From 83baa9bb740aab9f5fe34bc87d90ad2e0b11a2dd Mon Sep 17 00:00:00 2001
From: Sveinbjorn Thordarson <sveinbjorn@sveinbjorn.org>
Date: Tue, 19 Sep 2023 14:15:07 +0000
Subject: [PATCH 1/5] Don't load package version from source file

---
 setup.py                       | 8 ++------
 src/reynir_correct/__init__.py | 2 +-
 src/reynir_correct/version.py  | 1 -
 3 files changed, 3 insertions(+), 8 deletions(-)
 delete mode 100644 src/reynir_correct/version.py

diff --git a/setup.py b/setup.py
index 7c0b521..6e7c83a 100644
--- a/setup.py
+++ b/setup.py
@@ -2,7 +2,7 @@
 """
     Greynir: Natural language processing for Icelandic
 
-    Setup.py
+    setup.py
 
     Copyright (C) 2023 Miðeind ehf.
     Original author: Vilhjálmur Þorsteinsson
@@ -59,13 +59,9 @@ def read(*names: str, **kwargs: str):
         return ""
 
 
-# Load version string from file
-__version__ = "[missing]"
-exec(open(join("src", "reynir_correct", "version.py")).read())
-
 setup(
     name="reynir-correct",
-    version=__version__,
+    version="4.0.0",  # Remember to update in __init__.py as well
     license="MIT",
     description="A spelling and grammar corrector for Icelandic",
     long_description="{0}\n{1}".format(
diff --git a/src/reynir_correct/__init__.py b/src/reynir_correct/__init__.py
index 9afb89e..92f01cb 100644
--- a/src/reynir_correct/__init__.py
+++ b/src/reynir_correct/__init__.py
@@ -43,11 +43,11 @@
 from .errtokenizer import Correct_TOK, CorrectionPipeline, CorrectToken
 from .readability import FleschKincaidFeedback, FleschKincaidScorer, RareWordsFinder
 from .settings import Settings
-from .version import __version__
 from .wrappers import CorrectedSentence, CorrectionResult, GreynirCorrectAPI, ParseResultStats, check_errors
 
 __author__ = "Miðeind ehf"
 __copyright__ = "(C) 2023 Miðeind ehf."
+__version__ = "4.0.0"  # Remember to update in setup.py as well
 
 __all__ = (
     "Greynir",
diff --git a/src/reynir_correct/version.py b/src/reynir_correct/version.py
deleted file mode 100644
index ce1305b..0000000
--- a/src/reynir_correct/version.py
+++ /dev/null
@@ -1 +0,0 @@
-__version__ = "4.0.0"

From b8e477376da00ca0c9765cce2a7809abe55da466 Mon Sep 17 00:00:00 2001
From: Sveinbjorn Thordarson <sveinbjorn@sveinbjorn.org>
Date: Tue, 19 Sep 2023 16:20:39 +0000
Subject: [PATCH 2/5] Typing fixes + formatted excessively long lines in
 readability.py

---
 src/reynir_correct/errtokenizer.py | 2 +-
 src/reynir_correct/readability.py  | 7 +++++--
 src/reynir_correct/spelling.py     | 2 +-
 src/reynir_correct/wrappers.py     | 2 +-
 4 files changed, 8 insertions(+), 5 deletions(-)

diff --git a/src/reynir_correct/errtokenizer.py b/src/reynir_correct/errtokenizer.py
index 4663fc8..f57e50c 100644
--- a/src/reynir_correct/errtokenizer.py
+++ b/src/reynir_correct/errtokenizer.py
@@ -2972,7 +2972,7 @@ def __init__(
         self._suggest_not_correct = options.pop("suggest_not_correct", False)
         # Wordlist for words that should not be marked as errors or corrected
         self._ignore_wordlist = options.pop("ignore_wordlist", set())
-        self._ignore_rules = options.pop("ignore_rules", set())
+        self._ignore_rules = cast(frozenset, options.pop("ignore_rules", frozenset()))
         self.settings = settings
 
     def correct_tokens(self, stream: TokenIterator) -> TokenIterator:
diff --git a/src/reynir_correct/readability.py b/src/reynir_correct/readability.py
index 61b5b55..928b1fc 100644
--- a/src/reynir_correct/readability.py
+++ b/src/reynir_correct/readability.py
@@ -31,6 +31,7 @@
     This module implements the Flesch reading ease score for Icelandic text.
     A high score indicates that the text is easy to read, while a low score
     indicates that the text is difficult to read.
+
 """
 
 from __future__ import annotations
@@ -190,7 +191,8 @@ class RareWordsFinder:
 
     Rare words are defined as words which have a probability lower than the low_prob_cutoff.
     The probability of a word is calculated by looking up the word in an n-gram model.
-    The class is designed to be used with the tokenizer module and maintains an internal state which needs to be reset manually.
+    The class is designed to be used with the tokenizer module and maintains an internal
+    state which needs to be reset manually.
     """
 
     def __init__(self):
@@ -200,7 +202,8 @@ def __init__(self):
     def get_rare_words_from_stream(
         self, tok_stream: Iterable[tokenizer.Tok], max_words: int, low_prob_cutoff: float
     ) -> List[Tuple[str, float]]:
-        """Tracks the probability of each word in a token stream. This is done by yielding the tokens in the token stream."""
+        """Tracks the probability of each word in a token stream.
+        This is done by yielding the tokens in the token stream."""
         rare_words_dict: Dict[str, float] = {}
         for token in tok_stream:
             # Only consider words, not punctuation, numbers, etc.
diff --git a/src/reynir_correct/spelling.py b/src/reynir_correct/spelling.py
index dc31a5b..07311f3 100644
--- a/src/reynir_correct/spelling.py
+++ b/src/reynir_correct/spelling.py
@@ -661,7 +661,7 @@ def _case_of(text: str) -> Callable[[str], str]:
             # We don't use .istitle() and .title() because
             # they consider apostrophes to be word separators
             return lambda s: s[0].upper() + s[1:]
-        return str
+        return str  # noqa
 
     def _cast(self, word: str) -> str:
         """Cast the word to lowercase and correct accents"""
diff --git a/src/reynir_correct/wrappers.py b/src/reynir_correct/wrappers.py
index b9d0e9f..709924c 100644
--- a/src/reynir_correct/wrappers.py
+++ b/src/reynir_correct/wrappers.py
@@ -294,7 +294,7 @@ def _correct_spelling(
         # TODO: The pipeline needs a refactoring.
         # We use some hacks here to avoid having to rewrite the pipeline at this point.
         self.gc.pipeline._text_or_gen = text
-        self.gc.pipeline._ignore_rules = ignore_rules or set()
+        self.gc.pipeline._ignore_rules = cast(frozenset, ignore_rules or frozenset())
         self.gc.pipeline._suppress_suggestions = suppress_suggestions
         return self.gc.pipeline.tokenize()  # type: ignore
 

From 09089f5be0561752fb3dc35e932aa68f72fab697 Mon Sep 17 00:00:00 2001
From: Sveinbjorn Thordarson <sveinbjorn@sveinbjorn.org>
Date: Tue, 19 Sep 2023 20:29:07 +0000
Subject: [PATCH 3/5] Added package build settings and metadata to
 pyproject.toml

---
 pyproject.toml | 54 +++++++++++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 53 insertions(+), 1 deletion(-)

diff --git a/pyproject.toml b/pyproject.toml
index 12f05b0..6d6987f 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -1,3 +1,54 @@
+[project]
+name = "reynir-correct"
+version = "4.0.0"
+description = "Spelling and grammar correction for Icelandic"
+authors = [{ name = "Miðeind ehf", email = "mideind@mideind.is" }]
+readme = { file = "README.rst", content-type = "text/x-rst" }
+license = { file = "LICENSE.txt" }
+# For classifier list see: https://pypi.org/pypi?%3Aaction=list_classifiers
+classifiers = [
+    "Development Status :: 5 - Production/Stable",
+    "Intended Audience :: Developers",
+    "Intended Audience :: Science/Research",
+    "License :: OSI Approved :: MIT License",
+    "Operating System :: Unix",
+    "Operating System :: POSIX",
+    "Operating System :: Microsoft :: Windows",
+    "Operating System :: MacOS",
+    "Natural Language :: Icelandic",
+    "Programming Language :: Python",
+    "Programming Language :: Python :: 3",
+    "Programming Language :: Python :: 3.8",
+    "Programming Language :: Python :: 3.9",
+    "Programming Language :: Python :: 3.10",
+    "Programming Language :: Python :: 3.11",
+    "Programming Language :: Python :: Implementation :: CPython",
+    "Programming Language :: Python :: Implementation :: PyPy",
+    "Topic :: Software Development :: Libraries :: Python Modules",
+    "Topic :: Utilities",
+    "Topic :: Text Processing :: Linguistic",
+]
+requires-python = ">=3.8"
+dependencies = ["reynir>=3.5.3", "icegrams>=1.1.2", "typing_extensions"]
+
+[project.urls]
+Repository = "https://github.com/mideind/GreynirCorrect"
+
+[project.optional-dependencies]
+# Dev dependencies
+dev = []
+
+# *** Configuration of tools ***
+
+[tool.setuptools.packages.find]
+where = ["src"]
+
+[tool.pytest.ini_options]
+filterwarnings = [
+    # Ignore deprecation warnings in libraries, their problem not ours
+    "ignore::DeprecationWarning",
+]
+
 [tool.ruff]
 line-length = 120
 
@@ -5,6 +56,7 @@ line-length = 120
 line-length = 120
 
 [tool.isort]
-known_future_library = ["__future__", "typing", "typing_extensions"]  # This forces these imports to placed at the top
+# This forces these imports to placed at the top
+known_future_library = ["__future__", "typing", "typing_extensions"]
 profile = "black"
 line_length = 120

From d97a4d5957d8927d9fc7f86475ffef113d112bfc Mon Sep 17 00:00:00 2001
From: Sveinbjorn Thordarson <sveinbjorn@sveinbjorn.org>
Date: Wed, 20 Sep 2023 20:08:13 +0000
Subject: [PATCH 4/5] Reformatted excessively long lines of code

---
 src/reynir_correct/errtokenizer.py | 13 ++++++++++---
 src/reynir_correct/main.py         | 12 +++++++++---
 2 files changed, 19 insertions(+), 6 deletions(-)

diff --git a/src/reynir_correct/errtokenizer.py b/src/reynir_correct/errtokenizer.py
index f57e50c..863918c 100644
--- a/src/reynir_correct/errtokenizer.py
+++ b/src/reynir_correct/errtokenizer.py
@@ -1841,7 +1841,8 @@ def add_ritmyndir_error(token: CorrectToken) -> CorrectToken:
         return token
 
     def get_details(code: str, txt: str, correct: str, lemma: str) -> Tuple[str, str, List[str]]:
-        """Return short and detailed descriptions for the error category plus a link to grammar references where possible"""
+        """Return short and detailed descriptions for the error category
+        plus a link to grammar references where possible"""
         # text is the short version, about the category and the error.
         # details is the long version with references.
         try:
@@ -2497,7 +2498,12 @@ def number_error(token: CorrectToken, replace: str, code: str, instruction_txt:
                                 suggest=correct,
                             )
                         )
-            if suppress_suggestions and token.error_code == "Z001" and isinstance(token.val, list) and any(v.ordfl == "lo" for v in token.val):  # type: ignore
+            if (
+                suppress_suggestions
+                and token.error_code == "Z001"
+                and isinstance(token.val, list)
+                and any(v.ordfl == "lo" for v in token.val)
+            ):
                 orig = token.original.strip() if token.original else token.txt
                 token.remove_error(orig)
 
@@ -2968,7 +2974,8 @@ def __init__(
         self._generate_suggestion_list = options.pop("generate_suggestion_list", False)
         # Skip spelling suggestions
         self._suppress_suggestions = options.pop("suppress_suggestions", False)
-        # Only give suggestions, don't correct everything automatically. Currently only applies to lookup_unknown_words and check_wording.
+        # Only give suggestions, don't correct everything automatically.
+        # Currently only applies to lookup_unknown_words and check_wording.
         self._suggest_not_correct = options.pop("suggest_not_correct", False)
         # Wordlist for words that should not be marked as errors or corrected
         self._ignore_wordlist = options.pop("ignore_wordlist", set())
diff --git a/src/reynir_correct/main.py b/src/reynir_correct/main.py
index 45158b5..f16bea5 100644
--- a/src/reynir_correct/main.py
+++ b/src/reynir_correct/main.py
@@ -82,7 +82,11 @@
     nargs="?",
     type=str,
     default="text",
-    help="Determine output format.\ntext: Corrected text only.\ncsv: One token per line in CSV format.\njson: One token per line in JSON format.\nm2: M2 format, GEC standard.",
+    help="""Determine output format.
+text: Corrected text only.
+csv: One token per line in CSV format.
+json: One token per line in JSON format.
+m2: M2 format, GEC standard.""",
 )
 
 # Determines whether we supply only token-level annotations or also sentence-level annotations
@@ -127,7 +131,8 @@
 
 parser.add_argument(
     "--sentence_prefilter",
-    help="Run a heuristic filter on sentences to determine whether they are probably correct. Probably correct sentences will not go through the full parsing process.",
+    help="""Run a heuristic filter on sentences to determine whether they are probably
+correct. Probably correct sentences will not go through the full parsing process.""",
     action="store_true",
 )
 parser.add_argument(
@@ -145,7 +150,8 @@
     "--tov_config",
     nargs=1,
     type=str,
-    help="Add additional use-specific rules in a configuration file to check for custom tone-of-voice issues. Uses the same format as the default GreynirCorrect.conf file",
+    help="""Add additional use-specific rules in a configuration file to check for custom
+tone-of-voice issues. Uses the same format as the default GreynirCorrect.conf file""",
     default=None,
 )
 

From 5fbf2f13f1e323c94d7d8241902f7c2613dbe86d Mon Sep 17 00:00:00 2001
From: Sveinbjorn Thordarson <sveinbjorn@sveinbjorn.org>
Date: Fri, 22 Sep 2023 16:01:36 +0000
Subject: [PATCH 5/5] Updated pyproject.toml, rm setup.py, updated CI config

---
 .github/workflows/python-package.yml |   4 +-
 pyproject.toml                       |  13 ++-
 setup.py                             | 116 ---------------------------
 src/reynir_correct/__init__.py       |   2 +-
 4 files changed, 14 insertions(+), 121 deletions(-)
 delete mode 100644 setup.py

diff --git a/.github/workflows/python-package.yml b/.github/workflows/python-package.yml
index 1d85460..bc91e49 100644
--- a/.github/workflows/python-package.yml
+++ b/.github/workflows/python-package.yml
@@ -18,9 +18,9 @@ jobs:
         python-version: [ "3.8", "3.9", "3.10", "3.11", "pypy-3.9", "pypy-3.10"]
 
     steps:
-    - uses: actions/checkout@v2
+    - uses: actions/checkout@v4
     - name: Set up Python ${{ matrix.python-version }}
-      uses: actions/setup-python@v2
+      uses: actions/setup-python@v4
       with:
         python-version: ${{ matrix.python-version }}
     - name: Install GreynirCorrect
diff --git a/pyproject.toml b/pyproject.toml
index 6d6987f..795d9b4 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -35,14 +35,23 @@ dependencies = ["reynir>=3.5.3", "icegrams>=1.1.2", "typing_extensions"]
 Repository = "https://github.com/mideind/GreynirCorrect"
 
 [project.optional-dependencies]
-# Dev dependencies
-dev = []
+# dev dependencies
+dev = ["pytest"]
+# sentence_classifier dependencies
+sentence_classifier = ["transformers", "datasets", "torch"]
+
+[project.scripts]
+# 'correct' command line tool
+correct = "reynir_correct.main:main"
 
 # *** Configuration of tools ***
 
 [tool.setuptools.packages.find]
 where = ["src"]
 
+[tool.setuptools.package-data]
+where = ["src"]
+
 [tool.pytest.ini_options]
 filterwarnings = [
     # Ignore deprecation warnings in libraries, their problem not ours
diff --git a/setup.py b/setup.py
deleted file mode 100644
index 6e7c83a..0000000
--- a/setup.py
+++ /dev/null
@@ -1,116 +0,0 @@
-#!/usr/bin/env python3
-"""
-    Greynir: Natural language processing for Icelandic
-
-    setup.py
-
-    Copyright (C) 2023 Miðeind ehf.
-    Original author: Vilhjálmur Þorsteinsson
-
-    This software is licensed under the MIT License:
-
-        Permission is hereby granted, free of charge, to any person
-        obtaining a copy of this software and associated documentation
-        files (the "Software"), to deal in the Software without restriction,
-        including without limitation the rights to use, copy, modify, merge,
-        publish, distribute, sublicense, and/or sell copies of the Software,
-        and to permit persons to whom the Software is furnished to do so,
-        subject to the following conditions:
-
-        The above copyright notice and this permission notice shall be
-        included in all copies or substantial portions of the Software.
-
-        THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
-        EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
-        MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
-        IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
-        CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
-        TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
-        SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
-
-
-    This module sets up the GreynirCorrect package and installs the
-    'correct' command-line utility.
-
-    This package requires Python >= 3.8, and supports PyPy >= 3.8.
-
-"""
-
-import io
-import re
-import sys
-
-from glob import glob
-from os.path import basename, dirname, join, splitext
-
-from setuptools import find_packages  # type: ignore
-from setuptools import setup  # type: ignore
-
-
-if sys.version_info < (3, 8):
-    print("GreynirCorrect requires Python >= 3.8")
-    sys.exit(1)
-
-
-def read(*names: str, **kwargs: str):
-    try:
-        return io.open(join(dirname(__file__), *names), encoding=kwargs.get("encoding", "utf8")).read()
-    except (IOError, OSError):
-        return ""
-
-
-setup(
-    name="reynir-correct",
-    version="4.0.0",  # Remember to update in __init__.py as well
-    license="MIT",
-    description="A spelling and grammar corrector for Icelandic",
-    long_description="{0}\n{1}".format(
-        re.compile("^.. start-badges.*^.. end-badges", re.M | re.S).sub("", read("README.rst")),
-        re.sub(":[a-z]+:`~?(.*?)`", r"``\1``", read("CHANGELOG.rst")),
-    ),
-    author="Miðeind ehf",
-    author_email="mideind@mideind.is",
-    url="https://github.com/mideind/GreynirCorrect",
-    packages=find_packages("src"),
-    package_dir={"": "src"},
-    py_modules=[splitext(basename(path))[0] for path in glob("src/*.py")],
-    package_data={"reynir_correct": ["py.typed"]},
-    include_package_data=True,
-    zip_safe=True,
-    classifiers=[
-        # complete classifier list: http://pypi.python.org/pypi?%3Aaction=list_classifiers
-        "Development Status :: 5 - Production/Stable",
-        "Intended Audience :: Developers",
-        "Intended Audience :: Science/Research",
-        "License :: OSI Approved :: MIT License",
-        "Operating System :: Unix",
-        "Operating System :: POSIX",
-        "Operating System :: Microsoft :: Windows",
-        "Operating System :: MacOS",
-        "Natural Language :: Icelandic",
-        "Programming Language :: Python",
-        "Programming Language :: Python :: 3",
-        "Programming Language :: Python :: 3.8",
-        "Programming Language :: Python :: 3.9",
-        "Programming Language :: Python :: 3.10",
-        "Programming Language :: Python :: 3.11",
-        "Programming Language :: Python :: Implementation :: CPython",
-        "Programming Language :: Python :: Implementation :: PyPy",
-        "Topic :: Software Development :: Libraries :: Python Modules",
-        "Topic :: Utilities",
-        "Topic :: Text Processing :: Linguistic",
-    ],
-    keywords=["nlp", "parser", "icelandic", "spellchecker"],
-    setup_requires=[],
-    install_requires=["reynir>=3.5.3", "icegrams>=1.1.2", "typing_extensions"],
-    extras_require={
-        "sentence_classifier": ["transformers", "datasets", "torch"],
-    },
-    # Set up a 'correct' command ('correct.exe' on Windows),
-    # which calls main() in src/reynir-correct/main.py
-    entry_points={
-        "console_scripts": [
-            "correct=reynir_correct.main:main",
-        ],
-    },
-)
diff --git a/src/reynir_correct/__init__.py b/src/reynir_correct/__init__.py
index 92f01cb..49d6b3e 100644
--- a/src/reynir_correct/__init__.py
+++ b/src/reynir_correct/__init__.py
@@ -47,7 +47,7 @@
 
 __author__ = "Miðeind ehf"
 __copyright__ = "(C) 2023 Miðeind ehf."
-__version__ = "4.0.0"  # Remember to update in setup.py as well
+__version__ = "4.0.0"  # Remember to update in pyproject.toml as well
 
 __all__ = (
     "Greynir",