use a framework/standard API for quote updates
Brought to you by:
sstratos
Menu
▾
▴
#47 use a framework/standard API for quote updates
Currently, it's not easy to update quotes. It's very confusing, sometimes it does not update it properly and it requires many clicks. Today google provides real time quotations and there is room for improvment.
I suggest to use http://www.amichel.com/ydownloader/ framework (as example) for updates. It could real time mode as well, so user don't need to interact at all.
Discussion
Logged In: YES
user_id=643884
Originator: NO
Hmmm, i don't have problems with the existing "Quote" facility. See our documentation about the time-of-day that stuff becomes available from Yahoo.
Regarding the "amichel" downloader: Reading its installation notes, it seems to be for Windows, so that won't work for us.
However, i know that our Quote API does need to be tidied up to work better with Qt4 and to make it easier to create new quote plugins. I intend to investigate that soon.
Logged In: YES
user_id=835200
Originator: NO
I've found and fixed the mentioned problem with quotation update.
See the patch here:
https://sourceforge.net/tracker/index.php?func=detail&aid=2010375&group_id=25632&atid=384885
As a side note it still would be useful to have a scheduled updates (like 5min), so user wouldn't need to make 10 clicks in order to receive the updated data.
Logged In: YES
user_id=643884
Originator: NO
Please raise "help" issues on the Help Forum rather than in the Issue Tracker.
blshkv wrote:
> As a side note it still would be useful to have a scheduled updates
> (like 5min), so user wouldn't need to make 10 clicks in order to
> receive the updated data.
The CSV Plugin has an "Auto Reload" option. Does that work for you?
Also there have been various discussions on our Forums about automatically loading data. Try the SourceForge search system.
IMHO, should not we go in for standards based and, or universally accepted protocol, FIX comes in to mind.
I think, it is time to split qtstalker's functionality, in to a decent front-end client and proxy/caching back-end server. The later could also be a locally configured caching server for both the EOD and LIVE quotes, which will talk to upstream exchanges and, or other data providers with the help of gateway/conversion plug-ins. This I'm sure will definitely optimize the use of bandwidth, reduce the load on upstream servers and will improve the the data loading speed of qtstalker clients. The said server should be a self-maintaining daemon, possibly controlled by cron jobs.
Do we still need the Berkley DB?
BTW, How do we remove obsolete data from qtstalker's db tables say older than 7 years, 3 days and, or a week? I'm new to qtstalker so still exploring it.
Do we have any project roadmap, timeline and, or gnat chart for future developments to follow?
Last but not the least, qtstalker is written and maintained in a good object oriented programming language and is based on an object oriented toolkit, but seems far off from object oriented by design to me.