sql - Netezza conditionally count one record per ID -


i'm new netezza , need count number of ids have non-zero key. key transaction type, , want return count of ids have had transaction. want count of uids have non-zero key.

my data:

src uid key ...   118 3   ...   517 0   ...   517 1   ...   517 4   ... b   623 4   ... c   972 0   ... c   972 0   ...

what want return:

source  uids    uids_w_trans       2       2 b       1       1 c       1       0

here's code:

select     src source,     count(distinct(uid)) uids,     sum(case when key = 0 0 else 1) uids_w_trans      database  group     uid,     source

what i'm getting 

source  uids    uids_w_trans       2       3 b       1       1 c       1       0

you can see query counting every non-zero key. i've tried number of variations on above query, nothing has gotten me closer. how can count 1 non-zero key per uid?

i think looking conditional count(distinct):

select src source,        count(distinct(uid)) uids,        count(distinct case when key <> 0 uid end) uids_w_trans database group source

Comments

Post a Comment

Popular posts from this blog

'hasOwnProperty' in javascript -

i use obj.hasownproperty judge whether object has property, when replaced obj[prop] !== undefined , not normal implementation, ask, why behind method can not use it? object.hasownproperty(prop); object[prop] !== undefined; obj[prop] !== undefined wrong 2 reasons: you can explicitly set property undefined , obj[prop] = undefined; . obj.hasownproperty(prop) return true in case. obj[prop] follow prototype chain, return property that's inherited. obj.hasownproperty(prop) returns true if property exists directly in object, returns false inherited properties.
Read more

python - ValueError: No axis named 1 for object type <class 'pandas.core.series.Series'> -

i´m new programming. i´m trying use scipy minimize, had several issues , gotten through of them. right code, i'm not understanding why i´m getting error. par_opt = so.minimize(fun=fun_obj, x0=par_ini, method='nelder-mead', args=[series_pt_cal, dt, series_caudal_cal]) not enough info given op, somewhere in code it's specified operate data frame column (axis=1) on object pandas series. if code typically works occasional gives errors, check degenerative cases data frame may have 1 row. pandas has nasty habit of guessing want -- may decide reduce 1-row data frame series (e.g., apply() function; can disable using reduce=false in there). add line of code check object isinstance(df, pd.dataframe) or else convert offending pandas series data frame, s.to_frame().t problems had deal with.
Read more

java - How to provide dependency injections in Eclipse RCP 3.x? -

everyone! i'm trying deal eclipse rcp 3.x application. i've read lot of articles dependency injection in rcp e4 applications. can use di in rcp 3.x apps? i found perfect link di in e4. idea how use in rcp 3.x app? first of all, i'm interested in own objects injection (btw, how add own object application context?). best regards in 3.x code can ieclipsecontext necessary use contextinjectionfactory using ieclipsecontext workbenchcontext = platformui.getworkbench().getservice(ieclipsecontext.class); to workbench context. in view or editor can current part context using: ieclipsecontext partcontext = getsite().getservice(ieclipsecontext.class);
Read more