Użytkownik Ed Leafe napisał:
> It returns an integer, not a tuple. The duplication is because you are
> using enumerate(); the iterator returns the row number itself.
>
>
Your approach completely disagrees with Python iterators idea.
In Python documentation about 'next()' method you can read:
'Return the next item from the container.'
Accordingly to that definition, your bizobj is just integer object list,
where
enumerate(biz.bizItererator) = enumerate(range(biz.RowCount))
with one exception, pointer movements.
In your opinion, enumeration of list ["a", "b", "c"], should return [0,
1, 2],
and you are telling me, that functions like enumerate(), map(), etc.
are needless with iterators.
Sorry, but there is no sense in existence of such iterable object.
And I completely disagree with you in that matter.
>> Since bizobj is, in simplification, collection of rows,
>> don't you think that I should rather receive tuple of row number and row
>> itself,
>> like (0, row0_object)?
>>
>
> Well, it could, but it would have to be separate from the bizobj
> dataset. In other words, "biz.Record" is not a fixed set of data, but returns
> whatever record the bizobj is pointed at at the moment. If it were to return
> biz.Record instead of biz.RowNumber, you could not persist that reference.
> Maybe an example would make that clearer: let's assume we have a bizobj with
> 10 rows, and a column named 'century'. Each row's 'century' column is 100 x
> the row number: e.g., row 1 == 100, row 3 == 300, etc. Now consider this
> code, which assumes that the iterator returns the biz.Record instead:
>
> thirdrow = None
> for rownum, record in biz.bizIterator():
> if rownum == 3:
> thirdrow = record
> print thirdrow.century
>
> Since we only captured the record on the third row, we would expect
> 'thirdrow.century' to be equal to 300, right? But it would be equal to 1000,
> the value on the last record, since that's where the pointer is. Now consider
> we did:
>
> biz.first()
> print thirdrow.century
>
> Again, it would print 100, not 300.
>
> The other option is to return the dataset tuple for that row. That
> would fix the problem above, but would not allow you to modify that data and
> have it update the bizobj. IOW, if you changed a value in the returned data,
> it would not affect the data in the bizobj.
>
>
Yes!!! This is exactly what I expect from iterator.
And this is true for each iterator for mutable objects, not for Record only.
And I'm completely not interested what is the 'record' value outside the
loop.
> The whole purpose of the bizIterator is to avoid having to write code
> like this:
>
> biz.first()
> while True:
> try:
> biz.dosomething()
> biz.next()
> except dabo.dException.EndOfFileException:
> break
>
In other words, current iterator is equivalent to:
for i=0 to biz.RowCount:
biz.moveToRowNumber(i)
dosomething()
You tell me I must use something like:
for rownum, rownum_too in biz.bizIterator():
doSomething(biz.Record)
while I can just use:
map(doSomething, biz.bizIterator())
I know, I don't convince you Ed, so I even don't intend to do.
Thanks for time, Ed.
EOT
--
Regards
Jacek Kałucki
_______________________________________________
Post Messages to: [email protected]
Subscription Maintenance: http://leafe.com/mailman/listinfo/dabo-users
Searchable Archives: http://leafe.com/archives/search/dabo-users
This message: http://leafe.com/archives/byMID/[email protected]
