October 2006 - Cvs-nhc98

patch applied (hat): Corrected a bug in position of function declarations that define operators.
by Malcolm Wallace 10 Oct '06

10 Oct '06

Mon Aug 2 10:35:39 PDT 2004 olaf * Corrected a bug in position of function declarations that define operators. Also now handle positions of types correctly. The position of TypeCon still only described position of the type construtor, but getPos gives the position of the whole type. M ./src/compiler98/Extra.hs -2 +16 M ./src/compiler98/MkSyntax.hs -1 +2 M ./src/compiler98/Rename.hs -2 +2 M ./src/compiler98/SyntaxPos.hs -1 +3

1 0

patch applied (hat): Completed modification of parser etc to handle extended positions correctly.
by Malcolm Wallace 10 Oct '06

10 Oct '06

Mon Aug 2 07:45:34 PDT 2004 olaf * Completed modification of parser etc to handle extended positions correctly. Surely some cases have been missed, which will be noticed in some time as bugs. Note that the tricky problems mentioned last time (see copy below) still exist. Also, positions of types have been left unchanged. These positions do not appear in error messages and there is again a problem of missing position information: a TypeCon has a position for the type constructor, but not the whole type. Note that the type constructor may actually be positioned between the arguments (->). Some tricky things not yet solved: - some constructs may include optional semicolons or curly brackets, e.g. do { x <- myRead; putStr x; } these are currently not included in the position, so the do expression ends at the last "x". - the syntax does not include enough positions to give positions to everything. E.g. now the position of a function declaration DeclFun describes the full declaration, there is no position for the function variable itself. - some syntactic constructs are immediately desugared, for example list comprehensions; now every component of the translation gets the position of the full list comprehension. This makes it easy to map parts of the syntax tree to the source code, but impossible to map part of the source code uniquely to the part of the syntax tree that describes it completely. So hat-observe will find it hard to give for a marked list comprehension just the values of the full list comprehension, not intermediate internal values. M ./src/compiler98/MkSyntax.hs -7 +36 M ./src/compiler98/Parse.hs -33 +39 M ./src/compiler98/SyntaxPos.hs -1 +6

1 0

patch applied (hat): When printing source positions, avoid inelegant output like
by Malcolm Wallace 10 Oct '06

10 Oct '06

Mon Aug 2 05:20:20 PDT 2004 malcolm * When printing source positions, avoid inelegant output like 23:12-23:12 and 23:12-0:0 by collapsing both to just 23:12 M ./src/compiler98/Extra.hs -2 +4

1 0

patch applied (hat): Parts of parser and renamer modified to handle extended positions correctly.
by Malcolm Wallace 10 Oct '06

10 Oct '06

Fri Jul 30 11:57:25 PDT 2004 olaf * Parts of parser and renamer modified to handle extended positions correctly. This work is not yet complete. Some tricky things not yet solved: - some constructs may include optional semicolons or curly brackets, e.g. do { x <- myRead; putStr x; } these are currently not included in the position, so the do expression ends at the last "x". - the syntax does not include enough positions to give positions to everything. E.g. now the position of a function declaration DeclFun describes the full declaration, there is no position for the function variable itself. - some syntactic constructs are immediately desugared, for example list comprehensions; now every component of the translation gets the position of the full list comprehension. This makes it easy to map parts of the syntax tree to the source code, but impossible to map part of the source code uniquely to the part of the syntax tree that describes it completely. So hat-observe will find it hard to give for a marked list comprehension just the values of the full list comprehension, not intermediate internal values. M ./src/compiler98/Extra.hs +4 M ./src/compiler98/MkSyntax.hs -6 +8 M ./src/compiler98/Parse.hs -8 +13 M ./src/compiler98/Rename.hs -6 +10 M ./src/compiler98/SyntaxPos.hs -73 +63

1 0

patch applied (hat): Modified hat-trans to compile with the new extended position type Pos.
by Malcolm Wallace 10 Oct '06

10 Oct '06

Fri Jul 30 06:27:20 PDT 2004 olaf * Modified hat-trans to compile with the new extended position type Pos. However, the additional end line and column of a position are currently ignored. M ./src/hattrans/AuxFixity.hs -1 +1 M ./src/hattrans/TraceTrans.hs -3 +3

1 0

patch applied (hat): Changed output of positions to give both start and end.
by Malcolm Wallace 10 Oct '06

10 Oct '06

Thu Jul 29 10:27:15 PDT 2004 olaf * Changed output of positions to give both start and end. Thus noticed wrong ends of identifiers and numbers and corrected that. M ./src/compiler98/Extra.hs -3 +4 M ./src/compiler98/LexPre.hs -3 +3

1 0

patch applied (hat): Extended Pos in Extra to have start and end line & column.
by Malcolm Wallace 10 Oct '06

10 Oct '06

Thu Jul 29 09:49:05 PDT 2004 olaf * Extended Pos in Extra to have start and end line & column. Also modified lexer to generate token sequence with each token having its full position, i.e. including correct end line & column. However, parser not yet modified, so that positions in syntax tree only describe end of the first token of the syntactic construct, not the true one. Changes not yet noticeable, because error messages still give only start line:column. Hopefully no space leak introduced (positions are strict and their evaluation forced in lexer). (But size of a position obviously at least doubled.) M ./src/compiler98/Extra.hs -15 +36 M ./src/compiler98/LexPre.hs -38 +60 M ./src/compiler98/Lexical.hs -20 +20 M ./src/compiler98/ParseCore.hs +4

1 0

patch applied (hat): Just a few more comments and type annotations.
by Malcolm Wallace 10 Oct '06

10 Oct '06

Wed Jul 28 08:06:10 PDT 2004 olaf * Just a few more comments and type annotations. M ./src/compiler98/LexPre.hs +2 M ./src/compiler98/Lexical.hs -1 +10

1 0

patch applied (hat): Made the type for source code positions in Extra, Pos, an abstract data type.
by Malcolm Wallace 10 Oct '06

10 Oct '06

Wed Jul 28 06:10:00 PDT 2004 olaf * Made the type for source code positions in Extra, Pos, an abstract data type. This required changes of many type annotations where Pos and Int were confused. Also added some more type annotations. In Gcode the internal representation of Pos still appears. This is all the first step towards an extended Pos that contains line and column of both start (currently) and end (to do) of a syntactic construct. M ./src/compiler98/Derive.hs -1 +1 M ./src/compiler98/Extra.hs -8 +26 M ./src/compiler98/Gcode.hs -1 +1 M ./src/compiler98/IExtract.hs -10 +10 M ./src/compiler98/Need.hs -2 +2 M ./src/compiler98/NeedLib.hs +2 M ./src/compiler98/Parse2.hs +3 M ./src/compiler98/ParseI.hs +3 M ./src/compiler98/ParseLex.hs +2 M ./src/compiler98/PreImp.hs -7 +8 M ./src/compiler98/Rename.hs -2 +2 M ./src/compiler98/STGGcode.hs -6 +6 M ./src/compiler98/TypeCtx.hs -1 +1

1 0

patch applied (hat): Change ' make install' to ' make all' for all sub-components, to avoid
by Malcolm Wallace 10 Oct '06

10 Oct '06

Tue Jul 27 02:43:24 PDT 2004 malcolm * Change 'make install' to 'make all' for all sub-components, to avoid confusion. M ./src/compiler98/Makefile -1

1 0