Hi Paul,

Thank you for trying out the Excel Builder and for taking the time to report these issues — your feedback is really valuable and helps improve the tool for the whole community.

I've pushed updates that address all three of your questions:

1. Other replace text translations (#274564)

You're right — the other_replace_text column was not multi-language aware. I've now replaced it with per-language columns following the same pattern as text_xx and help_xx:

other_replace_text_en, other_replace_text_ro, other_replace_text_fr, other_replace_text_es

So for example, you can set other_replace_text_en = "Other (please specify)" and other_replace_text_fr = "Autre (veuillez préciser)". The scripts handle the rest — each language row in the TSV gets the correct translated label, with automatic fallback to the base language if a translation is missing.

Regarding country codes for column headers: you can use any ISO 639-1 language code that LimeSurvey supports. The most common ones are: en, fr, de, es, it, pt, nl, ro, ar, zh-Hans, ja, ko, pl, ru, sv, etc. Just add columns like text_de, help_de, other_replace_text_de and the scripts will auto-detect them. The full list of supported codes is available in LimeSurvey under Survey Properties > General Settings > Languages.

2. Randomization (#274572)

I've added two new columns to the template: random_order and random_group.


Set random_order = 1 on a Q row to randomize the display order of its answer options (for List types like L, !) or subquestion rows (for Array types like F, H).
Use random_group on individual SQ or A rows to group items — items with the same group name are shuffled only within that group. This is useful if you want to pin certain items (e.g. "None of the above" always last) while randomizing the rest.


The example survey now includes random_order=1 on two questions (exercise and dietfreq) so you can see it in action.

3. callr error on the website (#274573)

Fixed — the callr package was missing from the server. The TSV to Excel conversion on the web app should work now.

All changes are on GitHub and the web app has been updated. Please give it another try and let me know if you run into any other issues or have further questions.

Best regards,
Stefan

Hi Stefan,

Many thanks for implementing these changes so quickly. I will be working on the survey setup today and will report back if there are any other issues.

Thanks,
Paul

I think the quotas worksheet needs to have an extra column added for the URL. Can we also add in extra language columns on this sheet eg message_fr?

Hi Paul,

Both of those are already in the template!

The Quotas sheet has an autoload_url column (column 5) which controls whether the respondent is auto-redirected when a quota is full: set it to 0 for no redirect, or 1 to redirect. If you need to set the actual destination URL, that's configured in LimeSurvey itself after import (Survey Properties > Quotas > End URL) — the TSV format only carries the on/off flag.

As for language columns, they're already there too — the example template has message_en, message_ro, message_fr, and message_es (columns 6–9). If you need additional languages, just add more message_xx columns (e.g. message_de, message_it) and the scripts will pick them up automatically, same as with text_xx and help_xx on the Survey Design sheet.

Stefan

Hi Stefan,

I was thinking about the actual redirect url that each quota needs as this can be different for each quota. If we can't specify this in the Excel then we'd need to re-enter the urls every time we re-import the converted Excel file. Some of my surveys have more than 50 quotas!

Good point. You're right, with 50+ quotas re-entering URLs after every import would be a pain.

I've added two new columns to the Quotas sheet: quota_url (the redirect URL for each quota) and quota_url_description (the link text shown to the respondent). Each quota can now have its own URL, and it's preserved through the full round-trip.

The web app has been updated. please give it a try and let me know how it goes!

Best regards,
Stefan

Awesome! I probably won't have time today to test but will definitely give it a go in the morning.
Thanks for the quick update!