[DB-SIG] spec: named parameters: clarification needed

[Anthony Tuininga]

The problem is that the spec is not at all clear. Keyword arguments are
still arguments -- they are turned into a single dictionary which is the
"right" way in Python to pass arguments by name. Creating a dictionary,
populating it with arguments and then passing it directly is against
normal Python syntax (in my opinion, anyway). I realize that until
Python 2.0 you couldn't pass a dictionary that you had prepared in
advance (the rare case) as keyword arguments but do we really need to
consider compatibility with Python 1.5 still?

Perhaps what needs to be specified is:

execute(statement, *parameters) --> sequence (for non-named)
execute(statement, **parameters) --> mapping (for named)

And this last one for backwards compatibility if necessary

execute(statement, parameters) --> sequence or mapping depending on
module.paramstyle

Any thoughts on that?

On Sat, 2003-02-15 at 03:52, Harald Meland wrote:
> [M.-A. Lemburg]
> 
> > Anthony Tuininga wrote:
> >> When I wrote cx_Oracle I assumed 2 but I have lately been told that
> >> other drivers have assumed 1 instead, so I am intending to implement
> >> both ways. Perhaps this should be clarified in the API document?
> >
> > That's why I put up the question. Note that the .executeXXX()
> > APIs only define one parameter which is defined as sequence
> > or sequence of sequences, so both options conflict the spec
> > in some way.
> 
> Huh?  From the current spec, and I think this has been in there for
> quite some time:
> 
>     .execute(operation[,parameters])
>           
>         Prepare and execute a database operation (query or command).
>         Parameters may be provided as sequence or mapping and will be
>                                                ^^^^^^^^^^
>         bound to variables in the operation.
> 
> Similar wording is used for .executemany().
> 
> To me, that's pretty clear; am I missing something?
> 
> > Given that most drivers which use this param style seem
> > to implement option 2, I guess we should allow keyword
> > parameters to the .executeXXX() methods as well (provided
> > that .paramstyle is set to 'named').
> 
> Umm, if the spec is changes to allow leniency here, it will *further*
> complicate the matter of "my application needs to change database
> driver, and hence I need to transform lots of .execute*() calls for
> purely non-SQL differences".
> 
> Is that really wise?
> 
> If some drivers want to implement args as keyword params, they're
> obviously free to do so, but I'd prefer the spec to stay as it is
> (modulo better wording if necessary, of course).
-- 
Anthony Tuininga <anthony@computronix.com>