Python: Move constraints into pranch charpreds

For sequences and alternations, we require at least one child.
Otherwise, we wish to represent the term differently.
This avoids multiple representations.

Author: yoff

python/ql/src/semmle/python/RegexTreeView.qll

@@ -7,16 +7,29 @@ private import semmle.python.regex
  * An element containing a regular expression term, that is, either
  * a string literal (parsed as a regular expression)
  * or another regular expression term.
+ *
+ * For sequences and alternations, we require at least one child.
+ * Otherwise, we wish to represent the term differently.
+ * This avoids multiple representations of the same term.
  */
 newtype TRegExpParent =
   /** A string literal used as a regular expression */
   TRegExpLiteral(Regex re) or
   /** A quantified term */
   TRegExpQuantifier(Regex re, int start, int end) { re.qualifiedItem(start, end, _, _) } or
   /** A sequence term */
-  TRegExpSequence(Regex re, int start, int end) { re.sequence(start, end) } or
+  TRegExpSequence(Regex re, int start, int end) {
+    re.sequence(start, end) and
+    exists(seqChild(re, start, end, 1)) // if a sequence does not have more than one element, it should be treated as that element instead.
+  } or
   /** An alternatio term */
-  TRegExpAlt(Regex re, int start, int end) { re.alternation(start, end) } or
+  TRegExpAlt(Regex re, int start, int end) {
+    re.alternation(start, end) and
+    exists(int part_end |
+      re.alternationOption(start, end, start, part_end) and
+      part_end < end
+    ) // if an alternation does not have more than one element, it should be treated as that element instead.
+  } or
   /** A character class term */
   TRegExpCharacterClass(Regex re, int start, int end) { re.charSet(start, end) } or
   /** A character range term */
@@ -75,11 +88,7 @@ class RegExpTerm extends RegExpParent {
   int end;
 
   RegExpTerm() {
-    this = TRegExpAlt(re, start, end) and
-    exists(int part_end |
-      re.alternationOption(start, end, start, part_end) and
-      part_end < end
-    ) // if an alternation does not have more than one element, it should be treated as that element instead.
+    this = TRegExpAlt(re, start, end)
     or
     this = TRegExpBackRef(re, start, end)
     or
@@ -93,8 +102,7 @@ class RegExpTerm extends RegExpParent {
     or
     this = TRegExpQuantifier(re, start, end)
     or
-    this = TRegExpSequence(re, start, end) and
-    exists(seqChild(re, start, end, 1)) // if a sequence does not have more than one element, it should be treated as that element instead.
+    this = TRegExpSequence(re, start, end)
     or
     this = TRegExpSpecialChar(re, start, end)
   }
@@ -341,10 +349,7 @@ class RegExpRange extends RegExpQuantifier {
  * This is a sequence with the elements `(ECMA|Java)` and `Script`.
  */
 class RegExpSequence extends RegExpTerm, TRegExpSequence {
-  RegExpSequence() {
-    this = TRegExpSequence(re, start, end) and
-    exists(seqChild(re, start, end, 1)) // if a sequence does not have more than one element, it should be treated as that element instead.
-  }
+  RegExpSequence() { this = TRegExpSequence(re, start, end) }
 
   override RegExpTerm getChild(int i) { result = seqChild(re, start, end, i) }