* added access node to graph for template parameters, no icon so far
* don't reexpand function nodes with children when active before
* removed console warnings for edges
* replaced storage by a new version that utilizes sqlite for persistence.
* added SqliteStorage to manage direct access to the database.
* added testsuit for sqlite.
* slimmed down the protected interface of the storage.
* changed tests in StorageTestSuit to avoid using protected methods of the storage.
* added functions to begin and commit a transaction to the storage.
* removed GraphFilterTestSuite.
* removed GraphFilterConductorTestSuite.
* added Node and Edge constructor that takes id as parameter
* added constructor for TokenComponentNameCached that accepts NameHierarchy as parameter
* added TokenLocation constructor that takes locationId as parameter.
* created respective construction functions in TokenLocatonLine, -File and -Collection.
fortune cookie message = Recognition will come from unexpected sources.
This change bundles similar nodes in the graph view together. When the bundled node or it's aggregation edge are clicked,
then the bundle is split. This change also adds an arrow to the aggregation edge based on the direction of it's edges.
* fixed: implicit template specializations will be found even if the original template definition is located in a different file (that gets included).
* implemented correct extraction of template parameters, template arguments and template specialization edges.
* added token locations for template parameters of partial template specializations.
* graph view displays full names of template parameters and template arguments.
* added test code for features
This change adds the creation of aggregation edges to the StorageGraph, which save the connection counts between
children in their parents. Aggregation edges are displayed using a straight line, that is thicker than more connections
it represents with the connection count in the middle.
* Implemented basic parsing and storage of template parameter types, template classes and template functions
* Added template specialization edge.
* Implemented handling of partial specializations. Needs some more tweaking.
* Added a lot of test code for template stuff (still needs some cases to be tested)
* Removed the isTemplateParameterType from DataType, since it shouldn't be needed
* Function qualTypeToDataType uses clang PrintingPolicy to fetch the desired format.
* Merged a lot of code for getting the correct names and name hierarchies of definitions
* Removed TokenComponentDataType
* Merged both of the Storage::addTypeEdge(...) functions
* Added "-fno-delayed-template-parsing" flag to parser for parsing test code.
* fitting boxes to name length
* highlighting active edges and nodes
* hover for edges and nodes
* single click selection for edges and nodes
* edges have transparent second line for easier clicking
* sub structure class members by public/protected/private
This change only shows matches in the autocompletion list that are prefiltered by the already entered query. The
autocompletion now also shows up after an operator of type . or > has been entered to show an alphabetical list of
possible tokens.
This change stores each Token name in the separate singleton class Dictionary. The Dictionary gives each saved word an
Id and thereby avoids duplicated names. E.g if the constructor method "Graph::Graph" is stored then the word "Graph"
only appears once in memory.
The class SearchIndex is now responsible for the name hierarchy and is instantiated by the Storage. The SearchIndex
builds the name hierarchy using SearchNodes, each holding a Dictionary string reference of the name it holds. E.g if the
names "math::ceil" and "math::floor" are added to the SearchIndex then 3 nodes get created, the SearchNode "math" will
hold the two childs "ceil" and "floor".
The hierarchical graph creation functionality got split off from Graph into the new subclass StorageGraph. The
StorageGraph creates nodes with a passed SearchNode pointer of the name it represents in the SearchIndex. Thereby the
StorageGraph reuses the hierarchical information in the SearchIndex and can create nodes much quicker by avoiding node
searches and name comparisions.
The name information is now stored in the Nodes via the TokenComponentName class, which is subclassed into
TokenComponentNameReferenced and TokenComponentNameCached. The StorageClass creates nodes with the component
TokenComponentNameReferenced, which holds a pointer to the SearchNode instance holding the name. This allows for
retrieving the full name of the node, without using other Nodes int the Graph, which might not be present. If the Node
is copied then the component changes to a TokenComponentNameCached, which holds the full name as a string, so the
memory in the Storage doesn't have to be accessed anymore.
TokenComponentSignature is now only holding an Id of the signature string saved in the Dictionary, which speeds up the
signature comparison. A follow-up will change saving the whole signature as string to reusing the wordIds it is
consisting of.
Lastly the SearchIndex holds basic fuzzy search functionality. A passed query gets compared down the SearchNode
hierarchy as long as matches for each letter are found. Matches must contain all letters of the query. The search is
case-insensitive. If letters are found in front positions, next to each other or written in uppercase they are weighed
higher in the match ranking. The character ':' is also interpreted and found, although the '::' delimiter is not stored.
E.g. the query "m:l" used on the example above will return both "math::floor" and "math::ceil", but "floor" is ranked
higher because the 'l' appears closer to the start.
This change allows the user to use a simple query syntax in the search field to filter the results using names,
operators and predefined filters.
The class QueryTree is capable of parsing a simple query language consiting of QueryNodes, subclassed as
QueryOperators, QueryCommands and QueryTokens. QueryToken identifies a Token by name. QueryCommand represents predefined
filters. QueryOperator defines the syntactic relationship.
QueryNode examples:
"A" -> QueryToken that identifies the Token named A
class -> QueryCommand that filters all Nodes that are classes
. -> QueryOperator that concatenates filters
QueryExamples:
"A".class -> All Tokens named A that are classes
method.(private|protected) -> All methods that are private or protected
"A".member -> All members of A
"A":field.!const -> All fields of A that are not const (':' can be used instead of '.member')
QueryTree checks for correct operator precedence and validity of the query.
All QueryCommands can be found in data/query/QueryCommand.cpp
All QueryOperators can be found in data/query/QueryOperator.cpp
The parsed QueryTree is then passed to the class GraphFilterConductor, which is capable of applying the query on a
Graph. Each part of the query gets assigned a GraphFilter that filters the input Graph in order of node precedence. The
class SubGraph is used as intermediate container, only holding bare pointers to Nodes and Edges. The output Graph holds
all Nodes that match the query.
MessageActivateTokens is used to show all of the results in the CodeView. The GraphView currently only shows the first
Node of the results.
bug id = #6
This change moves the EdgeComponent system to Token and redefines them as TokenComponents.
* Performance tests on the side showed that using typeid() in getComponent() is faster than dynamically casting every
pointer.
* The hasComponent() method was deliberately left out in the implementation, because it used to look up the right
component and discard it again. This check for presence can also be achieved by just using getComponent().
* The method addComponent() is protected in Token, so that Node and Edge can guard which Component is set on which type
by implementing specific setters for each Component.
* The GraphTestSuite was extended to include tests for the Component implementation (and missing tests for locationIds
in Token and copying of Node and Edge were added).
Eberhard Graether