Top N within groups

I am trying to filter out data within my groups to the top 5 or 10.
Group 1 is the date group with subtotals
Group 2 is the department group with subtotals
I want to show all departments but only show the top 10 $ amounts within each group.
Does anyone have any solutions that might help me?

Use Group Sort Expert to get top 10 or 5 records.
You can find this in Report-->Group sort Expert
Raghavendra

Similar Messages

Top 3 customers group by product in OBIEE

Hi,
Can somebody assist me how can I display a report in oracle Bi answers consisting of three columns called Product, Customer and Revenue. In the report, I need to display Top 3 customers group by Product. How can I use the "group by" function in oracle BI? Please assist me.
Thanks,
Bk.

to get the top3 customers you have a specific function topN in oracle answers.
if you say you want to group by only product remove the customer column from answers and SUM Revenue....i assume revenue is summed in rpd.
-bifacts
http://www.obinotes.com
Edited by: bifacts on Oct 19, 2010 7:03 AM

Sort within group

I have several groups within groups, and on my last grouping:
<?for-each-group:current-group();./TASK?>
I'd like to sort alphabetically by task, but can only seem to sort on the highest group I have:
<?for-each-group:G_DEPT;./DEPT?><?sort:DEPT;'ascending';data-type='text'?>.
Is it possible to sort on the current group task? I just get a dump when I try to add the sort tag after for each statement.
thanks

I'm not sure about your <?for-each-group?> element because I haven't seen your data. This is an example from one of my reports.
<?for-each-group: current-group();ELEMENT_NAME?>
<?for-each:ELEMENT_NAME?>
<?sort:PROC_PRIORITY?>
HTH
- Dan Stober
Salt Lake City

Help! Can't hide layers within groups in Muse 2014.2

Muse 2014.2 is not letting me hide layers that are apart of a group. Is anyone else having this issue and is there a fix?

Sam,
Thanks for the reply.
I do not like that I cannot hide individual objects within a group. It is difficult to find good reason in only allowing top-level containers having this functionality. For objects that I may want to hide for later use my only option now is to delete (and lose all my work and formatting of the object) or bring the opacity down to 0% (time consuming).
It would be great to have the option to hide each object/layer just like all the other amazing Adobe softwares!

Since Last CC Update Photoshop has lost the ability to CTRL click on layer within group??

Weird little glitch is happening since the latest update of CC. Can anyone else confirm that they are having this issue:
1. Create a group with a bunch of different layers
2. Select any tool (besides the direct selection tool)
3. Now press CTRL to try and select a layer and move it within the group.
You will notice that it moves the whole group and not just the selected layer...it does not select the layer. My Direct Selection Tool is set to "Layer" not "group"....so that is not the issue here.
This is annoying...does adobe know about this?

Okay, found the fix.
With the Move tool selected, go to the Options toolbar at the top (below file menu) and change the drop down menu to Layer instead of Group. Voila!

How do I use just 1 click with move tool (command v) to select layer within group?

I'm on Snow Leopard 10.6.8, Photoshop CS3 10.0.
Is there a way to click just once with "command V" to select a layer within a group?
I used to be able to do it, and on my laptop I can do it now (same OS and Photoshop) but the mac I use for my design work won't do it...I have to also right click and select the layer that is highlighted. I have a many multilayered Photoshop document with lots of groups, and it's an extra time consuming step.
Is there some preference I can turn on? Or some other simple keystroke?
Thanks so much!

Glad I could help, Emily

Working with items within groups

With nearly all primary Adobe CS applications they offer a way to modify individual items within a group without ungrouping the item. Both Illustrator and InDesign in CS5.5 enable to doubleclick on the group which enters a mode were that item is isolated and only the contents within can be modified. As far as I can tell the only options to do this in Fireworks is to:
* Ungroup and then regroup, which is not ideal because you loose the name of the group and you have to reselect each item again afterwards to regroup
* Use the sub selection tool, which is also not ideal because it dose not allow you to target sub groups as items
Can anybody else confirm this behavior?

Try the Object extension by Aaron Beall http://fireworks.abeall.com/extensions/
You can also, instead of grouping the objects, convert them to a Symbol (Select all, then Modify > Symbol > Convert to Symbol > Graphic. The objects will then behave as a group. You can edit individual items in the group by opening the symbol in the Symbol editor. To open the Symbol editor, simply double click on it.

How to group: groups within group

Hi. I am resubmitting this question with some changes. Is it possible to group some values and then use those values to group by another group value? For exp:
Top level group is Month,i-e Jan, Feb, March etc
Next level group is Status,e.g OPEN, CLOSE
Third level group is Full_Name
Tables and Fields:
TableA:
Tkt_DateTime
Tkt_Status (open /closed )
Tkt_Created_By (FK)
Count(tkt_created_by) gives count of tickets open or closed
TableB:
Full_Name
Join:
TableA.Tkt_Created_By=TableB.Full_Name
Is it possible to do this all in a single Select statement?

Simply state the order in the group clause, separating each sub-group with a comma
e.g.
group by to_char(a.Tkt_DateTime, 'Mon'),
a.Tkt_Status,
b.Full_Name

Groups within groups broken in Mavericks Mail?

In Mavericks 10.9, it appears there's a problem if a Contacts group contains other Contacts groups.
Example:
Within Contacts, create Small Group with three contacts.
Within Contacts, create Big Group with six contacts.
Within Contacts, create Huge Group.
Drag Small Group and Big Group to Huge Group.
In Contacts, click on Huge Group. It appears to contain Small Group and Big Group.
Click on either Small Group or Big Group and Contacts reports "No Cards".
Open Mail.
Make a New Message.
In the To: field type "Huge Group".
No addresses appear.
In Contacts, a control-click on Huge Group shows a pop-up menu with command Send Email to "Huge Group" dimmed.
Is this happening to others?
mm

What does it say in the "charset=" header
charset="iso-8859-1"
But another mail has nothing there, only "Content-Type: text/html";
The important thing is not whether this charset is correct or not. The important thing is that Apple has made my mail experience worse by this "upgrade", for no good reason!
Earlier, I could change encoding of mails (perhaps a badly encoded mail), but Apple has taken away that posibility for no good reason. It only affects old mails, before UTF became common. It is typical of Apple not to care about history, but for me, old mails constitute valuable history.
Rather than taking away an option to change the encoding (or interpretation of an encoding) of a mail, I think Apple should have enhanced the feature: Sometimes, it happens that only the subject line is erronously interpretated (mime endoding of non-us-ascii characters), and I would like to fix that. It has never been possible. (This problem can even happen today with some new mails auto-generated by machines, such as auto-replies.)

How do I merge external drive images to desk top drive within the same master catalog?

I have about 900 raw files on an external drive. I edited those files using my master catalog using a laptop while still travelling. When I got home, I copied the master catalog from the external drive to the desk top, then opened the master catalog on my desktop. I can see the edited images on the external drive. So far so good.
I want to copy (add) the edited images to my desk top. I tried to export the images as a new catalog and then import it into the desire location on my desktop drive. However, LR5 does not allow that. I get this message: "There is nothing new in the catalog you have selected. All of the catalog's photos are up to date in this catalog." Yes, but the edited images are not on my desktop.....
I also tried to Import from the external drive and then Copy the files to the desktop, but then I lose all of the work done in the develop module and the keywords.
I watched Kost's LR 3 video on merging catalogs, but she dodges the question of how to deal with this issue since the "merge" is within the same catalog.
I'm using LR5 with Windows 7.
Any suggestions would be immensely appreciated as I've spent hours trying to resolve this problem.

Okay, so I would copy (not move) the folders/photos to the internal HD, keeping them in the same folder hierarchy as they are in now on the external HD. Yes, of course the drive name will be different, but other than that the folder path and file name should be the same.
Then, open the LR catalog that indexes the files and Update Folder Location (details are here: http://www.computer-darkroom.com/lr2_find_folder/find-folder.htm, scroll down to the "Update Folder Location" section)
Then, once that is complete and everything is working in your LR catalog, with no question marks on the photos or folders, optionally delete the photos on the external HD.

Help: SQL to select closest points within groups of records grouped by time

Hi,
I need help to find an SQL for efficient solution for the following problem:
- I have 20 buses circling on one bus route and their locations are reported every 10 seconds and recorded into a spatial table “bus_location” with “bus_loc” as SDO_GEOMETRY type.
- Each bus location record there has a msg_datetime column with the time when the location was recorded.
- Buses circle the bus route 8-12 times, from A to B, then from B back to A, then new loop from A to B, and so on.
- I have 3 timing points in spatial table “timing_point” with “tp_loc” as SDO_GEOMETRY type.
- A bus will pass each timing point 8-12 times in one direction and 8-12 times in the other direction, while making loops on the bus route.
My task is: for each timing point, for each bus, for each trip direction (A-B or B-A), find the closest bus location to that timing point.
Example:
Bus...*TimingPoint*......*Time*.....*Distance*
...........................................*between*
...........................................*bus and TP*
12......A................14:01:00......250 m
12......A................14:01:10......150 m
12......A................14:01:20......12 m <== This is the record closest to the TP for this pass
12......A................14:01:30......48 m
12......A................14:01:40......100 m
12......A................14:01:50......248 m
12......A................14:29:40......122 m
12......A................14:29:50......72 m
12......A................14:30:00......9 m <== This is the record closest to the TP for this pass
12......A................14:30:10......10 m
12......A................14:30:20......12 m
I tried to use SDO_NN and I can get closest bus locations, but how to find the closest bus location for each pass? Or how to identify a groups of bus locations which are together within 2-3 minutes and to find closest record for each of these groups of locations?
Thank you very much for any help.
Milan
Edited by: mburazor on 9-Feb-2012 2:41 PM

Milan,
Yes, the problem is one of analytics not spatial. Here is another go that should be independent of year/month/date/hour/minute
as it does all its math based on an absolute number of minutes since 1970. I round to 5 minutes but you could round/trunc to any
particular interval (5 minutes, 10 minutes, 15 minutes etc)
Note that I have created extra records in the same table that I built. (Tip: good idea to provide a sample dataset to the forum when
asking difficult SQL questions like this.)
drop table buslocns;
create table buslocns(
busno number,
timingpt varchar2(1),
pttime   timestamp,
distance number)
insert into buslocns values (12,'A',to_timestamp('2012/02/10 14:01:00','YYYY/MM/DD HH24:MI:SS'),250);
insert into buslocns values (12,'A',to_timestamp('2012/02/10 14:01:10','YYYY/MM/DD HH24:MI:SS'),150);
insert into buslocns values (12,'A',to_timestamp('2012/02/10 14:01:20','YYYY/MM/DD HH24:MI:SS'),12);
insert into buslocns values (12,'A',to_timestamp('2012/02/10 14:01:30','YYYY/MM/DD HH24:MI:SS'),48);
insert into buslocns values (12,'A',to_timestamp('2012/02/10 14:01:40','YYYY/MM/DD HH24:MI:SS'),100);
insert into buslocns values (12,'A',to_timestamp('2012/02/10 14:01:50','YYYY/MM/DD HH24:MI:SS'),248);
insert into buslocns values (12,'A',to_timestamp('2012/02/10 14:29:40','YYYY/MM/DD HH24:MI:SS'),122);
insert into buslocns values (12,'A',to_timestamp('2012/02/10 14:29:50','YYYY/MM/DD HH24:MI:SS'),72);
insert into buslocns values (12,'A',to_timestamp('2012/02/10 14:30:00','YYYY/MM/DD HH24:MI:SS'),9);
insert into buslocns values (12,'A',to_timestamp('2012/02/10 14:30:10','YYYY/MM/DD HH24:MI:SS'),10);
insert into buslocns values (12,'A',to_timestamp('2012/02/10 14:30:20','YYYY/MM/DD HH24:MI:SS'),12);
insert into buslocns values (12,'A',to_timestamp('2012/02/10 14:59:40','YYYY/MM/DD HH24:MI:SS'),53);
insert into buslocns values (12,'A',to_timestamp('2012/02/10 14:59:50','YYYY/MM/DD HH24:MI:SS'),28);
insert into buslocns values (12,'A',to_timestamp('2012/02/10 15:00:00','YYYY/MM/DD HH24:MI:SS'),12);
insert into buslocns values (12,'A',to_timestamp('2012/02/10 15:00:10','YYYY/MM/DD HH24:MI:SS'),73);
insert into buslocns values (12,'A',to_timestamp('2012/02/10 14:44:40','YYYY/MM/DD HH24:MI:SS'),53);
insert into buslocns values (12,'A',to_timestamp('2012/02/10 14:44:50','YYYY/MM/DD HH24:MI:SS'),28);
insert into buslocns values (12,'A',to_timestamp('2012/02/10 15:45:00','YYYY/MM/DD HH24:MI:SS'),12);
insert into buslocns values (12,'A',to_timestamp('2012/02/10 15:45:10','YYYY/MM/DD HH24:MI:SS'),73);
commit;
with tensOfMinutes as (
select row_number() over (order by busno,timingpt, pttime) as rid,
       busno,
       timingpt,
       pttime,
       round(((cast(pttime as date) - cast('01-JAN-1970' as date)) * 1440) / 5) * 5 as mins,
       distance
from BUSLOCNS
select busno,timingpt,to_char(to_date('01-JAN-1970','DD-MON-YYYY') + ( mins / 1440 ),'DD-MON-YYYY HH24:MI:SS') as pttime,minDist
from (select busno,timingpt,mins,min(distance) as minDist
          from tensOfMinutes a
         group by a.busno, a.timingpt, a.mins
         order by 1, 2, 3 ) ;
-- Result
BUSNO                  TIMINGPT PTTIME               MINDIST
12                     A        10-FEB-2012 14:00:00 12
12                     A        10-FEB-2012 14:30:00 9
12                     A        10-FEB-2012 14:45:00 28
12                     A        10-FEB-2012 15:00:00 12
12                     A        10-FEB-2012 15:45:00 12regards
Simon

3D Issue: background group showing up on top of foreground group

I have a group of clone layers that are positioned in the background that are inexplicably appearing on top of foreground elements. This first occurred when I applied a filter to the background clone group. Removing the filter put them back in the background, but now they are back on top and I don't know what is causing this.
Both groups are set to be 3D groups.
Please help! Thanks.

Howdy,
Applying a filter to a 3D group almost always causes the group to get rasterized (a little box appears around the group's icon in the Layers List), which means it will be comped in layer order, rather than depth order. If you search for "rasterize" in the 3D Supplemental manual (found in the Help menu), you'll find a more detailed explanation.

Conditions within groups

Hi,
I have a report which has a group sorted on column A. And it has some 5 other columns. Column B is a unique primary key and column C and D are begin and end dates respectively.
My report needs to list down only those records where in the begin date is less than the end date within that group and not for the same unique primary key row. The Begin Date is always less than the end date for any single record. However, I want the records where the begin date is lesser than the end date of the other records in the same group alone.
Example:
A B C D
100 1 01-JAN-07 02-JUN-07
100 2 03-JUL-07 08-SEP-08
200 3 03-DEC-07 29-DEC-07
200 4 30-DEC-07 31-DEC-09
200 5 28-JAN-08 05-FEB-08
Here the begin date "01-JAN-07" is less than the "02-JUN-07" in 1 row and "03-JUL-07" is less than "08-SEP-08" in 2 row. So, I do not want this displayed in the report. However, in 200 series, the discrepancy occurs.
Can anyone please throw me ideas on how to handle this in discoverer.
Your inputs are greatly appreciated.
Thanks!

Create a calculation like:
min(column_d) OVER (partition by column_b)and name the calculation group_end_date.
Then change your condition to something like:
column_d <= group_end_dateFor starters, include the calculation in the worksheet to verify that the MIN is working how you want, and then remove it from the worksheet (but don't delete it).
Also, substitute your actual item names for column_b and column_d.

Email multiple email addresses for one contact within Group email

I have a contact card within a Group that has 2 email addresses. When sending out mass emails I want to send it out to both of the addresses for that one contact card; however, when I put the persons name in the address bar in mail, it only defaults to one of them. How can I fix this without creating multiple contact cards for the same person? Thanks for your help!!!

Hi, I'm not positive, but I don't think you can.
I'd maybe keep a little text snippet of the 2 email addies to copy/paste in the TO: feild, or the BCC: feild is better.

Creation of Sub-Totals within Groups

Hi - hope this is me just being a plank but I've hit a brick wall whilst trying to recreate a set of management accounts. So far the report has just two Groups - call them A & B for ease. Group B summarizes the detail and Group A neatly collects the results from Group B.
Group A comprises 6 parts all doing their bit and all the standard Crystal Sub-Totals (Summaries) work just fine. The problem I encounter is when I need to insert another Sub-Total into Group A which summarizes multiple parts of Group A e.g. Part 1 is Sales, Part 2 is Cost of sales, Part 3 is Labour, Part 4 is Expenses, Part 5 is Other and Part 6 is Depreciation. I need to insert a Sub-Total between Parts 3 and 4 which basically gives a total of Sales minus Cost of Sales minus Labour and to then continue with Parts 4 to 6. This is obviously a huge simplification of the issue but an answer to this would let me proceed with everything else.
I've tried creating an line so I can add a formula for this by giving it a name of its own in Group A but it fails to appear on the report. I've also tried the 'Insert Section Below' option and whilst I can suppress it where necessary the formula just picks up values from the last entry on the Detail Section.
Any ideas where I am making some elementary mistakes and is there an easy solution? Many thanks

Hi,
Try to create a manual running total formula for summing the fields with a condition.
Create a @reset formula and place it report header:
formula:
whileprintingrecords;
numbervar i:=0;
Create a @eval formula and place it in group header A:
formula:
whileprintingrecords;
numbervar i;
if {groupname.field} = 'Labour' or {groupname.field} = 'Expenses' then
i:= i + {summary.field};
Create a @display formula and place it in report footer;
formula:
whileprintingrecords;
numbervar i;
Hope this helps!
Regards,
Vinay

Top N within groups

Similar Messages

Maybe you are looking for