From 8ed7640388cac8ba6d897739f5c8fe24eb87cc48 Mon Sep 17 00:00:00 2001
From: Abhaya Agarwal <abhaya@pothi.com>
Date: Mon, 10 Oct 2011 19:57:50 +0530
Subject: [PATCH] Fix the regex to capture all the words of the sentence.

---
 corenlp.py | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/corenlp.py b/corenlp.py
index 0b8328e..b624a3e 100644
--- a/corenlp.py
+++ b/corenlp.py
@@ -48,9 +48,10 @@ def parse_parser_results(text):
                 print line
                 raise Exception("Parse error. Could not find [Text=")
             tmp['words'] = [] 
-            exp = re.compile('\[([a-zA-Z0-9=. ]+)\]')
+            exp = re.compile('\[([^\]]+)\]')
             matches  = exp.findall(line)
             for s in matches:
+                print s
                 # split into attribute-value list 
                 av = re.split("=| ", s) 
                 # make [ignore,ignore,a,b,c,d] into [[a,b],[c,d]]
@@ -137,7 +138,7 @@ def __init__(self):
                 sys.exit(1)
         
         # spawn the server
-        self._server = pexpect.spawn("%s -Xmx3g -cp %s %s %s" % (java_path, ':'.join(jars), classname, props))
+        self._server = pexpect.spawn("%s -Xmx1800m -cp %s %s %s" % (java_path, ':'.join(jars), classname, props))
         
         print "Starting the Stanford Core NLP parser."
         self.state = "plays hard to get, smiles from time to time"