0.5.0 update to work with ghc 8.6.5, also tested with 8.4.4 and 8.2.1 rename some of the modules Lexer -> Lex, Parser -> Parse add a separate lexer to simplify code and speed up parsing replace SqlIndent with new tool, SimpleSqlParserTool (amazing name) which can indent, and parse and lex. experiments in new approach to dealing with fixities with separate pass after parsing dml :add support for insert, update, delete and truncate ddl: add limited support for create schema, plus drop schema create, alter and drop table with defaults and constraints create, alter and drop for domain, view, sequence create and drop for assertion access control: simple create and drop for role simple grant and revoke limited support for transaction management: start transation, rollback, commit, savepoint fix the precendence of operators which was following the weird postgresql 9.4 and earlier precendences instead of the standard refactor the syntax for names, identifiers and strings slightly refactor the dialect support, add some support for postgresql syntax change parsing of identifiers and strings to not unescape the identifier or string text during parsing add some explicit parse failures for probably ambiguous text */ without /* (outside quoted identifier, string) will fail .,e,E following a number without whitespace always fails three symbols together fails explicitly, instead of trying to lex and giving a less good error at parse time (applies to | and : in postgres dialect) fix parsing of functions whose name is a keyword (e.g. abs) add basic support for parsing odbc syntax ({d 'literals'} {fn app(something)} and {oj t1 left outer join ... } rename ValueExpr -> ScalarExpr (I think scalar expression is slightly less incorrect) rename CombineQueryExpr to QueryExprSetOp and CombineOp to SetOperatorName use explicit data type for sign in interval literals add comments to statement syntax (aimed at codegen) add support for oracle type size units 'char' and 'byte', example: varchar2(55 byte) updated the makefile to use cabal v2 commands fix for parsing window functions with keyword names 0.4.4 tested with ghc 8.2.1 and 8.4.3 0.4.3 tested with ghc 8.0.2 and 8.2.1 0.4.1 (commit c156c5c34e91e1f7ef449d2c1ea14e282104fd90) tested with ghc 7.4.2, 7.6.3, 7.8.4,7.10.0.20150123 simple demonstration of how dialects could be handled internally add ability to add comments to syntax tree to help with generating SQL code 0.4.0 (commit 7914898cc8f07bbaf8358d208469392346341964) now targets SQL:2011 update to ghc 7.8.2 remove dependency on haskell-src-exts derive Data and Typeable in all the syntax types improve the error messages a great deal sql features: parse schema qualified table names in from clause (thanks to Sönke Hahn) support multiline string literals support colon prefix host parameters and introducer support unique predicate support match predicate support array constructors and subscripting support character set literals support collate support escape for string literals as a postfix operator parse schema/whatever qualified ids in various places: identifiers (replaces equivalent functionality using '.' operator), function, aggregate, window function names, explicit tables and functions in from clauses, typenames support almost all typename syntax for SQL:2011 (just missing refs) support most multiset operations (missing some predicates only, likely to be added before next release) support two double quotes in a quoted identifier to represent a quote character in the identifier support filter and within group for aggregates support next value for parse special nullary functions annoying changes: replace Int with Integer in the syntax remove support for parsing clauses after the from clause if there is no from clause change the syntax representation of quantified comparison predicates change the hardcoded collate keyword in substring and trim to use the new collate postfix operator, this also changes the collation name to be an identifier instead of a string represent missing setquantifier as a literal default instead of as the actual default value (all in select, distinct in set operators) same for sort directions in order by implement complete interval literals (fixed the handling of the interval qualifier) make most of the standard reserved words actually reserved (still some gaps) change the natural in join abstract syntax to match the concrete syntax instead of combining natural, on and using into one field remove support for postgresql limit syntax bug fixes: fix some trailing whitespace issues in the keyword style functions, e.g. extract(day from x), dealing with trailing whitespace on the parens was fixed improve some cases of parsing chained prefix or postfix operators (still some issues here) fix bug where the 'as' was incorrectly optional in a 'with expression list item' fix bug in set operations where 'all' was assumed as the default instead of 'distinct', e.g. 'select * from t union select * from u' was parsed to 'select * from t union all select * from u' instead of 'select * from t union distinct select * from u'. fix corresponding bug where 'distinct' was being pretty printed in this case and 'all' was not since the assumed default was the wrong way round fix some trailing junk lexing issues with symbols and number literals fix number literals to accept upper case E 0.3.1 (commit 5cba9a1cac19d66166aed2876d809aef892ff59f) update to work with ghc 7.8.1 0.3.0 (commit 9e75fa93650b4f1a08d94f4225a243bcc50445ae) fix the basic operator fixity parsing swap the order in select item abstract syntax so it is now (expression, alias) which matches the order in the concrete syntax rename ScalarExpr -> ValueExpr rename Duplicates to SetQuantifier rename qeDuplicates to qeSetQuantifier rename OrderField to SortSpec rename InThing to InPredValue add support for ? for parameterized queries add new abstract syntax for special operators whose concrete syntax is a kind of limited named parameters syntax add more parsing for these operators: position, convert, translate, overlay, trim, and improve the substring parsing add support for multi keyword type names previously: double precision character varying now: double precision, character varying, char varying, character large object, char large object, national character, national char, national character varying, national char varying, national character large object, nchar large object, nchar varying, bit varying rename tools/PrettyIt to tools/SQLIdent and add to cabal file as optional executable (disabled by default) rename the qeFetch field in Select to qeFetchFirst change the pretty printer to use 'fetch first' instead of 'fetch next' 0.2.0 (commit 9ea29c1a0ceb2c3f3157fb161d1ea819ea5d64d4) '' quotes in string literal parse simple interval literal e.g. "interval '1 week'" support . in identifiers as a dot operator support quoted identifiers partial support for explicit window frames support multiple test expressions in when branches in case expressions rename CastOp to TypedLit support typenames with precision and scale in casts support nulls first and nulls last in order by support grouping expressions: group by (), grouping sets, cube, rollup and parens support with recursive support values table literal support 'table t' syntax rename fields qe1 and qe1 in combinequeryexpr to qe0 and qe1 add support for functions in from clause add support for lateral in from clause support column aliases in common table expressions refactor the tests and add lots more parse * in any scalar context instead of trying to restrict it support row ctor without 'row' e.g. (a,b) = (c,d) add % ^ | & ~ operators support ansi standard syntax for offset n rows and fetch first n rows only fix keyword parsing to be case insensitive 0.1.0.0 (commit 9bf4012fc40a74ad9a039fcb936e3b9dfc3f90f0) initial release