This toy problem demonstrates a problem I've stumbled across when using
ADODB to read in a tab separated value (TSV) file.

The problem is that double quotes within a particular recordset ends the
recordset's fields prematurely.

In the following examples, I've converted the tab characters to <TAB>.




Case 1: Quoted material at the start of the field

	1 one<TAB>2 two<TAB>"3" three<TAB>4 four<TAB>5 five

	Then the recordset will look like this:

	field 0: "1 one"
	field 1: "2 two"
	field 2: "3"
	Field 3: Null
	Field 4: Null



Case 2: Entire field is quoted

	1 one<TAB>2 two<TAB>"3 three"<TAB>4 four<TAB>5 five

	Then the recordset will look like this:

	field 0: "1 one"
	field 1: "2 two"
	field 2: "3 three"
	Field 3: "4 four"
	Field 4: "5 five"



Case 3: Quoted material is within the field

	1 one<TAB>2 two<TAB>A "3" B<TAB>4 four<TAB>5 five

	Then the recordset will look like this:

	field 0: "1 one"
	field 1: "2 two"
	field 2: "A"
	Field 3: "4 four"
	Field 4: "5 five"
