Re: [open-cobol-list] OpenCOBOL-ce-1.1 : repackaging libcob/common.h

[Brian Tiffin]

I like the ODBC angle Joe.  But it might be worth a google search, as
I seem to remember this being mentioned before.  It might only hold
hints, but John Ellis created a DBI handler that (I think) can be
configured for DBI:ODBC.  See if the article at
http://oldsite.add1tocobol.com/tiki-read_article.php?articleId=1 holds
any usable clues.

Cheers,
Brian

On 9/16/13, Joe Robbins <address@hidden> wrote:
> Thanks Brian. I hadn't seen the 2009-05-25  entry in ChangeLog. I'll reflect
> on the alternatives for common.h.
>
> Coincidentally (relating to your emails with Patrick on SQL), when writing
> my notes on the fileio rewrite (available soon!), amongst the "future
> possible enhancements" I had listed support for INDEXED ORGANIZATION using
> an ODBC database interface. This would provide the option to use MySQL,
> Microsoft SQL Server EXPRESS, etc as the store for INDEXED files as an
> alternative to VBISAM, BDB, etc. This is not a substitute for an SQL
> interface - the COBOL programmer would notice no difference. But it could
> provide a resilient, efficient, ... storage mechanism.
>
> Regards
>
> Joe Robbins
> address@hidden
>
>
> ________________________________________
> From: Brian Tiffin <address@hidden>
> Sent: 15 September 2013 23:40
> To: Joe Robbins
> Cc: address@hidden
> Subject: Re: [open-cobol-list] OpenCOBOL-ce-1.1 : repackaging
> libcob/common.h
>
>
> Not sure Joe.  But I found the entry below in the libcob/ChangeLog.
> Not much of a hint as to the rationale, but it does look like a design
> intent.
>
> 2009-05-25  Roger While <address@hidden>
>
>         * General : Remove the following extraneous include files:
>        byteswap.h, call.h, fileio.h, intrinsic.h, move.h, numeric.h,
>        screenio.h, strings.h, termio.h by integration into common.h
>        as first step
>
> I didn't find any comments on what step 2 was supposed to be at the time.
>
> I'm not sure how to best fix this for the CE cuts, but I'd be ok with
> a split back to 'extraneous' headers, with the caveat that I'll always
> defer to Simon for decisions on these commits.
>
> Cheers,
> Brian
>
>
>