!standard 7.3.2(15/3) 15-01-21 AI12-0146-1/01 !class presentation 15-01-21 !status Corrigendum 1-2012 15-01-28 !status WG9 Approved 15-06-26 !status ARG Approved 10-0-0 15-01-28 !status work item 15-01-21 !status received 15-01-15 !priority Low !difficulty Easy !qualifier Omission !subject Should say stream-oriented attribute !summary Read and Input are stream-oriented attributes. !question 7.3.2(15/3) says "After a successful call on the Read or Input stream attribute of the type T, the check is performed on the object initialized by the stream attribute;". However, there is no such thing (formally) as a "stream attribute". The title of 13.13.2 is "Stream-oriented attributes"; there isn't even an index entry for "stream attribute". Should we change this wording? (Yes.) !recommendation (See Summary.) !wording Modify 7.3.2(15/3): * After a successful call on the Read or Input stream{-oriented} attribute of the type T, the check is performed on the object initialized by the [stream] attribute; !discussion None needed. !corrigendum 7.3.2(15/3) @drepl @xbullet, the check is performed on the object initialized by the stream attribute;> @dby @xbullet, the check is performed on the object initialized by the attribute;> !ASIS No ASIS effect. !ACATS test None needed as this is just a wording change with no semantic impact. !appendix From: Randy Brukardt Sent: Thursday, January 15, 2015 5:18 PM 7.3.2(15/3) says "After a successful call on the Read or Input stream attribute of the type T, the check is performed on the object initialized by the stream attribute;". But we had a discussion in Portland in relation to some other wording that there is no such thing as a stream attribute. The RM defined these as "stream-oriented attributes" (see the title of 13.13.2). "Stream attribute" isn't even indexed. We changed a bunch of wording to reflect that. So we clearly need to make a similar change here: After a successful call on the Read or Input stream{-oriented} attribute of the type T, the check is performed on the object initialized by the stream{-oriented} attribute; or, maybe better to cut redundancy: After a successful call on the Read or Input stream{-oriented} attribute of the type T, the check is performed on the object initialized by the [stream] attribute; This obviously is a presentation issue, hopefully not worth much discussion. I'd put it into the presentation AI, except that we don't have one at the moment (having approved the last one as part of the Corrigendum). **************************************************************** From: Tucker Taft Sent: Thursday, January 15, 2015 6:36 PM I want this discussed ad nauseum at the next ARG meeting! **************************************************************** From: Randy Brukardt Sent: Thursday, January 15, 2015 6:55 PM I think we discussed it way too long in Portland (which pretty much is at all :-). But if someone felt strongly enough about it to bring it up there, it might as well be right elsewhere in the Standard, too. (At least the containers wording doesn't need to be fixed; it just talks about specific attributes and never mentions "stream" at all.) (I'm presuming that you forgot the smiley face in your message, 'cause I hope you're not serious!) **************************************************************** From: Erhard Ploedereder Sent: Thursday, January 15, 2015 7:26 PM > I want this discussed ad nauseum at the next ARG meeting! Surely we all do, stream-orientedly. **************************************************************** From: Tucker Taft Sent: Thursday, January 15, 2015 8:03 PM Here are a few smileys for future sprinkling as appropriate: ;-) ;-) ;-) ;-) **************************************************************** From: Randy Brukardt Sent: Thursday, January 15, 2015 8:25 PM Technically, those are winks. And when you wink that much, people will think that you've got Cottonwood cotton in your eye. :-) :-) :-) ****************************************************************