From 3fc53cc279ba5ab74e04cee9af693c05d9995fcd Mon Sep 17 00:00:00 2001
From: Paul Hummer <phummer@watchful.io>
Date: Thu, 20 Apr 2023 13:35:46 -0600
Subject: [PATCH] fix: remove cchardet as a dependency

`cchardet` hasn't had a commit in almost two years (8 days shy as
of this commit). [This issue](https://github.com/PyYoshi/cChardet/issues/81)
is a six month old bug showing that `cchardet` doesn't support
python 3.11, which is the most recent python.

`cchardet` is fast, certainly, though the usage here is not as
performance sensitive as would warrant its use over `chardet`,
e.g. a difference of 800ms is probably not something that is worth
keeping problematic dependencies around.
---
 requirements_dev.txt   |  1 -
 setup.cfg              |  1 -
 src/watchful/client.py | 18 +++++-------------
 3 files changed, 5 insertions(+), 15 deletions(-)

diff --git a/requirements_dev.txt b/requirements_dev.txt
index 2b029a2..a0d8643 100644
--- a/requirements_dev.txt
+++ b/requirements_dev.txt
@@ -5,7 +5,6 @@ pyflakes==2.4.0
 pylama==8.3.8
 pylint==2.17.2
 twine==4.0.1
-cchardet==2.1.7
 chardet==5.1.0
 flair==0.12.2
 psutil==5.9.2
diff --git a/setup.cfg b/setup.cfg
index 49a769a..37da639 100644
--- a/setup.cfg
+++ b/setup.cfg
@@ -22,7 +22,6 @@ license_files =
 package_dir =
     = src
 install_requires = 
-    cchardet>=2.1.7
     chardet>=5.1.0
     flair>=0.11.3
     psutil>=5.9.2
diff --git a/src/watchful/client.py b/src/watchful/client.py
index 024c86f..54aa801 100644
--- a/src/watchful/client.py
+++ b/src/watchful/client.py
@@ -17,6 +17,8 @@
 import urllib
 from typing import Callable, Dict, Generator, List, Literal, Optional, Union
 from uuid import uuid4
+
+import chardet
 import requests
 
 
@@ -1172,7 +1174,6 @@ def is_utf8(
     csv_bytes: bytes = None,
     filepath: str = None,
     threshold: float = 0.5,
-    is_fast: bool = True,
 ) -> bool:
     """
     This function attempts to detect if the encoding of the given bytes or the
@@ -1191,7 +1192,6 @@ def is_utf8(
     :type threshold: float, optional
     :param is_fast: Whether to use fast encoding detection with a lower
         accuracy, or not.
-    :type is_fast: bool, optional
     :return: `True` if the detected encoding is utf-8 and has a confidence of
         the given threshold or more, otherwise `False`.
     :rtype: bool
@@ -1207,11 +1207,6 @@ def is_utf8(
             "Only one of them needs to be specified."
         )
 
-    if is_fast:
-        import cchardet as chardet
-    else:
-        import chardet
-
     if csv_bytes:
         res = chardet.detect(csv_bytes)
     else:
@@ -1239,7 +1234,7 @@ def create_dataset(
     filename: str = "none",
     has_header: bool = True,
     threshold_detect: float = 0.5,
-    is_fast_detect: bool = True,
+    is_fast_detect: bool = True,  # pylint: disable=W0613
     force_load: bool = True,
 ) -> str:
     """
@@ -1259,8 +1254,7 @@ def create_dataset(
     :param threshold_detect: The minimum confidence required to accept the
         detected encoding.
     :type threshold_detect: float, optional
-    :param is_fast_detect: Whether to use fast encoding detection with a lower
-        accuracy, or not.
+    :param is_fast_detect: No longer used, but remains for API compatibility
     :type is_fast_detect: bool, optional
     :param force_load: The boolean indicating if the csv dataset will be loaded
         even when its encoding is detected to be non-utf-8, defaults to True.
@@ -1274,9 +1268,7 @@ def create_dataset(
     TODO: Add error handling.
     """
 
-    is_csv_bytes_utf8 = is_utf8(
-        csv_bytes, None, threshold_detect, is_fast_detect
-    )
+    is_csv_bytes_utf8 = is_utf8(csv_bytes, None, threshold_detect)
 
     if is_csv_bytes_utf8 or force_load:
         id_ = str(uuid4())