A division by zero error appeared in one indicator - page 10

New comment
 
The help forArrayBsearch doesn't say that it can work with an array of datetime type, maybe that's the problem? Maybe there are conflicts, or is it a full float counterpart and there should be no problem?
 
Hardly
 
Алексей Тарабанов:
Hardly

I don't know what to assume, because today the start and end dates were printed the same, which doesn't make sense, even if we assume that the start date is wrong, the end date should shift in relation to the start date, which hasn't happened.

 
Put the code on the front page, maybe someone can figure out the problem...
 
Check in the new 1974 build on MetaQuotes-Demo, please.
 
Sergey Savinkin:

https://www.mql5.com/ru/forum/262864/page5

The link is only to the page. It says #50 in the title of the post. Your #74 ))

Here #50 is the link to that particular post. PCM - copy the link, and paste it into your post.

Here's a link to your post #75, which I'm replying to now.

 
Renat Fatkhullin:
Check in the new1974 build on MetaQuotes-Demo, please.

Maybe1874?

 
Renat Fatkhullin:
Check in the new 1974 build on MetaQuotes-Demo, please.

Thank you for responding to the problem.

Downloaded the new build, will observe the results tomorrow when the trading session starts - I will report back after the tests.

 
Artyom Trishkin:

Here #50 is the link to that particular post. PCM - copy the link, and paste it into your post.

Here's a link to your post #75, which I'm responding to now.

Thank you, Artem.

 
Aleksey Vyazmikin:

Suppose we have

We know thatstart_time should be2018.04.28 23:00

So it turns out that the time is wrong here ?

Dac this. If you want time to be accurate to hours, you can do a simple thing: cut off minutes.

The algorithm is as follows:

1. You put the time in the structure;

2. Zero in on the minutes;

3. Convert the structure back to the time;

You get the time from 23:38 to 23:00. Well, if the error is repeated.

1...34567891011
New comment