oxfam error: Converted text is not utf-8

We get these a lot also, but we just tolerate them. But you're right, we need to either fix it or stop having it raise an alarm.

I'm assigning to Carlos but not until next week. I hope to get a full traceback from the next time it happens. Of course, our inability to get the source file producing this problem is an issue. We might have to put something in place that captures the offending file and copies it to var for analysis.

We need to slow everything down until we get clarity on the Q4 budget.

Carlos, think you can work on this one during the next 2 weeks (M120)?

This will happen anytime the text on an uploaded file does not match the encoding expected by the converter triggered by its file type. When this happens, text is coerced into ascii and processing continues. This text is used for indexing the file, so except for the non-compliant characters text search will work for this file.

This could probably be a warning instead of an error, but aside from that I don't think there's much that can be done, since users can upload whatever they want and there can't be a guarantee that they will always stick to some encoding.

Any suggestions?

I agree with your diagnosis. Thus, the real problem to solve is simple: get the stupid error monitor to shut up. :) Make it a warning and we'll close the ticket.

Done. Deployed to cguardia-1035308-converted-text-error

looks good.

I am going to teleport these into the future so we can test them. Breaking the rules a little.

Already tested and released in r3.99