Feature/pluralize

changes/82.feature.rst

@@ -0,0 +1,5 @@
+Add a new module for a PLURALIZE tag
+For a noun or a list of nouns, it will match any singular or plural word.
+
+Implemented by:
+Roland M. Mueller (https://github.com/rolandmueller)

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "rita-dsl"
-version = "0.5.8"
+version = "0.5.9"
 description = "DSL for building language rules"
 authors = [
     "Šarūnas Navickas <[email protected]>"

rita/__init__.py

@@ -10,7 +10,7 @@
 
 logger = logging.getLogger(__name__)
 
-__version__ = (0, 5, 8, os.getenv("VERSION_PATCH"))
+__version__ = (0, 5, 9, os.getenv("VERSION_PATCH"))
 
 
 def get_version():

rita/modules/pluralize.py

@@ -0,0 +1,41 @@
+import sys
+import logging
+try:
+    import inflect
+except ImportError:
+    logging.exception(
+        "Pluralize module requires 'inflect' package to be installed."
+        "Install it and try again"
+    )
+    sys.exit(1)
+
+from rita.macros import resolve_value
+from rita.utils import flatten
+
+
+def pluralizing(initial_list):
+    """"
+    For a list of nouns, it will return a list of the plurals and the initial nouns
+    """
+    p = inflect.engine()
+    plurals = [p.plural(word) for word in initial_list]
+    return initial_list + plurals
+
+
+def PLURALIZE(*args, config, op=None):
+    """
+    For a noun or a list of nouns, it will match any singular or plural word
+    Usage for a single word, e.g.:
+    PLURALIZE("car")
+    Usage for lists, e.g.:
+    vehicles = {"car", "bicycle", "ship"}
+    PLURALIZE(vehicles)
+    Will work even for regex or if the lemmatizer of spaCy is making an error
+    Has dependency to the Python inflect package https://pypi.org/project/inflect/
+    """
+    if type(args[0]) == list:
+        initial_list = [resolve_value(arg, config=config)
+                        for arg in flatten(args)]
+    else:
+        initial_list = [args[0]]
+    return "any_of", pluralizing(initial_list), op

rita/parser.py

@@ -26,7 +26,7 @@ def load_macro(name, config):
         pass
 
     def lazy_load(*args, **kwargs):
-        print(config.modules)
+        logger.info(config.modules)
         for mod in config.modules:
             try:
                 fn = getattr(mod, name)

tests/test_examples.py

@@ -281,3 +281,25 @@ def test_inlist_word_based(engine):
     results = parser(text)
     print(results)
     assert len(results) == 0
+
+
+@pytest.mark.parametrize('engine', [standalone_engine, spacy_engine])
+def test_pluralize(engine):
+    parser = engine("""
+    !IMPORT("rita.modules.pluralize")
+
+    vehicles={"car", "motorbike", "bicycle", "ship", "plane"}
+    {NUM, PLURALIZE(vehicles)}->MARK("VEHICLES")
+    """)
+
+    text = """
+    There were 7 cars, 2 motorbikes, 1 ship, 1 bicycle and 9 planes
+    """
+
+    results = set([text
+                   for text, label in parser(text)
+                   if label == "VEHICLES"])
+    print(results)
+
+    assert len(results) == 5
+    assert {"7 cars", "2 motorbikes", "1 ship", "1 bicycle", "9 planes"} == results

tox.ini

@@ -6,6 +6,7 @@ isolated_build = True
 deps =
     ply: ply==3.11
     spacy
+    inflect
     pytest
     pytest-mock
     pytest-benchmark
@@ -20,6 +21,7 @@ deps =
     codecov
     ply: ply==3.11
     spacy
+    inflect
     pytest
     pytest-benchmark
     pytest-cov