Unable to read/write/merge into nextcloud spreadsheet

Inspired by Typeform to Spreadsheet on NextCloud I would like, ultimately merge my HTTP request with a spreadsheet hosted in a Nextcloud instance.

Sadly it seems to hang at the merge and when I bypass the merge, with or without a spreadsheet node, the file is successfully created on the Nextcloud side, but it is empty.

On the Download side

it always say: Entries exist but they do not contain any JSON data.
even if tried the File Path with and without the slash (/) in front of file name and if I specify the sheet

On the Upload side, I tried:

  • with the file already created and without
  • with the column header (title and url) already defined and without
  • as xls and xlsx (which I would prefer since I would be able to read it with ONLYOffice)

my json in case

Hi @JOduMonT, the downloaded file should be visible under the “Binary” tab of your Nextcloud node:

This binary file would be what your spreadsheet node can read:

The upload should then work fine with the Binary Data option enabled:

When checking my Nextcloud instance, the file is available and contains the merged data:


image

Seeing you are saying the workflow is hanging, is there a chance your existing file does not contain any data yet? If a node outputs 0 items, execution would stop.

Yes I have like you but when I click on Show Binary Data it say This plugin is not supported;

but when I add Read As String like you it say my xlsx is a Corrupted zip file :stuck_out_tongue:

ERROR: End of data reached (data length = 12298, asked index = 470745088). Corrupted zip ?
 Details
Stack
Error: End of data reached (data length = 12298, asked index = 470745088). Corrupted zip ?
    at Uint8ArrayReader.checkIndex (/app/code/node_modules/xlsx/jszip.js:272:19)
    at Uint8ArrayReader.setIndex (/app/code/node_modules/xlsx/jszip.js:281:14)
    at ZipEntries.readCentralDir (/app/code/node_modules/xlsx/jszip.js:2267:21)
    at ZipEntries.load (/app/code/node_modules/xlsx/jszip.js:2345:14)
    at new ZipEntries (/app/code/node_modules/xlsx/jszip.js:2168:14)
    at JSZipSync.module.exports [as load] (/app/code/node_modules/xlsx/jszip.js:581:18)
    at new JSZipSync (/app/code/node_modules/xlsx/jszip.js:528:14)
    at zip_read (/app/code/node_modules/xlsx/xlsx.js:2974:38)
    at read_zip (/app/code/node_modules/xlsx/xlsx.js:21824:8)
    at Object.readSync [as read] (/app/code/node_modules/xlsx/xlsx.js:21894:69)

Awesome if I do abstraction of the import part yes this workflow work ;), thanks again!


So with your help, I could continue my project, but when it will be time to update the file via the cronjob it will not work

but when I add Read As String like you it say my xlsx is a Corrupted zip file :stuck_out_tongue:

Oh, I was using a CSV file in my example, sorry for the confusion. Reading an XLSX file also works fine for me though with a workflow like this:

1 Like

Actually it work perfectly now, and you make my day by helping me make my first workflow with n8n.
I was even being able to find the solution for my merge node :slight_smile:

I just wish discourse had a tipping system