* modified the TaskParseCxx to be able to run multiple times in parallel
* made FileRegister threadsafe and changed a lot of its mechanisms
* added TaskGroupParallel that runs all children in parallel
* added TaskParseWrapper that acts as a decorator to execute some code before and after parsing.
* implemented task setup in project with 4 parsing threads
* removed SimpleTask as it was only used as interface for the LambdaTask
* added SymbolType that can be used by ASTVisitor and ParserClient
* reordered some of the ParserClient's methods.
* removed return type of ParserClient methods.
* added recording of references to the ASTVisitor.
* refactored reference recording in ASTVisitor.
* merged all the onXxxUsageParsed methods of the ParserClient.
- added local symbols to storage
- added tests for local symbol parsing
- added MessageActivateLocalSymbols to activate local symbols in code view
- TokenLocations now store their LocationType instead of a plain bool "isScope"
- moved LocationType from TokenLocation to separate file, since it is also used in Annotations now.
- added local_symbol in stylesheet to define a style for hovering and activating these elements
- renamed location to token in stylesheet
* visit implicit declarations in ASTVisitor
* show implicit and undefined child nodes only when connected
* fixed some bundle edges couldn't be split
* add "implicit" to tooltip
* show undefiend and implicit nodes hatched
* storing the information if a node is defined implicitly in the database
* the analysis however does not visit implicit nodes directly. Currently they are just recorded when used somewhere in the code. This should be changed once implicit nodes are hidde in the UI.
* removed error logs for unnamed template parameters and function parameters since it is ok when they are anonymous.
* Added CDB option to project type selection
* show CDB source files in popup
* define header paths separately
* made generic content summary in wizzard and use it everywhere
* fixed file paths not canonical in analysis
* refactored wizzard content paths to use summary instead of subpaths
* refactored files popup to simpler communication with content
bug id = 35
* removed unused code: ParseFunction, ParseVariable and ParseTypeUsage. All this is handled using NameHierarchies now.
* Moved the getErrorCount method from ParserClient to the StorageAccess
* MessageFinishedParsing doesn't know the error count anymore.
* StatusBarController now has a StorageAccess to fetch the error count when needed.
* improved context detection of template parameters.
* fixed case where declaration and definition of template function has 2 nodes in graph.
* removed code duplications: replaced individual cache classes in ASTVisitor with template based
* unused member in undoredocontroller
* int - size_t comparion
* virtual destructor ASTVisitor added
* braces when && and || in same if
* 1 warning left: isNamedDeclUnnamed in ASTVisitor is unused
* increased performance. coati is now 1.55 times as fast as at version 0.4
* integrated sourceweb parsing code
* removed astbodyvisitor
* NameHierarchy now store signatures in a new format
* storage does not allow to re-set the type of a node, once it is marked as "defined".
* made onTemplateRecordSpecParsed just create the edge, not the node.
* removed distinction between template specialitzations for functions and classes
* when generating the name hierarchy of a template parameter the parameter name is now an extra element that follows the template class/struct/function element.
* Include paths are converted to a canonical version before they are tested for being in the project files. Therefore non-canonical include edges will be stored by Coati as well.
* re-added gui for c project setup
* ASTVisitor now visits RecordDecls instead of CXXRecordDecls
* prevented endless recursion when parsing struct that is defined in functions signature.
* improved the way of gettig a TemplateSpecializationTypeLoc from any TypeLoc to parse template arguments.
* fixed wrong forward mouse button code in app settings.
* functions will only display their names in the graph view. hovering the nodes shows the signature as a tooltip. However, nodes in the database will be created and used regarding their signature.
* improved lambda handling. All calls to generated lambda functions (constructor, call operator) will now point to the lambda class instead.
* implemented handling of the "auto" type.
* improved the CxxParserTestSuite to use function strings instead of function names for template related functions.
* reverted to old source location retrieval for CxxConstructExpr, because source locations were outside file
* removed exisiting source location check for better performance
* fixed CXXParserTestSuite to use c++11 standard
* fixed crash on missing signature retrieval
* implemented handling lambdas as functions
* functions can have child nodes now
* NameElements in NameHierarchy can have signatures now.
* fixed crash that occurred when message listener was used while being half way through its destruction process.
malte_langkabel