rodriguezum

hi i would like to create a global string variable that tells me the actual path being used, that is the actual workspace, so that i can export my jobs into linux solaris windows seamlessly. Can you help me?

Solved the problem by not sending any connection, but a paramenter of MONDRIAN_CONNECTION $P{MONDRIAN_CONNECTION} , has you said... didn't read it right the first time Post Edited by rodriguezum at 07/16/2009 08:50

i'm experiencing the same problems with ireport 3.5.0 in mode, on the crosstabs. It should be checked out.

doesn't work unfortunately, it throws me error evaluating expression Code: when i try to use subdataset instead of report it comes back and thorws me the same error, that i'm sending as a connection now is $P{MONDRIAN_CONNECTION} please advise Post Edited by rodriguezum at 07/16/2009 07:56

when i put printwhenexpression new Boolean($P{REPORT_CONNECTION}!=null) the subreport now shows but without the subreport, that is the report connection is null, but in the master report i do graph's and crosstabs, so that parameter shouldn't be null

i'm starting to use subreports and subdatasets, and i follow the instructions in the the ultimate manual, but it just doesn't seem to work, let me explain i'm using the same mondrian connection in the master and subreport, when i run them both separately they present perfectly, in the master report i reserve exactly the necessary space for the subreport to present, the subreport master report folder is in the classpath and i pass the report_connection parameter from one to the other, no input our output variables are needed. Finally when i run the report, it just doesn't show the subreport. Well instead i tried to use a subdataset with the subreport mdx query and necessary fields to create the chart that i want from the subreport, and pass the report_connection again, and i still can't make it work . don't know what else to try :(, i just follow the exact steps of the manual, when i try to do a report_connnection.toString() the result just comes out null, so in the end i'm just a little confused, i don't know what is the report_connection sending if anything, is the sql connection to my DB, is it the mondrian connection.... Please help :), Cheers here are both of the reports if you wana check it out

gonna post this in a new thread but, i have that property correctly set, actually i must tryed almost all combinations of that, even passing a static subreport and passing the connection and it still doesn't work

it does help thanks, did a workaround on it, just changed the mdx a little bit but i'm gonna try this, thanks another questions, i'm starting to use subreports and subdatasets, and i follow the instructions in the the ultimate manual, but it just doesn't seem to work, let me explain again of course i'm using a mondrian connection in the master and subreport, when i run them both separately they present perfectly, in the master report i reserve exactly the necessary space for the subreport to present, the subreport master report folder is in the classpath and i pass the report_connection parameter from one to the other, no input our output variables are needed. Finally when i run the report, it just doesn't show the subreport. Well instead i tried to use a subdataset with the subreport mdx query and necessary fields to create the chart that i want from the subreport, and pass the report_connection again, this time it gives me a null dataset argument. don't know what else to try :(, i just follow the exact steps of the manual, when i try to do a report_connnection.toString() the result just comes out null, so in the end i'm just a little confused, i don't know what is the report_connection sending if anything, is the sql connection to my DB, is it the mondrian connection.... Please help :), Cheers

i don't see the advantages in having different dimensions for a time dimension but, you problem is in a month week hierarquie, if you had weeks from 1 to 5 i imagine that would work, since the days are working right?, but since the weeks are unique from 1 to 50 something jasper analysis has no way of combining the months to the respective weeks, i don't know if this works but in your weeks dimension you could have the month week number and the year week number, and in the ordinal column have the month week number and in the column have the year week number, this in the mondrian schema. hope it helps, i'm struggling with jasper analysis, server, ireport also...

hey i'm running in java now an etl of just a datamart, it takes about 8,9 hours to completion, the thing is in the middle of the process it breaks downs and gives an error, always in the same spot, e checked with the dba and he doesn't end my connection, and when i try to run it from the job that failed till the end it goes smoothly again, and that particular job doesn't use global variables or anything, i'm using the jasperetl in java. Could this be a memory thing ?, help please.... :) and another thing, the first time i create it the toraclescd it works well, but in the next time i run it and all the others it throws me an error of the table not being there, but it stills inserts the data correctly where i want it, the exact number. So for nw i decided to ignore this error but still would like to know what is going on. And last but not least toraclebulk just works for my "small" table, tables containing millions of records are not inserted, i have to use toracleoutput for those, the big tables are stored in a csv file but then not inserted, i REALLY would like to have this feature on my etl sing the bulk is about 6 or 7 times faster, sorry to put this on this thread, but it seems like the only one that's "active" Cheers

hi, does anyone know how i can avoid the nulls in my dataset, i'm using mdx and for obvious reasons some crossjoin will have some nulls on it, but ireport charts wont take them any help?

if you want ot aggregate weeks in months you have to create a hierarchy wich has the level month and then the level week, you can also define a hierarchie that starts with the level year >> month >> week >> day lets say something like this <Hierarchy name="AQM" hasAll="true" allMemberName="AQM" primaryKey="DIM_TIME"> <Table name="****" schema="*****"> </Table> <Level name="YEAR" column="YEAR" type="Numeric" uniqueMembers="true" levelType="TimeYears" hideMemberIf="Never"> </Level> <Level name="QUARTER" column="QUARTER" type="Numeric" uniqueMembers="false" levelType="TimeQuarters" hideMemberIf="Never"> </Level> <Level name="MONTH" column="MONTH_NAME" ordinalColumn="MONTH_OF_YEAR" type="String" uniqueMembers="false" levelType="TimeMonths" hideMemberIf="Never"> </Level> <Level name="WEEK" column="WEEK_OF_MONTH" type="Numeric" uniqueMembers="false" levelType="TimeWeeks" hideMemberIf="Never"> </Level> <Level name="DAY" column="DAY_FULL_NAME" ordinalColumn="DAY_OF_WEEK" type="Numeric" uniqueMembers="true" levelType="TimeDays" hideMemberIf="Never"> </Level> </Hierarchy>

hi i have a limited knowledge on mdx but i want to display charts, the problem is that a couple of results contain null's,my charts work perfectly if i remove the problematic column, and i have searched the forums but the solution provided is to replace those nulls in the datasource (example in sql) and i think that is not posible in mdx (i think), is there any way to remove just those values from the charts (or set the values to 0) without having to remove the houl column. thanks

hey can you explain me what i am doing wrong? here's the jrxml

where can we get the new workbench?

i don't see that option in teltoracleoutput, what i use is just tcreatetable befor the toraclemap

hi i'm trying to retrieve a tables from my operational db and place them in my extraction phase in my etl, i wish i could do that with the bulk option (oracle db to oracle db), and that works fine with very smal tables, that is with up to 10000 lines, but when i try to bring bigger tables say 4000000 registers i see them pass in the run/statistics option but they don't get inserted in my destination table, although it is created through that component. In the run process it doesn't show any errors. and the information just stays in a csvfile in my workspace I REALLY need to speed up my extraction process and toracleoutput is 6 or 7 times slower, can anyone help?

hi, when i design my cube in my my clients and personnel dimensions i have their names in given_name, family_name and mother_name and i wish i could join them in jasperanalysis with mondrian. And wen i am using jpivot in JS it works and shows the full name, the problem is when i try to drill down, it give me an error of ambiguosly defined column. Can anyone help please

i think you have to create fields also for the headers, like Data(Rows .... , Measures(unitsales,?...)) don't know, just gessing, i'm also getting started in jr with mondrian

Hi, that did solve my problem of ireport i thinck (havent tested it, but it accepts the schema) thanks a lot, but for the conformed dimensions aproach that i was going for doesn't, since dimensionusage doesn't have an alias property, and since the caption isn't working either in them i guess i'll drop it for now and start working that way. Thanks

When i talk about 2 dimensions its not 2 hierarchies in the same dimensions, but two diferent Time dimensions, in each of the dimensions i have working perfectly 4 hierarchies, the thing is i'm designing a cube for a health plan analysis system, and i want there, a dimension time called entrance_date and another dimension called exit_date, both feeding from my DIM_TIME which is the table containing all de dates/months/quarter/years/holidays...., since that is supported by mondrian and jasperserver thats not a problem, i planned a development of conformed dimensions for maintenance reasons, then in my cubes i just use DIMENSIONUSAGE, but the caption in dimension usage doesnt seem to be working sadly, so i end up in my jpivot cube with two dim_time, instead of entrance_date and exit_date. having no clue wich one is entrance and wich is exit. Unless i use the mdx query of course. An even bigger problem is that ireport(3.5) doesn't admit my schema cube with 2 dimensions feeding of the same table. Can we do anything about that, i mean can i redesign my schema to make it work. the only other solution i see is to have in my cube two different time tables, but i REALLY don't wan't that, cause in the future cubes that i'm designing imay need 3,4 time dimensions....

Hi i'm new to ireport and this is my first post :) , and i'm using version 3.5, So i'm working a datawarehouse wich i access from a mondrian schema, i create a connection to the cube and it all connects perfect if i remove one of my time dimensions, i have 2 time dimensions in my mondrian schema entrance_date and exit_date , both feeding from the same table dim_time, i also have the similar case in time of day, a table wich contains the hours and minutes of a day. It is essential that i have 2 time dimensions in my cube, but i can only make it work if i remove one of them and rename the other to dim_time, i've been checking in ireport lib and it uses mondrian-2.3.2.8944.jar, exactly the same as jasperserverpro 3.5, but in the it does work. I'm guessing some of you had the same problem since multiple time dimensions is not uncommon, and i was hoping uo could help me find a way to design a schema that would be accepted in ireport anyway, in the caption isn't used in dimensionusage so i end up with the same name in both in jasperanalysis any thoughts?? Cheers

Hi i'm new to ireport and this is my first post :) , and i'm using version 3.5, So i'm working a datawarehouse wich i access from a mondrian schema, i create a connection to the cube and it all connects perfect if i remove one of my time dimensions, i have 2 time dimensions in my mondrian schema entrance_date and exit_date , both feeding from the same table dim_time, i also have the similar case in time of day, a table wich contains the hours and minutes of a day. It is essential that i have 2 time dimensions in my cube, but i can only make it work if i remove one of them and rename the other to dim_time, i've been checking in ireport lib and it uses mondrian-2.3.2.8944.jar, exactly the same as jasperserverpro 3.5, but in the it does work. I'm guessing some of you had the same problem since multiple time dimensions is not uncommon, and i was hoping uo could help me find a way to design a schema that would be accepted in ireport Cheers