Metatrader known bugs ... - page 42
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
Ooops, I just realized they redefined standard Windows constants. I can rename them in my code for Windows imports, but it is rather annoying. What a bright idea they use the same name, and even if they use, why the heck do they change their values?
#define FILE_SHARE_WRITE 0x00000002
#define FILE_SHARE_DELETE 0x00000004
[/CODE]
FILE_SHARE_WRITE 256
[/CODE]
Edit:
They seem to "help" us with some more Windows constants. The following fails as well.
[CODE]#define MB_ICONQUESTION 0x00000020
#define MB_ICONEXCLAMATION 0x00000030
#define MB_ICONSTOP 0x00000010
#define MB_YESNO 0x00000004while the following still works for unclear reason (but probably it won't last for long):
[CODE]#define FILE_END 2The real danger is that they do such (and others) changes without warning. And you cannot simply delete #define statement, because they apparently change values, sometimes.
It is just getting worse and worse
And then they change built in functions and old ex4 files will simply stop working or they will start doing some BS that nobody will know what they do. A bunch of "geniuses" that do not have a slightest idea what are they doing
New problems are just getting more and more instead less and less
What they are doing in build 660 is an absolute nonsense - a lot of code will have to be changed (even newly written) again
They are "equalizing" metatrader 4 and 5. And they decided to do it not by correcting the bs from metatrader 5 but by adjusting metatrader 4. That way they are going to make a lot of metatrader 4 code not working. They simply do not know to count : on 1 metatrader 5 user there is 100 metatrader 4 users. Their logic is fantastic. They still do not get it why metatrader 5 is not accepted
They are "equalizing" metatrader 4 and 5. And they decided to do it not by correcting the bs from metatrader 5 but by adjusting metatrader 4. That way they are going to make a lot of metatrader 4 code not working. They simply do not know to count : on 1 metatrader 5 user there is 100 metatrader 4 users. Their logic is fantastic. They still do not get it why metatrader 5 is not accepted
I do not get it either. It is a while since the MT5 started, so the code base should not be the reason any longer. Perhaps they overshot the price, but I do not know anything about their products pricing.
If the MT5 maintenance became expensive for them, they probably looked for some solution. However their decision of breaking the stable MT4 was a failure. It would have been quite acceptable to keep the old MQL implementation, while adding the new language as a (beta) add-on. They ruined world of occasionally coding traders by breaking the old syntax, and professionals by unfinished buggy implementation of MQL5 features.
I do not get it either. It is a while since the MT5 started, so the code base should not be the reason any longer. Perhaps they overshot the price, but I do not know anything about their products pricing. If the MT5 maintenance became expensive for them, they probably looked for some solution. However their decision of breaking the stable MT4 was a failure. It would have been quite acceptable to keep the old MQL implementation, while adding the new language as a (beta) add-on. They ruined world of occasionally coding traders by breaking the old syntax, and professionals by unfinished buggy implementation of MQL5 features.
For some time I thought that they did it for greed. But then what happened to people that made versions that were working just fine? Something else is going on there and it is becoming obvious that the current coding stuff there is not capable of making a workable version
The same is with stderror.mqh. If you are using the stderror.mqh file in your code (it is used in 90% of EAs - so they are useless now), do not try to recompile the code. This is what you get :
The best is that you DO NOT DOWNLOAD build 660 - it is a complete messThey had no idea about these errors at all
Seems that nobody at all is testing their beta versions among those the develop it. No other way how errors like that can happen
They had no idea about these errors at all Seems that nobody at all is testing their beta versions among those the develop it. No other way how errors like that can happen
I saw the post where they are asking if it was reported to service desk. But it was reported here and it was reported in the Russian part of their forum too. Either they do not know Russian any more, or they do not want to know about the errors
They had no idea about these errors at all Seems that nobody at all is testing their beta versions among those the develop it. No other way how errors like that can happen
Actually I wonder what kind of tests they perform before any release. The weird thing is that the bugs keep reappearing, which never happens in a standard unit testing.
Actually I wonder what kind of tests they perform before any release. The weird thing is that the bugs keep reappearing, which never happens in a standard unit testing.
The fact that the errors keep reappearing after they already have been solved made me think that there is a kind of a "sabotage" among the coders working on that project. Or their project manager is lost in combining individual works of coders. But what is sure : there is no normal testing of what they update or "upgrade" at all, otherwise, even after a couple of minutes, they would know what is wrong