Ingesting Semicolon-Separated CSV Files into Dataverse

[Tutasi]

Hello,

I am currently facing an issue while attempting to ingest a CSV file into our Dataverse instance. We got the following message of a CSV file that uses semicolons (;) as delimiters instead of commas (,) during the ingestion process.:

• Error Message: "Tabular data ingest failed. Reading mismatch, line 2185 of the Data file: 1 delimited values expected, 2 found."

Example of the CSV file in plain text:

Journal;ISSN;eISSN;Articles;ISSN_DOAJ;eISSN_DOAJ;APC;APC_DOAJ;APC_EUR ATTIC-REVISTA D INNOVACIO EDUCATIVA;1989-3477;NA;1;NA;1989-3477;No;NA;0,00 € 452 F-REVISTA DE TEORIA DE LA LITERATURA Y LITERATURA COMPARADA;NA;2013-3294;11;NA;2013-3294;No;NA;0,00 € CEDILLE-REVISTA DE ESTUDIOS FRANCESES;1699-4949;NA;9;NA;1699-4949;No;NA;0,00 € ...

Based on feedback from the community, it seems that Dataverse expects CSV files to be comma-separated by default, and this discrepancy might be causing the ingestion process to fail.

Questions:

1. Is there a way to configure Dataverse to accept semicolon-separated CSV files?
2. Are there specific settings or parameters that need to be adjusted in the ingestion process to accommodate semicolon delimiters?
3. If the system does not support semicolon-separated files by default, are there any recommended workarounds or best practices to resolve this issue?
4. Has anyone else faced a similar issue, and if so, how was it resolved?

[Dimitri Szabo]

Hello,

Sadly Dataverse can't accept semicolon-separated CSV files as of now. It is a problem several installation faced and reported, here is the related issue : https://github.com/IQSS/dataverse/issues/8990 if you want to add on it :).
I am not aware of any recommended workarounds, in our installation we simply advertise as a good practice to use commas rather than semicolons.

Best Regards,

Dimitri