>> least start:length, where both start and length are positive integers.
>You mean start:end I hope.
>The proposal originally had start:length but was changed to start:end. I
I believe it was changed to allow for negative numbers as the end
position (ala slicing, rather than substr). If we remove the
wrap-around, then we should go back to length as that's more consistent
with other programming and query languages. I feel we should use 1 as
the first position as that's consistent with other query languages (SQL
and XPath) as Jakob (IIRC) pointed out.
To write it out, then:
Both start and length are positive integers. Start selects a start
point in the string, with the first character being position 1. Length
gives the number of characters to select from that start position. If
length is omitted, then the entire string after the start position is
marc.008 =/substring=1:6 970723
# first six characters
dc.identifier =/substring=10: "ms 123-a"
# characters from position 10 to end