More cleanup.

[python_utils.git] / string_utils.py

diff --git a/string_utils.py b/string_utils.py
index 3ae9eeae96141fce9d4bb17516a5bace6ceb8b6f..4bec031738e989d10507992387e18aa47996da8e 100644 (file)
--- a/string_utils.py
+++ b/string_utils.py
@@ -40,7 +40,17 @@ import string
 import unicodedata
 import warnings
 from itertools import zip_longest
-from typing import Any, Callable, Dict, Iterable, List, Optional, Sequence, Tuple
+from typing import (
+    Any,
+    Callable,
+    Dict,
+    Iterable,
+    List,
+    Literal,
+    Optional,
+    Sequence,
+    Tuple,
+)
 from uuid import uuid4
 
 import list_utils
@@ -142,16 +152,16 @@ MARGIN_RE = re.compile(r"^[^\S\r\n]+")
 ESCAPE_SEQUENCE_RE = re.compile(r"\e\[[^A-Za-z]*[A-Za-z]")
 
 NUM_SUFFIXES = {
-    "Pb": (1024 ** 5),
-    "P": (1024 ** 5),
-    "Tb": (1024 ** 4),
-    "T": (1024 ** 4),
-    "Gb": (1024 ** 3),
-    "G": (1024 ** 3),
-    "Mb": (1024 ** 2),
-    "M": (1024 ** 2),
-    "Kb": (1024 ** 1),
-    "K": (1024 ** 1),
+    "Pb": (1024**5),
+    "P": (1024**5),
+    "Tb": (1024**4),
+    "T": (1024**4),
+    "Gb": (1024**3),
+    "G": (1024**3),
+    "Mb": (1024**2),
+    "M": (1024**2),
+    "Kb": (1024**1),
+    "K": (1024**1),
 }
 
 
@@ -1075,13 +1085,13 @@ def to_date(in_str: str) -> Optional[datetime.date]:
     """
     Parses a date string.  See DateParser docs for details.
     """
-    import dateparse.dateparse_utils as dp  # type: ignore
+    import dateparse.dateparse_utils as du
 
     try:
-        d = dp.DateParser()
+        d = du.DateParser()  # type: ignore
         d.parse(in_str)
         return d.get_date()
-    except dp.ParseException:
+    except du.ParseException:  # type: ignore
         msg = f'Unable to parse date {in_str}.'
         logger.warning(msg)
     return None
@@ -1094,10 +1104,10 @@ def valid_date(in_str: str) -> bool:
     import dateparse.dateparse_utils as dp
 
     try:
-        d = dp.DateParser()
+        d = dp.DateParser()  # type: ignore
         _ = d.parse(in_str)
         return True
-    except dp.ParseException:
+    except dp.ParseException:  # type: ignore
         msg = f'Unable to parse date {in_str}.'
         logger.warning(msg)
     return False
@@ -1110,9 +1120,9 @@ def to_datetime(in_str: str) -> Optional[datetime.datetime]:
     import dateparse.dateparse_utils as dp
 
     try:
-        d = dp.DateParser()
+        d = dp.DateParser()  # type: ignore
         dt = d.parse(in_str)
-        if type(dt) == datetime.datetime:
+        if isinstance(dt, datetime.datetime):
             return dt
     except ValueError:
         msg = f'Unable to parse datetime {in_str}.'
@@ -1208,7 +1218,7 @@ def sprintf(*args, **kwargs) -> str:
     return ret
 
 
-class SprintfStdout(object):
+class SprintfStdout(contextlib.AbstractContextManager):
     """
     A context manager that captures outputs to stdout.
 
@@ -1228,10 +1238,10 @@ class SprintfStdout(object):
         self.recorder.__enter__()
         return lambda: self.destination.getvalue()
 
-    def __exit__(self, *args) -> None:
+    def __exit__(self, *args) -> Literal[False]:
         self.recorder.__exit__(*args)
         self.destination.seek(0)
-        return None  # don't suppress exceptions
+        return False
 
 
 def capitalize_first_letter(txt: str) -> str:
@@ -1371,7 +1381,7 @@ def make_contractions(txt: str) -> str:
             for second in second_list:
                 # Disallow there're/where're.  They're valid English
                 # but sound weird.
-                if (first == 'there' or first == 'where') and second == 'a(re)':
+                if (first in ('there', 'where')) and second == 'a(re)':
                     continue
 
                 pattern = fr'\b({first})\s+{second}\b'
@@ -1457,11 +1467,11 @@ def shuffle_columns_into_list(
     # Column specs map input lines' columns into outputs.
     # [col1, col2...]
     for spec in column_specs:
-        chunk = ''
+        hunk = ''
         for n in spec:
-            chunk = chunk + delim + input_lines[n]
-        chunk = chunk.strip(delim)
-        out.append(chunk)
+            hunk = hunk + delim + input_lines[n]
+        hunk = hunk.strip(delim)
+        out.append(hunk)
     return out
 
 
@@ -1487,11 +1497,11 @@ def shuffle_columns_into_dict(
     # Column specs map input lines' columns into outputs.
     # "key", [col1, col2...]
     for spec in column_specs:
-        chunk = ''
+        hunk = ''
         for n in spec[1]:
-            chunk = chunk + delim + input_lines[n]
-        chunk = chunk.strip(delim)
-        out[spec[0]] = chunk
+            hunk = hunk + delim + input_lines[n]
+        hunk = hunk.strip(delim)
+        out[spec[0]] = hunk
     return out
 
 
@@ -1516,9 +1526,9 @@ def to_ascii(x: str):
     b'1, 2, 3'
 
     """
-    if type(x) is str:
+    if isinstance(x, str):
         return x.encode('ascii')
-    if type(x) is bytes:
+    if isinstance(x, bytes):
         return x
     raise Exception('to_ascii works with strings and bytes')