lazuli | Honest batch generated N-of-N ECDSA | Blockchain library

It looks like buyPrice is a single float value, corresponding to 12.7 in the data you give us. Then, in your template, you're trying to iterate over it by doing for price in buyPrice, which throws the error.

Depending on what you want to do, you should either extract the pricePerUnit values into an array that you can then use in the template, or modify the template to expect only a single float value.

First off, PowerShell has the built in ability to parse and manipulate CSV documents, so that would be a better option. But I will stick with batch processing.

Regular expressions are no good to a pure native batch solution for two reasons:

• It is impossible to alter FOR /F behavior to parse tokens by regular expressions - it is what it is - very limited.
• To parse your file with FOR /F you would need to manipulate each line prior to parsing. Batch does not have any regex utility that can alter content. It only has FINDSTR which can do very crude regex searches, but it always returns the original matching line. On top of that, the FINDSTR regex is so crippled, I'm not sure you could properly parse a CSV anyway.

You could use custom JScript or VBScript via CSCRIPT to preprocess the file with a regular expression search and replace in such a way that FOR /F could then parse the file. I have already written a hybrid JScript/batch regular expression processing utility called JREPL.BAT that works well for this.

A quoted CSV field can contain quote literals, in which case the quote liberals are doubled. The following regex would match any CSV token (not including the comma delimiter) ("(?:""|[^"])*"|[^,"]*). It looks for a quote followed by any number of non-quote characters and/or doubled quotes, followed by a closing quote or any number of characters not including quote or comma. But your CSV does not contain any doubled quote literals, so the regex can be simplified to ("[^"]*"|[^,"]*).

CSCRIPT has no mechanism to pass quote literals within arguments, so JREPL has an /XSEQ option to enable extended escape sequence support, including \q to represent ". The other option is to use the standard \x22 sequence. JREPL "(\q[^\q]*\q|[^,\q]*)," "$1;" /XSEQ /F "test.csv" will match any token (possibly empty) followed by a comma delimiter, and preserve the token and replace the comma with a semicolon.

But that still leaves empty tokens, and FOR /F does not properly parse empty tokens. So I can throw a bit of JSCRIPT into the replacement term to remove any existing quotes, and then surround each token with quotes (except for the last one, where it isn't needed)
JREPL "(\q[^\q]*\q|[^,\q]*)," "$txt='\q'+$1.replace(/'\q'/,'')+'\q;'" /JQ /XSEQ /F "test.csv"

Here is a demonstration showing how it could be used to parse your CSV:

1st variant

You should store the selected cell's index path and remove its checkmark accessory in didSelectRowAt.