Hi, I know your Period is probably a date type in Oracle, but in OBIEE terms, as the field has accuracy that includes time, the data type will be datetime. i.e. sysdate in Oracle DB is 02-JUN-2017 09:02:35 GMT So if you cast your Period as date in your formula you may find that fixes your issue. Hope this helps, Robert.

Hi, fairly sure I have seen page breaks work when exporting to pdf, but that may have been 11g.....

Hi, you can also use hidden dashboard prompts to 'catch' the drill down on specific values in your report and set presentation variables, which you then use as the basis of filters in the target, using this technique you can get around drill limitations, it no longer has to be based on the same subject area content. Hope…

Wouldn't the simple solution be to actually be included on the distribution list, or is that too simple?

Hi, I concur with my much more learned colleagues on the advisability of what you are trying to achieve, but on any subject area you can deem that any fact measure is implicitly (Implicit Fact Column) written into any query on that subject area that has no fact content. If you have a PK in the fact folder that will…

Hi Gianni, I take it you could add a text area and write some jscript in it to make legends not display? A question, not an answer....

Hi, provided the data group electrical / mechanical / hydraulic exists in your data set, or can be calculated using case / bin logic or whatever then it should just be a case of dragging the field into the columns area, same as the existing switch, plug, cord etc

Hi Christian, I would discourage people using this as a standard for 12c as it is all too easy to save the local back to the server which corrupts the rpd password, which depending on your rpd discipline can be painful.... the download / upload functionality with 12c is fast and robust and encourages version control or at…

Hi Pedro, many thanks, yes - change the property to SUM and it does exactly what it says on the tin, many thanks! (and please pass my regards on to the Peak crowd). Robert.

Thanks for all your input, I did not think I was missing anything in OBIEE for this, and I was aware of evaluate but try not to resort to it wherever possible for performance reasons.

You might want to think about passing your user to Apex at this point. Really writeback is extremely limited cf above.

There is a special application role which is automatically assigned, 'authenticated user' What you can do is remove 'authenticated user' from BIConsumer Application roles. Thanks to Gianna Ceresa (if I remember correctly) who originally answered this question for me. Hope this helps, Robert.

Hi Christian, sorry web browser crashed earlier and it was a lengthy response... but I owe you a response for your courteous and considered responses. Dimension X - a normal dimension that joins to the detail level of the combined fact Dimension Y (based on fact details, with 3 tiers in hierarchy, Total, Header and Detail…

On the face of a report this would still yield the wrong answer if the measure was aggregated, wouldn't it? i.e. Dimension X Header PK (not displayed) Header Measure A 1 5 --correct in isolation B 1 5 --ditto C 1 5 --ditto D 2 20 Total 35 Desired Result - sum of distinct PK measure Total 25 My example above assumes that…

Double check your content levels fact to dimension, both should have a dimension hierarchy level set for the content that you are using as a combination, so per your example I would expect whichever dimension provides the location to have content level set as detail and the fact table providing revenue to have the…

Thanks, I have used the fact as the dimension previously so I 'get' exactly what you are saying. However, your caveat - so long as you don't need to factor header-level measures across detail rows is the bottom line, and the crux of my question, so thanks you confirmed my thinking for me and others, and you confirmed my…

Thanks Christian. On a. - agree totally! On b, I kind of had this in mind but you don't have a dimension hierarchy for a fact, which left me with c. - where reading between the lines I guess you are suggesting that I treat the header fact like it is an agg fact(?) but I am not 'getting' the rest of it? On joins you are…

And you have checked that there are no other filters that are already on the report before you apply the new filter? If so, go into design mode, import the new filter CONTENT and if it does not return data then remove the elements of the filter one at a time until you see data, then you will have your answer. Or, add the…

Thanks all, I guess in simple terms if my extra column makes the volume of rows 500000 instead of 500 then the pivot still has to handle, group, summarise etc the extra volume of data, yes?

Provided the filter is compatible with the subject area of the graph then it should work. Are you sure that it is for the same subject area, and are you also sure that the combination of that filter and any filters that were already on the graph that you should still get data?