Access the stemmer of Terrier from PyTerrier (#382)

* Update index.py

* added test cases

* added documentation

* fixes and testing other stemmers

---------

Co-authored-by: Sean MacAvaney <sean.macavaney@gmail.com>

pyterrier/index.py

@@ -261,11 +261,18 @@ class IndexingType(enum.Enum):
     SINGLEPASS = 2 #: A single-pass indexing regime, which builds an inverted index directly. No direct index structure is created. Typically is faster than classical indexing.
     MEMORY = 3 #: An in-memory index. No persistent index is created.
 
+_stemmer_cache = {}
 class TerrierStemmer(Enum):
     """
         This enum provides an API for the stemmers available in Terrier. The stemming configuration is saved in the index
         and loaded at retrieval time. `Snowball <https://snowballstem.org/>`_ stemmers for various languages 
         `are available in Terrier <http://terrier.org/docs/current/javadoc/org/terrier/terms/package-summary.html>`_.
+
+        It can also be used to access the stemmer::
+
+            stemmer = pt.TerrierStemmer.porter
+            stemmed_word = stemmer.stem('abandoned')
+
     """
     none = 'none' #: Apply no stemming
     porter = 'porter' #: Apply Porter's English stemmer
@@ -316,7 +323,22 @@ def _to_class(this):
 
         if isinstance(this, str):
             return this
-
+
+    def stem(self, tok):
+        if self not in _stemmer_cache:
+            clz_name = self._to_class(self)
+            if clz_name is None:
+                class NoOpStem():
+                    def stem(self, word):
+                        return word
+                _stemmer_cache[self] = NoOpStem()
+            else:
+                if '.' not in clz_name:
+                    clz_name = f'org.terrier.terms.{clz_name}'
+                 # stemmers are termpipeline objects, and these have chained constructors
+                 # pass None to use the appropriate constructor
+                _stemmer_cache[self] = autoclass(clz_name)(None)
+        return _stemmer_cache[self].stem(tok)
 
 class TerrierStopwords(Enum):
     """

tests/test_terrier_wrappers.py

@@ -0,0 +1,29 @@
+import pandas as pd
+import pyterrier as pt
+import os
+import unittest
+from .base import BaseTestCase
+import warnings
+
+class TestTerrierWrappers(BaseTestCase):
+
+    def test_stemming(self):
+        stemmer = pt.TerrierStemmer.porter
+        TESTS = {
+            "abandoned": "abandon",
+    		"abandon": "abandon",
+	    	"abergavenny": "abergavenni"
+        }
+        for i,j in TESTS.items():
+            self.assertEqual(j, stemmer.stem(i))
+
+        stemmer = pt.TerrierStemmer.none
+        for i,j in TESTS.items():
+            self.assertEqual(i, stemmer.stem(i))
+
+        stemmer = pt.TerrierStemmer.portugese
+        for i,j in TESTS.items():
+            self.assertTrue(len(stemmer.stem(i)) > 0)
+
+if __name__ == "__main__":
+    unittest.main()