Question - non-consistent result in referenced formula

Aurelien
Google Developer Expert
Google Developer Expert

Hi there

Iโ€™m experiencing something unusual, Iโ€™m pretty sure my formula used to work, butโ€ฆmaybe not.

I have various tables:

  • File
  • Detail_File (children of File)
  • Businesses

Each Detail_File has 1 โ€œrelated businessโ€ column.
Each business owns properties, and among these: business managers1 and business manager2 in particular (letโ€™s call them BM1 and BM2)

My need is to have a list of ref at the File level, referencing all business managers from the whole related detail_file items.

My formula is:

[Related key_Managers]=
[Related key_Business][key_BM1]+[Related key_Business][key_BM2]

[Related key_Managers] is of type List of Ref, referring to employees table.
[Related key_Business] is of type List of Ref, referring to businesses table.

โ€ฆNo manager comes back to the File level, although [Related key_Business] displays a list of businesses as expected.

In addition, when I look to the table of businesses, the BM1 and BM2 are well reported (as well as all the other businesses properties).

Any clue on what may have gone wrong ?

Many thanks in advance

1 18 265
18 REPLIES 18

Steve
Platinum 4
Platinum 4

I havenโ€™t tinkered with using = with lists recently, but my recollection is that lists are compared as Text values, meaning each list is converted to Text first, then those values are compared. Youโ€™ll run into problems if the lists are in different orders and/or contain duplicates. So, ideally, youโ€™d do this:

SORT(list1 - LIST("")) = SORT(list2 - LIST(""))

or, if your lists may legitimately have blank elements (both of the following are equivalent):

    SORT(list1 - LIST()) = SORT(list2 - LIST())
    SORT(UNIQUE(list1)) = SORT(UNIQUE(list2))

Assuming your goal is the determine whether the two lists contain the exact same distinct items.

Thanks @Steve,

I just realize I explained poorly my idea.
Just an old habit to describe, like when I was a young student: A=2+3.
I will give this habit away.

The purpose was not to make a comparison here.
Expressed differently, for the column [Related key_Managers] which is supposed to collect all referenced manager owning businesses, the formula I use is:

[Related key_Business][key_BM1]+[Related key_Business][key_BM2]

Is it clearer this way ?
So, expressed differently again, I want to get a list summarizing all related managers to my file, according to the related details in the file.

Let me know if this is more accurate.
And thanks again for your help !

The way youโ€™re doing it is appropriate. You might want to wrap the result in UNIQUE() to avoid possible duplicates.

I might also add

- LIST("")
to remove any blank values that may appear

If I have your stuff correct:

[Related key_Business][key_BM1] would be??? A reference to the employees table?

[key_BM1] is what in the business table?

@MultiTech_Visions

[Related key_Business][key_BM1] would be a list of ref to employees table, indeed.
[key_BM1] is the key-column in Employees table.

I should have written : [related_key_BM1] instead of [key_BM1] actually.
And [related_key_BM1] belong to business table.

Aurelien
Google Developer Expert
Google Developer Expert

Thank you to both of you.
I will improve my formula to avoid any blank or duplicate items, thanks for these tips.

In the meantime, I think something else keeps going wrong : if may formula is right indeed, then there is no way Iโ€™m having conflicting results.
here is a video trying to explain my misunderstanding.
(please excuse my accent )

Please post screenshots of the Related Detail column configuration and its complete App formula expression.

Good to ask, there is a step in between the file and the detail.
I did not mention it as I didnโ€™t think that was useful.

When you talk about column configuration : are you thinking about that: key, label, show, editable, etcโ€ฆ?

Related Detail App Formula:

SPLIT(
     CONCATENATE(
          [Related Detail_Fiches][Related Detail_Jours]), 
    ","

)

Related Detail_Fiches App Formula:

REF_ROWS("Detail_Fiche", "Related_Key_FH")

Related Detail_Jours App Formula:

REF_ROWS("Detail_Jour", "Related_Key_Detail_FH")

No. I mean the column configuration screen:

SPLIT() and CONCATENATE() are likely the problem. Try removing them both:

[Related Detail_Fiches][Related Detail_Jours]


Here is the configuration.

I just tried removing the split/concatenate, but got a list of list of ref, and other formulas depending on it made an error to appear.
As itโ€™s late here, I will give it a rest till tomorrow in the morning.

Iโ€™ll keep you updated on my progress.

Hi @Steve ,

Iโ€™ve been working on the whole data set, cleaning display to better understand what happens.
Now I realize there is a difference into data process, cannot explain why.

Here is what I found:

App Formula for Related Technicien

UNIQUE( 
      [Related Chantiers][Barcode Technicien Responsable1]
      +[Related Chantiers][Barcode Technicien Responsable2] 
) 
-LIST("")

Result:

Any clue ?

What are the column types of the Barcode Technicien Responsable1 and Barcode Technicien Responsable2 columns of the Chantiers table?

Type Ref, source Table : Employees
Both of them.

Note that in order to update the computed value stored in the spreadsheet, you must open the row in a form view and Save it (even without making any changes). Have you tried that?

Iโ€™m not sure of understanding what you mean.
In the app, some workflow will write datas, and other datas will be written depending on form saved.
In the capture I showed, nothing was written by other mean than by the app.
Are you saying that, in some case such as โ€œquick editโ€ (even if, at this stage, this did not happen), there might be error at updating values ?
In addition, nothing in computed in the sheet : only in the app.

Do you think I should engage support on my app ?

That might be easier. The problem might be readily evident to someone there who can look at your app configuration.

OK, letโ€™s do it then.

Thank you very much anyway for the whole time you spent on my case !

Top Labels in this Space