30151 2017-07-19 02:24:19 +0000 [FO31] The function fn:boolean does not have properties like determinism defined 2017-12-20 10:51:16 +0000 1 1 1 Unclassified XPath / XQuery / XSLT Functions and Operators 3.1 Recommendation PC Windows NT RESOLVED FIXED P2 normal --- 1 abel.braaksma mike andrew_coleman public-qt-comments oldest_to_newest 128725 0 abel.braaksma 2017-07-19 02:24:19 +0000 Section 7.3.1 fn:boolean misses a Properties section. As a result it is not directly clear whether it is context-independent or not (it isn't). I think the properties should be: deterministic, context-independent, focus-independent It also looks like that sometimes properties are listed for operators and sometimes not. We do say in 1.7 Properties of functions: "All functions defined in this specification are ·deterministic· unless otherwise stated. Exceptions include the following: " But we don't have a similar statement on functions (and operators) being context-independent (unless otherwise noted) or focus-independent (unless otherwise noted). While it may seem obvious that if it isn't listed, they are context-independent, I don't see that stated. There aren't many operators defined that have a Properties section, but still, I'd guess we ought to be consistent here, or have a general statement about it (I sought for it, but couldn't find it). 128858 1 mike 2017-08-28 22:14:12 +0000 fn:boolean appears to be the only function in the fn namespace for which the properties section is absent. For op: functions I think properties are given only in the cases where there is some context-dependency to report. 128864 2 abel.braaksma 2017-08-29 20:09:14 +0000 (In reply to Michael Kay from comment #1) > fn:boolean appears to be the only function in the fn namespace for which the > properties section is absent. Yes, I think so too, it's missing only there. > For op: functions I think properties are given only in the cases where there > is some context-dependency to report. I'd expect a line somewhere that says this. Maybe there is, but I haven't found it. The logical place would be in 1.7 Properties of Functions. Suggestion: All operators defined in this spec with the "op" prefix are deterministic, context-independent and focus-independent, unless stated otherwise. (if this is indeed correct, of course) 129020 3 mike 2017-12-19 15:21:23 +0000 Erratum FO31.E2 was drafted to fix this problem. 129024 4 andrew_coleman 2017-12-20 10:51:16 +0000 resolved in latest commit