开云体育

ctrl + shift + ? for shortcuts
© 2025 开云体育
Date

Re: twsapi: Re: Timing of events

Michel
 

开云体育

Thanks Scott, this is as I understand it.
So there is no way to detect various trades at the same price and size?
This could be a very bad issue because it is important (at least for me, and I think for many people) to compute volume along the day.
?
Michel.

-----Mensaje original-----
De: marinindextrader [mailto:marinindextrader@...]
Enviado el: miércoles, 19 de junio de 2002 19:38
Para: twsapi@...
Asunto: twsapi: Re: Timing of events

Both tickPrice and tickSize are autonamous of one another. Further
they only fire when there is a change.

An example would be 10 trades going off at the same price at
different volumes would fire zero price events, and 10 size events.

and vice versa

10 trades at different price but the same size would fire 10 price
events and zero size events.

and ten trades going off the same price and size would have the same
effect as if no trade had occured at all.

That is my understanding and experience with tickPrice and tickSize.

Scott


--- In twsapi@y..., "migurull" wrote:
> Hello,
>
> I'm evaluating TWS ActiveX component in Visual Basic and I have
some
> questions/problems regarding 'tickPrice' and 'tickSize' events (I'm
> using the web based Demo system for this evaluations):
>
> 1- When last price AND last size change at the same time(new
> quotation), is there some guarantee about the order in which
> tickPrice and tickSize events are rised? (ALWAYS tickPrice BEFORE
> tickSize?)
>
> 2- When quotation repeats at the same price and the same size, it
> seems that there is no tickPrice, nor tickSize events fired. In
this
> case, how to detect that a new quotation had taken place (for
> example, in order to compute total traded volume by adding all last
> sizes)?
>
> Thanks for your help.
>
> Michel.



To unsubscribe from this group, send an email to:
twsapi-unsubscribe@...



Your use of Yahoo! Groups is subject to the .


Re: Timing of events

marinindextrader
 

Both tickPrice and tickSize are autonamous of one another. Further
they only fire when there is a change.

An example would be 10 trades going off at the same price at
different volumes would fire zero price events, and 10 size events.

and vice versa

10 trades at different price but the same size would fire 10 price
events and zero size events.

and ten trades going off the same price and size would have the same
effect as if no trade had occured at all.

That is my understanding and experience with tickPrice and tickSize.

Scott


--- In twsapi@y..., "migurull" <migurull@t...> wrote:
Hello,

I'm evaluating TWS ActiveX component in Visual Basic and I have
some
questions/problems regarding 'tickPrice' and 'tickSize' events (I'm
using the web based Demo system for this evaluations):

1- When last price AND last size change at the same time(new
quotation), is there some guarantee about the order in which
tickPrice and tickSize events are rised? (ALWAYS tickPrice BEFORE
tickSize?)

2- When quotation repeats at the same price and the same size, it
seems that there is no tickPrice, nor tickSize events fired. In
this
case, how to detect that a new quotation had taken place (for
example, in order to compute total traded volume by adding all last
sizes)?

Thanks for your help.

Michel.


Timing of events

migurull
 

Hello,

I'm evaluating TWS ActiveX component in Visual Basic and I have some
questions/problems regarding 'tickPrice' and 'tickSize' events (I'm
using the web based Demo system for this evaluations):

1- When last price AND last size change at the same time(new
quotation), is there some guarantee about the order in which
tickPrice and tickSize events are rised? (ALWAYS tickPrice BEFORE
tickSize?)

2- When quotation repeats at the same price and the same size, it
seems that there is no tickPrice, nor tickSize events fired. In this
case, how to detect that a new quotation had taken place (for
example, in order to compute total traded volume by adding all last
sizes)?

Thanks for your help.

Michel.


C# .NET + TWS ActiveX = 'Catastrophic failure' exception?

pjslim99
 

I have connected fine using Java and C++ API to TWS, however using
Visual Studio.NET and C#,
I get

'An unhandled exception of type:
System.Runtime.InteropServices.COMException
occured in mscorlib.dll: Catastrophic failure'

I am doing the following

- Start TWS running locally.
- TWS: Settings->Enable ActiveX and Socket Clients
- Create new C# project
- Solution Explorer: Add Reference: Tws ActiveX Control module
(Tws.ocx)
- Add Button with handler and code:

using TWSLib;

.
.

TwsClass myTws = new TwsClass();
myTws.connect("", 7496);

.
.

- and run the program.

If I call any function or retrieve a property from the ActiveX
control I get the same exception.

Any thoughts? Anyone succesfully got C# + TWS API working?

thanks,
PJ


Re: twsapi: Java API

gu6526
 

There are basically two methods to do this:

1. wait() / notifyAll() communication

In your main program use wait() to go to sleep without polling.
Catch the interruptedException and go on.

In the wrapper functions that receive the data do some checking that
all your requested data have arrived and reactivate your main with
notifyAll(). The wrapper functions are called asynchronously by the
reader thread that was spawned from your main with the connect() call.

2. If you use a GUI, e.g. some Swing objects you can use the
appropriate event handling to update your presentation objects with
the asynchrounously received ticket data. The TestJavaClient
application from IB is an example.

Gernot

In twsapi@y..., usernew <no_reply@y...> wrote:

I need to get info on n tickers, so this is what I do at the moment:

tws.connect();
for(i=0; i<n; i++){
tws.requestdata(...)
}
tws.wait_trans(); // to make sure all the prices have arrived
tws.disconnect();

// prices have arrived
// so do stuff here

Looking at the docs for request data, because I need info on n
tickers, seems I would have to call requestdata n times

My concern was making tws.wait_trans() not consume cycles.


Re: twsapi: Java API

usernew
 

I need to get info on n tickers, so this is what I do at the moment:

tws.connect();
for(i=0; i<n; i++){
tws.requestdata(...)
}
tws.wait_trans(); // to make sure all the prices have arrived
tws.disconnect();

// prices have arrived
// so do stuff here

Looking at the docs for request data, because I need info on n
tickers, seems I would have to call requestdata n times

My concern was making tws.wait_trans() not consume cycles.

--- In twsapi@y..., "David" <kotomo@p...> wrote:
usernew,
maybe i do not know what you want or misunderstand TWS control but
you need to send requestData() only once - you can do it before the
session.
David
----- Original Message -----
From: usernew
To: twsapi@y...
Sent: Wednesday, June 19, 2002 3:56 AM
Subject: Re: twsapi: Java API


So you are saying tickPrice, tickSize etc. are guaranteed to
happen
by the time requestData() returns ? I thought tickPrice etc. were
completely asynchronous.

--- In twsapi@y..., Todd Turner <todd_a_turner@y...> wrote:
> I believe that the callbacks only occur during a call
> to requestData().
>
> In my initial testing I used Sleep with no problem. I
> now have Windows sending a WM_TIMER message to my apps
> message queue every 250 msecs now.
> --- usernew <no_reply@y...> wrote:
> > Yeah, I tried that and got exceptions. The problem
> > might be that when
> > you sleep the whole thread is put to sleep, so the
> > event tickprice
> > etc. callbacks can't happen. So, a regular sleep
> > might not work well
> > in this case.
> >
> > --- In twsapi@y..., Todd Turner <todd_a_turner@y...>
> > wrote:
> > > I am not familiar with Java, but you need some
> > sort of
> > > system call Sleep(int timeout_msecs), so that the
> > > process blocks until the timeout occurs. This
> > > functionality is provided by the OS.
> > >
> > >
> > > --- usernew <no_reply@y...> wrote:
> > > > My program currently is structured as follows:
> > > >
> > > > class TWS implements EWrapper {
> > > > public void requestdata(...)
> > > > public void tickPrice(...)
> > > > public void tickSize(...)
> > > > ...
> > > > public void wait_trans() {
> > > > for(i=0; i < xxxxx; i++);
> > > > }
> > > > }
> > > >
> > > > public class Main {
> > > > tws.connect();
> > > > for(i=0; i<n; i++)
> > > > tws.requestdata(...)
> > > > tws.wait_trans();
> > > > tws.disconnect();
> > > > }
> > > >
> > > > Is there a better way to structure this program
> > ?
> > > >
> > > > My concern is wait_trans(). It is continuously
> > > > eating up cycles. My
> > > > program does not need to do anything while it is
> > > > waiting for the
> > > > arrays to be populated, but I would still like
> > to
> > > > wait without
> > > > consuming cycles.
> > > >
> > > > Thanks in advance
> > > >
> > > >
> > >
> > >
> > > __________________________________________________
> > > Do You Yahoo!?
> > > Yahoo! - Official partner of 2002 FIFA World Cup
> > >
> >
> >
>
>
> __________________________________________________
> Do You Yahoo!?
> Yahoo! - Official partner of 2002 FIFA World Cup
>


Yahoo! Groups Sponsor
ADVERTISEMENT




To unsubscribe from this group, send an email to:
twsapi-unsubscribe@y...



Your use of Yahoo! Groups is subject to the Yahoo! Terms of
Service.


Re: twsapi: Java API

David
 

开云体育

usernew,
maybe i do not know what you want or misunderstand TWS control but you need to send requestData() only once - you can do it before the session.
David

----- Original Message -----
From: usernew
Sent: Wednesday, June 19, 2002 3:56 AM
Subject: Re: twsapi: Java API

So you are saying tickPrice, tickSize etc. are guaranteed to happen
by the time requestData() returns ? I thought tickPrice etc. were
completely asynchronous.

--- In twsapi@y..., Todd Turner <todd_a_turner@y...> wrote:
> I believe that the callbacks only occur during a call
> to requestData().
>
> In my initial testing I used Sleep with no problem.? I
> now have Windows sending a WM_TIMER message to my apps
> message queue every 250 msecs now.
> --- usernew wrote:
> > Yeah, I tried that and got exceptions. The problem
> > might be that when
> > you sleep the whole thread is put to sleep, so the
> > event tickprice
> > etc. callbacks can't happen. So, a regular sleep
> > might not work well
> > in this case.
> >
> > --- In twsapi@y..., Todd Turner > > wrote:
> > > I am not familiar with Java, but you need some
> > sort of
> > > system call Sleep(int timeout_msecs), so that the
> > > process blocks until the timeout occurs.? This
> > > functionality is provided by the OS.
> > >
> > >
> > > --- usernew wrote:
> > > > My program currently is structured as follows:
> > > >
> > > > class TWS implements EWrapper {
> > > >?? public void requestdata(...)
> > > >?? public void tickPrice(...)
> > > >?? public void tickSize(...)
> > > >?? ...
> > > >?? public void wait_trans() {
> > > >???? for(i=0; i < xxxxx; i++);
> > > >?? }
> > > > }
> > > >
> > > > public class Main {
> > > >??? tws.connect();
> > > >??? for(i=0; i> > > >?????? tws.requestdata(...)
> > > >??? tws.wait_trans();
> > > >??? tws.disconnect();
> > > > }
> > > >
> > > > Is there a better way to structure this program
> > ?
> > > >
> > > > My concern is wait_trans(). It is continuously
> > > > eating up cycles. My
> > > > program does not need to do anything while it is
> > > > waiting for the
> > > > arrays to be populated, but I would still like
> > to
> > > > wait without
> > > > consuming cycles.
> > > >
> > > > Thanks in advance
> > > >
> > > >
> > >
> > >
> > > __________________________________________________
> > > Do You Yahoo!?
> > > Yahoo! - Official partner of 2002 FIFA World Cup
> > >
> >
> >
>
>
> __________________________________________________
> Do You Yahoo!?
> Yahoo! - Official partner of 2002 FIFA World Cup
>



To unsubscribe from this group, send an email to:
twsapi-unsubscribe@...



Your use of Yahoo! Groups is subject to the .


Re: twsapi: Java API

gu6526
 

You should look at EClientSocket.java:
It spawns an asynchronous thread with

tm_reader = new EReader( this, dis);

The wrapper functions are called from this thread whenever a message
is received by the client. So there is no need to wait! It's done
all asynchronously.

Gernot

--- In twsapi@y..., Todd Turner <todd_a_turner@y...> wrote:
I believe that the callbacks only occur during a call
to requestData().

In my initial testing I used Sleep with no problem. I
now have Windows sending a WM_TIMER message to my apps
message queue every 250 msecs now.
--- usernew <no_reply@y...> wrote:
Yeah, I tried that and got exceptions. The problem
might be that when
you sleep the whole thread is put to sleep, so the
event tickprice
etc. callbacks can't happen. So, a regular sleep
might not work well
in this case.

--- In twsapi@y..., Todd Turner <todd_a_turner@y...>
wrote:
I am not familiar with Java, but you need some
sort of
system call Sleep(int timeout_msecs), so that the
process blocks until the timeout occurs. This
functionality is provided by the OS.


--- usernew <no_reply@y...> wrote:
My program currently is structured as follows:

class TWS implements EWrapper {
public void requestdata(...)
public void tickPrice(...)
public void tickSize(...)
...
public void wait_trans() {
for(i=0; i < xxxxx; i++);
}
}

public class Main {
tws.connect();
for(i=0; i<n; i++)
tws.requestdata(...)
tws.wait_trans();
tws.disconnect();
}

Is there a better way to structure this program
?

My concern is wait_trans(). It is continuously
eating up cycles. My
program does not need to do anything while it is
waiting for the
arrays to be populated, but I would still like
to
wait without
consuming cycles.

Thanks in advance


__________________________________________________
Do You Yahoo!?
Yahoo! - Official partner of 2002 FIFA World Cup

__________________________________________________
Do You Yahoo!?
Yahoo! - Official partner of 2002 FIFA World Cup


Re: twsapi: Java API

Todd Turner
 

yep, I am pretty sure that all that requestData() does
is read data that has arrived from TWS, and call the
appropriate EWrapper member function for tickPrice
etc.

If it was async, there would be no need to call
requestData() every second, or ever.


--- usernew <no_reply@...> wrote:
So you are saying tickPrice, tickSize etc. are
guaranteed to happen
by the time requestData() returns ? I thought
tickPrice etc. were
completely asynchronous.

--- In twsapi@y..., Todd Turner <todd_a_turner@y...>
wrote:
I believe that the callbacks only occur during a
call
to requestData().

In my initial testing I used Sleep with no
problem. I
now have Windows sending a WM_TIMER message to my
apps
message queue every 250 msecs now.
--- usernew <no_reply@y...> wrote:
Yeah, I tried that and got exceptions. The
problem
might be that when
you sleep the whole thread is put to sleep, so
the
event tickprice
etc. callbacks can't happen. So, a regular sleep
might not work well
in this case.

--- In twsapi@y..., Todd Turner
<todd_a_turner@y...>
wrote:
I am not familiar with Java, but you need some
sort of
system call Sleep(int timeout_msecs), so that
the
process blocks until the timeout occurs. This
functionality is provided by the OS.


--- usernew <no_reply@y...> wrote:
My program currently is structured as
follows:

class TWS implements EWrapper {
public void requestdata(...)
public void tickPrice(...)
public void tickSize(...)
...
public void wait_trans() {
for(i=0; i < xxxxx; i++);
}
}

public class Main {
tws.connect();
for(i=0; i<n; i++)
tws.requestdata(...)
tws.wait_trans();
tws.disconnect();
}

Is there a better way to structure this
program
?

My concern is wait_trans(). It is
continuously
eating up cycles. My
program does not need to do anything while
it is
waiting for the
arrays to be populated, but I would still
like
to
wait without
consuming cycles.

Thanks in advance


__________________________________________________
Do You Yahoo!?
Yahoo! - Official partner of 2002 FIFA World
Cup

__________________________________________________
Do You Yahoo!?
Yahoo! - Official partner of 2002 FIFA World Cup

__________________________________________________
Do You Yahoo!?
Yahoo! - Official partner of 2002 FIFA World Cup


AW: twsapi: Java API

 

开云体育

You should look at EClientSocket.java:

?

It spawns an asynchronous thread with

tm_reader = new EReader( this, dis);

?

The wrapper functions are called from this thread whenever a message is received

by the client.? So there is no need to wait!? It’s done all asynchronously.

?

Gernot

?

-----Ursprüngliche Nachricht-----
Von: usernew [mailto:no_reply@...]
Gesendet: Tuesday, June 18, 2002 4:24 PM
An: twsapi@...
Betreff: twsapi: Java API

?

My program currently is structured as follows:

class TWS implements EWrapper {
? public void requestdata(...)
? public void tickPrice(...)
? public void tickSize(...)
? ...
? public void wait_trans() {
??? for(i=0; i < xxxxx; i++);
? }
}

public class Main {
?? tws.connect();
?? for(i=0; i
????? tws.requestdata(...)
?? tws.wait_trans();
?? tws.disconnect();
}

Is there a better way to structure this program ?

My concern is wait_trans(). It is continuously eating up cycles. My
program does not need to do anything while it is waiting for the
arrays to be populated, but I would still like to wait without
consuming cycles.

Thanks in advance



To unsubscribe from this group, send an email to:
twsapi-unsubscribe@...



Your use of Yahoo! Groups is subject to the .



Re: twsapi: Java API

usernew
 

So you are saying tickPrice, tickSize etc. are guaranteed to happen
by the time requestData() returns ? I thought tickPrice etc. were
completely asynchronous.

--- In twsapi@y..., Todd Turner <todd_a_turner@y...> wrote:
I believe that the callbacks only occur during a call
to requestData().

In my initial testing I used Sleep with no problem. I
now have Windows sending a WM_TIMER message to my apps
message queue every 250 msecs now.
--- usernew <no_reply@y...> wrote:
Yeah, I tried that and got exceptions. The problem
might be that when
you sleep the whole thread is put to sleep, so the
event tickprice
etc. callbacks can't happen. So, a regular sleep
might not work well
in this case.

--- In twsapi@y..., Todd Turner <todd_a_turner@y...>
wrote:
I am not familiar with Java, but you need some
sort of
system call Sleep(int timeout_msecs), so that the
process blocks until the timeout occurs. This
functionality is provided by the OS.


--- usernew <no_reply@y...> wrote:
My program currently is structured as follows:

class TWS implements EWrapper {
public void requestdata(...)
public void tickPrice(...)
public void tickSize(...)
...
public void wait_trans() {
for(i=0; i < xxxxx; i++);
}
}

public class Main {
tws.connect();
for(i=0; i<n; i++)
tws.requestdata(...)
tws.wait_trans();
tws.disconnect();
}

Is there a better way to structure this program
?

My concern is wait_trans(). It is continuously
eating up cycles. My
program does not need to do anything while it is
waiting for the
arrays to be populated, but I would still like
to
wait without
consuming cycles.

Thanks in advance


__________________________________________________
Do You Yahoo!?
Yahoo! - Official partner of 2002 FIFA World Cup

__________________________________________________
Do You Yahoo!?
Yahoo! - Official partner of 2002 FIFA World Cup


Re: twsapi: Java API

Todd Turner
 

btw,

Have you been able to call connect, and then call
disconnect and then call connect again? The second
call to connect always seems to cause an exception
inside the API, for me anyway.


--- Todd Turner <todd_a_turner@...> wrote:
I believe that the callbacks only occur during a
call
to requestData().

In my initial testing I used Sleep with no problem.
I
now have Windows sending a WM_TIMER message to my
apps
message queue every 250 msecs now.
--- usernew <no_reply@...> wrote:
Yeah, I tried that and got exceptions. The problem
might be that when
you sleep the whole thread is put to sleep, so the
event tickprice
etc. callbacks can't happen. So, a regular sleep
might not work well
in this case.

--- In twsapi@y..., Todd Turner
<todd_a_turner@y...>
wrote:
I am not familiar with Java, but you need some
sort of
system call Sleep(int timeout_msecs), so that
the
process blocks until the timeout occurs. This
functionality is provided by the OS.


--- usernew <no_reply@y...> wrote:
My program currently is structured as follows:

class TWS implements EWrapper {
public void requestdata(...)
public void tickPrice(...)
public void tickSize(...)
...
public void wait_trans() {
for(i=0; i < xxxxx; i++);
}
}

public class Main {
tws.connect();
for(i=0; i<n; i++)
tws.requestdata(...)
tws.wait_trans();
tws.disconnect();
}

Is there a better way to structure this
program
?

My concern is wait_trans(). It is continuously
eating up cycles. My
program does not need to do anything while it
is
waiting for the
arrays to be populated, but I would still like
to
wait without
consuming cycles.

Thanks in advance


__________________________________________________
Do You Yahoo!?
Yahoo! - Official partner of 2002 FIFA World Cup

__________________________________________________
Do You Yahoo!?
Yahoo! - Official partner of 2002 FIFA World Cup


__________________________________________________
Do You Yahoo!?
Yahoo! - Official partner of 2002 FIFA World Cup


Re: twsapi: Java API

Todd Turner
 

I believe that the callbacks only occur during a call
to requestData().

In my initial testing I used Sleep with no problem. I
now have Windows sending a WM_TIMER message to my apps
message queue every 250 msecs now.
--- usernew <no_reply@...> wrote:
Yeah, I tried that and got exceptions. The problem
might be that when
you sleep the whole thread is put to sleep, so the
event tickprice
etc. callbacks can't happen. So, a regular sleep
might not work well
in this case.

--- In twsapi@y..., Todd Turner <todd_a_turner@y...>
wrote:
I am not familiar with Java, but you need some
sort of
system call Sleep(int timeout_msecs), so that the
process blocks until the timeout occurs. This
functionality is provided by the OS.


--- usernew <no_reply@y...> wrote:
My program currently is structured as follows:

class TWS implements EWrapper {
public void requestdata(...)
public void tickPrice(...)
public void tickSize(...)
...
public void wait_trans() {
for(i=0; i < xxxxx; i++);
}
}

public class Main {
tws.connect();
for(i=0; i<n; i++)
tws.requestdata(...)
tws.wait_trans();
tws.disconnect();
}

Is there a better way to structure this program
?

My concern is wait_trans(). It is continuously
eating up cycles. My
program does not need to do anything while it is
waiting for the
arrays to be populated, but I would still like
to
wait without
consuming cycles.

Thanks in advance


__________________________________________________
Do You Yahoo!?
Yahoo! - Official partner of 2002 FIFA World Cup

__________________________________________________
Do You Yahoo!?
Yahoo! - Official partner of 2002 FIFA World Cup


Re: twsapi: Java API

usernew
 

Yeah, I tried that and got exceptions. The problem might be that when
you sleep the whole thread is put to sleep, so the event tickprice
etc. callbacks can't happen. So, a regular sleep might not work well
in this case.

--- In twsapi@y..., Todd Turner <todd_a_turner@y...> wrote:
I am not familiar with Java, but you need some sort of
system call Sleep(int timeout_msecs), so that the
process blocks until the timeout occurs. This
functionality is provided by the OS.


--- usernew <no_reply@y...> wrote:
My program currently is structured as follows:

class TWS implements EWrapper {
public void requestdata(...)
public void tickPrice(...)
public void tickSize(...)
...
public void wait_trans() {
for(i=0; i < xxxxx; i++);
}
}

public class Main {
tws.connect();
for(i=0; i<n; i++)
tws.requestdata(...)
tws.wait_trans();
tws.disconnect();
}

Is there a better way to structure this program ?

My concern is wait_trans(). It is continuously
eating up cycles. My
program does not need to do anything while it is
waiting for the
arrays to be populated, but I would still like to
wait without
consuming cycles.

Thanks in advance


__________________________________________________
Do You Yahoo!?
Yahoo! - Official partner of 2002 FIFA World Cup


Re: updateExtended(), extended properties, etc, etc

marinindextrader
 

Mick,

I havn't gotten to that point in my work, but perhaps ultimately OCA
is not server sided? In other words the TWS manages internally a
marker...that ostensibly is sent back and forth, but all of the OCA
work is done on the client side?

For me the point will be moot in a sense, because I intend to expand
greatly on the OCA concept.

Perhaps you might consider programming your own OCA procedures. That
way in the future, you too can also expand on them. Its simple enough
to add a flag to the dynamic array structure that tracks orders.

Just a thought...

Scott

--- In twsapi@y..., mickjenks <no_reply@y...> wrote:
Thanks Elliot,

However, I have read the TWS docs and understand the purpose of
OCA.
That was never the question. The placeOrder() method doesn't
include
in its declaration a variable for sending an OCA identifier, nor
does
it allow for TIF or Transmit to be sent with the individual orders.
Could someone explain how to implement these properties on an order
to order basis? Thanks


Re: updateExtended(), extended properties, etc, etc

bee_jay_61
 

--- In twsapi@y..., mickjenks <no_reply@y...> wrote:
Thanks Elliot,

However, I have read the TWS docs and understand the purpose of
OCA.
That was never the question. The placeOrder() method doesn't
include
in its declaration a variable for sending an OCA identifier, nor
does
it allow for TIF or Transmit to be sent with the individual orders.
Could someone explain how to implement these properties on an order
to order basis? Thanks
Once Again. Use the ActiveX properties to set OCA, TIF and Transmit.
The ActiveX is written in C++ and uses then WinSocket API. Look at
the C++ API for more information.


Re: updateExtended(), extended properties, etc, etc

mickjenks
 

Thanks Elliot,

However, I have read the TWS docs and understand the purpose of OCA.
That was never the question. The placeOrder() method doesn't include
in its declaration a variable for sending an OCA identifier, nor does
it allow for TIF or Transmit to be sent with the individual orders.
Could someone explain how to implement these properties on an order
to order basis? Thanks


Re: twsapi: Java API

Todd Turner
 

I am not familiar with Java, but you need some sort of
system call Sleep(int timeout_msecs), so that the
process blocks until the timeout occurs. This
functionality is provided by the OS.


--- usernew <no_reply@...> wrote:
My program currently is structured as follows:

class TWS implements EWrapper {
public void requestdata(...)
public void tickPrice(...)
public void tickSize(...)
...
public void wait_trans() {
for(i=0; i < xxxxx; i++);
}
}

public class Main {
tws.connect();
for(i=0; i<n; i++)
tws.requestdata(...)
tws.wait_trans();
tws.disconnect();
}

Is there a better way to structure this program ?

My concern is wait_trans(). It is continuously
eating up cycles. My
program does not need to do anything while it is
waiting for the
arrays to be populated, but I would still like to
wait without
consuming cycles.

Thanks in advance


__________________________________________________
Do You Yahoo!?
Yahoo! - Official partner of 2002 FIFA World Cup


Java API

usernew
 

My program currently is structured as follows:

class TWS implements EWrapper {
public void requestdata(...)
public void tickPrice(...)
public void tickSize(...)
...
public void wait_trans() {
for(i=0; i < xxxxx; i++);
}
}

public class Main {
tws.connect();
for(i=0; i<n; i++)
tws.requestdata(...)
tws.wait_trans();
tws.disconnect();
}

Is there a better way to structure this program ?

My concern is wait_trans(). It is continuously eating up cycles. My
program does not need to do anything while it is waiting for the
arrays to be populated, but I would still like to wait without
consuming cycles.

Thanks in advance


Re: twsapi: Re: VB and ActiveX....updateExtended() Method????

Elliot Lee
 

On 2002-06-18 9:13, "BrianL_98" <brianl_98@...> wrote:

The IB documentation for OCA orders via the TWS API is pretty thin.
The only thing I can find is that the string OCA property "identifies
an OCA group."

Does anyone know how to go about actually placing OCA orders using
this property? Am I supposed to submit two orders, and then change
the OCA property to include the order id of each of the orders?
As with many features, OCA functionality is parallel in the API and the GUI,
so reading the TWS docs on this would be a good idea.

The OCA string is an identifier assigned by you to group orders together. If
an order with an OCA string set is executed or cancelled, all other orders
with the same string will be cancelled.
-- Elliot