ëÏÒÏÌÅ× éÌØÑ wrote:
The advantage is that a list of field names is more flexible than a
single string listing the fields. This advantage matters if you don't
need all of the fields, or you need to have them in multiple formats. I
would only let you change this as you suggest if you can prove that all
uses of the field list objects only need the fields in a single format,
and they need all of them, always.

--
MySQL++ Mailing List
For list archives: http://lists.mysql.com/plusplus
To unsubscribe: http://lists.mysql.com/plusplus?unsu...ie.nctu.edu.tw

??????? ???? wrote:
I'm going to have to take some time to study this. I hope to get back
to you with more commentary soon.

It's fine to prototype the mechanism with this, but I don't think we can
get away with this for the final version. Windows partisans seem to
have an aversion to installing interpreters. We'll need to replace it
with a C++ implementation for the final version, I think.

--
MySQL++ Mailing List
For list archives: http://lists.mysql.com/plusplus
To unsubscribe: http://lists.mysql.com/plusplus?unsu...ie.nctu.edu.tw

ëÏÒÏÌÅ× éÌØÑ wrote:
The closest thing I can think of is XSLT. This would make sense if, as
I originally suggested, we declare SSQLSv2 structures using some form of
XML. Then we don't even have to build our own translator tool: we can
use xsltproc, or something like that. The hard part is in maintaining
the XSL code.

If you don't like that, the best thing I can suggest is to try one of
the new programmers' search engines that have popped up recently. I
haven't had a good reason to use one, though, so I don't have any of
their names in mind right now.

--
MySQL++ Mailing List
For list archives: http://lists.mysql.com/plusplus
To unsubscribe: http://lists.mysql.com/plusplus?unsu...ie.nctu.edu.tw