flatgraph

https://flatgraph.joern.io/
Admin
flatgraph is a fast and memory-efficient columnar graph database with a framework to generate...
Tokens:8,006
Snippets:51
Trust Score:8.6
Update:1 month ago
Show doc for...
Tokens:
Raw
### Create and query a GratefulDead graph

Source: https://flatgraph.joern.io/index.html

Uses the generated domain classes to create an empty graph, build a diff graph with two artists and one song, add edges, apply the diff, and run type-safe queries. Demonstrates that nodes referenced in edges do not need separate addNode calls.

```scala
val gratefulDead = GratefulDead.empty

// create a simple graph with two nodes and two edges
val diffGraph    = GratefulDead.newDiffGraphBuilder
val garcia       = NewArtist().name("Garcia")
val boDiddley    = NewArtist().name("Bo_Diddley")
val heyBoDiddley = NewSong().name("HEY BO DIDDLEY").songtype("cover").performances(5)
// n.b. if nodes are referenced in edges, they don't need to be added separately via `diffGraph.addNode`
diffGraph.addEdge(src = heyBoDiddley, dst = boDiddley, Writtenby.Label)
diffGraph.addEdge(src = heyBoDiddley, dst = garcia, Sungby.Label)
DiffGraphApplier.applyDiff(gratefulDead.graph, diffGraph)

// now query it in the typesafe domain-specific query language
gratefulDead.artist.name.sorted shouldBe List("Bo_Diddley", "Garcia")
gratefulDead.artist.name("Garcia").sang.name.l shouldBe List("HEY BO DIDDLEY")
gratefulDead.song.writtenBy.name.l shouldBe List("Bo_Diddley")
```

--------------------------------

### Define GratefulDead schema with SchemaBuilder

Source: https://flatgraph.joern.io/index.html

Defines the GratefulDead domain schema: properties 'name' and 'songType', node types 'artist' and 'song', and an edge type 'sungBy' with cardinality One. The schema is used by the code generator to produce domain-specific classes.

```scala
val builder  = new SchemaBuilder("GratefulDead", basePackage = "testdomains.gratefuldead")
val name     = builder.addProperty("name", ValueType.String).mandatory(default = "")
val songType = builder.addProperty("songType", ValueType.String)
val artist   = builder.addNodeType("artist").addProperty(name)
val song     = builder.addNodeType("song").addProperty(name).addProperty(songType)
val sungBy   = builder.addEdgeType("sungBy")
song.addOutEdge(
  sungBy, 
  inNode = artist, 
  cardinalityOut = Cardinality.One, 
  stepNameOut = "sungBy", 
  stepNameIn = "sang")
```

--------------------------------

### Map Step for Property Extraction

Source: https://flatgraph.joern.io/traversals/index.html

Uses the map step to extract the IS_EXTERNAL and NAME properties from METHOD nodes in the Code Property Graph.

```scala
joern> cpg.method.map(node => (node.isExternal, node.name)).toList
res6: List[(Boolean, String)] = List(
  (false, "main"),
  (true, "fprintf"),
  (true, "exit"),
  (true, "<operator>.logicalAnd"),
  (true, "<operator>.equals"),
  (true, "<operator>.greaterThan"),
  (true, "strcmp"),
  (true, "<operator>.indirectIndexAccess"),
  (true, "printf")
)
```

--------------------------------

### Filter METHOD nodes by IS_EXTERNAL property

Source: https://flatgraph.joern.io/traversals/index.html

These three queries return the names of METHOD nodes with IS_EXTERNAL set to false. The first uses the filter step with a lambda predicate; the second uses the property-filter step isExternal(false); the third rewrites the same query using the where step with a traversal predicate. All produce the same result: List("main").

```scala
joern> cpg.method.filter(_.isExternal == false).name.toList 
res11: List[String] = List("main")
```

```scala
joern> cpg.method.isExternal(false).name.toList 
res11: List[String] = List("main")
```

```scala
joern> cpg.method.where(_.isExternal(false)).name.toList 
res24: List[String] = List("main")
```
Tokens:
Raw