LOL no. “The Excel-guy” does not say anything, because they do not see the problem at all: the document malfunctions compared to what its author did. Validation is part of the document; and you can’t do assumptions about importance of the function. The correct allegory would be the Excel-guy falling into the hole in the night, while Calc-guy would see the block on the road, and call the service.
This helps! Thanks!
Yes the Excel-guy can die, but he does not have to.
The Log-guy has to die.
Because this
is no valid answer to the question: pass the hole with or without safety.
You decided to play children’s game, with silly rules; OK.
No.
They will survive when they see the block, and have a backdor/windows (can use the original application, here: Google Sheets, where the file was created). Excel-guy “can die”.
They will survive when they can repair the problem (they can modify the validation). Excel-guy “can die”.
They will survive when they can wait the repairmen service (notify the author of the file, and have them fix it). Excel-guy “can die”.
And only in a highly unnatural, artificially chosen circumstances of a stupid person who had waited without food long enough to run out if time, trying only now to exit the house, and finding out that there’s a problem, and being unable to repair or workaround … then a problem would arise.
After all, more Ecxel-guys will likely “die” than Calc users.
But anyway: analogies are bad things. They don’t prove anything, don’t even illustrate anything. Seeing a problem is much better than not even realizing of its existence; and only if you can statistically prove that downsides outweigh upsides, there’s something to discuss.
What/Who is that “Log-guy”?
.
Please study the list of the existing functions an operators of the spreadsheet softwares instead, and use the common and compatible ones - if you want to edit a specific document with different office suites.
Or study the Macro feature (as I suggested above). Then you will able to create the missing functions.
A while ago the question has been marked as solved. Thanks once more for the solution.
But this:
is an invalid simplification of someone who does not understand what happens out there in the real world.
There is not one user, if you do data validation. There are at least two: the one, who rules and the one who get ruled. The ruler does the type definition (in this case the validation expression) and one who get ruled creates objects conforming to the type (in this case table cells conforming the validation expression).
For the person, who writes validation expressions, it is a useful advice to tell him to use only those function, which are available on all platforms. But for my wife, who can just write clock times below the headlines “start” and “end”, it is not a useful advice, because she was never in the position of the ruler and probably will never be. She has no idea at all what a validation function is. My wife has learned how to look after kids. And I guess that the majority of users of spread sheet programs know as much about validation functions as my wife.
The fault of LibreOffice Calc (Loc) is to not distinguish this correctly. After a user has entered some data Loc says: “the data is invalid”. But this is just wrong. The correct statement would be: “the data validation had a malfunction”. But even this is not perfect, because the information, that an unknown function has been used in the file, is known right after the file has been loaded. Long before the user enters any data, the program knows already: what ever the user will enter in this cell will cause a validation malfunction, because a function required for this validation is missing. So a useful implementation of Loc would show an error message right after the file has been loaded.
If Loc would have done this, it would be instantly obvious to my wife: this file is broken, I have to ask my boss for a new file. Instead Loc opens the Excel file, very well knowing the platform problems of incompatible function sets, and waits “perfidiously” until the user enters some data, in order to present him a wrong error message confusing more than explaining anything. This needs to be fixed.
And BTW I know how to program, and I know why Loc says: “the data is invalid” instead of “the validation has failed”. This is just, because a programmer was too lazy to write the proper exception handling. When you look into the code of Loc, you will very likely find such code:
try {
evaluate (cell)
}
catch (Throwable Iamtoolazytocareaboutthis) {
return false
}
And the result is that the continuation can not distinguish anymore between wrong data and wrong rules.
May it also be important to understand that in the “real reality” the ruled user shouldn’t trust in a ruling user (the one who designed the sheets, and defined the validity settings therein in our case) who obviously ignores the real world by not considering the availability of a tool or a function in a slightly different environment or when using an otherwise compatible software?
I understand that there are cases where ensuring compatibility is difficult or impossible. The present case must look very simple to anyone who passes on sheets for use by others: Avoid the anyway superfluous functions, and use the comparison operators that any spreadsheet software has implemented.
Or is the formula =(first<=second) invalid in GoogleSheets?
What would that say about GoogleSheets?
[Since my native language is not English, I use the suggestions of the translation service of DeepL.com for parts of my English-language postings. The errors are mine, nonetheless.]
Actually not true. Code can be loaded afterwards by statements and it is an disadvantage of all interpreted languages to stumble on first evaluation of a line of code, not on loading the file.
.
So it would be better to have a big (red?) message telleing “abandon all hope, you enter the lair of incompatibility” or better refuse to load as perfection is not possible?
This would be a reasonable enhancement request.
…a reasonable enhancement export Google Sheets to other formats?
I mean, emitting a proper message when the validation formula couldn’t be calculated.
Indeed, the bold statement that “the information, that an unknown function has been used in the file, is known right after the file has been loaded” is wrong, and does not consider possible UDF (including those that require loading of their libraries, which could be triggered by some event, which can’t be known at loading stage)… but having a proper error message at the validation time is reasonable.
1 Like