{"id": "807", "author": "Chris Lauwers", "date": "2020-08-18T23:01:00Z", "comment": "(From Tal): Do we want the comparison constraints to work for strings? E.g. should \"greater_than\" do a sorting-based comparison? I'll just point that it is non-trivial to sort Unicode strings. The most common way is to use the Unicode Collation Algorithm, which involves a database of information. There is a reference implementation in ICU. Good and proper Unicode libraries will support it (e.g. here is Go's), but I do imagine it may be a burden for some implementations. I suggest we discuss this in the ad hoc and consider the pros and cons.", "target": "Notes"}
These comments were hidden in csd06 section 10.2.2.2 and moved here as part of https://github.com/oasis-tcs/tosca-specs/issues/61
{"id": "807", "author": "Chris Lauwers", "date": "2020-08-18T23:01:00Z", "comment": "(From Tal): Do we want the comparison constraints to work for strings? E.g. should \"greater_than\" do a sorting-based comparison? I'll just point that it is non-trivial to sort Unicode strings. The most common way is to use the Unicode Collation Algorithm, which involves a database of information. There is a reference implementation in ICU. Good and proper Unicode libraries will support it (e.g. here is Go's), but I do imagine it may be a burden for some implementations. I suggest we discuss this in the ad hoc and consider the pros and cons.", "target": "Notes"}