Skip to content

Graph Markup Language

  • Create graphs using configuration files
  • Define a regular expression that represents the graph markup language
  • Allow external database tools to ingest graph configuration files

Adjacency List Markup (Labels Only)

"NODE_1": ("<NODE_1>", "<NODE_1>")
"NODE_1": ("<NODE_1>", "<NODE_1>", "<NODE_2>")
"NODE_1": ("<NODE_1>", "<NODE_1>")
"NODE_1": ("<NODE_1>", "<NODE_1>")
  • This would be a config if you want to sort by label name
  • Should except multi line comments with this syntax: /* commment */
  • Regular expression contains the following for each line
    • A node is defined by having it's label followed by a colon
    • List of neighbors are surrounded by parenthesis
    • Each element is surrounded by quotations
    • Each element is separated by commas 
      [ID, "NODE_2"]: ([ID, "<NODE_1>"], [ID, "<NODE_1>"])
"NODE_1": ("<NODE_1>", "<NODE_1>", "<NODE_2>")
"NODE_1": ("<NODE_1>", "<NODE_1>")
"NODE_1": ("<NODE_1>", "<NODE_1>")

Matrix Markup

rows: <NUMBER>
cols: <NUMBER>

<INT>,<INT>,<INT>
<INT>,<INT>,<INT>
<INT>,<INT>,<INT>
  • Two regular expressions will be defined since not all lines are the same
  • The first regular expression will follow this pattern
    • The rows/cols variable will be defined and have a colon after it
    • The next line can take in any digit except 0
  • The second regular expression is for the matrix values
    • Each line is separated by comma values
    • Each element must be a series of digits

Adjacency List Markup

[0,"B"]: ([1,"A",0],[2,"C",0])
[1,"A"]: ([0,"B",0],[2,"C",0],[3,"D",0])
[2,"C"]: ([1,"A",0],[0,"B",0],[4,"E",0])
[3,"D"]: ([1,"A",0])
[4,"E"]: ([2,"C",0])
  • Weights and id's of nodes are now include in config file
  • id is included so calling methods to get unique nodes is not needed
  • Regular expression is defined with the following criteria
  • Extract integer values or any thing with quotes in between it