@l-andorrà , I don't like tagging people, but on one of my other posts, you mentioned if I was willing to put in serious effort, you'd be willing to assist me where needed. I hope this was a strong enough attempt on my part ahaha. In advance, I appreciate your time. You seem very committed to this community, and I want you to know you're appreciated. Idk if you hear it enough ahaha.
First off, if I create a shared project link to this, and then I delete the project, will the link still work? I don't want future people to lose the benefit of studying it if they need help. (If someone says they get deleted, I'll immediately upload a source code below for future people, because I will probably delete this project eventually). (Project link at the bottom of this post)
I created a test project just to see if I understand/grasp how to use Arrays correctly. I've never used them before and I don't generally work with code, so if you give advice pertaining to code, treat me like I'm five. Hahaha. Feel free to comment on anything in the project, but I mainly care about the Array stuff specifically. So, send me whatever pointers you wish, but don't feel like you're required to comment on everything if you don't feel up to it. The arrays are really the heart here.
While there's a lot of blocks here, what I'm trying to achieve is simple. I will explain the premise as well as a brief description of the block by block in case something confuses you. Again, don't get hung up on stuff. I just care about the Arrays, but for completeness, I want to share the rest.
All I'm attempting here is to, when above a 500MA, go find the highest and lowest candle in the last 100. And if the highest price is closer than the lowest price (we're likely making a pullback off of the high right now in this case), I want to loop from the highest high back to the lowest low, count how many "down closed candles" there were between the high and the low, take that count and set my array size to it, then I loop back through the candles again, and I attempt to store the opening price and candle ID of all valid candles in the proper index slot for the proper array. At the end, display "done" (I'm aware this doesn't do anything with the values right now. I'm just trying to make sure I'm doing this correctly before I make something absolutely massive with it that no one can hope to understand). I know some of the "useful" logic isn't here, but I would do that kinda stuff on my own. I really wanted to make this simple.
Block by block, if needed:
On Init, I try to declare my two arrays. I think I did it right?
Price above MA, true, we go left, false we go right.
Store the CandleIDs for the highest and lowest candle in the last 100 candles.
If the high is closer, then the low is further, and I assume there's a bullish leg. I check this by seeing if the candle ID is lower for the "high" candle.
So, I want to find "down closed candles" in the assumed leg.
I set the loop_id equal to that of the High Candle. This is because I will use loop_id for candle ID, and I only want to start at the highest candle. I also set "ValidCandlesFound" to 0. This number will be the ArrayResize value I try to use. Before starting, there is zero valid candles, so my array should ideally be zero at the end of this if no valid candles are found.
Check if my loop_id value is equal to or less than my "low_candleID" because I want the loop to stop at that candle.
If the loop_id allows it to pass true, we check if it's a down closed candle. If we get a true output, in the variables tab, I add one to the ValidCandlesFound variable. And whether true or false, I exit to a pass block, just to visually loop the loop up in a friendly way. I also add +1 onto the loop_id at this point, just because I can.
Once the loop terminates, I attempt to change the size of both of my arrays to the size of "ValidCandlesFound". I've never worked with Arrays before, but I assumed I needed to set the array at a size before I can start cramming values in? If I'm wrong, can I just skip that step and move into the next section? (Which you'll better be able to answer after reading it ahah)
At this point, I set the loop_id to the ID of the high candle again, and I also set the ArrayIndexID to -1. I heard the first index on an array is "zero," so, setting it to negative one made sense here shortly.
I again check if the loop_id is within tolerance. If it is:
I check for down closed candles again. If there is a down closed candle, in the same block, I add one onto the ArrayIndexID.
In the next block, I store the opening price and the candle ID of the candle in temporary variables, for the next block.
The next block, I attempt to fill the value of each index with the values stored in the temporary variables.
If the candle wasn't a down closed candle, I pass the array logic. After exiting the array logic (so either of these), I add one onto my loop_id again and back to the beginning.
I repeat this until the loop_id exceeds my swing I'm referencing, at which point, it hopefully prints a message on the screen that says "done!" (I've not worked with comments either, so idk if I did that correctly).
Naturally, the same type of logic is true for the other side, but I want to spare you. Haha. 
So, the biggest things I need to know are:
Are the On Init declarations correct and necessary?
Did I need to do the loop to set the array sizes like I did in blocks 25 and 19?
And, did I fill my arrays correctly in 30/32?
Furthermore. I didn't know how to attempt it, but say for instance later, one of these points became "invalid" for whatever reason and I wished to remove them from the array? And it was in the "middle" of the array somewhere. Is there a way to remove it and resize the array? Or would it make more sense to pursue an alternative...like repopulating the entire array again from scratch, without that one. Or would it make more sense to store a zero in it or something so that I can ignore it?
Earlier Link:
https://fxdreema.com/shared/H017dd00b
Final Link:
https://fxdreema.com/shared/kuEfhkS8d
Thank you so much in advance!
@sktsec Thanks, bud! I kind of expect that, too.
And then after running the following blocks, it goes back to the "run block" block and runs any connected blocks, I assume?
May sound derpy, but if I use "Run Block" to run some block elsewhere in my project...
It will also run the connected blocks, right? Lol. I mean this in two ways, also.
So, if "Run Block" is number 6 (doesn't matter), and through the logic, it gets hit, and it is commanded to run Block 14....
Will we go run block 14 AND all that is connected to 14 (following normal rules) or do we ONLY run 14?
Furthermore, after finishing up with whatever happens with 14, we should return back to our "run block" and run its outputs then, correct?
Something like this?
I've still not learned any actual "coding" yet really (hence why I'm using Dreema).
How similar is this in MQL5? I don't really know how different their syntax is, but it seems like these functions would be so basic that they might possibly be the same in both languages?
@l-andorrà Not necessarily. Dreema is very approachable and gives me the opportunity to better learn how to think like a programmer/coder anyways.
Furthermore, I won't be going from nothing to final iteration in one pass. It's just too much to ask of someone earlier in this coding process. Dreema gives me a stepping stone to get the results I need, one iteration at a time, and move into coding in bite sized pieces instead of one overwhelming wave.
Ultimately, I believe dreema is the right place to start. But, I'm also, again, asking the question simply because I'm not at that destination yet. How am I supposed to know if Dreema can deliver everything I need it to? That's why I'm asking the question I am. I don't even know if the workflow I described is the most correct approach for what I'm trying to achieve and I don't know what the backend looks like on all of that.
I just want to know if it's possible, what skills I'm going to need to acquire, and any other words of caution I might need between here and there.
Thanks again for the reply!
@roar Gotcha! Thanks again. That makes sense.
@roar AH! Okay! Thanks!
So, it sets it to -1 on the initial hit, it obviously exits on the "true" side of the block, which takes it to 0, which is used for the first index of the array. And then each time the loop completes, it comes all the way back to the "loop" block, adds one, and keeps going until the loop breaks?
I'm okay with the loop idea, but I'm still new with arrays. Will this "Loop" block (I've never used it) break (give a false output) if we get to an index one larger than the array? Or does this specific loop actually break on the "Set Market" block when we hit the snag of trying to load an index that doesn't exist?
@l-andorrà Thanks for the reply. I'm assuming you mean a recent MT5 update? Do they frequently break a lot of stuff? 
I'm new on MT5 for this stuff (I mostly did MT4 previously).
I also am seeking the specific approach I am (a biasing EA separate from the entry stuff) because I'm not an indicator trader. I'm not opposed to using them as needed, but I will be likely using loops and arrays (loops I can do, arrays I'm still reading the MQL5 site on) to scan through candlestick formations and time based points of interest throughout historical data, storing it in the arrays and such, and then using these reference points across a multitude of other EAs basically.
I imagine coding this exact same logic into a bunch of different algos would be quite bloated and redundant. Furthermore, I know "advancing" up the ranks of programming and stuff means getting better at not being so repetitive and redundant. What do you call it? DRY? Or something like that.
Anyways, I'm a programming nub, but very rational and logical and the way of thinking, I'm okay with. Just lack the experience. My point, I see myself having lots of inefficient code/block layouts that GRADUALLY over the years, I will be able to reduce. I just imagine, up front, I'm going to be inefficient and the CPU demands per tick are going to be stupid if I tried to do this type of logic too many times in too many inefficient ways.
The point, I know it's probably logistically more complicated, but I know I also only have to learn it once. But I imagine it being lighter on resources and I imagine it being easier for me to manage and troubleshoot smaller chunks of stuff instead of one big monster trying to do it "all."
@l-andorrà I appreciate the reply!
I would like to learn to code as needed. I'm not "passionate" about programming outright, but I AM passionate about mechanical trading systems and algos in general.
I'm willing to learn what I need to learn. This is my career and my bachelor's and master's are both in investments and I've been in the markets for more than a decade. This isn't a get rich quick or whim thing for me. I have models that I use quite regularly, but I have so many of them that I would like to start the process of gradually transitioning all of them into fully mechanized things.
I appreciate the warning about this being a very ambitious aim for a newby programmer. I'm willing to take as many years as it takes at this point. This isn't a get rich quick or anything like that for me. It's my career for life basically ahaha. So. It's "required learning" for me at this point. I'm doing it with or without anyone's help. I would just appreciate the guidance and to have a community to be involved in. Trading can be a quite lonely journey.
Hey, @roar !
I'm using your example here as a learning opportunity for myself.
I have a question (I'm trying to learn arrays). In this instance, what was the purpose of subtracting one off of the loop id pre block, but then adding it back on after?
Wouldn't this mean that it would keep doing the same index in the array repeatedly and never move on to another?
@mentor123 Why the heck are you spamming posts and private messaging me for things not even related to the topic at hand?
I'm all about helping when I can, because we all need help at times, but this post and my DM's aren't the right venue.
Make your own post in the proper place with the proper title and the proper questions and you might get some engagement.
If it genuinely isn't mechanically feasible (since you're suggesting to maybe do it all in one EA instead), then I can accept that, too.
I just know Dreema doesn't make it very easy for me to copy/paste blocks from all of my different projects into one big one and I'm concerned about my ability to manage and troubleshoot such a robust, singular EA, whereas I know I can handle a lot of smaller ones. I just don't know how I would easily import a partial project into the "whole" when done with it. If that's the route I need to go, then so be it. I just genuinely imagined there was "reasonable" ways to do share information between EAs meaningfully.
@TipsyWisdom (I'm sure this place is drowning in people who don't like to do the work. I get why you'd be guarded about it. Haha. I've seen plenty of trading discords and stuff. I'm not trying to exploit anyone here).
@TipsyWisdom I appreciate your reply!
And of course I don't want anyone to build my EAs for me. I'm competent and happy to do it myself.
What I don't understand is what the workflow looks like for this.
My opposition to doing it in "one" EA is to reduce a lot of redundancy, share information across pairs, simplify troubleshooting it, and to give me the flexibility to plug and play an abundance of different models I have in.
My point, though, I don't understand what the backend looks like on all of this. I don't know if I'm storing things in files or global variables or if one is more capable than the other, if one prevents me from using a VPS, etc.
Again, I don't want anyone to do the work for me. I just have no idea how all of these things work behind the scenes.
I'm fine to look on the forums for examples, too, but I don't even know what I'm looking for examples of! Ahaha. This is why I'm inquiring.
But as stated, if this can be explained a bit or illustrated for me, I'd appreciate it. I'm not trying to get off easy. I'm ready to work and learn. I just don't know where my efforts need to go.
@TipsyWisdom I appreciate your quick reply and your contributions to the community. Thanks again!
(Also, if you have any tips on how one would even be able to test any of this other than on live market data via demo account. I suppose using the strategy tester is off limits other than manually feeding in my data to the low timeframe ones, confirming they understand and interpret it, or outputting data for the higher timeframe one to acknowledge it's doing it correctly).