Is this a better .csv?
ccsv.ioCSV with special seperation characters is nothing new. It won't generally solve the problem, because the choice of the seperation characters is always related to the data supposed to be hold by the table. Just think about a CCSV file holding CCSV files.
If you want to avoid escaping, you need to declare the length of your fields, so it says for instance "now come 20 bytes of field value" and no termination symbol is necessary. This is nothing new, read about the pros and cons of null terminated strings or old and traditional standards such as ASN.1 if you want to learn more about that.
I don't like the name TSV (Tab seperated values), because it is just CSV, and the name should be agnostic about the used seperators. See for instance https://pandas.pydata.org/pandas-docs/stable/reference/api/p... for a versatile parsing library in a widespread data science library. It just calls all these data formats CSV.
The SQLite command shell supports the format described there. I have also independently used the same format too. But now, they have a name for it, which can be useful.
One variant I have sometimes used is DLE to escape the next character if it is a control character that should be part of the data.
Unfortunately, I was unable to get the .separator command to accept the control characters. It always threw an error stating that multi-character separators are not allowed. Since those characters are in the regular ASCII table, I wouldn't have thought they were multi-character/byte characters, but maybe sqlite just doesn't allow the use of non-printing characters.
To use control characters as separators, type:
.mode ascii