Re: Java compiler development tools

Scott Stanchfield <thetick@scruz.net>
18 Jul 1997 12:33:25 -0400

          From comp.compilers

Related articles
Java compiler development tools popx@ermine.ox.ac.uk (Jocelyn Paine) (1997-07-13)
Re: Java compiler development tools sreeni@csc.albany.edu (1997-07-16)
Re: Java compiler development tools thetick@scruz.net (Scott Stanchfield) (1997-07-16)
Re: Java compiler development tools gruban@bnl.gov (1997-07-16)
Re: Java compiler development tools thetick@scruz.net (Scott Stanchfield) (1997-07-18)
Re: Java compiler development tools johnm@non.net (1997-07-21)
| List of all articles for this month |

From: Scott Stanchfield <thetick@scruz.net>
Newsgroups: comp.compilers,comp.lang.java.programmer
Date: 18 Jul 1997 12:33:25 -0400
Organization: scruz-net
References: 97-07-068 97-07-083
Keywords: Java, parse, tools

gruban@bnl.gov wrote:
>
> JavaCC, from http://www.suntest.com/JavaCC/
> has the advantage that it is developed by a branch of Sun, and so is
> slightly less likely than most other free tools to go the way of the
> diplodocus and the dodo. The developers are also very friendly and
> responsive.
Hardly a fair statement -- just because a big company is behind it now
doesn't mean the product could not get dumped and completely vanish.


The advantage in this light to something like ANTLR 2.0 is that it's
completely public domain, _source_code_and_all. ANTLR (formerly PCCTS)
has been around for quite some time and is going strong. (Read this as
"Terence Parr has been working on language analysis tools for many more
years than JavaCC has existed." There's something to be said for
experience.) Even if Ter decides to stop support anyone else could pick
it up. (I'd do it in a flash...)


> Other tools are also mentioned in the FAQ:
> http://www.best.com/~pvdl
> but have the classic problem of being supported only when the
> developer gets "a round tuit" - so common to most free tools.
Again, after initial development, a tool from a large company may take
quite some time to incorporate requested changes. (Especially a tool
that company provides free.) Sun is much better than M$ is providing
timely releases (and IMHO they usually produce higher-quality stuff) but
there's something to be said for lack of beauracracy wrt speed of
getting a product out the door. Not to mention having source code so
anyone can suggest a solution to a bug.




Take a look at both tools and see what you think -- they're both nice
tools (I prefer ANTLR) but judge them on the value of the tools, not on
speculation about "companies vs. individuals."


-- Scott


--
Scott Stanchfield
Santa Cruz, CA
--


Post a followup to this message

Return to the comp.compilers page.
Search the comp.compilers archives again.