Added leetspeak function encoding to encode_text

augly/tests/assets/expected_metadata/text_tests/expected_metadata.json

@@ -82,18 +82,20 @@
             "src_length": 1
         }
     ],
-    "encode_base64": [
+    "encode_text": [
         {
             "dst_length": 1,
             "input_type": "list",
             "intensity": 100.0,
-            "name": "encode_base64",
+            "name": "encode_text",
             "src_length": 1,
-            "granularity": "all",
             "aug_min": 1,
-            "aug_max": 10,
-            "aug_p": 0.3,
-            "n": 1
+            "aug_max": 1,
+            "aug_p": 1.0,
+            "n": 1,
+            "p": 1.0,
+            "encoder": "base64",
+            "method": "sentence"
         }
     ],
     "get_baseline": [

augly/tests/text_tests/functional_unit_test.py

@@ -10,7 +10,9 @@
 import unittest
 
 from augly import text as txtaugs
+from augly.text.augmenters.utils import Encoding
 from augly.utils import FUN_FONTS_GREEK_PATH
+from nlpaug.util import Method
 
 
 class FunctionalTextUnitTest(unittest.TestCase):
@@ -36,6 +38,24 @@ def test_apply_lambda(self) -> None:
         augmented_apply_lambda = txtaugs.apply_lambda(self.texts)
         self.assertTrue(augmented_apply_lambda[0] == self.texts[0])
 
+    def test_base64_sentence(self) -> None:
+        augmented_words = txtaugs.encode_text(
+            "Hello, world!", 1, 1, 1.0, Method.SENTENCE, Encoding.BASE64
+        )
+        self.assertEqual(augmented_words[0], "SGVsbG8sIHdvcmxkIQ==")
+
+    def test_base64_word(self) -> None:
+        augmented_words_word = txtaugs.encode_text(
+            "Hello, world!", 1, 1, 1.0, Method.WORD, Encoding.BASE64
+        )
+        self.assertEqual(augmented_words_word[0], "SGVsbG8=, world!")
+
+    def test_base64_char(self) -> None:
+        augmented_words_char = txtaugs.encode_text(
+            "Hello, world!", 1, 1, 1.0, Method.CHAR, Encoding.BASE64
+        )
+        self.assertEqual(augmented_words_char[0], "SA==ello LA== dw==orld IQ==")
+
     def test_change_case(self) -> None:
         augmented_words = txtaugs.change_case(self.texts[0], cadence=3.0, case="upper")
         self.assertTrue(
@@ -51,35 +71,6 @@ def test_contractions(self) -> None:
             augmented_words[0] == "I would call him but I don't know where he's gone"
         )
 
-    def test_encode_base64_all(self) -> None:
-        augmented_words = txtaugs.encode_base64("Hello, world!")
-        self.assertTrue(augmented_words[0] == "SGVsbG8sIHdvcmxkIQ==")
-
-    def test_encode_base64_word(self) -> None:
-        random.seed(42)  # Set seed for reproducibility
-        augmented_words_word = txtaugs.encode_base64(
-            "Hello, world!", granularity="word", aug_min=1, aug_max=1, aug_p=1.0
-        )
-        self.assertEqual(augmented_words_word[0], "SGVsbG8=, world!")
-
-    def test_encode_base64_char(self) -> None:
-        random.seed(42)
-        augmented_words_char = txtaugs.encode_base64(
-            "Hello, world!", granularity="char", aug_min=1, aug_max=2, aug_p=1.0
-        )
-        self.assertEqual(augmented_words_char[0], "SA==ellbw== LA== wbw==rlZA== IQ==")
-
-    def test_encode_base64_general(self) -> None:
-        random.seed(42)
-        augmented_words_low_p = txtaugs.encode_base64(
-            "Hello, world!", granularity="word", aug_min=1, aug_max=2, aug_p=0.1
-        )
-        random.seed(42)
-        augmented_words_high_p = txtaugs.encode_base64(
-            "Hello, world!", granularity="word", aug_min=1, aug_max=2, aug_p=0.9
-        )
-        self.assertTrue(len(augmented_words_high_p[0]) > len(augmented_words_low_p[0]))
-
     def test_get_baseline(self) -> None:
         augmented_baseline = txtaugs.get_baseline(self.texts)
         self.assertTrue(
@@ -281,6 +272,18 @@ def test_insert_zero_width_chars(self) -> None:
             ],
         )
 
+    def test_leetspeak_sentence(self) -> None:
+        augmented_words = txtaugs.encode_text(
+            "Hello, world!", 1, 1, 1.0, Method.SENTENCE, Encoding.LEETSPEAK
+        )
+        self.assertEqual(augmented_words[0], "h3110, w0r1d!")
+
+    def test_leetspeak_word(self) -> None:
+        augmented_words = txtaugs.encode_text(
+            "Hello, world!", 1, 1, 1.0, Method.WORD, Encoding.LEETSPEAK
+        )
+        self.assertEqual(augmented_words[0], "h3110, world!")
+
     def test_merge_words(self) -> None:
         augmented_split_words = txtaugs.merge_words(self.texts, aug_word_p=0.3, n=1)
         self.assertTrue(

augly/tests/text_tests/transforms_unit_test.py

@@ -14,7 +14,9 @@
 from typing import Any, Dict, List
 
 from augly import text as txtaugs
+from augly.text.augmenters.utils import Encoding
 from augly.utils import TEXT_METADATA_PATH
+from nlpaug.util import Method
 
 
 def are_equal_metadata(
@@ -136,57 +138,68 @@ def test_Compose(self) -> None:
             are_equal_metadata(self.metadata, self.expected_metadata["compose"]),
         )
 
-    def test_EncodeBase64(self) -> None:
-        augmented_text = txtaugs.EncodeBase64(
-            granularity="all", aug_min=1, aug_max=10, aug_p=0.3, n=1, p=1.0
+    def test_Base64_Sentence(self) -> None:
+        augmented_text = txtaugs.EncodeTextTransform(
+            aug_min=1,
+            aug_max=1,
+            aug_p=1.0,
+            method=Method.SENTENCE,
+            encoder=Encoding.BASE64,
+            n=1,
+            p=1.0,
         )(
             ["Hello, world!"],
             metadata=self.metadata,
         )
 
         self.assertTrue(augmented_text[0] == "SGVsbG8sIHdvcmxkIQ==")
+        self.expected_metadata["encode_text"][0]["encoder"] = Encoding.BASE64
         self.assertTrue(
-            are_equal_metadata(self.metadata, self.expected_metadata["encode_base64"])
+            are_equal_metadata(self.metadata, self.expected_metadata["encode_text"])
         )
 
-    def test_EncodeBase64_Word(self) -> None:
+    def test_Base64_Word(self) -> None:
         self.metadata = []
 
-        random.seed(42)
-        augmented_text = txtaugs.EncodeBase64(
-            granularity="word", aug_min=1, aug_max=1, aug_p=1.0, n=1, p=1.0
+        augmented_text = txtaugs.EncodeTextTransform(
+            aug_min=1,
+            aug_max=1,
+            aug_p=1.0,
+            method=Method.WORD,
+            encoder=Encoding.BASE64,
+            n=1,
+            p=1.0,
         )(
             ["Hello, world!"],
             metadata=self.metadata,
         )
         self.assertEqual(augmented_text[0], "SGVsbG8=, world!")
 
-        expected_metadata = deepcopy(self.expected_metadata["encode_base64"])
-        expected_metadata[0]["granularity"] = "word"
-        expected_metadata[0]["aug_p"] = 1.0
-        expected_metadata[0]["aug_max"] = 1
-        expected_metadata[0]["intensity"] = 100.0
+        metadata_expected = deepcopy(self.expected_metadata["encode_text"])
+        metadata_expected[0]["method"] = "word"
+        metadata_expected[0]["encoder"] = Encoding.BASE64
+        self.assertTrue(are_equal_metadata(self.metadata, metadata_expected))
 
-        self.assertTrue(are_equal_metadata(self.metadata, expected_metadata))
-
-    def test_EncodeBase64_Char(self) -> None:
+    def test_Base64_Char(self) -> None:
         self.metadata = []
 
-        random.seed(42)
-        augmented_text = txtaugs.EncodeBase64(
-            granularity="char", aug_min=1, aug_max=2, aug_p=1.0, n=1, p=1.0
+        augmented_text = txtaugs.EncodeTextTransform(
+            aug_min=1,
+            aug_max=1,
+            aug_p=1.0,
+            method=Method.CHAR,
+            encoder=Encoding.BASE64,
+            n=1,
+            p=1.0,
         )(
             ["Hello, world!"],
             metadata=self.metadata,
         )
-        self.assertEqual(augmented_text[0], "SA==ebA==lo LA== wbw==rlZA== IQ==")
-
-        expected_metadata = deepcopy(self.expected_metadata["encode_base64"])
-        expected_metadata[0]["granularity"] = "char"
-        expected_metadata[0]["aug_p"] = 1.0
-        expected_metadata[0]["aug_max"] = 2
-        expected_metadata[0]["intensity"] = 100.0
+        self.assertEqual(augmented_text[0], "SA==ello LA== wocg==ld IQ==")
 
+        expected_metadata = deepcopy(self.expected_metadata["encode_text"])
+        expected_metadata[0]["method"] = "char"
+        expected_metadata[0]["encoder"] = Encoding.BASE64
         self.assertTrue(are_equal_metadata(self.metadata, expected_metadata))
 
     def test_GetBaseline(self) -> None:
@@ -278,6 +291,48 @@ def test_InsertZeroWidthChars(self) -> None:
             ),
         )
 
+    def test_LeetSpeak_Sentence(self) -> None:
+        augmented_text = txtaugs.EncodeTextTransform(
+            aug_min=1,
+            aug_max=1,
+            aug_p=1.0,
+            method=Method.SENTENCE,
+            encoder=Encoding.LEETSPEAK,
+            n=1,
+            p=1.0,
+        )(
+            ["Hello, world!"],
+            metadata=self.metadata,
+        )
+
+        self.assertTrue(augmented_text[0] == "h3110, w0r1d!")
+        self.expected_metadata["encode_text"][0]["encoder"] = Encoding.LEETSPEAK
+        self.assertTrue(
+            are_equal_metadata(self.metadata, self.expected_metadata["encode_text"])
+        )
+
+    def test_Leetspeak_Word(self) -> None:
+        self.metadata = []
+
+        augmented_text = txtaugs.EncodeTextTransform(
+            aug_min=1,
+            aug_max=1,
+            aug_p=1.0,
+            method=Method.WORD,
+            encoder=Encoding.LEETSPEAK,
+            n=1,
+            p=1.0,
+        )(
+            ["Hello, world!"],
+            metadata=self.metadata,
+        )
+        self.assertEqual(augmented_text[0], "h3110, world!")
+
+        metadata_expected = deepcopy(self.expected_metadata["encode_text"])
+        metadata_expected[0]["method"] = "word"
+        metadata_expected[0]["encoder"] = Encoding.LEETSPEAK
+        self.assertTrue(are_equal_metadata(self.metadata, metadata_expected))
+
     def test_MergeWords(self) -> None:
         aug_merge_words = txtaugs.MergeWords(aug_word_p=0.3)(
             self.texts, metadata=self.metadata

augly/text/__init__.py

@@ -12,7 +12,7 @@
     apply_lambda,
     change_case,
     contractions,
-    encode_base64,
+    encode_text,
     get_baseline,
     insert_punctuation_chars,
     insert_text,
@@ -32,9 +32,10 @@
 )
 from augly.text.intensity import (
     apply_lambda_intensity,
+    base64_intensity,
     change_case_intensity,
     contractions_intensity,
-    encode_base64_intensity,
+    encode_text_intensity,
     get_baseline_intensity,
     insert_punctuation_chars_intensity,
     insert_text_intensity,
@@ -56,7 +57,7 @@
     ApplyLambda,
     ChangeCase,
     Contractions,
-    EncodeBase64,
+    EncodeTextTransform,
     GetBaseline,
     InsertPunctuationChars,
     InsertText,
@@ -81,7 +82,7 @@
     "ApplyLambda",
     "ChangeCase",
     "Contractions",
-    "EncodeBase64",
+    "EncodeTextTransform",
     "GetBaseline",
     "InsertPunctuationChars",
     "InsertText",
@@ -101,7 +102,7 @@
     "apply_lambda",
     "change_case",
     "contractions",
-    "encode_base64",
+    "encode_text",
     "get_baseline",
     "insert_punctuation_chars",
     "insert_text",
@@ -119,9 +120,10 @@
     "split_words",
     "swap_gendered_words",
     "apply_lambda_intensity",
+    "base64_intensity",
     "change_case_intensity",
     "contractions_intensity",
-    "encode_base64_intensity",
+    "encode_text_intensity",
     "get_baseline_intensity",
     "insert_punctuation_chars_intensity",
     "insert_text_intensity",

augly/text/augmenters/__init__.py

@@ -7,31 +7,36 @@
 
 # pyre-unsafe
 
+from augly.text.augmenters.base64 import Base64
 from augly.text.augmenters.baseline import BaselineAugmenter
 from augly.text.augmenters.bidirectional import BidirectionalAugmenter
 from augly.text.augmenters.case import CaseAugmenter
 from augly.text.augmenters.contraction import ContractionAugmenter
-from augly.text.augmenters.encode_base64 import EncodeBase64
+from augly.text.augmenters.encode_text_context import EncodeText
+from augly.text.augmenters.encode_text_strategy import EncodeTextAugmentation
 from augly.text.augmenters.fun_fonts import FunFontsAugmenter
 from augly.text.augmenters.insert_text import InsertTextAugmenter
 from augly.text.augmenters.insertion import InsertionAugmenter
+from augly.text.augmenters.leetspeak import LeetSpeak
 from augly.text.augmenters.letter_replacement import LetterReplacementAugmenter
 from augly.text.augmenters.text_replacement import TextReplacementAugmenter
 from augly.text.augmenters.typo import TypoAugmenter
 from augly.text.augmenters.upside_down import UpsideDownAugmenter
 from augly.text.augmenters.word_replacement import WordReplacementAugmenter
 from augly.text.augmenters.words_augmenter import WordsAugmenter
 
-
 __all__ = [
+    "Base64",
     "BaselineAugmenter",
     "BidirectionalAugmenter",
     "CaseAugmenter",
     "ContractionAugmenter",
-    "EncodeBase64",
+    "EncodeText",
+    "EncodeTextAugmentation",
     "FunFontsAugmenter",
     "InsertTextAugmenter",
     "InsertionAugmenter",
+    "LeetSpeak",
     "LetterReplacementAugmenter",
     "WordsAugmenter",
     "TextReplacementAugmenter",