Stanfordparser (or rjb) error on Snow Leopard

I’m having trouble getting stanfordparser to work under Snow Leopard. It
appears to be an rjb issue. I tried the approach suggested at this post
without success:

The error I’m getting is the following:

ruby-1.9.2-p180 :001 > require ‘stanfordparser’
=> true
ruby-1.9.2-p180 :002 > preproc =
StanfordParser::DocumentPreprocessor.new
RuntimeError: Constructor not found
from
/Users/boris/.rvm/gems/ruby-1.9.2-p180/gems/stanfordparser-2.2.0/lib/java_object.rb:40:in
new' from /Users/boris/.rvm/gems/ruby-1.9.2-p180/gems/stanfordparser-2.2.0/lib/java_object.rb:40:in initialize’
from
/Users/boris/.rvm/gems/ruby-1.9.2-p180/gems/stanfordparser-2.2.0/lib/stanfordparser.rb:239:in
initialize' from (irb):2:in new’
from (irb):2
from /Users/boris/.rvm/rubies/ruby-1.9.2-p180/bin/irb:16:in
`’

If I don’t follow the approach suggested at the above post, I get a
different error:

ruby-1.9.2-p180 :001 > require ‘stanfordparser’
[BUG] cross-thread violation on rb_gc()
(null)

Abort trap

Any ideas?

I also followed the instructions of the blog post you mentioned, and
also got the same results. The gem I’m using is the “stanfordparser”
v2.2.0 (GitHub - jcwilk/stanfordparser: Github upload of Bill McNeal's ruby wrapper for Stanford's NLP library)

From the changelog at the Stanford Parser website, I figured the last
version (1.6.6) changed APIs that the ruby wrapper does not understand.


Version 1.6.6 2011-04-20 Internal code and API changes (ArrayLists
rather than Sentence; use of CoreLabel objects) to match tagger and
CoreNLP.

So, I tried the previous version, which is available at.

http://nlp.stanford.edu/software/stanford-parser-2010-11-30.tgz

I got it to work with this version, but I could not make the simple test
work:

ruby-1.8.7-p334 :003 > require ‘stanfordparser’
=> true
ruby-1.8.7-p334 :004 > preproc =
StanfordParser::DocumentPreprocessor.new
=>
ruby-1.8.7-p334 :005 > puts preproc.getSentencesFromString(“This is a
sentence. So is this.”)
This
is
a
sentence
.
So
is
this
.
=> nil

I got the StanfordParser::LexicalizedParser, but not the
DocumentPreprocessor, which was also what I was hoping to use.

I also installed the gem “stanfordparser-infochimps”, version 2.2.2
(GitHub - mrflip/stanfordparser: Github upload of Bill McNeal's ruby wrapper for Stanford's NLP library), instead of the 2.2.0 version
above, and got the same results.

EDIT: in order to make the alternative gem to work, I had to install
version 1.2.5 of the rjb gem.

Did you have made any further progress than this?