Fixes #2088. Allow partial but not actual esoteric interpolation sequ…

…ences to be regexp values
jruby · Nov 3, 2014 · 8801d55 · 8801d55
1 parent 78999e2
commit 8801d55
Show file tree

Hide file tree

Showing 2 changed files with 86 additions and 15 deletions.
diff --git a/core/src/main/java/org/jruby/lexer/yacc/RubyYaccLexer.java b/core/src/main/java/org/jruby/lexer/yacc/RubyYaccLexer.java
@@ -1596,7 +1596,20 @@ private int dollar() throws IOException {
             return identifierToken(Tokens.tGVAR, tokenBuffer.toString().intern());
         }
     }
-
+
+    // FIXME: I added number gvars here and they did not.
+    public boolean isGlobalCharPunct(int c) {
+        switch (c) {
+            case '_': case '~': case '*': case '$': case '?': case '!': case '@':
+            case '/': case '\\': case ';': case ',': case '.': case '=': case ':':
+            case '<': case '>': case '\"': case '-': case '&': case '`': case '\'':
+            case '+': case '1': case '2': case '3': case '4': case '5': case '6':
+            case '7': case '8': case '9': case '0':
+                return true;
+        }
+        return isIdentifierChar(c);
+    }
+
     private int dot() throws IOException {
         int c;
 

diff --git a/core/src/main/java/org/jruby/lexer/yacc/StringTerm.java b/core/src/main/java/org/jruby/lexer/yacc/StringTerm.java
@@ -81,6 +81,74 @@ private int endFound(RubyYaccLexer lexer, LexerSource src) throws IOException {
             return Tokens.tSTRING_END;
     }
 
+    // Return of 0 means failed to find anything.  Non-zero means return that from lexer.
+    private int parsePeekVariableName(RubyYaccLexer lexer, LexerSource src) throws IOException {
+        int c = src.read(); // byte right after #
+        int significant = -1;
+        switch (c) {
+            case '$': {  // we unread back to before the $ so next lex can read $foo
+                int c2 = src.read();
+
+                if (c2 == '-') {
+                    int c3 = src.read();
+
+                    if (c3 == RubyYaccLexer.EOF) {
+                        src.unread(c3); src.unread(c2);
+                        return 0;
+                    }
+
+                    significant = c3;                              // $-0 potentially
+                    src.unread(c3); src.unread(c2);
+                    break;
+                } else if (lexer.isGlobalCharPunct(c2)) {          // $_ potentially
+                    src.unread(c2); src.unread(c);
+                    lexer.setValue(new Token("#" + c2, lexer.getPosition()));
+                    return Tokens.tSTRING_DVAR;
+                }
+
+                significant = c2;                                  // $FOO potentially
+                src.unread(c2);
+                break;
+            }
+            case '@': {  // we unread back to before the @ so next lex can read @foo
+                int c2 = src.read();
+
+                if (c2 == '@') {
+                    int c3 = src.read();
+
+                    if (c3 == RubyYaccLexer.EOF) {
+                        src.unread(c3); src.unread(c2);
+                        return 0;
+                    }
+
+                    significant = c3;                                // #@@foo potentially
+                    src.unread(c3); src.unread(c2);
+                    break;
+                }
+
+                significant = c2;                                    // #@foo potentially
+                src.unread(c2);
+                break;
+            }
+            case '{':
+                lexer.setValue(new Token("#" + c, lexer.getPosition()));
+                return Tokens.tSTRING_DBEG;
+            default:
+                // We did not find significant char after # so push it back to
+                // be processed as an ordinary string.
+                src.unread(c);
+                return 0;
+        }
+
+        if (significant != -1 && Character.isAlphabetic(significant) || significant == '_') {
+            src.unread(c);
+            lexer.setValue(new Token("#" + significant, lexer.getPosition()));
+            return Tokens.tSTRING_DVAR;
+        }
+
+        return 0;
+    }
+
     public int parseString(RubyYaccLexer lexer, LexerSource src) throws IOException {
         boolean spaceSeen = false;
         int c;
@@ -105,22 +173,12 @@ public int parseString(RubyYaccLexer lexer, LexerSource src) throws IOException
             lexer.getPosition();
             return ' ';
         }
-
-        ByteList buffer = createByteList(lexer);
 
+        ByteList buffer = createByteList(lexer);
         if ((flags & RubyYaccLexer.STR_FUNC_EXPAND) != 0 && c == '#') {
-            c = src.read();
-            switch (c) {
-            case '$':
-            case '@':
-                src.unread(c);
-                lexer.setValue(new Token("#" + c, lexer.getPosition()));
-                return Tokens.tSTRING_DVAR;
-            case '{':
-                lexer.setValue(new Token("#" + c, lexer.getPosition())); 
-                return Tokens.tSTRING_DBEG;
-            }
-            buffer.append((byte) '#');
+            int token = parsePeekVariableName(lexer, src);
+
+            if (token != 0) return token;
         }
         src.unread(c);