Errors, bugs, questions - page 853

New comment
 

1. Why does the editor close when I press view-tools? The tool window doesn't show up at all.

2. Why does it show the content in the help, but not the articles themselves? Updating the help does not help.

3.Terminal log says OpenCl not found, install the latest videodriver for better perfomance. Also it says no comments.

I tried to restore my Windows and install/update MT5, it did not help. I don't want to tear it all down and install it again.

 
Virty:

1. Why does the editor close when I press view-tools? The tool window does not show up at all.

Does it mean that the editor is down? What version of the editor, 705?


2. Why does it show the contents of the help file but not the articles themselves? Updating the help does not help.

Can you provide a screenshot?


3.Terminal log says OpenCl not found, install the latest videodriver for better perfomance. Also it says no comments.

It is desirable to update the video drivers to the latest ones. What does it say? Copy the lines from the logbook - this can easily be done from the menu.


I tried to restore my Windows and install/update MT5, it did not help. I don't want to tear it all down and install it again.

What is the exact version of the operating system? Maybe a very old Windows XP without service packs?

To update to the latest terminal version, connect to MetaQuotes-Demo server and open a demo account on it.

 

I have encountered an unknown glitch. On Friday, close to midnight, the Expert Advisor stopped testing on the NordFX-Server, although a couple of hours ago everything was fine. By printing, etc. I have received:

NN      0       Core 1  13:38:13        2009.10.01 08:00:00   SYMBOL_TRADE_EXECUTION_INSTANT
FH      0       Core 1  13:38:13        2009.10.01 08:00:00   _price:2.2168; price:2.2168
OL      0       Core 1  13:38:13        2009.10.01 08:00:00   0.1;50.0;0.1
HD      0       Core 1  13:38:13        2009.10.01 08:00:00   tp:2.0517; sl:2.2616
PM      0       Core 1  13:38:13        2009.10.01 08:00:00   OrderType:ORDER_TYPE_SELL_LIMIT
HL      2       Core 1  13:38:13        2009.10.01 08:00:00   failed sell limit 0.10 GBPNZD at 2.2168 sl: 2.2616 tp: 2.0517 [Invalid order type]

_price - adjusted price (after checking for stop and free levels);

price - initial price;

in the third line - min. lot, max. lot and trade volume.

I did everything I could - zeroed the fields and reinvented the help, and pogatilos the forum - the problem could not be solved. By the way, I use this function in my Championship and it was used in the previous one too, i.e. I'm more or less sure about it.

I have tried GoMarkets-Demo server - the same thing.

I have tried Metaquotes-Demo server - it works as it should (!?).

I have now run the same Expert Advisor on NordFX-Server at market opening (though the terminal has already been updated):

JM      0       Core 1  01:36:07        2009.10.01 08:00:00   SYMBOL_TRADE_EXECUTION_INSTANT
RI      0       Core 1  01:36:07        2009.10.01 08:00:00   _price:2.2168; price:2.2168
CK      0       Core 1  01:36:07        2009.10.01 08:00:00   0.1;50.0;0.1
LE      0       Core 1  01:36:07        2009.10.01 08:00:00   tp:2.0517; sl:2.2616
LM      0       Core 1  01:36:07        2009.10.01 08:00:00   OrderType:ORDER_TYPE_SELL_LIMIT
PP      0       Core 1  01:36:07        GBPUSD: symbol to be synchronized
RE      0       Core 1  01:36:07        GBPUSD: symbol synchronized, 3464 bytes of symbol info received
QL      0       Core 1  01:36:07        GBPUSD: load 4310 bytes of history data to synchronize
FJ      0       Core 1  01:36:07        GBPUSD: history synchronized from 2000.01.03 to 2012.10.05
MS      0       Core 1  01:36:07        2009.10.01 08:00:08   sell limit 0.10 GBPNZD at 2.2168 sl: 2.2616 tp: 2.0517 (2.2148 / 2.2150 / 2.2148)

As you can see - everything is the same, only one small difference (I don't know, is it important or not) - GBPUSD was pumped again and my trade was successful.

At least I forgot about everything and had a rest during weekend, instead of thinking what was wrong :)

 

New warnings in 705 are nice:

possible use of uninitialized variable 'openp'  mvpUtils.mqh    47      146
possible use of uninitialized variable 'cnt2'   mvpUtils.mqh    93      20
possible use of uninitialized variable 'RetF'   mvpUtils.mqh    97      19
Not only all variables are exactly initialized(openp is initialized at in transactions, and if all history is queried, then clearly in will be before out or inout - but I agree that in this case, the compiler is not a telepath; cnt2 - I think there is a small flaw:
int cnt2;
for (int f = 1 ; f <= 100; f++) {
    tempHPR = 1;
    cnt2 = 0;
//------------------какой-то код--------
}
Print(cnt2); // <- на эту строчку выдаёт warning

It's obvious that the loop will be executed at least once; RetF - similarly to openp, i.e. I admit that the compiler had a right to issue a warning)

But the question is different: errors in the inlider are produced only when compiling something which uses this inlider. If you compile the inline itself, the warning won't show up.

 
notused:

New warnings in 705 are nice:

It is not enough that all the variables are initialized exactly(openp is initialized at in transactions, and if the whole history is queried, then clearly in will be earlier than out or inout - but I agree that in this case the compiler is not a telepath; cnt2 - I think there is a small defect:

It's obvious that the loop will be executed at least once; RetF - similarly to openp, i.e. I admit that the compiler had a right to issue a warning)

This is the first version of the analyzer for now, we will tune it up.


But the question is different: errors in the inlider are generated only when compiling something which uses this inlider. If you compile the inline itself, you won't get warnings.

The inluders don't pass through code generation, so some of the checks don't work.

This is because inluders are not standalone programs and their functions are almost completely removed during optimization, as they have no entry points for analysis of execution threads.

 
Renat:
The inluders don't pass the code generation stage and that's why some of the checks don't work.

This is because inluders are not a standalone program and their functions are almost completely removed when optimised, as they have no entry points for analysis of execution threads.

OK, I see.
 
Renat:
You mean the editor is down? What version of the editor, 705?

Do you mean the editor is crashing? What version of the editor, 705? - MetaEditor 5.00 Build 705 (05 Oct 2012) The editor can compile and run an old EA, which is in good condition. But if there is an error in the Expert Advisor, it chews it and nothing happens - the Toolbox with the error does not appear. Attached is a screenshot before the crash

 
Renat:
Can you give me a screenshot?
I am attaching a screenshot. Windows help doesn't work in other programs either. In the past, this could only be fixed by reinstalling XP SP3 completely. I can not fix this error. I will go to your site for help. By the way, the help is much better on the website than on the computer.
 
Virty:

You mean the editor is crashing? What version of the editor, 705? - MetaEditor 5.00 Build 705 (05 Oct 2012) The editor can compile and run an old EA in good condition. But if there is an error in the Expert Advisor, it chews it and nothing happens - the Toolbox with the error does not appear. Here is a screenshot before the crash

Let's send with all details (axis version, bit rate, logging, IE version) to servicedesk. We tried it on our side - it does not repeat.

 

The news in Russian is like this:

windows 8 64 bit 705 build, liteforex.

Where do I go with this? To the dc, to the service desk?

1...846847848849850851852853854855856857858859860...3190
New comment