[QFJ-612] BigDecimal for stock prices Created: 21/Jun/11 Updated: 15/Nov/12 Resolved: 21/Jun/11 | 
    |
| Status: | Closed | 
| Project: | QuickFIX/J | 
| Component/s: | Message Generation | 
| Affects Version/s: | Future Releases | 
| Fix Version/s: | 1.2.0 | 
| Type: | New Feature | Priority: | Major | 
| Reporter: | Chintan Shah | Assignee: | Unassigned | 
| Resolution: | Fixed | Votes: | 0 | 
| Labels: | QuickfixJ | ||
| Issue Links: | 
                
  | 
        ||||||||
| Description | 
| 
             Hi Folks, I am trying to use quickfixj for our needs. At this point I am not worried about the performance and want to be able to send orders with precise stock prices using quickfixj 1.5 I see the type NewOrderSingle takes a Price object which takes a double value. So I am getting the market data from our provider and convert those floating point values to BigDecimal. Should I just keep them in double and send orders with double values? What is the chance of raising my bid by a cent due to lack of precision in that case? What is the quickest possible way to achive my objective? Thanks a lot,  | 
    
| Comments | 
| Comment by Jörg Thönnes [ 21/Jun/11 ] | 
| 
             This is definitely not a blocker. But you are correct – if you want to do serious processing of financial data, BigDecimal values are much better. See http://www.quickfixj.org/quickfixj/usermanual/1.5.0/usage/codegen.html  | 
| Comment by Jörg Thönnes [ 21/Jun/11 ] | 
| 
             This issue was already solved in QF/F 1.2.  | 
| Comment by Jörg Thönnes [ 21/Jun/11 ] | 
| 
             Please re-open if anything is missing.  | 
| Comment by Toli Kuznets [ 21/Jun/11 ] | 
| 
             There's an example of how to use the flags to switch on BigDecimal at http://quickfixj.org/quickfixj/usermanual/1.5.0/installation.html#build  | 
| Comment by Jörg Thönnes [ 21/Jun/11 ] | 
| 
             Thanks, Toli. Did not find this in the first browse.  |