Formula Wizard limitations

[mbuta]

The Wizard (a very helpful tool) has several limitions, or I don't know how to use it.
Here are a few examples:
1. Only && accepted, no || or other logical operations.
2. No way to declare variables in premain.
3. No way to construct nested expressions.
4. If ANY modification is made through the editor the Wizard rejects the formula.

Is there a way to circumvent these shortcommings?
Are they documented any place?
Thank you.

[DionLoy]

Hi mbuta,

The features that you are requesting are probably beyond the scope of the wizard. The wizard is meant to generate very basic formula code that covers most of the requests that we get. For more complicated code, nested expressions, etc.. it is much better to go ahead and use the regular editor and do it manually.

We do plan on adding the Function Editor to the regular formula editor at some point in the future, so you can gain the advantage of having a function documentation within the editor.

With too many options in the Formula Wizard, it will eventually evolve down to a regular text editor =).

[Lancer]

Re: "it is much better to go ahead and use the regular editor and do it manually." Yeah, if the user IS the formula wizard.

I've tried, but I'm still completely clueless when it comes to writing EFS formulas. The problem I run into is that there is no one central, organized, searchable reference manual to study and learn EFS in progression; no progressive tutorials either. All I find are bits and pieces in no particular order, spread all over. I've just given up on it until there's a proper manual.

I vote for the Formula Wizard with too many options.

[DionLoy]

Yeah, a step by step EFS tutorial would definitely help new EFS programmers.

So what sort of additional things would you find useful in the Formula Wizard? How would we present a nested expression that would be intuitive for users to use?

mbuta: each set is OR'd (||) with the other set, while the expression lines within each set are AND'd together. Does this solve your OR problem? Can you give me an example of an expression that you want the Formula Wizard to handle?

Also, what sort of variables do you want defined in preMain()? I assume you want global variables rather than preMain() variables as preMain() variables are no longer available outside of preMain().

Thanks guys!

[ACM]

Dion

With too many options in the Formula Wizard, it will eventually evolve down to a regular text editor =).

That is true never the less there are still some basic things that the FW should be able to do. For example the ability to create some simple expressions as in the attached efs (line 48).

Admittedly, I am not the brightest bulb in the chandelier when it comes to efs, but this is one simple expression.
Yet I had to jump through hoops to get it done in FW. And even though it is correct FW continues to reject it if I try to open the efs with it.

How would we present a nested expression that would be intuitive for users to use?

Continuing to use the attached efs as an example this is the way I envision FW work.
In the section "Returned Data Points" one would be able to list all the variables that will be used in the expression ie in this case ADXDMStudy.PDI and ADXDMStudy.NDI and give them a name ie value1, value2, etc and then in the expression line one can simply write value1-value2.

Another solution would be that the first time one calls the EDIT in the expression line it brings up the list as it does now. At that point one clicks on the ADXDMStudy.PDI then types in the operator (in this case -) then calls the EDIT again and selects the other variable. Here instead the FW overwrites the first variable whereas in my opinion it should add it to the existing expression.

Just an opinion

Alex

[ACM]

continued from my prior message


Note that in the section "Generated EFS code" it says the following

function main() {
/**
* The main() function is called once per bar on all previous bars, once per
* each incoming completed bar, and if you don't have 'setComputeOnClose(true)'
* in your preMain(), it is also called on every tick.
*/

Yet there is no way that I can see to insert "setComputeOnClose" with the FW.

Alex

[ACM]

Dion

Another thing that IMHO would create a better logic flow would be to reverse the Position of the conditions (ie Set1/Set2/etc) and the Returned Data Points.
In this way you would have

• Formula Configuration
  Studies Used
  Returned Data Points
  Set1/Set2/etc

Alex

[ACM]

further thoughts on prior messages

In one message I write
In the section "Returned Data Points" one would be able to list all the variables that will be used in the expression ie in this case ADXDMStudy.PDI and ADXDMStudy.NDI and give them a name ie value1, value2, etc and then in the expression line one can simply write value1-value2.

It would be preferable if these variables could also be expressions themselves. For example (this is just a simple case to illustrate the point)
value1 = High
value2 = Low
value3 = (value1-value2)*0.25

At that point one could use value3 in an expression in Returned Data Points. For example a line would be plotted at

vEMA28_of_High.getValue(MAStudy.MA)+value3

With regards to having the Set1/Set2/etc conditions positioned after the Returned Data Points the point is that one would then be able to also use the values of the Returned Data Points expression in the conditions. For example

IF High > Expression THEN paint bar blue AND Sound Alert ELSE do nothing

Again all these suggestions are seen from the point of view of someone who is totally inept when it comes to efs but still has the ability to formulate a logical construct.

Alex

[DionLoy]

Great feedback Alexis. I was thinking of simplifying the 'returned data points' section as well, by adding some check boxes next to the each study to see if you want to return it, thus saving a few steps.. Will provide of course, additional ways of returning custom expressions.