dot: fix parsing of IDs

mtf90 · mtf90 · commit 407dc29eb485 · 2025-05-15T10:27:04.000+02:00
closes #89
diff --git a/serialization/dot/src/main/javacc/DOT.jj b/serialization/dot/src/main/javacc/DOT.jj
@@ -186,15 +186,16 @@ TOKEN:
 |	< COMMA: "," >
 |	< EQUALS: "=" >
 |	< EDGEOP: "--" | "->" >
-|	< ID: <NUMERAL> | <LETTER> (<LETTER> | <NUMBER>)*>
-|	< LETTER: ["a"-"z","A"-"Z","\200"-"\377","_"] >
-|	< NUMBER: ["0"-"9"] >
-|	< NUMERAL: ["-"]( ("."(<NUMBER>)+) | (<NUMBER>)+ ("." (<NUMBER>)+)?)>
+|	< ALPHABETIC : <LETTER> (<LETTER> | <NUMBER>)*>
+|	< NUMERAL: ("-")?( ("."(<NUMBER>)+) | (<NUMBER>)+ ("." (<NUMBER>)+)?)>
+|	< #LETTER: ["a"-"z","A"-"Z","\200"-"\377","_"] >
+|	< #NUMBER: ["0"-"9"] >
 }
 
 SKIP:
 {
 	< BEGIN_QID: "\"" > : IN_QID
+|	< BEGIN_HTML: "<" > : IN_HTML
 }
 
 <IN_QID> TOKEN:
@@ -207,6 +208,18 @@ SKIP:
 	< END_QID: "\"" > : DEFAULT
 }
 
+<IN_HTML> TOKEN:
+{
+	// we don't check for well-matchedness of tags since we are only interested in finding the final >
+	<HTML : ("<" (<CONTENT>)+ ">" | (<CONTENT>)+ )+ >
+|	<#CONTENT : ~["<", ">"]>
+}
+
+<IN_HTML> SKIP:
+{
+	<END_HTML : ">" > : DEFAULT
+}
+
 SKIP:
 {
 	< BEGIN_LINE_COMMENT1: "//" > : IN_LINE_COMMENT
@@ -397,13 +410,15 @@ private void compass_pt():
 	"n" | "ne" | "e" | "se" | "s" | "sw" | "w" | "nw" | "c" | "_"
 }
 
-
 private String identifier():
 {
 	Token t;
 }
 {
-	t=<ID> { return t.toString(); }
-|	"\"\"" { return ""; }
+	t=<ALPHABETIC> { return t.toString(); }
+|	t=<NUMERAL> { return t.toString(); }
 |	t=<QID> { return StringUtil.unescapeQuotes(t.toString()); }
+|	t=<HTML> { return t.toString(); }
+|	"\"\"" { return ""; }
+|	"<>" { return ""; }
 }
diff --git a/serialization/dot/src/test/java/net/automatalib/serialization/dot/DOTDeserializationTest.java b/serialization/dot/src/test/java/net/automatalib/serialization/dot/DOTDeserializationTest.java
@@ -25,6 +25,7 @@
 import java.util.Comparator;
 import java.util.HashSet;
 import java.util.List;
+import java.util.Objects;
 import java.util.Queue;
 import java.util.Set;
 
@@ -189,6 +190,15 @@ public void testDuplicateTransitions() throws IOException, FormatException {
         checkIsomorphism(dfa, parsed, dfa.getInputAlphabet());
     }
 
+    @Test
+    public void testIDs() throws IOException, FormatException {
+        final DFA<?, String> parsed =
+                DOTParsers.dfa().readModel(DOTSerializationUtil.getResource(DOTSerializationUtil.ID_RESOURCE)).model;
+
+        Assert.assertEquals(parsed.size(), 6);
+        Assert.assertTrue(parsed.accepts(Word.fromString("abcd").transform(Objects::toString)));
+    }
+
     private static <S1, S2, I, T1, T2, SP, TP> void checkIsomorphism(UniversalAutomaton<S1, I, T1, SP, TP> source,
                                                                      UniversalAutomaton<S2, I, T2, SP, TP> target,
                                                                      Alphabet<I> alphabet) {
diff --git a/serialization/dot/src/test/java/net/automatalib/serialization/dot/DOTSerializationUtil.java b/serialization/dot/src/test/java/net/automatalib/serialization/dot/DOTSerializationUtil.java
@@ -56,6 +56,7 @@
 
 final class DOTSerializationUtil {
 
+    static final String ID_RESOURCE = "/id.dot";
     static final String EMPTY_RESOURCE = "/empty.dot";
     static final String DFA_RESOURCE = "/dfa.dot";
     static final String NFA_RESOURCE = "/nfa.dot";
diff --git a/serialization/dot/src/test/resources/id.dot b/serialization/dot/src/test/resources/id.dot
@@ -0,0 +1,19 @@
+digraph g {
+
+	<<TABLE><TR><TD BGCOLOR="yellow">class</TD></TR><TR><TD BGCOLOR="lightblue">qualifier</TD></TR></TABLE>> [shape="doublecircle"];
+	13 [shape="circle"];
+	-5.13 [shape="circle"];
+	"id with \" quotes" [shape="circle"];
+	aLpHa_B3T1C [shape="doublecircle"];
+	<> [label="empty"];
+	"" [label="empty2"]; // same id as <>, so redundant!
+
+	13 -> -5.13 [label="a"];
+	-5.13 -> "id with \" quotes" [label="b"];
+	"id with \" quotes" -> aLpHa_B3T1C [label="c"];
+	aLpHa_B3T1C -> <<TABLE><TR><TD BGCOLOR="yellow">class</TD></TR><TR><TD BGCOLOR="lightblue">qualifier</TD></TR></TABLE>> [label="d"];
+
+__start0 [label="" shape="none" width="0" height="0"];
+__start0 -> 13;
+
+}

Original file line number	Diff line number	Diff line change
`@@ -186,15 +186,16 @@ TOKEN:`
`186`	`186`	`\| < COMMA: "," >`
`187`	`187`	`\| < EQUALS: "=" >`
`188`	`188`	`\| < EDGEOP: "--" \| "->" >`
`189`		`-\| < ID: <NUMERAL> \| <LETTER> (<LETTER> \| <NUMBER>)*>`
`190`		`-\| < LETTER: ["a"-"z","A"-"Z","\200"-"\377","_"] >`
`191`		`-\| < NUMBER: ["0"-"9"] >`
`192`		`-\| < NUMERAL: ["-"]( ("."(<NUMBER>)+) \| (<NUMBER>)+ ("." (<NUMBER>)+)?)>`
	`189`	`+\| < ALPHABETIC : <LETTER> (<LETTER> \| <NUMBER>)*>`
	`190`	`+\| < NUMERAL: ("-")?( ("."(<NUMBER>)+) \| (<NUMBER>)+ ("." (<NUMBER>)+)?)>`
	`191`	`+\| < #LETTER: ["a"-"z","A"-"Z","\200"-"\377","_"] >`
	`192`	`+\| < #NUMBER: ["0"-"9"] >`
`193`	`193`	`}`
`194`	`194`
`195`	`195`	`SKIP:`
`196`	`196`	`{`
`197`	`197`	`< BEGIN_QID: "\"" > : IN_QID`
	`198`	`+\| < BEGIN_HTML: "<" > : IN_HTML`
`198`	`199`	`}`
`199`	`200`
`200`	`201`	`<IN_QID> TOKEN:`
`@@ -207,6 +208,18 @@ SKIP:`
`207`	`208`	`< END_QID: "\"" > : DEFAULT`
`208`	`209`	`}`
`209`	`210`
	`211`	`+<IN_HTML> TOKEN:`
	`212`	`+{`
	`213`	`+ // we don't check for well-matchedness of tags since we are only interested in finding the final >`
	`214`	`+ <HTML : ("<" (<CONTENT>)+ ">" \| (<CONTENT>)+ )+ >`
	`215`	`+\| <#CONTENT : ~["<", ">"]>`
	`216`	`+}`
	`217`	`+`
	`218`	`+<IN_HTML> SKIP:`
	`219`	`+{`
	`220`	`+ <END_HTML : ">" > : DEFAULT`
	`221`	`+}`
	`222`	`+`
`210`	`223`	`SKIP:`
`211`	`224`	`{`
`212`	`225`	`< BEGIN_LINE_COMMENT1: "//" > : IN_LINE_COMMENT`
`@@ -397,13 +410,15 @@ private void compass_pt():`
`397`	`410`	`"n" \| "ne" \| "e" \| "se" \| "s" \| "sw" \| "w" \| "nw" \| "c" \| "_"`
`398`	`411`	`}`
`399`	`412`
`400`		`-`
`401`	`413`	`private String identifier():`
`402`	`414`	`{`
`403`	`415`	`Token t;`
`404`	`416`	`}`
`405`	`417`	`{`
`406`		`- t=<ID> { return t.toString(); }`
`407`		`-\| "\"\"" { return ""; }`
	`418`	`+ t=<ALPHABETIC> { return t.toString(); }`
	`419`	`+\| t=<NUMERAL> { return t.toString(); }`
`408`	`420`	`\| t=<QID> { return StringUtil.unescapeQuotes(t.toString()); }`
	`421`	`+\| t=<HTML> { return t.toString(); }`
	`422`	`+\| "\"\"" { return ""; }`
	`423`	`+\| "<>" { return ""; }`
`409`	`424`	`}`