!standard 3.5.1(14) 20-07-30 AI12-0386-1/02 !standard 4.2.1(19/5) !standard 5.5.2(28/5) !standard A.18.32(20/3) !class presentation 20-06-19 !status Amendment 1-2012 20-06-19 !status ARG Approved 15-0-0 20-07-30 !status work item 20-06-19 !status received 20-06-15 !priority Low !difficulty Easy !qualifier Omission !subject Still More Presentation issues !summary [Editor's note: These changes are included in the current draft of the Standard, even though they are not yet approved.] (1) Add 'Image to the example 5.5.2(25/5). (2) Remove an extra closing parenthesis in 4.2.1(19/5). (3) Correct the array component subtype in 4.2.1(19/5). (4) Add a definition for Month_Name to 3.5.1(14). (5) Add Prepend (The_Path, N); to A.18.33(20/3). !question (1) 5.5(28/5) says: Put_Line("Total=" & Partial_Sum'Reduce("+", 0) & ", Min=" & Partial_Min'Reduce(Natural'Min, Natural'Last) & ", Max=" & Partial_Max'Reduce(Natural'Max, 0)); end; The component subtype of the Partial_ arrays is Natural, so the result of the reduction is also Natural. We can't concatenate a Natural to a String, so this call is illegal. Should this be repaired? (Yes.) (2) 4.2.1(19/5) is the body of To_Roman_Number (given as an expression function): function To_Roman_Number (S : String) return Roman_Number is (declare R : constant array (Integer range <>) of Integer := (for D in S'Range => Roman_Digit'Enum_Rep (Roman_Digit'Value (''' & S(D) & '''))); -- See 3.5.2 and 13.4 begin [for I in R'Range => (if I < R'Last and then R(I) < R(I + 1) then -1 else 1) * R(I))] 'Reduce("+", 0) ); The value_sequence in the square brackets has 4 opening parentheses and 5 closing parentheses. Should the last of these be removed? (Yes.) (3) Also in 4.2.1(19/5), the array R has components of subtype Integer, so the reduction expresion is also returning type Integer. But the result of the function is type Roman_Number, which is a different integer type. Should this be fixed? (Yes.) (4) type Date declared in 3.8(27) uses type Month_Name. However, Month_Name is not declared anywhere. Should this be fixed? (Yes.) (5) The example in A.18.33 produces a path that does not include the last step to the Target node. That is conventionally included in the path. Should the example be fixed? (Yes.) !response (1) It appears the intent is to display the Image of these results. We can just add 'Image to the end of each 'Reduce, as an attribute_reference is a syntactic name and thus can be the prefix to 'Image. (2) It appears that someone miscounted the parentheses in this example. (3) Since attribute Enum_Rep is a universal integer attribute, we can change the component subtype of array R to Roman_Number without any other change. This seems to be the minimum change that fixes the type mismatch. Alternatively, we could have surrounded the entire reduction with a type conversion to Roman_Number. But this is more obtrusive and also fails to illustrate the value of properly selecting types. (4) There has been a principle in the Ada Reference Manual that if one combined all of the declarations of all of the examples together, the result would be compilable (that is, all types and objects used would be defined). [Editor's note: I've added an AARM Note describing this principle following the definition of the Example header in 1.1.2. We've referred to it for a long time without it ever appearing in the AARM.] While Month_Name itself is only used in type Date in 3.8(27), type Date is used in a number of aggregate examples, including 4.3.1(25), 4.3.1(27)), 4.3.4(23/5), and 4.3.4(28/5). These examples use the enumeration literals "July" and "Apr", which are inconsistent. Thus we also change 4.3.4(28/5) to use "April". (5) The result path should include the Target. !wording (1) Replace 5.5(28/5) with: Put_Line ("Total=" & Partial_Sum'Reduce("+", 0)'Image & ", Min=" & Partial_Min'Reduce(Natural'Min, Natural'Last)'Image & ", Max=" & Partial_Max'Reduce(Natural'Max, 0)'Image); end; (2) & (3) Replace 4.2.1(19/5) with: function To_Roman_Number (S : String) return Roman_Number is (declare R : constant array (Integer range <>) of Roman_Number := (for D in S'Range => Roman_Digit'Enum_Rep (Roman_Digit'Value (''' & S(D) & '''))); -- See 3.5.2 and 13.4 begin [for I in R'Range => (if I < R'Last and then R(I) < R(I + 1) then -1 else 1) * R(I))] 'Reduce("+", 0) ); (4) Replace 3.5.1(14) with: type Day is (Mon, Tue, Wed, Thu, Fri, Sat, Sun); type Month_Name is (January, February, March, April, May, June, July, August, September, October, November, December); type Suit is (Clubs, Diamonds, Hearts, Spades); type Gender is (M, F); type Level is (Low, Medium, Urgent); type Color is (White, Red, Yellow, Green, Blue, Brown, Black); type Light is (Red, Amber, Green); -- Red and Green are overloaded Replace 4.3.4(28/5) with: Tomorrow := ((Yesterday with delta Day => 12) with delta Month => April); -- see 3.8 (5) Replace A.18.33(20/3) with: declare N : Node := Target; begin Prepend (The_Path, N); while N /= Source loop N := Reached_From(N); Prepend (The_Path, N); end loop; end; !discussion None. !corrigendum 3.5.1(14) @drepl @xcode<@b Day @b (Mon, Tue, Wed, Thu, Fri, Sat, Sun); @b Suit @b (Clubs, Diamonds, Hearts, Spades); @b Gender @b (M, F); @b Level @b (Low, Medium, Urgent); @b Color @b (White, Red, Yellow, Green, Blue, Brown, Black); @b Light @b (Red, Amber, Green); --@ft<@i< Red and Green are overloaded>>> @dby @xcode<@b Day @b (Mon, Tue, Wed, Thu, Fri, Sat, Sun); @b Month_Name @b (January, February, March, April, May, June, July, August, September, October, November, December); @b Suit @b (Clubs, Diamonds, Hearts, Spades); @b Gender @b (M, F); @b Level @b (Low, Medium, Urgent); @b Color @b (White, Red, Yellow, Green, Blue, Brown, Black); @b Light @b (Red, Amber, Green); --@ft<@i< Red and Green are overloaded>>> !corrigendum 4.2.1(0) @dinsc See the conflict file for the changes. !corrigendum 4.3.4(0) @dinsc See the conflict file for the changes. !corrigendum 5.5(21) @dinsa @xcode Next /= Head @b -- @ft<@i> Sum := Sum + Next.Value; Next := Next.Succ; @b Summation;> @dinss @i See the conflict file for the changes. !corrigendum A.18.33(20/3) @drepl @xcode< @b N : Node := Target; @b @b N /= Source @b N := Reached_From(N); Prepend (The_Path, N); @b; @b;> @dby @xcode< @b N : Node := Target; @b Prepend (The_Path, N); @b N /= Source @b N := Reached_From(N); Prepend (The_Path, N); @b; @b;> !ASIS No ASIS effect. !ACATS test No ACATS tests needed for presentation issues. !appendix !topic Shortest_Paths example error !reference Ada 202x RM18.32(20/3) !from John J Cupak Jr 20-06-09 !keywords Container Example Error !discussion The Shortest_Paths generic package presented in A.18.32 Example of Container Use (pages 580-581) does not return the complete path. Using the tinyEWD.txt data file from Sedwick’s Algorithms, the correct returned path from node 0 to node 6 should be: 0 -> 2 -> 7 -> 3 -> 6 However, the actual returned path is: 0 -> 2 -> 7 -> 3 The error lies in the Rebuild path from target to source code block, which is published as: 19/3 -- Rebuild path from target to source. 20/3 declare N : Node := Target; begin while N /= Source loop N := Reached_From(N); Prepend (The_Path, N); end loop; end; return The_Path; After analysis, it was determined that a Prepend (The_Path, N); statement was missing between the begin and the while loop. The corrected version should be: 19/3 -- Rebuild path from target to source. 20/3 declare N : Node := Target; begin Prepend (The_Path, N); — MISSING while N /= Source loop N := Reached_From(N); Prepend (The_Path, N); end loop; end; return The_Path; ****************************************************************