs unreadable. But if we
> >translate to EBCDIC early, the UTF8 detection/encoding/etc. will
> be screwed.
> >
> >(Just started looking at porting 2.1 to MVS.)
>
> In chatting with our zOS Directory folks, they say that
> porting of SWD to zOS (to create zOS Directory) required
> changing of 90% of the code.

I'm beginning to see that.

The most centralized place to handle things is in liblber, but the schema
info
is needed to determine what to translate. On the client side there's no
schema
to reference, so the only approach seems to translate all strings except for
attributes specifically tagged with ";binary". Obviously there's plenty of
holes
in this approach.

On the server side I'm considering what it would take to leave the internals
in UTF8 and just translate DNs specified in the config file. Could use
something
like BSD's xstr to extract the string constants from the source and
ASCII-encode
the ones that are ber_print'd as a preprocessing step. Still not sure, the
debug logs would be ugly.

  -- Howard Chu
  Chief Architect, Symas Corp.       Director, Highland Sun
  http://www.symas.com               http://highlandsun.com/hyc
  Symas: Premier OpenSource Development and Support