wiki:FeatureList

Version 12 (modified by phdmakk, 10 years ago) (diff)

--

FeatureList xText EBNF Version 2.0

Core Features:

  1. BNF front-end (Parser, Lexer, Linker, etc.)
  2. translates a textual representation into an Ecore model
  3. Editor - provides IDE features, e.g. auto-completion, syntax highlighting, etc.

Validation Rules:

These rules will give warnings in the editor if they find find consistencies in the code.

  1. checkUpdateGrammarConsistency: checks, if the grammar is an updated grammar, for consistencies between the originals and the new grammar

    status: not migrated
  1. !CheckReferencedOnlyOnce: this should warn if there is a rule that is only referenced once, this is an inlining hint e.g.:

    status: migrated
  1. checkPassthroughRule: e.g.:this tests if a rule gets just passed through a ::= b b ::= "literal" <=> a ::= "literal" , which is better for EBNF

    status: migrated
  1. checkEqualAlternative: this tests if a rule got equal alternatives e.g.: a::= d | d status: still some problems

    status: migrated
  1. checkDuplicateRules: checks if there is another rule in the EtsiBnf?, that got the same behaviour e.g.: a ::= "literal" | b c ::= "literal" | b

    status: migrated
  1. checkSubruleDuplicates:

status: not migrated

  1. check unused rule: it checks if a rule is used (exept for the rule with the number 1)

status: migrated

  1. checkNameIsUnique: checks if the nam of a rule is already in use e.g.: a ::= "end" a ::= "fin"

status: migrated

Generation:

Generating an intermediate xsl-fo representation for all BNF grammars (merge or delta grammars not supported). This can be rendered to PDF using apache-fop

status: in progress

Formatting:

  • old: had some but they are got commented out in the main project
  • new: line-break after every rule, white-spaces after all literals, possible update linebreak and double tab after all >|< bars

Refactoring and Quick-fixing:

Automated restructuring support.

  1. auto Numbering of rules
    - not migrated
  2. remove unused rules
    - migrated
  3. replacement of passthroughRules
    - not migrated
  4. refactor uppercasetokenRules
    - not migrated
  5. find references - provided by xtext2
  6. renaming - provided by xtext2

Grammar Composition:

  • 3 grammartypes:
    • grammar: normal grammar
    • delta grammar: different notation for a extension grammar,
      • adds new rules to a grammar (regular rules) and adds changes to rules of the extended grammar (extension rules) (can be automatically created)
    • merge grammar: for merging extensions
  • old version had an automatic composition process for merge grammars with a small pop-up-menu.
  • at the moment no support in the new version

Deployment:

  • export project as Plugin: Plug-in Development/deployable plug-ins and fragments
  • RCP project with a minimal eclipse editor: TBD