with cat construction cache

empath/core.py

@@ -11,6 +11,7 @@ def __init__(self, backend_url="http://54.148.189.209:8000"):
         self.staging = {}
         self.backend_url = backend_url
         self.base_dir = os.path.dirname(util.__file__)
+        self.inv_cache = {}
         self.load(self.base_dir+"/data/categories.tsv")
         for f in os.listdir(self.base_dir+"/data/user/"):
             if len(f.split(".")) > 1 and f.split(".")[1] == "empath":
@@ -42,8 +43,13 @@ def analyze(self,doc,categories=None,tokenizer="default",normalize=False):
         if not categories:
             categories = self.cats.keys()
         invcats = defaultdict(list)
-        for k in categories:
-           for t in self.cats[k]: invcats[t].append(k)
+        key = tuple(sorted(categories))
+        if key in self.inv_cache:
+            invcats = self.inv_cache[key]
+        else:
+            for k in categories:
+                for t in self.cats[k]: invcats[t].append(k)
+            self.inv_cache[key] = invcats
         count = {}
         tokens = 0.0
         for cat in categories: count[cat] = 0.0

setup.py

@@ -3,12 +3,12 @@
 setup(
   name = 'empath',
   packages = ['empath'], # this must be the same as the name above
-  version = '0.088',
+  version = '0.089',
   description = 'A tool for text analysis',
   author = 'Ethan Fast',
   author_email = '[email protected]',
   url = 'https://github.com/Ejhfast/empath-client', # use the URL to the github repo
-  download_url = 'https://github.com/Ejhfast/meta/empath-client/0.39',
+  download_url = 'https://github.com/Ejhfast/meta/empath-client/0.40',
   keywords = ['social science', 'lexicon', 'text analysis'], # arbitrary keywords
   package_data= { 'empath': ['data/categories.tsv', "data/user/blank"]},
   classifiers = [],

test/test.py

@@ -3,3 +3,5 @@
 empath = Empath()
 
 print(empath.analyze(["this","that"]))
+print(empath.analyze(["this","that"]))
+print(empath.analyze("the person typed on the keyboard"))