If I understand your patch correctly, IFS does not need to be reset because it uses IFS="foo" read to only set it locally for the read command. Do we know if that works everywhere, since read is afterall a shell builtin?
I don't understand why you changed the RET extraction to use [0-9] but did not change the next line to also use that. I also don't understand what you mean by this sentence: > Also with the complete line now available in $DB_LINE, the return > code can always be stripped from $RET. Once the above is clarified, I'll be ready to apply this patch. -- see shy jo
signature.asc
Description: Digital signature
