Errors, bugs, questions - page 1983

 
STARIJ:
From your broker. This question comes up every week...

And then how do you link the BROKER mt4 to your site? Is there an article or video?

 
Vladimir Pastushak:

A note for the future: http://orf.textologia.ru/definit/bezindikatorniy/?q=532&n=7113

 
Vitaly Muzichenko:

A note for the future


Yes, thank you.

 
pavlick_:

Colleagues, has anyone encountered attempts to implement STL c++ library (correcting for language capabilities)? Preferably with codobase placement, search says no, but vrdug . . Certainly not all, but some basic stuff: a few normal containers via templates, and that auto_ptr (I certainly wrote my own, but self-writing such primitives is silly).
I don't understand why std µl is non-template and why we had to invent new names for methods/classes analogues in STL.

According to some information from MQ, they are preparing something along these lines as part of their standard library. What exactly and how complete it is, I am not aware of. I found this out because I have been writing my own version for some time now. But in the light of this information suspended the case, I want to see what will be released officially.

 
Vladimir Pastushak:

It's spelled correctly. In Russian, Y is changed to Y after consonantal prefixes. For example, prefixes. So syndicatorless, oddly enough, is correct.

 

I have encountered such a glitch in MT5 as an indicator buffer remaining.

The situation is as follows: I write an indicator with 4 buffers, compile it, put it on a chart, think about it.

Then I decide to bypass two buffers and leave two buffers. I partially remove them, i.e. I do not fill them with data.

I look at the chart where the indicator has been applied and there is trash in the zeroed buffer (in one buffer at that) - change of TF and restart of the indicator do not help.

The indicator is launched on a separate chart without glitches, i.e. there is no rubbish.

Is it meant to be like this, or it is a bug after all?

 
Aleksey Vyazmikin:

I have encountered such a glitch in MT5 as an indicator buffer remaining.

The situation is as follows: I write an indicator with 4 buffers, compile it, put it on a chart, think about it.

Then I decide to bypass two buffers and leave one buffer. I partially remove it, i.e. I do not fill it with data.

I look at the chart where the indicator has been applied and there is trash in the zeroed buffer (in one buffer at that) - change of TF and restart of the indicator do not help.

The indicator is launched on a separate chart without glitches, i.e. there is no rubbish.

Is it meant to be like this, or it is a bug after all?

It was designed that way. The cleanness of the array and / or the indicator buffer is the concern of the programmer.

 

Yes, and it's funny how the calculation-deprived buffers have swapped the first one to become the second one...

 
Alexey Viktorov:

That's how it's designed. The cleanliness of the array and/or indicator buffer is the programmer's concern.


Shouldn't the buffer disappear after the indicator disappears?

However, maybe the calculations need to be de-initialised in some way, but how?

 

My point is that since the calculations remain, running a different indicator, with the same buffer name, may produce an unpleasant result... I have to check this hypothesis, of course...

Reason: