You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I realise this might be too much scope creep, but I think it would be nice if this tool were adapted to work on CSV files.
functionality in the extension could handle a CSV url + mimetype
a user could upload a CSV in the local instance for discovery
This could be implemented by a "wrapper" that converts CSV content to a list of JSON objects and then passes that JSON to the discovery tool. a max size limit could be applied.
The text was updated successfully, but these errors were encountered:
I had similar thoughts and even looked into this possibility. The main problem with CSV is that it’s not a strict format, making it extremely difficult to distinguish CSV from other content automatically without false positives. Additionally, the format is highly variable in terms of delimiters, character escaping methods, and so forth.
It might be implemented someday, but it’s far from our current priorities. Most likely, this feature will be opt-in, enabled in the settings at the user’s own risk. In the near future, our plans include adding support for JSONL.
The main problem with CSV is that it’s not a strict format, making it extremely difficult to distinguish CSV from other content automatically without false positives
I bet 99% of the time it's a .csv (or .tsv) extension. Detecting delimiters should be relatively easy in most cases (best guess match comma or tab based on extension, then if no matches, non-word symbol character).
I see what you mean, but it seems doable.
I realise this might be too much scope creep, but I think it would be nice if this tool were adapted to work on CSV files.
This could be implemented by a "wrapper" that converts CSV content to a list of JSON objects and then passes that JSON to the discovery tool. a max size limit could be applied.
The text was updated successfully, but these errors were encountered: