Quote from: wlargou on 01 Nov 2024 01:53:56 PMHi,

Our regulations requires we record every user activity for retrieval of data, I know that we can record a user session but needs to be enabled each time by the user, do you know any other way we can record detailed user activity on Cognos ?

Thank you

Cognos Extended Audit has this feature but that has to be installed. It captures every click.

Try the COGIPF_ACTIONS table in the audit database.

Hi,

You can define navigation paths within a data module, and/or within an individual dashboard. If you want them to be generally available, a data module is the way to go. They are not something you can define in Framework Manager (although that supports Dimensionally Modeled Relational objects, which permit drill-down/up).

Cheers!

MF.

Hi,

Are the navigation path only available on Data Modules as of Cognos 12 ?

Thank you

Hi,

Our regulations requires we record every user activity for retrieval of data, I know that we can record a user session but needs to be enabled each time by the user, do you know any other way we can record detailed user activity on Cognos ?

Thank you

I have used R to process open datasets that were in CSV.  Metadata was at the top.  I had to tell R to start reading the file at the line where the data actually begins.  It's a valid format (especially since CSV has a guideline, not a standard).  Maybe Cognos Analytics should be able to read and write CSV files with arbitrary text for an arbitrary number of lines at the beginning of the file.  Consider submitting an RFE.

There is no way to do this - By definition a .csv output will only have column headers and data included.

That then leaves only "hokey" solutions that dougp indicates.
These include either adding a dummy row of output which could include text like "Filters: A, B, C,..." (this is something along the lines of what dougp is suggesting).
Or, perhaps, adding an additional text column at the end/beginning which would be your filter values in the Column Header Name with all Null values in the data rows.

Just my thoughts.

Quote from: MFGF on 30 Oct 2024 03:09:39 PMHi,

I'm not quite clear on the rules you want to implement to focus on just 14 of the 48+ combinations of members? Can you achieve this using filter() functions on the relevant sets, then use option 2 above? Or maybe just use the relevant members in a set() function for each?

Cheers!

MF.

Thanks for the speedy reply!

So, the sets for each dimension have already been filtered down to only ones that will ultimately be used in at least one combination. However, not all combinations of should make it into the final report. consider a more simplistic example than my real one to illustrate what I mean here.

Imagine that I have two dimensions:

[product]: 30 total cube members filtered down to only 3 for my report ([popcorn],[candy],[soda])
and [facility] with 50 total members filtered down to only 3 for my report ([facility 1],[facility 2],[facility 3])

In my final report, I only care about popcorn sales for facility 1, candy sales for facility 2 and soda sales for facility 3. so out of 9 possible tuple sets, I only care about three.

Quote from: jbarberio on 30 Oct 2024 02:59:45 PMI have a scenario where I need to replicate a single crosstab on many pages where the same data and visual layout gets repeated, but three separate dimensional parameters are changing between each page. I know I could simply copy/paste my page contents and queries with different slicers on each query, but I see this as a total last resort (There will by about 14 tabs in total and having to replicate any changes on every single tab/query is very sub-optimal). I've tried a variety of ways to accomplish this, but can't seem to get any of them working. Any help would be greatly appreciated!

Things I've tried:

1) Tuple-based grouping in a page set: I built a prompt with string "display" values and tuple() statements with the members of the three dimensions for the "use" values. I then created a page break query and report query with data items pointed at the prompt selection(s) using #promptMany(), set up a page set grouping on that data item and a master/detail relationship on my page linking the same value. Here, I get the error "Tuple-based value expressions are not supported as grouping items in list reports"

2) Nested page sets: I created a separate page set grouped on each of the three sets and nested them inside one-another. This somewhat works, but it includes far more permutations than I want (All combinations of possible values results in 48+ permutations instead of the subset of only 14 that I care about). and I wasn't able to figure out how to alter the naming of the tabs as desired.

3) Report Booklet with multiple references to a single base report with views sitting on top of that base (one view for each eventual tab with saved parameters for the three values on each). In the view properties, I set the parameter values and set "Prompt for Values" to No. Then, on the report references in my report booklet, I disabled "show prompt pages" and set "saved parameter values" to "merged". When I run my report booklet, it still prompts me for values instead of using the ones saved in the view properties. Furthermore, because the report references all ultimately use a single base report, they share parameter names and it only prompts once and uses the selection for ALL references.

Hi,

I'm not quite clear on the rules you want to implement to focus on just 14 of the 48+ combinations of members? Can you achieve this using filter() functions on the relevant sets, then use option 2 above? Or maybe just use the relevant members in a set() function for each?

Cheers!

MF.

I have a scenario where I need to replicate a single crosstab on many pages where the same data and visual layout gets repeated, but three separate dimensional parameters are changing between each page. I know I could simply copy/paste my page contents and queries with different slicers on each query, but I see this as a total last resort (There will by about 14 tabs in total and having to replicate any changes on every single tab/query is very sub-optimal). I've tried a variety of ways to accomplish this, but can't seem to get any of them working. Any help would be greatly appreciated!

Things I've tried:

1) Tuple-based grouping in a page set: I built a prompt with string "display" values and tuple() statements with the members of the three dimensions for the "use" values. I then created a page break query and report query with data items pointed at the prompt selection(s) using #promptMany(), set up a page set grouping on that data item and a master/detail relationship on my page linking the same value. Here, I get the error "Tuple-based value expressions are not supported as grouping items in list reports"

2) Nested page sets: I created a separate page set grouped on each of the three sets and nested them inside one-another. This somewhat works, but it includes far more permutations than I want (All combinations of possible values results in 48+ permutations instead of the subset of only 14 that I care about). and I wasn't able to figure out how to alter the naming of the tabs as desired.

3) Report Booklet with multiple references to a single base report with views sitting on top of that base (one view for each eventual tab with saved parameters for the three values on each). In the view properties, I set the parameter values and set "Prompt for Values" to No. Then, on the report references in my report booklet, I disabled "show prompt pages" and set "saved parameter values" to "merged". When I run my report booklet, it still prompts me for values instead of using the ones saved in the view properties. Furthermore, because the report references all ultimately use a single base report, they share parameter names and it only prompts once and uses the selection for ALL references.