
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
A question was raised about FileReadArray, and in the explanation I pointed out that it has a feature that reads the wrong type of data.
hence the question: how does this feature fit in with language security?
If it's ok, why not make a standard function to convert a bit field to the right type?
In general, either put things in order and remove this feature or let people implement long arithmetic.
Here's an example of this feature through a wipe file:
Paranoia detected.
You should think it over.
I'm actually in favour of deeming it safe and making its counterpart without using a file, but directly.
But if MQ recognizes it as unsafe, then it should be removed from the function as well.
ZZZY By the way maybe I really lug something wrong today, I have not answered two questions, you enlighten me, point the finger where stupid.
Here's an example of this fix via file with wiping the screw:
When testing on crosses there is accurate modelling and other rates for correct conversion of profits and margin requirements.
Try running the visualisation and you will immediately understand the amount of calculations based on the number of background characters in the marketwatch.
Develop the thought.
I'm actually in favour of deeming it safe and making its counterpart without using a file, but directly.
But if MQ recognizes it as unsafe, it should be removed from the function as well.
ZZZY By the way maybe I'm really babbling today something not right, I have not answered two questions, you really enlighten me, point the finger where dumb.
The file itself is impersonal. It is not known beforehand which way this file was written. In other words, the functionality described above cannot be banned. Of course, you can organoleptically detect a pure text file, and there are some issues with Unicode and Ansi-encodings.
So you can read any file at will. And it won't lead to any dangerous situation, because the sizes of reads (and writes too) are controlled. You won't be able to break the stack. You will not be able to get an address into process memory in any way.
I can understand that, but a 13.7 times speed difference... Well, 2x is okay. And the puppyish joy of testing speed at opening prices on majors has been replaced by tearful dejection on crosses. And instead of using 30 native remote agents, we'll have to bribe the cludes again... Abyss!
The file itself is an impersonal file. It is not known in advance how the file was recorded. In other words, the functionality described above cannot be prohibited. It is of course possible to organoleptically detect a pure text file, and there are nuances with Unicode and Ansi encodings.
So you can read any file at will. And it won't lead to any dangerous situation, because the sizes of reads (and writes too) are controlled. You won't be able to break the stack. You cannot get an address into process memory in any way.
Then I see no reason not to introduce a direct read function without type conversion, ala
Put it in a function and everything will be safe.
Then I see no reason not to introduce a direct data reading function without type conversion, ala
Put it in a function and everything will be safe.
struct __long { long v; }
__double a; __long b;
a.v=123.456;
b=a;
b.v=4638373815016729713;
not hard.
+ read besides the top links https://www.mql5.com/ru/articles/364
And conversion from one type to another can be done either using macros or functions.
but there's nothing really complicated about it.