Fixed issue #18693: Using checkbox question for yes/no loses data when data entry view

[gabrieljenik]

No description provided.

[Shnoulle]

Since unseen can be used as subquetiontitle, maybe use a different separator :unseen can work i think.

See : https://www.w3.org/TR/html401/types.html#type-name

[sonarcloud]

SonarCloud Quality Gate failed.   

0 Bugs
0 Vulnerabilities
0 Security Hotspots
4 Code Smells

0.0% Coverage
0.0% Duplication

[Shnoulle]

The question still exist «why not {$fname['fieldname']}:unseen » ?

[gabrieljenik]

The question still exist «why not {$fname['fieldname']}:unseen » ?

I believe the : can be error prone or problematic.
The current naming schema works. So, don't want to add an extra layer of "risk" (as a way to say it) :)

[Shnoulle]

I believe the : can be error prone or problematic.
The current naming schema works. So, don't want to add an extra layer of "risk" (as a way to say it) :)

You think i write : without controlling it's OK ?

ID and NAME tokens must begin with a letter ([A-Za-z]) and may be followed by any number of letters, digits ([0-9]), hyphens ("-"), underscores ("_"), colons (":"), and periods (".").

Source : https://www.w3.org/TR/html401/types.html#type-name

[gabrieljenik]

You think i write : without controlling it's OK ?

No, of course not. Why do you take it like that.
I am just saying it is not very usual to use : on names, so prefer to not use it.

We can disagree sometimes, right? If at the end it works....

[Shnoulle]

PS : strange when i try to test : seems to be a 5.X

[Shnoulle]

No, of course not. Why do you take it like that.

Because you don't tell why you don't want it …

I am just saying it is not very usual to use : on names, so prefer to not use it.

Its one of the reason to use it, you have less chance to have it in a name. And it's make clear that is a new name.

In LimeSurvey : we already use _ as separator between code and subquestion code, the when see _ : a new xdev can think it's a separator between code and subcode.

Another : there are discussion since years to use QuestionCode in place of sidXgidXqid, with unseen at start with _ : you disable usage of unseen.

We can disagree sometimes, right? If at the end it works....

Yes, but maybe discuss on idea is a good solution to have something work for long term and more understandable

[olleharstedt]

Right, both underscore and hash are used in question code generation? Or is that SGQA code only?

Where in the code is the current naming schema decided? Can't find it in the DataEntry class.

@Shnoulle You have an example that could break the current schema?

[olleharstedt]

name='unseen_{$questionInputField}'

This one right? Pja, can it break if someone creates a question with code "unseen" and then subquestions?

Double underscore is another alternative, maybe. Or name=unseen[$questionCode]

[gabrieljenik]

Waiting on approavl of #3109 before updating this one.

[gabrieljenik]

Picked up. Ready for re-reviewal

[Shnoulle]

Please : don't loose data sent

(i didn't review all currently)

[gabrieljenik]

Please : don't loose data sent

We set JS validation, just like the other branch...

[Shnoulle]

Please : don't loose data sent

We set JS validation, just like the other branch...

Don't send a 400 error in case JS broken !

If JS broken : save data but send an flash alert.

[Shnoulle]

Please : don't loose data sent

We set JS validation, just like the other branch...

About other branch : this one is the 1st done, i didn't see the 400 error before sorry.

[sonarcloud]

SonarCloud Quality Gate failed.   

0 Bugs
0 Vulnerabilities
0 Security Hotspots
25 Code Smells

0.0% Coverage
10.3% Duplication