jcpp : C 预编译器的 java 实现

http://www.anarres.org/projects/jcpp/

JCPP - A Java C Preprocessor
Introduction

The C Preprocessor is an interesting standard. It appears to be derived from the de-facto behaviour of the first preprocessors, and has evolved over the years. Implementation is therefore difficult. JCPP is a complete, compliant, standalone, pure Java implementation of the C preprocessor. It is intended to be of use to people writing C-style compilers in Java using tools like sablecc, antlr, JLex, CUP and so forth. This project has has been used to successfully preprocess much of the source code of the GNU C library. As of version 1.2.5, it can also preprocess the Apple Objective C library.

The license is Apache-2.0. Versions prior to 1.2.2 were licensed under GPL-2.0, but on the basis of representations received, the license has now changed. Subversion access is available on request. Less restrictive licenses are freely available on request, but please ask, because I want people to use this software, but I'm curious to know who is using it and for what purpose.



=================================================
jcpp : C 预编译器的 java 实现。

=================================================

代码中使用了模板。通过 vpp 任务将模板生成源代码的。

JCpp\etc\targets\global-vpp.xml
        <vppcopy todir="${global.dir.build.java}">
            <fileset dir="${global.dir.src.java}">
<include name="**/*.java" />
            </fileset>

=================================================

通过 ant 编译后，build\java 目录下的代码是普通的 java 代码。

=================================================

程序运行后，可进行 c 语言预处理。如：
>#define aa 2
>aa

将输出 2。

=================================================
org/anarres/cpp/Main.java
if (g.getOptind() == args.length)
pp.addInput(new InputLexerSource(System.in));

通过这两行，将 System.in 做为输入。

=================================================
语法分析的代码在 LexerSource.java

	public Token token()
						throws IOException,
								LexerException {
		Token	tok = null;

		int		_l = line;
		int		_c = column;

		int		c = read();
		int		d;

		switch (c) {
			case '\n':
				//throw new RuntimeException("token() 582");
				//System.out.println("token() 582");
				if (ppvalid) {
					bol = true;
					if (include) {
						tok = new Token(NL, _l, _c, "\n");
					}
					else {
						int	nls = 0;
						do {
							nls++;
							d = read();
						} while (d == '\n');
						unread(d);
						char[]	text = new char[nls];
						for (int i = 0; i < text.length; i++)
							text[i] = '\n';
						// Skip the bol = false below.
						tok = new Token(NL, _l, _c, new String(text));
					}
					if (DEBUG)
						System.out.println("lx: Returning NL: " + tok);
					return tok;
				}
				/* Let it be handled as whitespace. */
				break;

			case '!':
				tok = cond('=', NE, '!');
				break;

			case '#':
				//System.out.println("token() 613");
				if (bol)
					tok = new Token(HASH);
				else
					tok = cond('#', PASTE, '#');
				break;

			case '+':
				d = read();
				if (d == '+')
					tok = new Token(INC);
				else if (d == '=')
					tok = new Token(PLUS_EQ);
				else
					unread(d);
				break;
			case '-':
				d = read();
				if (d == '-')
					tok = new Token(DEC);
				else if (d == '=')
					tok = new Token(SUB_EQ);
				else if (d == '>')
					tok = new Token(ARROW);
				else
					unread(d);
				break;

			case '*':
				tok = cond('=', MULT_EQ, '*');
				break;
			case '/':
				d = read();
				if (d == '*')
					tok = ccomment();
				else if (d == '/')
					tok = cppcomment();
				else if (d == '=')
					tok = new Token(DIV_EQ);
				else
					unread(d);
				break;

			case '%':
				d = read();
				if (d == '=')
					tok = new Token(MOD_EQ);
				else if (digraphs && d == '>')
					tok = new Token('}');	// digraph
				else if (digraphs && d == ':') PASTE: {
					d = read();
					if (d != '%') {
						unread(d);
						tok = new Token('#');	// digraph
						break PASTE;
					}
					d = read();
					if (d != ':') {
						unread(d);	// Unread 2 chars here.
						unread('%');
						tok = new Token('#');	// digraph
						break PASTE;
					}
					tok = new Token(PASTE);	// digraph
				}
				else
					unread(d);
				break;

			case ':':
				/* :: */
				d = read();
				if (digraphs && d == '>')
					tok = new Token(']');	// digraph
				else
					unread(d);
				break;

			case '<':
				if (include) {
					tok = string('<', '>');
				}
				else {
					d = read();
					if (d == '=')
						tok = new Token(LE);
					else if (d == '<')
						tok = cond('=', LSH_EQ, LSH);
					else if (digraphs && d == ':')
						tok = new Token('[');	// digraph
					else if (digraphs && d == '%')
						tok = new Token('{');	// digraph
					else
						unread(d);
				}
				break;

			case '=':
				tok = cond('=', EQ, '=');
				break;

			case '>':
				d = read();
				if (d == '=')
					tok = new Token(GE);
				else if (d == '>')
					tok = cond('=', RSH_EQ, RSH);
				else
					unread(d);
				break;

			case '^':
				tok = cond('=', XOR_EQ, '^');
				break;

			case '|':
				d = read();
				if (d == '=')
					tok = new Token(OR_EQ);
				else if (d == '|')
					tok = cond('=', LOR_EQ, LOR);
				else
					unread(d);
				break;
			case '&':
				d = read();
				if (d == '&')
					tok = cond('=', LAND_EQ, LAND);
				else if (d == '=')
					tok = new Token(AND_EQ);
				else
					unread(d);
				break;

			case '.':
				d = read();
				if (d == '.')
					tok = cond('.', ELLIPSIS, RANGE);
				else
					unread(d);
				/* XXX decimal fraction */
				break;

			case '0':
				/* octal or hex */
				d = read();
				if (d == 'x' || d == 'X')
					tok = number_hex((char)d);
				else {
					unread(d);
					tok = number_octal();
				}
				break;

			case '\'':
				tok = character();
				break;

			case '"':
				tok = string('"', '"');
				break;

			case -1:
				close();
				tok = new Token(EOF, _l, _c, "<eof>");
				break;
		}

		if (tok == null) {
			if (Character.isWhitespace(c)) {
				tok = whitespace(c);
			}
			else if (Character.isDigit(c)) {
				tok = number_decimal(c);
			}
			else if (Character.isJavaIdentifierStart(c)) {
				tok = identifier(c);
			}
			else {
				tok = new Token(c);
			}
		}

		if (bol) {
			switch (tok.getType()) {
				case WHITESPACE:
				case CCOMMENT:
					break;
				default:
					bol = false;
					break;
			}
		}

		tok.setLocation(_l, _c);
		if (DEBUG)
			System.out.println("lx: Returning " + tok);
		// (new Exception("here")).printStackTrace(System.out);
		return tok;
	}