On Sun, 09 Jan 2000 12:58:45 +1100, the world broke into rejoicing as
Robert Graham Merkel <[EMAIL PROTECTED]>  said:
> Christopher Browne writes:
> <snip>
>  > > QIF import is unaffected by date formats, AFAIK, as QIF encodes dates
>  > > into a common format.
>  > 
>  > This isn't *quite* true; it appears that different versions of Quicken
>  > encode dates differently, with differences including different
>  > delimiters (e.g. - "/" versus "-" versus ".").
> 
> You're kidding, right?  Nobody would be silly enough to encode dates
> in a version-specific way in a file format specifically designed to 
> exchange data?

You're kidding, right?  Nobody would be silly enough to expect a company
to produce a portable encoding, right?  :-)

>  > QIFIO.c now does support the various delimiters, and has a *limited*
>  > ability to recognize which date component is which.
> 
> Thanks for the correction.  I'm truly flabbergasted it would be
> necessary to do so.
> 
> BTW, have you had any time to think some more about borders?

It seems to me that there may need to be a separate "channel" for border
information, outside of the "style" info.

Otherwise, we wind up multiplying the number of style names by the number
of ways we might want to draw borders.

It may be enough, for the time being, to merely pass in a "borders"
argument for each field, distinct from the "style."

Russell Nelson's comment that we may be trying to get too sophisticated
is probably fair.  My tendancy would be to add in "borders" as a sort of
placeholder, and have the value normally be #f.  

The other possibility is to attach borders both to the style and to have
a separate value passed in.  And have the separate value override the
style, if it's present.  That way, we could have the "total" style include
a double-underline and a single overline, and then, if doing something
fancy, override that.
--
One man's constant is another man's variable.  -- Alan Perlis
[EMAIL PROTECTED] - <http://www.ntlug.org/~cbbrowne/lsf.html>

--
Gnucash Developer's List 
To unsubscribe send empty email to: [EMAIL PROTECTED]

Reply via email to