While working on the SniffHeaderByDateType sniffer helper, I noticed that if there are spaces around the delimiter character, they interfere with the code used to split a line into fields. You end up with inconsistent results (spaces where they shouldn't be, delimiter char, quote char in the field, etc.). This happened when I was tinkering with the "tiv_2011","tiv_2012" headers within the "headerCommaQuoteNonnumeric" test CSV data. Try to reproduce and then fix.
While working on the
SniffHeaderByDateTypesniffer helper, I noticed that if there are spaces around the delimiter character, they interfere with the code used to split a line into fields. You end up with inconsistent results (spaces where they shouldn't be, delimiter char, quote char in the field, etc.). This happened when I was tinkering with the "tiv_2011","tiv_2012" headers within the "headerCommaQuoteNonnumeric" test CSV data. Try to reproduce and then fix.