It appears the file is not set up uniformly. If I run it with the default settings, which scans the first 10000 rows for the schema, I get a lot of warnings. This is due to a " character in one of the values.

Setting the Field Qualifier Character and Field Qualifier Escape Character parameters to empty solves the issue and all rows are being read.

Thanks Red that did the trick :)

I'd just chime in to mention that there is a CSVValidator custom transformer on the FME Hub that helps for this sort of issue. I ran your data through this and it found the issues you were experiencing (qualifier characters, etc). The only problem is that it assumed a comma delimiter. I'll see if I can change it to be user-defined.