MetaTrader 5 Strategy Tester 3D Visualization - page 4

 
falkov:

Yeah, well, not printing out the highlighted section is a sign of squalor, isn't it? (I need to print one function from a listing of 2,500 lines - how do I do that?)

Not being able to edit headers (date, print time, file name) is a sign of squalor, isn't it? (I need to know if this is the latest version of the indicator or where the file I printed a week ago lies)

Absolutely not. You're confusing a code editor with a general-purpose editor.

In reality there is no problem with printing selected fragments - you just invented a function that is not used in reality.

Not being able to highlight text(https://www.mql5.com/ru/forum/2582/page2#comment_41664) properly is a sign of squalor, isn't it? (I need to highlight two hundred lines of text for something)

The highlighting is fine, maybe you gave the wrong link?


The lack of folding and a normal project manager is not very good either, is it?

Folding is not a critical function. The project manager will come a little later, just along with resources and preparation for the shop.

It would be understandable if other code editors were just as flawed, but it seems to me that you must be writing MT in Visual Studio? Compare that editor and your own :)

Instead of giving programmers a normal editor, you waste your time on stupid (in my opinion!) styling, on snippets.

But I assure you, one folding and a normal project manager is much more convenient to a programmer (i.e. ten times more) than styling and all the snippets taken together.

You're just wishful thinking.

MetaEditor is quite good and better in some respects than Visual Studio (no one knows this better than me, I've been using it for the last 13 years). In fact, without editor extenders like Visual Assist, the studio is frankly weak in code editing.

Intellisense, snippets, styling, built-in help, integration with support resources - all these are necessary elements of the development environment. And MetaEditor has them.


They have refused to handle exceptions, which is a super mechanism for reliable programming, under the most far-fetched pretexts.

And this was explained by the same concern about the programmers, such as you are so unqualified that it will be worse for you, so check all conceivable and unthinkable in advance.

Although it is called EXECUTION handling and allows you just concentrate on competent error handling in one place within a program unit.

Instead of dispersing all these numerous checks over the code and repeating the same checks dozens of times which inevitably leads to decrease of software reliability.

Come on, what am I saying, as if you don't know it :)

Out of two versions of error handling: by return code or by an action, we have chosen return codes. Our opinion: the action codes are an outright evil, while any talk of increasing reliability by using them is a direct lie.

The topic "return codes vs escapes" is flaming and useless - no one will change their mind. That's why it's better not to bring it up.

 

Dear!

I already have a project with 15 thousand lines and more than 30 include files, here's an old scheme (didn't have time to tweak it):


Here are the actual screenshots of the project:


I don't have any difficulties with coding, except one thing: there is no possibility to search text in only open files.

I asked the developers to make this simple option, but for some reason they don't want to do it....

 

I have freely pored over respected Renat's answer:

Renat:

In reality, there is no problem with printing selected fragments - you just invented a function that is not used in reality.

I'll say it again: I need to print a code fragment from a listing of, for example, 2.5 thousand lines - how do I do it?

The way I do it now is to find the pages where the fragment I need is printed in the preview and then specify these page numbers when printing. Which is not very convenient :) And this checkbox "Selected fragment" is in the dialogue box "Print", but it is just not active. If not difficult, name at least one code editor with such feature :)

The selection is OK, maybe you gave a wrong link?

To quote a part of my post:

you put the mouse on the page numbers, press the mouse button and move the mouse down, the lines are highlighted.

As soon as the mouse hits the horizontal scrollbar (to still highlight the lines below), it immediately scrolls to the beginning of the listing and highlights from the beginning of the listing to the current location.

Of course, I've adapted - I need to lead the mouse a bit to the left so it doesn't hit the horizontal scrollbar, but you have to agree, it's somehow wrong! :)


Folding is not a critical feature. The project manager will be a bit later, just along with the resources and preparation for the shop.

Agreed. So will the styler and snippets, ok?

If you have to choose between folding and project manager, I'll choose the latter. Only it should reflect not only file names, but also included functions, classes, etc.

You are wishful thinking.

MetaEditor is rather good and better in some respects than Visual Studio (I should know this, as I have been using it for 13 years). In fact, without editor extenders like Visual Assist, the studio is frankly weak in code editing.

Intellisense, snippets, styling, built-in help, integration with support resources - all these are necessary elements of the development environment. And MetaEditor has them.

Well, you tell me. For me, snippets, styling, interfacing with support resources are not essential.

Of the two options of error handling: return code or action, we chose return codes. Our opinion: escapes are an absolute evil, and talks about increasing reliability of using them are direct lie.

It seems to me that you have chosen on the principle of either sour or sweet. What if it's just tasty?

In my opinion, there is no need to pit exclusions and return codes against each other. They are two excellent mechanisms, each for its own purpose. Wouldn't it be better to keep both?

The topic "return codes vs escapes" is flamey and useless - no one will change their mind. That's why it's better not to bring it up.

Yes I realised that back then - but there's hope, maybe time has passed and you've changed your mind :)

In any case, thanks for your answers.

Ideological disagreements are not yet a reason for swearing :)

 
falkov:
Well, you know best. To me, snippets, styling, integration with support resources is not essential.

You've made a very accurate point.

We actually know better, because we have to think about the final result. And you just expressed your opinion, and the blatantly wrong one - "snippets, styling, integration with support resources are not necessary".

The task of mass involvement and training of developers to a new language without intellisense, styling, helps, integration, wizards and our general direction of development is almost a failure. And folding has no effect on the final result.



The Editor is developing further and a new version of the more advanced MQL5 wizard that can generate very complex Expert Advisors is about to be released. New features will be added actively and support for projects is on the way.

 
AlexSTAL:

I don't have any difficulties with coding, except one:
there is no possibility to search for text in only open files.

I asked the developers to make this simple option, but for some reason they don't want to do it....

This will all be done for sure - there will be a lot of changes related to project management.
 
Renat:

You are very accurately expressed.

We actually know better, because we have to think about the final result. You have only given your opinion, and the blatantly wrong "snippets, styling, and integration with support resources are not necessary".

The task of mass attraction and education of developers to a new language without intelligence, styling, helps, integration, wizards and our general direction of development is almost a failure. And the folding has no effect on the final result.

The editor is developing further and a new version of the more advanced MQL5 wizard, which can generate very complex EAs, is about to be released. New features will be added actively and support for projects is coming.

The manual contains very few examples of application of functions. And yet the basic principle of any training is "do as I do".

The ME4 manual is just an example. It's a very good one, and it's precisely because of the simple examples.

I understand that there are primary tasks, but until there are new application examples for each function, the efficiency of this help file for newbies will remain low.

 
Urain:

There are very few examples of how to use the functions in the manual. But the basic principle of any training is "do as I do".

The ME4 manual is just an example. It is a very good idea because of the simple examples.

I understand that there are primary tasks, but until there are application examples for each function, the efficiency of the help for newbies will remain low.

In our opinion, the MQL5 documentation is many times more detailed and certainly exceeds the MQL4 documentation both in size and examples.

Of course, you can never have enough examples. This is why we put so much effort in developing the documentation sections, articles and the MQL5 code base. Especially for the convenience of developers, we have integrated in the editor access to articles/code and implemented a quick search function for the entire MQL5.community:


To get automatic hints from all over MQL5.com, just press Ctrl+F1 on a phrase of interest (you can select several words in advance) and get quick search results, where you can get more details or ask a question:



This is one of our global tasks and we are investing a lot of effort in this direction.

 
Urain:

There are very few examples of how to use the functions in the manual. And the basic principle of any training is "do as I do".

The ME4 manual is just an example. It's a very good thing, and it's precisely because of the simple examples.

I understand that there are primary tasks, but until there are examples of application for each function, the efficiency of the guide for beginners will remain low.

The language will more or less settle down and examples are likely to appear.

And as I understand the developers are hoping for articles since there is no tutorial...

 
falkov:
I'll quote part of my post:

you put the mouse on the page numbers, press the mouse button and move the mouse downwards, the rows are highlighted.

As soon as the mouse hits the horizontal scrollbar (to still highlight the lines below), the listing scrolls to the beginning and the selection is made from the beginning of the listing to the current location.

Of course, I've adapted - I have to move mouse a bit to the left to avoid hitting the horizontal scrollbar, but, you must agree, it's somehow wrong! :)
We'll deal with that.
 
Renat:

In our opinion MQL5 documentation is many times more detailed and certainly exceeds MQL4 documentation both in size and examples.

Of course, you can never have enough examples. This is why we put so much effort in developing the documentation sections, articles and the MQL5 code base. Especially for the convenience of developers, we have integrated in the editor access to articles/code and implemented a quick search function for the entire MQL5.community:


To get automatic hints from all over MQL5.com, just press Ctrl+F1 on a phrase of interest (you can select several words in advance) and get quick search results, where you can get more details or ask a question:



The MQL5 language training for developers is one of our global tasks and we strongly invest into this direction.

I'll give you a short answer.

HZZY Why would a beginner need 250 examples of a function, he needs one simple example, and immediately, not a minute of searching.

You can find all the information you need in the FAQ, preferably right in the manual and not on a server somewhere:


Reason: