modernization of proselint

.ruff.toml

@@ -55,7 +55,7 @@ preview = true
 [lint.per-file-ignores]
 "tests/**" = ["S", "N8", "SLF001", "PLR2004"]
 "proselint/checks/**" = ["RUF001", "RUF002", "RUF003", "TID252"]
-"scripts/**" = ["E501"]
+"scripts/**" = ["E501", "DTZ005"]
 
 "plugins/**" = ["INP001", "RUF012"]
 "scripts_web/**" = ["S1", "SIM", "RET", "TCH"]

proselint/checks/__init__.py

@@ -24,7 +24,8 @@
 ResultCheck: TypeAlias = tuple[int, int, str, str, Optional[str]]
 # content: start_pos, end_pos, check_name, message, replacement)
 # NOTE1: NewType() is too strict here
-# NOTE2: py312 can use -> type ResultCheck = tuple[int, int, str, str, Optional[str]]
+# NOTE2: py312 can use ->
+# type ResultCheck = tuple[int, int, str, str, Optional[str]]
 
 
 ###############################################################################
@@ -39,26 +40,28 @@ def get_checks(options: dict) -> list[Callable[[str, str], list[ResultCheck]]]:
     # TODO: benchmark consecutive runs of this
     #       config should only translate once to check-list
     checks = []
-    check_names = [key for (key, val) in options["checks"].items() if val]
+    check_names = ["." + key for (key, val) in options["checks"].items() if val]
 
     for check_name in check_names:
         try:
-            module = importlib.import_module("." + check_name, "proselint.checks")
+            module = importlib.import_module(check_name, "proselint.checks")
         except ModuleNotFoundError:
             log.exception(
                 "requested config-flag '%s' not found in proselint.checks",
                 check_name,
             )
             continue
-        checks += [getattr(module, d) for d in dir(module) if re.match(r"^check", d)]
+        checks += [
+            getattr(module, d) for d in dir(module) if re.match(r"^check", d)
+        ]
 
     log.debug("Collected %d checks to run", len(checks))
     return checks
 
 
 def run_checks(_check: Callable, _text: str, source: str = "") -> list:
-    # TODO: frozenset is hashable (list without duplicates) -> check-list, result-lists
-    # padding
+    # TODO: frozenset is hashable (list without duplicates)
+    # -> check-list, result-lists padding
     # -> some checks expect words in text and need something around it
     # -> this prevents edge-cases
     _text = f" \n{_text}\n "  # maybe not the fastest OP
@@ -136,7 +139,9 @@ def find_ranges(_text: str) -> list[tuple[int, int]]:
             pc = c
         return ranges
 
-    def position_in_ranges(ranges: list[tuple[int, int]], _position: int) -> bool:
+    def position_in_ranges(
+        ranges: list[tuple[int, int]], _position: int
+    ) -> bool:
         return any(start <= _position < end for start, end in ranges)
 
     return position_in_ranges(find_ranges(text), position)
@@ -258,7 +263,13 @@ def existence_check(  # noqa: PLR0913, PLR0917
         if any(re.search(exception, txt) for exception in exceptions):
             continue
         errors.append(
-            (m.start() + 1 + offset, m.end() + offset, err, msg.format(txt), None),
+            (
+                m.start() + 1 + offset,
+                m.end() + offset,
+                err,
+                msg.format(txt),
+                None,
+            ),
         )
         # TODO: doesn't the padding alter the start+end?
 
@@ -314,9 +325,7 @@ def topics(text: str) -> list[str]:
 
 def context(text, position, level="paragraph"):
     """Get sentence or paragraph that surrounds the given position."""
-    if level == "sentence":
-        pass
-    elif level == "paragraph":
+    if level in {"sentence", "paragraph"}:
         pass
 
     return ""

proselint/checks/cliches/misc.py

@@ -922,6 +922,7 @@ def check_cliches_gnu_diction(text: str) -> list[ResultCheck]:
 
     return existence_check(text, items, err, msg, ignore_case=True)
 
+
 def check_cliches_nigel(text: str) -> list[ResultCheck]:
     """Check the text.
 
@@ -1080,7 +1081,6 @@ def check_cliches_nigel(text: str) -> list[ResultCheck]:
         "when the chips are down",
         "wind of change",
         "yawning gap",
-
     ]
 
     return existence_check(text, cliches, err, msg, ignore_case=True)

proselint/checks/cursing/filth.py

@@ -21,7 +21,10 @@
 def check(text: str) -> list[ResultCheck]:
     """Check the text."""
     err = "cursing.filth"
-    msg = "Nobody ever tells you this as a kid, but you're supposed to avoid this word."
+    msg = (
+        "Nobody ever tells you this as a kid,",
+        "but you're supposed to avoid this word.",
+    )
 
     items = [
         "shit",

proselint/checks/dates_times/am_pm.py

@@ -22,7 +22,9 @@ def check_lowercase_periods(text: str) -> list[ResultCheck]:
     err = "dates_times.am_pm.lowercase_periods"
     msg = "With lowercase letters, the periods are standard."
 
-    return existence_check(text, [r"\d{1,2} ?[ap]m"], err, msg, ignore_case=False)
+    return existence_check(
+        text, [r"\d{1,2} ?[ap]m"], err, msg, ignore_case=False
+    )
 
 
 def check_spacing(text: str) -> list[ResultCheck]:
@@ -36,7 +38,9 @@ def check_spacing(text: str) -> list[ResultCheck]:
 def check_midnight_noon(text: str) -> list[ResultCheck]:
     """Check the text."""
     err = "dates_times.am_pm.midnight_noon"
-    msg = "12 a.m. and 12 p.m. are wrong and confusing. Use 'midnight' or 'noon'."
+    msg = (
+        "12 a.m. and 12 p.m. are wrong and confusing. Use 'midnight' or 'noon'."
+    )
 
     return existence_check(text, [r"12 ?[ap]\.?m\.?"], err, msg)
 

proselint/checks/misc/greylist.py

@@ -45,7 +45,7 @@ def check(text: str) -> list[ResultCheck]:
 
     bad_words = {
         "obviously": "This is obviously an inadvisable word to use.",
-        "utilize": "Do you know anyone who *needs* to utilize the word utilize?",
+        "utilize": "Do you know anyone who needs to utilize the word utilize?",
     }
 
     results = []

proselint/checks/misc/phrasal_adjectives.py

@@ -30,7 +30,9 @@ def check_ly(text: str) -> list[ResultCheck]:
 
     items = [r"\s[^\s-]+ly-"]
 
-    return existence_check(text, items, err, msg, padding=Pd.disabled, offset=-1)
+    return existence_check(
+        text, items, err, msg, padding=Pd.disabled, offset=-1
+    )
 
 
 def check(text: str) -> list[ResultCheck]:
@@ -64,7 +66,10 @@ def check(text: str) -> list[ResultCheck]:
         ["hit-and-run statute", ["hit and run statute"]],
         ["HIV-negative person", ["HIV negative person"]],
         ["HIV-positive person", ["HIV positive person"]],
-        ["information-technology personnel", ["information technology personnel"]],
+        [
+            "information-technology personnel",
+            ["information technology personnel"],
+        ],
         ["intellectual-property rights", ["intellectual property rights"]],
         ["interest-group pressures", ["interest group pressures"]],
         ["joint-stock company", ["joint stock company"]],

proselint/checks/misc/preferred_forms.py

@@ -79,7 +79,10 @@ def check(text: str) -> list[ResultCheck]:
         ["lent me their", ["loaned me their"]],
         ["lent me your", ["loaned me your"]],
         ["linguist", ["linguistician"]],
-        ["matzo-ball", ["matzoh-ball", "matza-ball", "matzah-ball", "matsah-ball"]],
+        [
+            "matzo-ball",
+            ["matzoh-ball", "matza-ball", "matzah-ball", "matsah-ball"],
+        ],
         ["mayoralty", ["mayorality"]],
         ["mealy-mouthed", ["mealymouthed"]],
         ["mean-spirited", ["meanspirited"]],

proselint/checks/redundancy/misc.py

@@ -257,7 +257,11 @@ def check_atd_1(text: str) -> list[ResultCheck]:
         ["earlier", ["earlier in time"]],
         [
             "eliminate",
-            ["completely eliminate", "eliminate altogether", "entirely eliminate"],
+            [
+                "completely eliminate",
+                "eliminate altogether",
+                "entirely eliminate",
+            ],
         ],
         ["ember", ["glowing ember"]],
         ["embers", ["burning embers"]],
@@ -272,7 +276,11 @@ def check_atd_1(text: str) -> list[ResultCheck]:
         ["essential", ["absolutely essential"]],
         [
             "estimated at",
-            ["estimated at about", "estimated at approximately", "estimated at around"],
+            [
+                "estimated at about",
+                "estimated at approximately",
+                "estimated at around",
+            ],
         ],
         ["etc.", ["and etc."]],
         ["evolve", ["evolve over time"]],
@@ -366,7 +374,10 @@ def check_atd_2(text: str) -> list[ResultCheck]:
         # ["input",             ["input into"]],
         ["instinct", ["natural instinct", "naturally instinct"]],
         ["integrate", ["integrate together", "integrate with each other"]],
-        ["interdependent", ["interdependent on each other", "mutually interdependent"]],
+        [
+            "interdependent",
+            ["interdependent on each other", "mutually interdependent"],
+        ],
         ["introduced", ["introduced for the first time"]],
         ["invention", ["new invention"]],
         ["kneel", ["kneel down"]],

proselint/checks/typography/symbols.py

@@ -13,7 +13,8 @@
 
 
 def check(text: str) -> list[ResultCheck]:
-    # NOTE: these tests are pretty fast, so they are combined for a more even load
+    # NOTE: these tests are pretty fast, so they are combined for a more even
+    # load
 
     @limit_results(3)
     def check_ellipsis(_text: str) -> list[ResultCheck]:

proselint/command_line.py

@@ -5,7 +5,7 @@
 import json
 import os
 import signal
-import subprocess
+import subprocess  # noqa: S404
 import sys
 import time
 from pathlib import Path
@@ -43,9 +43,7 @@ def run_benchmark(corpus: str = "0.1.0") -> float:
     # force a clean slate
     cache.clear()
     # corpus was removed in https://github.com/amperser/proselint/pull/186
-    log.error(
-        "Benchmarking the corpus does not work for the time being -> will use demo",
-    )
+    log.error("Corpus is unavailable for the time being -> using demo")
     corpus_path = demo_file.parent  # proselint_path.parent / "corpora" / corpus
     for _type in ["uncached", "cached"]:
         start = time.time()
@@ -93,10 +91,14 @@ def run_benchmark(corpus: str = "0.1.0") -> float:
     help="Override config to change format.",
 )
 @click.option("--dump-config", is_flag=True, help="Prints current config.")
-@click.option("--dump-default-config", is_flag=True, help="Prints default config.")
+@click.option(
+    "--dump-default-config", is_flag=True, help="Prints default config."
+)
 @click.option("--version", is_flag=True)
-@click.argument("paths", nargs=-1, type=click.Path(exists=True, resolve_path=True))
-def proselint(  # noqa: PLR0913, PLR0917, C901
+@click.argument(
+    "paths", nargs=-1, type=click.Path(exists=True, resolve_path=True)
+)
+def proselint(  # noqa: PLR0912, PLR0913, PLR0917, C901
     paths: Union[list[Path], Path, None],
     config: Optional[Path] = None,
     clean: bool = False,  # TODO: this should be a subcommand
@@ -105,7 +107,7 @@ def proselint(  # noqa: PLR0913, PLR0917, C901
     benchmark: bool = False,
     demo: bool = False,
     dump_config: bool = False,  # TODO: this should be a subcommand
-    dump_default_config: bool = False,  # TODO: this should be a switch in dump-cmd
+    dump_default_config: bool = False,  # TODO: this should be a switch in dump
     version: bool = False,
 ):
     """Create the CLI for proselint, a linter for prose."""

proselint/config_paths.py

@@ -10,7 +10,7 @@
 
 def _get_xdg_path(variable_name: str, default_path: Path) -> Path:
     _value = os.environ.get(variable_name)
-    if _value is None or _value == "":
+    if not _value:
         return default_path
     return Path(_value)
 

proselint/memoizer.py

@@ -3,7 +3,7 @@
 import contextlib
 import functools
 import hashlib
-import pickle
+import pickle  # noqa: S403
 import shutil
 from concurrent.futures import Executor
 from concurrent.futures import Future
@@ -46,7 +46,8 @@ def __new__(cls, *_args: tuple, **_kwargs: Unpack[TypedDict]) -> Self:
     def __init__(self) -> None:
         self._data: dict[str, list[ResultLint]] = {}
         self._age: dict[str, int] = {}
-        self._ts_now: int = round(datetime.now().timestamp())
+        # TODO: make timezone aware, possibly use UTC?
+        self._ts_now: int = round(datetime.now().timestamp())  # noqa: DTZ005
         self._modified: bool = False
         self.fname2key: dict[str, str] = {}
 
@@ -67,7 +68,8 @@ def __del__(self):
         Cache._instance = None
 
     def __getitem__(self, key: str) -> list[ResultLint]:
-        """Allows dict access -> instance["key"], in addition to instance.data["key"]"""
+        """Allows direct dict access.
+        This allows instance["key"], in addition to instance.data["key"]"""
         return self._data[key]
 
     def __setitem__(self, key: str, value: list[ResultLint]):
@@ -109,7 +111,11 @@ def from_file(cls) -> Self:
                 data = pickle.load(fd, fix_imports=False)  # noqa: S301
                 # TODO: consider replacing pickle with something faster
                 # only restore if data fits and package-version matches
-                if isinstance(data, list) and len(data) >= 3 and data[2] == version:
+                if (
+                    isinstance(data, list)
+                    and len(data) >= 3
+                    and data[2] == version
+                ):
                     _inst._data = data[0]  # noqa: SLF001
                     _inst._age = data[1]  # noqa: SLF001
                     log.debug(" -> found & restored cache")
@@ -129,7 +135,9 @@ def calculate_key(text: str, checks: list[Callable]) -> str:
 
         text_hash = hashlib.sha224(text.encode("utf-8")).hexdigest()[:50]
         chck_list = [f"{c.__module__}.{c.__name__}" for c in checks]
-        chck_hash = hashlib.sha224(" ".join(chck_list).encode("utf-8")).hexdigest()[:10]
+        chck_hash = hashlib.sha224(
+            " ".join(chck_list).encode("utf-8")
+        ).hexdigest()[:10]
         # NOTE: Skip hashing config!
         #   -> Valid assumption that (current) config has
         #      no influence on result below this level

proselint/score.py

@@ -5,7 +5,7 @@
 
 import os
 import re
-import subprocess
+import subprocess  # noqa: S404
 from pathlib import Path
 
 from .config_paths import proselint_path