WI #2584 Implement SORT table statement from Cobol v6.3 (#2586)

TypeCobol.Test/Misc/LanguageLevel/tcQualifiedNames.Mix.txt

@@ -47,6 +47,7 @@ Line 45[20,24] <27, Error, Syntax> - Syntax error : extraneous input 's1::b' exp
            PERFORM s1::b
       *KO qualifiedDataName1
 Line 47[42,67] <27, Error, Syntax> - Syntax error : mismatched input 'WORK-STUDENT::STUDENT-ID-W' expecting user defined word
+Line 47[12,15] <27, Error, Syntax> - Syntax error : SORT file statement requires at least one sorting KEY.
            SORT FILE-03 ON ASCENDING KEY WORK-STUDENT::STUDENT-ID-W
            USING FILE-01 GIVING FILE-02
       *KO qualifiedConditionName

TypeCobol.Test/Parser/CodeElements/SORT_ko.cbl

@@ -1,13 +1,7 @@
-* neither USING not GIVING
-SORT x    ASCENDING      d.
-SORT x ON DESCENDING KEY d.
-SORT x    ASCENDING      d1 d2 d3.
-SORT x    DESCENDING     d1 d2 d3.
-SORT x ASCENDING d1 d2 ASCENDING d3 DESCENDING d4 ASCENDING d5 d6 DESCENDING d7 d8 d9.
-* missing USING
+* missing USING: invalid but not checked by ANTLR grammar
 SORT x DESCENDING d GIVING filename1 filename2 filename3.
 SORT x ASCENDING  d1 d2 d3 GIVING filename.
-* missing GIVING
+* missing GIVING: invalid but not checked by ANTLR grammar
 SORT x DESCENDING d USING filename1 filename2 filename3.
 SORT x ASCENDING  d1 d2 d3 USING filename.
 * either USING or INPUT PROCEDURE

TypeCobol.Test/Parser/CodeElements/SORT_ok.CodeElements.txt

@@ -95,3 +95,73 @@
 
 [[SentenceEnd]] [196,196+:.]<PeriodSeparator> --> [196,196+:.]<PeriodSeparator>
 
+[[SortStatement]] [1,4:SORT]<SORT> --> [6,6:x]<UserDefinedWord>
+
+[[SortStatement]] [1,4:SORT]<SORT> --> [26,26:d]<UserDefinedWord>
+
+[[SentenceEnd]] [27,27+:.]<PeriodSeparator> --> [27,27+:.]<PeriodSeparator>
+
+[[SortStatement]] [1,4:SORT]<SORT> --> [26,26:d]<UserDefinedWord>
+
+[[SentenceEnd]] [27,27+:.]<PeriodSeparator> --> [27,27+:.]<PeriodSeparator>
+
+[[SortStatement]] [1,4:SORT]<SORT> --> [32,33:d3]<UserDefinedWord>
+
+[[SentenceEnd]] [34,34+:.]<PeriodSeparator> --> [34,34+:.]<PeriodSeparator>
+
+[[SortStatement]] [1,4:SORT]<SORT> --> [32,33:d3]<UserDefinedWord>
+
+[[SentenceEnd]] [34,34+:.]<PeriodSeparator> --> [34,34+:.]<PeriodSeparator>
+
+[[SortStatement]] [1,4:SORT]<SORT> --> [84,85:d9]<UserDefinedWord>
+
+[[SentenceEnd]] [86,86+:.]<PeriodSeparator> --> [86,86+:.]<PeriodSeparator>
+
+[[SortStatement]] [1,4:SORT]<SORT> --> [27,31:ORDER]<ORDER>
+
+[[SentenceEnd]] [32,32+:.]<PeriodSeparator> --> [32,32+:.]<PeriodSeparator>
+
+[[SortStatement]] [1,4:SORT]<SORT> --> [27,31:ORDER]<ORDER>
+
+[[SentenceEnd]] [32,32+:.]<PeriodSeparator> --> [32,32+:.]<PeriodSeparator>
+
+[[SortStatement]] [1,4:SORT]<SORT> --> [27,31:ORDER]<ORDER>
+
+[[SentenceEnd]] [32,32+:.]<PeriodSeparator> --> [32,32+:.]<PeriodSeparator>
+
+[[SortStatement]] [1,4:SORT]<SORT> --> [24,25:IN]<IN>
+
+[[SentenceEnd]] [32,32+:.]<PeriodSeparator> --> [32,32+:.]<PeriodSeparator>
+
+[[SortStatement]] [1,4:SORT]<SORT> --> [27,31:ORDER]<ORDER>
+
+[[SentenceEnd]] [32,32+:.]<PeriodSeparator> --> [32,32+:.]<PeriodSeparator>
+
+[[SortStatement]] [1,4:SORT]<SORT> --> [13,22:DUPLICATES]<DUPLICATES>
+
+[[SentenceEnd]] [32,32+:.]<PeriodSeparator> --> [32,32+:.]<PeriodSeparator>
+
+[[SortStatement]] [1,4:SORT]<SORT> --> [24,25:IN]<IN>
+
+[[SentenceEnd]] [32,32+:.]<PeriodSeparator> --> [32,32+:.]<PeriodSeparator>
+
+[[SortStatement]] [1,4:SORT]<SORT> --> [13,22:DUPLICATES]<DUPLICATES>
+
+[[SentenceEnd]] [32,32+:.]<PeriodSeparator> --> [32,32+:.]<PeriodSeparator>
+
+[[SortStatement]] [1,4:SORT]<SORT> --> [30,41:alphabetname]<UserDefinedWord>
+
+[[SentenceEnd]] [42,42+:.]<PeriodSeparator> --> [42,42+:.]<PeriodSeparator>
+
+[[SortStatement]] [1,4:SORT]<SORT> --> [30,41:alphabetname]<UserDefinedWord>
+
+[[SentenceEnd]] [42,42+:.]<PeriodSeparator> --> [42,42+:.]<PeriodSeparator>
+
+[[SortStatement]] [1,4:SORT]<SORT> --> [30,41:alphabetname]<UserDefinedWord>
+
+[[SentenceEnd]] [42,42+:.]<PeriodSeparator> --> [42,42+:.]<PeriodSeparator>
+
+[[SortStatement]] [1,4:SORT]<SORT> --> [30,41:alphabetname]<UserDefinedWord>
+
+[[SentenceEnd]] [42,42+:.]<PeriodSeparator> --> [42,42+:.]<PeriodSeparator>
+

TypeCobol.Test/Parser/CodeElements/SORT_ok.cbl

@@ -29,4 +29,25 @@ SORT x ASCENDING  d1 d2 d3 INPUT PROCEDURE IS procedurename1 THRU procedurename2
 SORT x ASCENDING  d1 d2 d3 INPUT PROCEDURE    procedurename1 THRU procedurename2 OUTPUT PROCEDURE IS procedurename3 THRU procedurename4.
 * everything
 SORT x DESCENDING d             DUPLICATES                    SEQUENCE    alphabetname INPUT PROCEDURE    procedurename1                     OUTPUT PROCEDURE    procedurename2 THRU procedurename3.
-SORT x ASCENDING  d1 d2 d3 WITH DUPLICATES IN ORDER COLLATING SEQUENCE IS alphabetname INPUT PROCEDURE IS procedurename1 THRU procedurename2 OUTPUT PROCEDURE IS procedurename3 THRU procedurename4.
+SORT x ASCENDING  d1 d2 d3 WITH DUPLICATES IN ORDER COLLATING SEQUENCE IS alphabetname INPUT PROCEDURE IS procedurename1 THRU procedurename2 OUTPUT PROCEDURE IS procedurename3 THRU procedurename4.
+* Table SORT format
+SORT x
+SORT x    ASCENDING      d.
+SORT x ON DESCENDING KEY d.
+SORT x    ASCENDING      d1 d2 d3.
+SORT x    DESCENDING     d1 d2 d3.
+SORT x ASCENDING d1 d2 ASCENDING d3 DESCENDING d4 ASCENDING d5 d6 DESCENDING d7 d8 d9.
+* WITH DUPLICATES clause
+SORT x WITH DUPLICATES IN ORDER.
+SORT x      DUPLICATES IN ORDER.
+SORT x WITH DUPLICATES    ORDER.
+SORT x WITH DUPLICATES IN      .
+SORT x      DUPLICATES    ORDER.
+SORT x WITH DUPLICATES         .
+SORT x      DUPLICATES IN      .
+SORT x      DUPLICATES         .
+* COLLATING SEQUENCE clause
+SORT x COLLATING SEQUENCE IS alphabetname.
+SORT x           SEQUENCE IS alphabetname.
+SORT x COLLATING SEQUENCE    alphabetname.
+SORT x           SEQUENCE    alphabetname.

TypeCobol.Test/Parser/Programs/Cobol85/SortStatement.Mix.txt

@@ -0,0 +1,63 @@
+       IDENTIFICATION DIVISION.
+       PROGRAM-ID. TCOMFL06.
+       ENVIRONMENT DIVISION.
+       INPUT-OUTPUT SECTION.
+       FILE-CONTROL.
+           Select Out001        Assign To SOR001.
+           Select AmbiguousName Assign To SOR001.
+       DATA DIVISION.
+       FILE SECTION.
+       SD Out001.
+       01 Out001Data PIC X(200).
+       SD AmbiguousName.
+       01 AmbiguousNameData PIC X(200).
+       WORKING-STORAGE SECTION.
+       01 tab1.
+          05 a OCCURS 20.
+             10 item-a PIC X.
+       01 AmbiguousName PIC X.
+       PROCEDURE DIVISION.
+      *KO unable to resolve SORT target
+Line 21[17,37] <27, Error, Syntax> - Syntax error : Unable to resolve reference to SORT target 'something-not-defined'.
+           SORT something-not-defined
+      *KO ambiguous reference
+Line 23[17,29] <27, Error, Syntax> - Syntax error : Ambiguous reference to SORT target 'AmbiguousName'.
+           SORT AmbiguousName
+      *KO SORT file without KEY
+Line 25[12,15] <27, Error, Syntax> - Syntax error : SORT file statement requires at least one sorting KEY.
+           SORT Out001 INPUT PROCEDURE  inputProc
+                       OUTPUT PROCEDURE outputProc
+      *KO SORT file without input
+Line 28[12,15] <27, Error, Syntax> - Syntax error : Missing input definition in SORT file statement: either add INPUT PROCEDURE phrase or USING file phrase to define required input.
+           SORT Out001 ASCENDING Out001Data
+                       OUTPUT PROCEDURE outputProc
+      *KO SORT file without output
+Line 31[12,15] <27, Error, Syntax> - Syntax error : Missing output definition in SORT file statement: either add OUTPUT PROCEDURE phrase or GIVING file phrase to define required output.
+           SORT Out001 ASCENDING Out001Data
+                       INPUT PROCEDURE  inputProc
+      *KO SORT table with input
+Line 34[12,15] <27, Error, Syntax> - Syntax error : SORT table statement does not allow input definition.
+           SORT a ASCENDING item-a
+                  INPUT PROCEDURE  inputProc
+      *KO SORT table with output
+Line 37[12,15] <27, Error, Syntax> - Syntax error : SORT table statement does not allow output definition.
+           SORT a ASCENDING item-a
+                  OUTPUT PROCEDURE outputProc
+      *KO SORT table without KEY
+Line 40[12,15] <27, Error, Syntax> - Syntax error : SORT table statement has no sorting KEY and the sorted table 'a' does not define any KEY clause.
+           SORT a
+      *Ok, this is a valid SORT file
+           SORT Out001 ASCENDING Out001Data
+                       INPUT PROCEDURE  inputProc
+                       OUTPUT PROCEDURE outputProc
+      *Ok, this is a valid SORT table
+           SORT a ASCENDING item-a
+           GOBACK
+           .
+       inputProc.
+           DISPLAY "input proc"
+           .
+       outputProc.
+           DISPLAY "output proc"
+           .
+       END PROGRAM TCOMFL06.

TypeCobol.Test/Parser/Programs/Cobol85/SortStatement.rdz.cbl

@@ -0,0 +1,55 @@
+       IDENTIFICATION DIVISION.
+       PROGRAM-ID. TCOMFL06.
+       ENVIRONMENT DIVISION.
+       INPUT-OUTPUT SECTION.
+       FILE-CONTROL.
+           Select Out001        Assign To SOR001.
+           Select AmbiguousName Assign To SOR001.
+       DATA DIVISION.
+       FILE SECTION.
+       SD Out001.
+       01 Out001Data PIC X(200).
+       SD AmbiguousName.
+       01 AmbiguousNameData PIC X(200).
+       WORKING-STORAGE SECTION.
+       01 tab1.
+          05 a OCCURS 20.
+             10 item-a PIC X.
+       01 AmbiguousName PIC X.
+       PROCEDURE DIVISION.
+      *KO unable to resolve SORT target
+           SORT something-not-defined
+      *KO ambiguous reference
+           SORT AmbiguousName
+      *KO SORT file without KEY
+           SORT Out001 INPUT PROCEDURE  inputProc
+                       OUTPUT PROCEDURE outputProc
+      *KO SORT file without input
+           SORT Out001 ASCENDING Out001Data
+                       OUTPUT PROCEDURE outputProc
+      *KO SORT file without output
+           SORT Out001 ASCENDING Out001Data
+                       INPUT PROCEDURE  inputProc
+      *KO SORT table with input
+           SORT a ASCENDING item-a
+                  INPUT PROCEDURE  inputProc
+      *KO SORT table with output
+           SORT a ASCENDING item-a
+                  OUTPUT PROCEDURE outputProc
+      *KO SORT table without KEY
+           SORT a
+      *Ok, this is a valid SORT file
+           SORT Out001 ASCENDING Out001Data
+                       INPUT PROCEDURE  inputProc
+                       OUTPUT PROCEDURE outputProc
+      *Ok, this is a valid SORT table
+           SORT a ASCENDING item-a
+           GOBACK
+           .
+       inputProc.
+           DISPLAY "input proc"
+           .
+       outputProc.
+           DISPLAY "output proc"
+           .
+       END PROGRAM TCOMFL06.

TypeCobol/AntlrGrammar/CobolCodeElements.g4

@@ -6684,12 +6684,24 @@ setStatementForConditions:
 // segment or wholly contained within the same independent segment as that SORT
 // statement.
 
+// Table SORT statements can appear anywhere in the PROCEDURE DIVISION. This format of the SORT
+// statement can be used with programs that are compiled with the THREAD option.
+// data-name-2
+// Specifies a table data-name that is subject to the following rules:
+// - data-name-2 must have an OCCURS clause in the data description entry.
+// - data-name-2 can be qualified.
+// - data-name-2 can be subscripted. The rightmost or only subscript of the table must be omitted or
+// replaced with the word ALL.
+// The number of occurrences of table elements that are referenced by data-name-2 is determined
+// by the rules in the OCCURS clause. The sorted table elements are placed in the same table that is
+// referenced by data-name-2.
+
 sortStatement:
-	SORT fileNameReference onAscendingDescendingKey+ 
+	SORT dataItemReferenceOrFileName onAscendingDescendingKey*
 	(WITH? DUPLICATES IN? ORDER?)?
 	collatingSequence?
-	(usingFilenames  | inputProcedure)
-	(givingFilenames | outputProcedure);
+	(usingFilenames  | inputProcedure)?
+	(givingFilenames | outputProcedure)?;
 
 // Rules shared with mergeStatement
 

TypeCobol/AntlrGrammar/CobolExpressions.g4

@@ -40,6 +40,10 @@ dataItemReferenceOrConditionReferenceOrIndexName:
 // [Type ambiguity] : fileName cannot be distinguished from identifier at this parsing stage
 dataItemReferenceOrConditionReferenceOrFileName: 
 	qualifiedDataNameOrQualifiedConditionNameOrFileName (LeftParenthesisSeparator subscript+ RightParenthesisSeparator)?;
+
+// [Type ambiguity] : fileName cannot be distinguished from identifier at this parsing stage
+dataItemReferenceOrFileName: 
+	qualifiedDataNameOrFileName (LeftParenthesisSeparator subscript+ RightParenthesisSeparator)?;
 
 // [Type ambiguity] : className cannot be distinguished from identifier at this parsing stage
 dataItemReferenceOrConditionReferenceOrClassName: 

TypeCobol/AntlrGrammar/CobolWords.g4

@@ -1715,6 +1715,10 @@ qualifiedDataNameOrQualifiedConditionNameOrIndexName:
 qualifiedDataNameOrQualifiedConditionNameOrFileName:
 	dataNameReferenceOrConditionNameReferenceOrConditionForUPSISwitchNameReferenceOrFileNameReference | qualifiedDataNameOrQualifiedConditionName1;
 
+// [Type ambiguity] at this parsing stage
+qualifiedDataNameOrFileName:
+	dataNameReferenceOrFileNameReference | qualifiedDataName1;
+
 // [Type ambiguity] at this parsing stage
 qualifiedDataNameOrQualifiedConditionNameOrClassName:
 	dataNameReferenceOrConditionNameReferenceOrConditionForUPSISwitchNameReferenceOrClassNameReference | qualifiedDataNameOrQualifiedConditionName1;

TypeCobol/Compiler/CodeElements/CobolLanguageLevelVisitor.cs

@@ -336,7 +336,7 @@ bool IsStopVisitingChildren
         bool Visit([NotNull] DataSection dataSection);
         bool Visit([NotNull] DataDivision dataDivision);
         bool Visit([NotNull] FileSection fileSection);
-        bool Visit([NotNull] FileDescriptionEntryNode fileDescriptionEntryNode);
+        bool Visit([NotNull] FileDescription fileDescription);
         bool Visit([NotNull] GlobalStorageSection globalStorageSection);
         bool Visit([NotNull] WorkingStorageSection workingStorageSection);
         bool Visit([NotNull] LocalStorageSection localStorageSection);
@@ -1355,7 +1355,7 @@ public virtual bool Visit(JsonParse jsonParse)
             return true;
         }
 
-        public virtual bool Visit(FileDescriptionEntryNode fileDescriptionEntryNode) {
+        public virtual bool Visit(FileDescription fileDescription) {
             return true;
         }
 

TypeCobol/Compiler/CodeElements/Statement/SortStatement.cs

@@ -1,6 +1,4 @@
-using System.Collections.Generic;
-
-namespace TypeCobol.Compiler.CodeElements
+namespace TypeCobol.Compiler.CodeElements
 {
     /// <summary>
     /// p422:
@@ -21,17 +19,24 @@ public SortStatement() : base(CodeElementType.SortStatement, StatementType.SortS
         { }
 
         /// <summary>
-        /// p422:
         /// file-name-1
         /// The name given in the SD entry that describes the records to be sorted.
-        ///
-        /// p423:
-        /// No pair of file-names in a SORT statement can be specified in the same SAME
-        /// SORT AREA clause or the SAME SORT-MERGE AREA clause. File-names
-        /// associated with the GIVING clause (file-name-3, ...) cannot be specified in the SAME
-        /// AREA clause; however, they can be associated with the SAME RECORD AREA clause.
+        /// No pair of file-names in a SORT statement can be specified in the same SAME SORT AREA clause or the
+        /// SAME SORT-MERGE AREA clause. File-names associated with the GIVING clause (file-name-3, ...) cannot
+        /// be specified in the SAME AREA clause; however, they can be associated with the SAME RECORD AREA
+        /// clause.
+        /// 
+        /// data-name-2
+        /// Specifies a table data-name that is subject to the following rules:
+        /// - data-name-2 must have an OCCURS clause in the data description entry.
+        /// - data-name-2 can be qualified.
+        /// - data-name-2 can be subscripted. The rightmost or only subscript of the table must be omitted or
+        /// replaced with the word ALL.
+        /// The number of occurrences of table elements that are referenced by data-name-2 is determined
+        /// by the rules in the OCCURS clause. The sorted table elements are placed in the same table that is
+        /// referenced by data-name-2.
         /// </summary>
-        public SymbolReference FileName { get; set; }
+        public DataOrConditionStorageArea FileNameOrTableName { get; set; }
 
         /// <summary>
         /// p423:
@@ -255,7 +260,7 @@ public SortStatement() : base(CodeElementType.SortStatement, StatementType.SortS
         public override bool VisitCodeElement(IASTVisitor astVisitor)
         {
             return base.VisitCodeElement(astVisitor) && astVisitor.Visit(this)
-                   && this.ContinueVisitToChildren(astVisitor, FileName, WithDuplicates, CollatingSequence, 
+                   && this.ContinueVisitToChildren(astVisitor, FileNameOrTableName, WithDuplicates, CollatingSequence, 
                    InputProcedure, ThroughInputProcedure, OutputProcedure, ThroughOutputProcedure)
                    && this.ContinueVisitToChildren(astVisitor, SortingKeys, InputFiles, OutputFiles);
         }

TypeCobol/Compiler/CodeModel/Method.cs

@@ -73,7 +73,7 @@ public class Method
         /// More than one record description entry can be specified; each is an alternative description of the same record storage area.
         /// Data areas described in the FILE SECTION are not available for processing unless the file that contains the data area is open.
         /// </summary>
-        public IDictionary<SymbolDefinition, FileDescription> FileDescriptions { get; set; }
+        public IDictionary<SymbolDefinition, FileDescriptionEntry> FileDescriptions { get; set; }
 
         /// <summary>
         /// A single copy of the WORKING-STORAGE for a method is statically allocated on the first invocation of the method and persists in a last-used state for the duration of the run unit.