Can you calculate a hidden field removing the underscore from a select_multiple
Accepted Solutions
I have found a very roundabout way to remove the underscores from a select_multiple response. I would love to hear if there is a different way to accomplish this, but here is how I did it:
1) For each option in the 'select_multiple' question, add a hidden question below. Use the calculate column to test whether an option was selected, then to populate that hidden question with either the corrected text (with spaces) or null if that option was not selected.
1.5) If you don't mind some extra commas in your list, you can simply add another text question and use the calculate column to "join" all of the hidden questions. Your result is a list of corrected text, but with some extra commas in the list if you didn't choose a name in the middle of the list.
2) Let's assume that you want no extra commas in the list. Then we need a few more steps. First, go back and add the separator characters (comma-space) to each of the corrected text strings.
3) Then join those using an empty string as the separator:
4) Because you don't know which option will be the last selected, your list will now include a trailing comma-space. We can use substring to remove the trailing comma-space if you want.
(BONUS) 5) But what if we want to include an option to add other names, not in the pre-defined list? First, add 'other' as an option in your select_multiple list on the choices tab.
6) Then add a text question for the user to fill in other names. Use the 'relevant' column to make it visible only when 'Other' has been selected. Add it to the list of questions in the join.
Now we have another problem: the 'other' entry might not include a trailing separator, and therefore might delete the last two characters that we typed:
7) To get around this problem, we could test whether the final two characters are 'comma-space' and only remove them if they are 'comma-space':
Of course, people can enter anything they want into the 'other' text box, and it will delete a trailing comma-space even if they intended for it to stay...
😎 Now you can use 'bind Esri field type' set to null to prevent the layer from saving all of your hidden questions.
Conclusion: it's possible to hack your way to a list of values with spaces, but it takes a lot of setup!
I have found a very roundabout way to remove the underscores from a select_multiple response. I would love to hear if there is a different way to accomplish this, but here is how I did it:
1) For each option in the 'select_multiple' question, add a hidden question below. Use the calculate column to test whether an option was selected, then to populate that hidden question with either the corrected text (with spaces) or null if that option was not selected.
1.5) If you don't mind some extra commas in your list, you can simply add another text question and use the calculate column to "join" all of the hidden questions. Your result is a list of corrected text, but with some extra commas in the list if you didn't choose a name in the middle of the list.
2) Let's assume that you want no extra commas in the list. Then we need a few more steps. First, go back and add the separator characters (comma-space) to each of the corrected text strings.
3) Then join those using an empty string as the separator:
4) Because you don't know which option will be the last selected, your list will now include a trailing comma-space. We can use substring to remove the trailing comma-space if you want.
(BONUS) 5) But what if we want to include an option to add other names, not in the pre-defined list? First, add 'other' as an option in your select_multiple list on the choices tab.
6) Then add a text question for the user to fill in other names. Use the 'relevant' column to make it visible only when 'Other' has been selected. Add it to the list of questions in the join.
Now we have another problem: the 'other' entry might not include a trailing separator, and therefore might delete the last two characters that we typed:
7) To get around this problem, we could test whether the final two characters are 'comma-space' and only remove them if they are 'comma-space':
Of course, people can enter anything they want into the 'other' text box, and it will delete a trailing comma-space even if they intended for it to stay...
😎 Now you can use 'bind Esri field type' set to null to prevent the layer from saving all of your hidden questions.
Conclusion: it's possible to hack your way to a list of values with spaces, but it takes a lot of setup!
