tuples as tuples; digits as ints

dasmith · dasmith · commit 9e2feedbd617 · 2011-02-27T16:21:01.000-05:00
diff --git a/README.md b/README.md
@@ -44,17 +44,17 @@ Assuming you are running on port 8080, the code in `client.py` shows an example
 Produces a list with a parsed dictionary for each sentence:
 
     Result [{"text": "hello world", 
-            "tuples": [["amod", "world", "hello"]], 
-            "words": {"world": {"NamedEntityTag": "O", 
-                                "CharacterOffsetEnd": "11", 
+            "tuples": [("amod", "world", "hello")], 
+            "words": {"world": {"NamedEntityTag": O, 
+                                "CharacterOffsetEnd": 11, 
                                 "Lemma": "world", 
                                 "PartOfSpeech": "NN", 
-                                "CharacterOffsetBegin": "6"}, 
-                      "hello": {"NamedEntityTag": "O", 
-                                "CharacterOffsetEnd": "5", 
+                                "CharacterOffsetBegin": 6}, 
+                      "hello": {"NamedEntityTag": O, 
+                                "CharacterOffsetEnd": 5, 
                                 "Lemma": "hello", 
                                 "PartOfSpeech": "JJ", 
-                                "CharacterOffsetBegin": "0"}}}]
+                                "CharacterOffsetBegin": 0}}}]
 
 
 <!--
diff --git a/server.py b/server.py
@@ -42,11 +42,15 @@ def parse_parser_results(text):
                 raise Exception("Parse error. Could not find [Text=")
             tmp['words'] = {} 
             exp = re.compile('\[([a-zA-Z0-9=. ]+)\]')
-            m = exp.findall(line)
-            for s in m:
-                av = re.split("=| ", s) # attribute-value tuples
-                tmp['words'][av[1]] = dict(zip(*[av[2:][x::2] for x in (0, 1)])) 
-            print tmp
+            matches  = exp.findall(line)
+            for s in matches:
+                # split into attribute-value list 
+                av = re.split("=| ", s) 
+                # make [ignore,ignore,a,b,c,d] into [[a,b],[c,d]]
+                av = zip(*[av[2:][x::2] for x in (0, 1)]) 
+                # save as attr-value dict, convert numbers into ints
+                tmp['words'][av[1]] = dict(map(lambda x: (x[0], x.isdigit(x[1])
+                    and int(x[1]) or x[1]), av))
             state = 3
         elif state == 3:
             # skip over parse tree
@@ -60,7 +64,7 @@ def parse_parser_results(text):
                 split_entry = re.split("\(|, ", line[:-1]) 
                 if len(split_entry) == 3:
                     rel, left, right = map(lambda x: remove_id(x), split_entry)
-                    tmp['tuples'].append((rel,left,right))
+                    tmp['tuples'].append(tuple(rel,left,right))
                     print "\n", rel, left, right
             elif "Coreference links" in line:
                 state = 5
