Marshall Schor commented on UIMA-1524:

Some of this discussion revolves around keyword and positional args. By 
"keywords", I'm thinking keywords are when you use separate builder methods 
(whose names are like keywords) which then act like keys for their args.  
Positionals have the benefit of allowing more concise expression.

The value of having methods "following/preceding(...)" to me is when multiple 
positional args are used; without them, following(10) is simply equivalent to 
the shorter and more standard stream method "limit(10)", and preceding(10) is 
equivalent to .towardBegin().limit(10).  

The select(...) currently takes 0 or 1 argument (the type); so there's "room" 
for more positionals.  One proposal from Richard seems nice: have the 2nd (or 
2nd and additional) argument(s) be a position or bound spec (4 forms:  fs, 2 
ints representing begin/end, or either of these plus a 3rd int: the "offset").  
I think that the "position or bound specification" seems to be the most 
frequently useful thing to have as the next positional arg for select.

> JFSIndexRepository should be enhanced with new generic methods
> --------------------------------------------------------------
>                 Key: UIMA-1524
>                 URL: https://issues.apache.org/jira/browse/UIMA-1524
>             Project: UIMA
>          Issue Type: Improvement
>          Components: Core Java Framework
>    Affects Versions: 2.3
>            Reporter: Joern Kottmann
> Existing methods should be overloaded with an additional Class argument to 
> specify the exact return type. This changes make down casting of returned 
> objects unnecessary. 

This message was sent by Atlassian JIRA

Reply via email to