Re: [guppi-list] text import glitch

[Havoc Pennington <rhp zirx pair com>]

On Tue, 30 Mar 1999, Asger K. Alstrup Nielsen wrote:
> 
> My Visual C++ 6.0 compiled version accepts this just fine...
> What is the problem?
>

It parses it incorrectly, but the exact error is at work. It must be some
kind of egcs artifact so I'll look into it myself. Thanks for checking on
MS C++.
 
> I see that you have updated the tree with some new stuff.
> So is the problem still there?
> 

Yes, I didn't change any of the parsing code. (I hopefully didn't break
anything and it should all be source compatible... that was the intent
anyway.:-)

> Regarding the ownership of the DataSets and the DataConverters:
> Yes, the user of the AsciiImport will own these structures.
> In my code, I have a use for the DataConverters that the
> Ascii Import produces, because I use the same ones for the
> parsing and presentation when the user does interactive
> editing of the data.
> 

Cool.

> Regarding the new type tags:  In my code, I explicitly manage
> the DataSet and DataConverters outside of goose scope.  I thought
> that Jon preferred to make the DataSets DataConverter ignorant,
> and that makes sense to me.
> So, I keep a pair<DataSet *, DataConverter *> around in order to
> solve this problem.  It works great for me, so I'm not sure we
> really need the type tags, but it's up to you to decide.
> 

I felt that I needed them to determine what kinds of DataSets were in 
the vector<DataSet*> returned from AsciiImport. Did you have a different
mechanism in mind?

Havoc