gptc/__init__.py

@@ -5,6 +5,7 @@
 from gptc.compiler import compile as compile
 from gptc.classifier import Classifier as Classifier
 from gptc.pack import pack as pack
+from gptc.tokenizer import has_emoji as has_emoji
 from gptc.model import Model as Model, deserialize as deserialize
 from gptc.exceptions import (
     GPTCError as GPTCError,

gptc/classifier.py

@@ -29,6 +29,7 @@ class Classifier:
         self.model = model
         model_ngrams = model.max_ngram_length
         self.max_ngram_length = min(max_ngram_length, model_ngrams)
+        self.has_emoji = gptc.tokenizer.has_emoji and model.has_emoji
 
     def confidence(self, text: str) -> Dict[str, float]:
         """Classify text with confidence.
@@ -48,7 +49,9 @@ class Classifier:
 
         model = self.model.weights
 
-        tokens = gptc.tokenizer.tokenize(text, self.max_ngram_length)
+        tokens = gptc.tokenizer.tokenize(
+            text, self.max_ngram_length, self.has_emoji
+        )
         numbered_probs: Dict[int, float] = {}
         for word in tokens:
             try:

gptc/model.py

@@ -12,10 +12,14 @@ class Model:
         weights: Dict[int, List[int]],
         names: List[str],
         max_ngram_length: int,
+        has_emoji: Union[None, bool] = None,
     ):
         self.weights = weights
         self.names = names
         self.max_ngram_length = max_ngram_length
+        self.has_emoji = (
+            gptc.tokenizer.has_emoji if has_emoji is None else has_emoji
+        )
 
     def serialize(self) -> bytes:
         out = b"GPTC model v4\n"
@@ -24,16 +28,7 @@ class Model:
                 {
                     "names": self.names,
                     "max_ngram_length": self.max_ngram_length,
-                    "has_emoji": True,
-                    # Due to an oversight in development, version 3.0.0 still
-                    # had the code used to make emoji support optional, even
-                    # though the `emoji` library was made a hard dependency.
-                    # Part of this code checked whether or not the model
-                    # supports emoji; deserialization would not work in 3.0.0
-                    # if the model was compiled without this field. Emoji are
-                    # always supported with 3.0.0 and newer when GPTC has been
-                    # installed correctly, so this value should always be True.
-                    # Related: #11
+                    "has_emoji": self.has_emoji,
                 }
             ).encode("utf-8")
             + b"\n"
@@ -62,11 +57,14 @@ def deserialize(encoded_model: bytes) -> Model:
     try:
         names = config["names"]
         max_ngram_length = config["max_ngram_length"]
+        has_emoji = config["has_emoji"]
     except KeyError:
         raise InvalidModelError()
 
     if not (
-        isinstance(names, list) and isinstance(max_ngram_length, int)
+        isinstance(names, list)
+        and isinstance(max_ngram_length, int)
+        and isinstance(has_emoji, bool)
     ) or not all([isinstance(name, str) for name in names]):
         raise InvalidModelError()
 
@@ -88,4 +86,4 @@ def deserialize(encoded_model: bytes) -> Model:
         for code in weight_codes
     }
 
-    return Model(weights, names, max_ngram_length)
+    return Model(weights, names, max_ngram_length, has_emoji)

gptc/tokenizer.py

@@ -3,10 +3,22 @@
 from typing import List, Union
 import hashlib
 import base64
+
+try:
     import emoji
 
+    has_emoji = True
+except ImportError:
+    has_emoji = False
 
-def tokenize(text: str, max_ngram_length: int = 1) -> List[int]:
+
+def tokenize(
+    text: str, max_ngram_length: int = 1, use_emoji: bool = True
+) -> List[int]:
+    """Convert a string to a list of lemmas."""
+    converted_text: Union[str, List[str]] = text.lower()
+
+    if has_emoji and use_emoji:
         text = text.lower()
         parts = []
         highest_end = 0
@@ -22,7 +34,7 @@ def tokenize(text: str, max_ngram_length: int = 1) -> List[int]:
     for char in converted_text:
         if char.isalpha() or char == "'":
             tokens[-1] += char
-        elif emoji.is_emoji(char):
+        elif has_emoji and emoji.is_emoji(char):
             tokens.append(char)
             tokens.append("")
         elif tokens[-1] != "":

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "gptc"
-version = "3.0.1"
+version = "3.0.0"
 description = "General-purpose text classifier"
 readme = "README.md"
 authors = [{ name = "Samuel Sloniker", email = "sam@kj7rrv.com"}]