TK wrote:
Alan Wrote
The procedure at the linked site, which uses a Collection to load a
ListBox, is about 8 or 9 times slower than using a Dictionary instead
for that purpose, and about 7 times slower than using the ArrayUniques
function, which I referred to earlier in this thread, for that purpose
Alan out of 21 and still counting suggestions
On review, it seems to be more like 4 suggestions: Advanced
Filter|Uniques Only, some sort of query (that I don't understand),
specific code, and a pivot table (which I also don't understand).
you take exception
to mine.
Nothing personal. By the time I posted the above comment there was no
point in "taking exception to" the Advanced Filter|Uniques Only
approach, the query approach, or the specific code -- the OP had already
rejected them. I couldn't very well "take exception to" the pivot table
suggestion because I don't understand it enough to do so, or to compare
its operation to other approaches.
Your suggestion was the only one that involved using a Collection Object
for generating a unique set of data from a larger set, rather than using
a Dictionary Object. It's an area I have only recently become interested
in, and I felt (and feel) that the general info about the superiority of
the Dictionary Object for this purpose was worth mentioning. If
downloading, modifying and implementing John Walkenbach's procedure was
not beyond the OP's understanding, or that of any other particular user
following this thread, then certainly using a Dictionary Object in that
procedure instead of a Collection Object was also not beyond his or
their understanding, and would almost certainly be a better choice; I
thought that was worth pointing out.
And I thought that other users for whom the use of the ArrayUniques
function might be less daunting might be interested in the fact that
that function seems significantly faster than the Walkenbach procedure
(which it is primarily because of its use of the Dictionary Object).
I didnt reject your suggestion the O.P. did.
I was not evaluating these procedures I was suggesting
an alternative that he may have a better understanding
of.
What dont you understand about his reply to you?
Alan,. . .your site you sent me to is way over
my head. Sorry
Well, e.g., exactly what about the site that it is that he finds
way over his head. If his reaction is representative, then knowing
more about that reaction might provide an opportunity to simplify
or clarify the site for him and others like him. That was part of the
reason I provided a follow-up post outlining more precisely the steps
that could be utilized to use the data from the site.
Sorry for any aspects of my post that caused you to take offense. None
was intended.
Alan Beban
...........
Jason
TK
"Alan Beban" wrote:
TK wrote:
Hi Jason:
"Jason@Simcon" wrote:
Boiled down the problem is this: from a column of 50, non sorted, entries
only 10 are unique, the rest are repetitions. Return in another column the 10
unique entries.
Is it me or does that not sound like an iimpressively simple problem to solve?
obviously not :)
John Walkenbach the author of the Bible, no not
that Bible, the Excel Bible has written a procedure
and made it available for download on his site.
http://www.j-walk.com/ss
search for "nodupes" without the quotes
Good Luck
TK