开云体育

ctrl + shift + ? for shortcuts
© 2025 开云体育
Date
Re: status of com.ib.controller.ApiController API
Nick, in the context of Java, I think my usage of 'atomic' was correct. See the following:
By Richard L King · #47093 ·
Re: Do I need to create a GUI for using the IBAPI sample program TestCppClient. Thanks.
Ibapi code can be run from terminal. You don’t need a GUI.
By fchiesadoc@... · #47092 ·
Do I need to create a GUI for using the IBAPI sample program TestCppClient. Thanks.
Thanks.
By jnyusf · #47091 ·
Re: status of com.ib.controller.ApiController API
Yeah, (out) is a good idea.
By Nick <news1000@...> · #47090 ·
Re: status of com.ib.controller.ApiController API
[resent due to formatting issues] . No disagreement here. It is just that m_reqId++ is not one step but (at least) four: load, copy, increment, and store. In Java possibly unboxing, boxing, and
By Jürgen Reinold · #47089 ·
Re: status of com.ib.controller.ApiController API
. No disagreement here. It is just that m_reqId++ is not one step but (at least) four: load, copy, increment, and store. In Java possibly unboxing, boxing, and various value conversions as well. So
By Jürgen Reinold · #47088 ·
Re: status of com.ib.controller.ApiController API
There is a difference of perspective here. This is what I mean by atomic. From https://preshing.com/20130618/atomic-vs-non-atomic-operations also
By Nick <news1000@...> · #47087 ·
Re: status of com.ib.controller.ApiController API
Richard is right on this one. * In Java, C, C++, ... no guarantees are given that a post increment on an intrinsic is atomic, so m_reqId++ definitely is not. * two or more threads can get the same
By Jürgen Reinold · #47086 ·
Re: status of com.ib.controller.ApiController API
The increment is atomic. A thread will either see the pre or post increment value. It will never see some bytes of the pre-increment value and some bytes of the post-increment value. The increment is
By Nick <news1000@...> · #47085 ·
Re: status of com.ib.controller.ApiController API
Artyom, yes that's true of request IDs, but not of order ids. I haven't looked at your pull request, as I have no interest in the ApiController, so you may well have dealt with any synchronization
By Richard L King · #47084 ·
Re: Time in force of child order
It's TWS time by default but timezone can be specified explicitly either. ср, 21 апр. 2021 г., 22:19 Bruce B <bruceb444@...>:
By ds-avatar · #47083 ·
Re: Time in force of child order
There is a time parameter you receive at one of account update request which might be the time you need to use. This is probably provided other ways too. -Bruce
By Bruce B · #47082 ·
Re: Time in force of child order
Thanks, Richard! Now I saw I can also use GTD and specify an exact point in time when I want the order to be canceled automatically. The time shall be given in the format YYYYMMDD hh:mm:ss. Now I'm
By Despair · #47081 ·
Re: status of com.ib.controller.ApiController API
Richard, from EClient documentation it's not apparent that reqId parameters must be in any order. The only requirement I can read is 'unique identifier'. Artyom
By Artyom T. · #47080 ·
Re: status of com.ib.controller.ApiController API
Dmitry All the API request methods are declared as synchronized, so there will be no parallel usage of the socket if they are called from different threads. Managing request ids, order ids etc
By Richard L King · #47079 ·
Re: Time in force of child order
Yes, you need to set tif to GTC for the child order ( I hit the same thing some time back). And no, the trailing order (or any kind of stop-loss order) will not be cancelled automatically if you
By Richard L King · #47078 ·
Re: Time in force of child order
You ignored the first question. I take this as confirmation? So setting the tif for the trail order to GTC will have it in the market until it is executed or canceled. Then I think you also
By Despair · #47077 ·
adding timestamp as a field in the tickPrice object
I am streaming options quotes using the tickprice object, which is in line 734 of program.py. The options quotes stream nicely, but I notice there is no "timestamp" field there. Is there a good way
By ebtrader · #47076 ·
Re: Time in force of child order
One order has nothing to do with the other. In the scenario you explained, you might end up shorting the product - for example - if you have a margin (when all three of your orders go
By Bruce B · #47075 ·
Time in force of child order
I'm wondering the following, I have a buy limit order with an attached trailing stop order as child order. When the order is executed the trailing stop starts working, all fine. But the next day the
By Despair · #47074 ·