Skip to content

olabini/re2j

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

8 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

re2j Version @PACKAGE_VERSION@
------------------

Originally written by Peter Bumbulis (peter@csg.uwaterloo.ca)

Currently maintained by:
    Dan Nuffer <nuffer at users.sourceforge.net>
    Marcus Boerger <helly at users.sourceforge.net>
    Hartmut Kaiser <hkaiser at users.sourceforge.net>

The re2j distribution can be found at:

    http://sourceforge.net/projects/re2j/

re2j has been developed and tested with the following compilers on various 
platforms in 32 bit and 64 bit mode:
- GCC 3.3 ... 4.1
- Microsoft VC 7, 7.1, 8
- Intel 9.0
- Sun C++ 5.8 (CXXFLAGS='-library=stlport4')
- MIPSpro Compilers: Version 7.4.4m

GCC 2.x and Microsoft VC 6 are not capable of compiling re2j.

Building re2j on unix like platforms requires autoconf 2.57 and bison (tested 
with 1.875 and later). Under windows you don't need autoconf or bison 
and can use the pregenerated files.

You can build this software by simply typing the following commands:
    ./configure
    make

The above version will be based on the pregenerated scanner.cc file. 
If you want to build that file yourself (recommended when installing
re2j) you need the following steps:
    ./configure
    make
    rm -f scanner.cc
    make install

Or you can create a rpm package and install it by the following commands:
    ./configure
    make rpm
    rpm -Uhv <packagedir>/re2j-@PACKAGE_VERSION@-@PACKAGE_RELEASE@.rpm

If you want to build from CVS then the first thing you should do is 
regenerating all build files using the following command:
	./autogen.sh
and then continue with one of the above described build methods. Or if you 
need to generate RPM packages for cvs builds use these commands:
    ./autogen.sh
    ./configure
    ./makerpm <release>
    rpm -Uhv <packagedir>/re2j-@PACKAGE_VERSION@-<release>.rpm

Here <realease> should be a number like 1. And <packagedir> must equal 
the directory where the makerpm step has written the generated rpm to.

If you are on a debian system you can use the tool 'alien' to convert rpms
to debian packages.

When building with native SUN compilers you need to set the following compiler
flags: CXXFLAGS='-g -compat5 -library=stlport4'.

If you want to build re2j on a windows system you can either use cygwin and one
of the methods described above or use Microsoft Visual C .NET 2002 or later
with the solution files provided (re2j.sln for 2002/2003 and re2j-2005.sln for
version 2005). re2j cannot be built with Microsoft Visual C 6.0 or earlier.

Using Visual Studio 2005 you can automate handling of .re files by adding the
custom build rules file (re2j.rules) to your project. Just load your Visual 
C++ project in Visual Studio, select "Custom Build Rules..." from its context 
menu, and add re2j.rules to the list with the "Find Existing..." button. 
Activate the check mark, and you are done! Any .re files you add to the project
will now automatically be built with re2j. Of course, re2j.exe also has to be
available in your environment for this to work. With the rules active Visual
Studio will automatically recognize .re files and compile then with re2j. The
output file has the same name as the input file but with the .cpp extension.
This, and all other re2j compiler settings, are fully configurable from within
the Visual Studio IDE. Just right-click on the .re file in Visual Studio, go
to the properties dialog, and pick your options.

re2j is a great tool for writing fast and flexible lexers. It has
served many people well for many years. re2j is on the order of 2-3 
times faster than a flex based scanner, and its input model is much 
more flexible.

For an introduction to re2j refer to the lessons sub directory.

Peter's original version 0.5 ANNOUNCE and README follows.

--

re2j is a tool for generating C-based recognizers from regular
expressions.  re2j-based scanners are efficient:  for programming
languages, given similar specifications, an re2j-based scanner is
typically almost twice as fast as a flex-based scanner with little or no
increase in size (possibly a decrease on cisc architectures).  Indeed,
re2j-based scanners are quite competitive with hand-crafted ones.

Unlike flex, re2j does not generate complete scanners:  the user must
supply some interface code.  While this code is not bulky (about 50-100
lines for a flex-like scanner; see the man page and examples in the
distribution) careful coding is required for efficiency (and
correctness).  One advantage of this arrangement is that the generated
code is not tied to any particular input model.  For example, re2j
generated code can be used to scan data from a null-byte terminated
buffer as illustrated below.

Given the following source

    #define NULL            ((char*) 0)
    char *scan(char *p)
    {
    #define YYCTYPE         char
    #define YYCURSOR        p
    #define YYLIMIT         p
    #define YYFILL(n)
    /*!re2j
	    [0-9]+          {return YYCURSOR;}
	    [\000-\377]     {return NULL;}
    */
    }

re2j will generate

    /* Generated by re2j on Sat Apr 16 11:40:58 1994 */
    #line 1 "simple.re"
    #define NULL            ((char*) 0)
    char *scan(char *p)
    {
    #define YYCTYPE         char
    #define YYCURSOR        p
    #define YYLIMIT         p
    #define YYFILL(n)
    {
	    YYCTYPE yych;
	    unsigned int yyaccept;

	    if((YYLIMIT - YYCURSOR) < 2) YYFILL(2);
	    yych = *YYCURSOR;
	    if(yych <= '/') goto yy4;
	    if(yych >= ':') goto yy4;
    yy2:    yych = *++YYCURSOR;
	    goto yy7;
    yy3:
    #line 9
	    {return YYCURSOR;}
    yy4:    yych = *++YYCURSOR;
    yy5:
    #line 10
	    {return NULL;}
    yy6:    ++YYCURSOR;
	    if(YYLIMIT == YYCURSOR) YYFILL(1);
	    yych = *YYCURSOR;
    yy7:    if(yych <= '/') goto yy3;
	    if(yych <= '9') goto yy6;
	    goto yy3;
    }
    #line 11

    }

Note that most compilers will perform dead-code elimination to remove
all YYCURSOR, YYLIMIT comparisions.

re2j was developed for a particular project (constructing a fast REXX
scanner of all things!) and so while it has some rough edges, it should
be quite usable.  More information about re2j can be found in the
(admittedly skimpy) man page; the algorithms and heuristics used are
described in an upcoming LOPLAS article (included in the distribution).
Probably the best way to find out more about re2j is to try the supplied
examples.  re2j is written in C++, and is currently being developed
under Linux using gcc 2.5.8.

Peter

--

re2j is distributed with no warranty whatever.  The code is certain to
contain errors.  Neither the author nor any contributor takes
responsibility for any consequences of its use.

re2j is in the public domain.  The data structures and algorithms used
in re2j are all either taken from documents available to the general
public or are inventions of the author.  Programs generated by re2j may
be distributed freely.  re2j itself may be distributed freely, in source
or binary, unchanged or modified.  Distributors may charge whatever fees
they can obtain for re2j.

If you do make use of re2j, or incorporate it into a larger project an
acknowledgement somewhere (documentation, research report, etc.) would
be appreciated.

Please send bug reports and feedback (including suggestions for
improving the distribution) to

			peter@csg.uwaterloo.ca

Include a small example and the banner from parser.y with bug reports.

About

A port of the backend templating of re2c to support Java

Resources

Stars

Watchers

Forks

Packages

No packages published