JSqlParser Versions Save

• added options to EXPLAIN statement
• preserved input of && and avoid translation to AND
• added multi values in IN expression
• added COMMENT support for ALTER TABLE statement
• added some FOREIGN KEY definition improvments
• allow VALIDATE as column name
• first support for CREATE SCHEMA and DROP SCHEMA
• allow ON as a value in a set statement (SET myvalue = ON)
• support for ALTER TABLE ONLY mytable ...
• allow foreign key definition in alter statements without referenced columns specification
• allow datatype binary for column definitions
• support for ALTER TABLE COLUMN DROP NOT NULL
• allow order as column name
• support for table function in IN expression
• allow complex expressions within all of a case when statement
• support for parameters in create index statement
• support for integer parameters for ->> and -> JSON expressions
• support for more keywords in following parts of multipart object names (e.g. mytab.select, mytab.create, ...)
• support for OUTER APPLY
• support for WITH(NOLOCK)
• support for VIEW as object name
• functions hold now multipart names
• support for DISABLE as object name
• API change in ColumnDefinition.setColumnSpecStrings to ColumnDefinition.setColumnSpecs to better allow the merge between AlterExpression.ColumnDataType and ColumnDefinition

• support for cross apply
• allowed keyword character as object name
• allowed keyword size as object name
• introduced FILTER expression for window functions
• allow more complex expressions for CASE.
• allowed start as object name as column name or table name
• introduced more positions for ! instead of NOT
• allowed more complex expressions within if function
• introduced multicolumn alias like select * from mytab as tab(c1, c2)
  • additional type definition is possible (issue #849)

• support for array constructs using square brackets. This collides with square bracket quotation of SqlServer. The parser has now a flag to enable this quotation again (CCJSqlParser.withSquareBracketQuotation).
• support for update table1 inner join table2 ... (API change)
• support for declare statement
• allow empty double quotes
• allow year, month ... as column data type for create table
• allow duplicate as object name
• make it JDK 11 buildable
• switched to minimum JDK 8
• avoid buffer copy of input data if a normal String input is used
• allow limit and offset as keywords in specific places
• DIV operator
• improved performance for SQLCondition production
• support for full text search (MATCH..AGAINST)

• support for similar to statement
• support for show statment (API - change)
• introduces Statement streaming support by CCJSqlParserUtil.streamStatements
• improved nested bracket parsing performance by a large scale

• support for GROUPING SETS
• first support for db date arithmetic
• support for chained functions
• first support for FOR XML PATH
• support for NEXTVAL FOR
• changed all source code license headers to reflect the dual license of JSqlParser more correctly
• support of OPTIMIZE FOR 20 ROWS like expressions
• allowed conditions within then and else of a case statement
  • SELECT * FROM mytable WHERE CASE WHEN a = 1 THEN b IN (1,2,3) ELSE c IN (1,2,3) END
• change of parsing for not within condition: outer not is represented now by NotExpression
• support of named parameters for execute: EXEC procedure @param = 'foo'
• support multivalue set statement
• support of describe
• support of explain
• support of prefix _utf8'strings'
• allow top keyword as table / column / alias name

• support of substring(col from 2), position('test' in col), .. and more (pull request #702)
• support of db2 VALUES syntax (issue #561)
• bugfixes

This is the last 1.x release of JSqlParser.

JSqlParser 1.3

• support for with - selects statements in create view definitions
• support for block statements (begin ... end)
• support for additional raw string and byte prefixes (issue #659)
• support for special oracle type syntax varchar2(255 BYTE) (issue #273)
• introduced dotted multipart names for uservariables (issue #608)
• changed behaviour of dotted multipart names for tables and columns to accept ORM class names (issue #163) ** the parser allows now empty inner names, to still accept missing schema names for SQLServer (db..col) ** methods like getDatabase will still work but have no sense using it for classnames
• named parameter for OFFSET (issue #612)
• corrected ISNULL regression (issue #610)
• refactored statement test classes to the class corresponding packages
• allowed nested postgresql casts (e.g. col::bigint::int)

• support for isnull
• support for single quoted aliases select col1 as 'alias'
• support for Unicode string literal like N'test' or U'test'
• support for bitwise not ~
• support for drop view
• support for indexed JDBC parameters at multiple places
• allowed index as object name
• switched to JavaCC modern template
• switched to JDK 1.7
• introduced more AST node links
• support for aliased table in insert into statement
• SQL_CALC_FOUND_ROWS support
• support for more complex expressions within case expr when expr then expr end.
• support for << and >> left and right shift operations
• breaking API change: merge of within group and over (window expressions)
• first support for use statements
• first support for call statements
• create table allows now quoted identifiers within primary key definition
• introduced alias for subquery in combination with a pivot definition (this changes alias handling within the library for pivot sqls)
• force the parser to fully parse a String using parseCondExpression or parseExpression

• support for UPSERT syntax
• absolute token positions in addition to column/line
• common normal form transformer for expressions (https://en.wikipedia.org/wiki/Conjunctive_normal_form)
• checkstyle integration to force first souce code conventions
  • checkstyle is activated by default, it can be deactivated by an environment property skipCheckSources
  • pull requests should follow this style settings

• support FOR UPDATE WAIT
• support for simple expressions within case when
• rewrite of SelectBody - production, reduce of needed lookaheads results in huge parser performance improvements at least in some cases
  • please test it due to possible changes in the parse tree
• released as version 1.0