To: J3 J3/19-235 From: Malcolm Cohen Subject: Image selectors and established coarrays. Date: 2019-October-14 1. Introduction Paper 19-180 contains three purported interpretation requests glommed into a single paper. This paper discusses the third of them. 2. The alleged problem From 19-180: "Section 9.6, paragraph 3 explains the semantics of image selectors that specify TEAM=, TEAM_NUMBER=, or that specify neither. There are several problems with this paragraph. ... "Finally, if neither TEAM= nor TEAM_NUMBER= is specified, nothing is said as to whether the coarray object has to be established." 3. Response from /INTERP /INTERP believes there is only a single case where a coarray can be "not established" in the current team or an ancestor thereof: that is, when it is an unallocated allocatable. An unallocated allocatable is not permitted to be referenced or defined. Using an image selector for a subobject of an allocatable coarray that is not allocated is thus already not conforming. Therefore there can be no problem. /INTERP will entertain assertions to the contrary if such an assertion is accompanied by otherwise conforming Fortran code demonstrating the asserted problem. 4. Response from the Editor The editor agrees that the paragraph in question is large and unwieldy, and deserves to be more clearly written. In particular, splitting it into three paragraphs, each of which describes one of the three cases (TEAM=, TEAM_NUMBER=, and neither) would almost certainly be an improvement. Adding a sentence to the "neither" case noting that the coarray needs to be established in the current team or an ancestor thereof, and/or noting that if the coarray is allocatable it needs to be allocated, would likely also be an improvement. ===END===