P.S. Usually financial companies like to use PyTables for this purpose. On Wednesday, 9 April 2014 18:03:58 UTC-5, Massimo Di Pierro wrote: > > It is simply a key/value store built on top of SQLite. > Would you be able to share any example code about accessing Bloomberg Data > Terminal and ICE? > > Massimo > > On Wednesday, 9 April 2014 16:15:36 UTC-5, Trent Telfer wrote: >> >> I am actually pulling the data from a Bloomberg Data Terminal, Bank of >> Canada, ICE and NGX. All have various ways to access the data. Some >> providers have fairly nice API's (bloomberg) and others like NGX only >> provide Excel files. >> >> I'm interested in this persistent dictionary concept and am curious if it >> at the least can be adapted to be used with the Bloomberg terminal data. >> >> -Trent >> >> On Wednesday, April 9, 2014 3:08:28 PM UTC-6, Massimo Di Pierro wrote: >>> >>> Are you getting from Yahoo finance? >>> Look into github.com/mdipierro/nlib >>> >>> from nlib import * >>> symbol = 'AAPL' >>> d = PersistentDictionary() >>> if symbol in d: >>> h = d[symbol] >>> else >>> h = d[symbol] = YStock(symbol).historical() >>> print d[0].adjusted_close >>> >>> PersistentDictionary() is like shelve but uses sqlite and therefore is >>> thread safe. >>> >>> >>> >>> >>> On Wednesday, 9 April 2014 10:29:14 UTC-5, Trent Telfer wrote: >>>> >>>> Brian M, >>>> >>>> Thanks for the reply. I am looking into doing the following with >>>> historical market information >>>> >>>> a) Base Table >>>> >>>> ID,Exchange,Description,Base Currency >>>> Char(30),Char(10),Varchar(256),Char(5) >>>> >>>> b) Market Data >>>> >>>> ID,Date,HighPrice,LowPrice,OpenPrice,ClosePrice,Volume >>>> Char(30),Date,Float,Float,Float,Float,Long >>>> >>>> I am unsure if that table setup is the best choice or if there is a >>>> better way to approach it? >>>> >>>> I've also been wondering if I should jump into the world of NoSQL >>>> (specifically cassandra) as I may need to have smaller pricing intervals >>>> than days in the future. >>>> >>>> -Trent >>>> >>>> >>>> On Tuesday, April 8, 2014 6:48:50 PM UTC-6, Brian M wrote: >>>>> >>>>> Assuming each source needs the same data fields, how about just using >>>>> one table and including an extra field to specify which source each >>>>> record >>>>> came from? >>>>> Or if you really want a separate table for each timeseries, you could >>>>> look into using table inheritance. >>>>> http://web2py.com/books/default/chapter/29/06/the-database-abstraction-layer#Table-inheritance >>>>> >>>>> A little more about how you are planning to use or display the data >>>>> might help. 38 separate tables, one table with 38 columns or rows? As a >>>>> graph with each series being a line? >>>>> >>>>> ~Brian >>>>> >>>>>
-- Resources: - http://web2py.com - http://web2py.com/book (Documentation) - http://github.com/web2py/web2py (Source code) - https://code.google.com/p/web2py/issues/list (Report Issues) --- You received this message because you are subscribed to the Google Groups "web2py-users" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
