fix location off-by-ones with regexp parsing

Author: erik-krogh

javascript/extractor/src/com/semmle/js/extractor/ASTExtractor.java

@@ -589,7 +589,7 @@ public Label visit(Literal nd, Context c) {
 
       trapwriter.addTuple("literals", valueString, source, key);
       Position start = nd.getLoc().getStart();
-      com.semmle.util.locations.Position startPos = new com.semmle.util.locations.Position(start.getLine(), start.getColumn(), start.getOffset());
+      com.semmle.util.locations.Position startPos = new com.semmle.util.locations.Position(start.getLine(), start.getColumn() + 1 /* Convert from 0-based to 1-based. */, start.getOffset());
 
       if (nd.isRegExp()) {
         OffsetTranslation offsets = new OffsetTranslation();
@@ -867,7 +867,7 @@ private void extractRegxpFromBinop(BinaryExpression nd, Context c) {
       }
       OffsetTranslation offsets = concatResult.snd();
       Position start = nd.getLoc().getStart();
-      com.semmle.util.locations.Position startPos = new com.semmle.util.locations.Position(start.getLine(), start.getColumn(), start.getOffset());
+      com.semmle.util.locations.Position startPos = new com.semmle.util.locations.Position(start.getLine(), start.getColumn() + 1 /* Convert from 0-based to 1-based. */, start.getOffset());
       SourceMap sourceMap = SourceMap.legacyWithStartPos(SourceMap.fromString(nd.getLoc().getSource()).offsetBy(0, offsets), startPos);
       regexpExtractor.extract(foldedString, sourceMap, nd, true);
       return;

javascript/extractor/src/com/semmle/js/extractor/RegExpExtractor.java

@@ -121,10 +121,12 @@ private Label extractTerm(RegExpTerm term, Label parent, int idx) {
   }
 
   public void emitLocation(SourceElement term, Label lbl) {
-    int sl = sourceMap.getStart(term.getLoc().getStart().getColumn()).getLine();
-    int sc = sourceMap.getStart(term.getLoc().getStart().getColumn()).getColumn() + 1; // convert to 1-based
-    int el = sourceMap.getEnd(term.getLoc().getEnd().getColumn()).getLine();
-    int ec = sourceMap.getEnd(term.getLoc().getEnd().getColumn()).getColumn() - 1; // convert to inclusive
+    int start = term.getLoc().getStart().getColumn();
+    int sl = sourceMap.getStart(start).getLine();
+    int sc = sourceMap.getStart(start).getColumn();
+    int end = term.getLoc().getEnd().getColumn();
+    int el = sourceMap.getStart(end).getLine();
+    int ec = sourceMap.getStart(end).getColumn() - 1; // convert to inclusive
     locationManager.emitSnippetLocation(lbl, sl, sc, el, ec);
   }
 

javascript/extractor/tests/regexp/input/multipart.js

@@ -11,4 +11,21 @@ var reg4 = new RegExp(
     "bar" + 
     "baz" + 
     "qux"
-);
+);
+
+var bad95 = new RegExp(
+    "(a" + 
+    "|" + 
+    "aa)*" + 
+    "b$"
+);
+
+var bad96 = new RegExp("(" + 
+    "(c|cc)*|" + 
+    "(d|dd)*|" +
+    "(e|ee)*" +
+")f$");
+
+var bad97 = new RegExp(
+    "(g|gg" + 
+    ")*h$");