Created
March 12, 2014 16:13
-
-
Save Simn/9510238 to your computer and use it in GitHub Desktop.
LexerGraph for hxparse
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| import hxparse.Ruleset; | |
| import hxparse.State; | |
| import dot.Graph; | |
| import dot.Node; | |
| import dot.Attribute; | |
| using Lambda; | |
| class LexerGraph<T> { | |
| var graph:Graph; | |
| var ruleset:Ruleset<T>; | |
| var map:Map<State, Node>; | |
| public function new(ruleset:Ruleset<T>) { | |
| this.ruleset = ruleset; | |
| this.graph = new Graph(); | |
| map = new Map(); | |
| processState(ruleset.state); | |
| sys.io.File.saveContent("graph.dot", graph.getDotCode()); | |
| } | |
| function processState(state:State) { | |
| if (map.exists(state)) { | |
| return map[state]; | |
| } | |
| var attrs = [Label("")]; | |
| if (state.final > -1) { | |
| attrs.push(Shape(Doublecircle)); | |
| } | |
| var node = graph.node(attrs); | |
| map[state] = node; | |
| var targets = new Map(); | |
| for (i in 0...256) { | |
| if (state.trans[i] == null) { | |
| continue; | |
| } | |
| var target = state.trans[i]; | |
| if (!targets.exists(target)) { | |
| targets[target] = [i]; | |
| } else { | |
| targets[target].push(i); | |
| } | |
| } | |
| for (target in targets.keys()) { | |
| var il = targets[target]; | |
| var targetNode = processState(target); | |
| var edgeLabel = getRangeString(il); | |
| graph.edge(node, targetNode, [Label(edgeLabel)]); | |
| } | |
| return node; | |
| } | |
| function getRangeString(il:Array<Int>) { | |
| if (il.length > 240) { | |
| return "[^" + getRangeString(complementOf(il)) + "]"; | |
| } else if (il.length == 1) { | |
| return printCode(il[0]); | |
| } | |
| var ranges = []; | |
| var i = 0; | |
| var last = -1; | |
| var start = -1; | |
| function addRange() { | |
| if (start == last) { | |
| ranges.push(printCode(start)); | |
| } else { | |
| ranges.push(printCode(start) + "-" +printCode(last)); | |
| } | |
| } | |
| while (i < il.length) { | |
| var cur = il[i]; | |
| if (start == -1) { | |
| start = cur; | |
| ++i; | |
| } else if (cur != last + 1) { | |
| addRange(); | |
| start = -1; | |
| } else { | |
| ++i; | |
| } | |
| last = cur; | |
| } | |
| if (start != -1) { | |
| addRange(); | |
| } | |
| return ranges.join(" "); | |
| } | |
| function printCode(i:Int) { | |
| if (i >= 32) { | |
| if (i == '"'.code) { | |
| return '\\"'; | |
| } | |
| return String.fromCharCode(i); | |
| } else { | |
| return "\\" +i; | |
| } | |
| } | |
| function complementOf(il:Array<Int>) { | |
| var ret = []; | |
| for (i in 0...256) { | |
| if (!il.has(i)) { | |
| ret.push(i); | |
| } | |
| } | |
| return ret; | |
| } | |
| } |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment