import math
from numbers import Number
-from typing import Generic, Iterable, List, Optional, Tuple, TypeVar
+from typing import Dict, Generic, Iterable, List, Optional, Tuple, TypeVar
-T = TypeVar("T", bound=Number)
+T = TypeVar("T", int, float)
+Bound = int
+Count = int
class SimpleHistogram(Generic[T]):
-
# Useful in defining wide open bottom/top bucket bounds:
POSITIVE_INFINITY = math.inf
NEGATIVE_INFINITY = -math.inf
- def __init__(self, buckets: List[Tuple[T, T]]):
+ def __init__(self, buckets: List[Tuple[Bound, Bound]]):
from math_utils import RunningMedian
- self.buckets = {}
+ self.buckets: Dict[Tuple[Bound, Bound], Count] = {}
for start_end in buckets:
if self._get_bucket(start_end[0]) is not None:
raise Exception("Buckets overlap?!")
self.buckets[start_end] = 0
- self.sigma = 0
- self.median = RunningMedian()
- self.maximum = None
- self.minimum = None
- self.count = 0
+ self.sigma: float = 0.0
+ self.median: RunningMedian = RunningMedian()
+ self.maximum: Optional[T] = None
+ self.minimum: Optional[T] = None
+ self.count: Count = 0
@staticmethod
def n_evenly_spaced_buckets(
min_bound: T,
max_bound: T,
n: int,
- ) -> List[Tuple[T, T]]:
- ret = []
+ ) -> List[Tuple[int, int]]:
+ ret: List[Tuple[int, int]] = []
stride = int((max_bound - min_bound) / n)
if stride <= 0:
raise Exception("Min must be < Max")
- for bucket_start in range(min_bound, max_bound, stride):
+ imax = math.ceil(max_bound)
+ imin = math.floor(min_bound)
+ for bucket_start in range(imin, imax, stride):
ret.append((bucket_start, bucket_start + stride))
return ret
- def _get_bucket(self, item: T) -> Optional[Tuple[T, T]]:
+ def _get_bucket(self, item: T) -> Optional[Tuple[int, int]]:
for start_end in self.buckets:
if start_end[0] <= item < start_end[1]:
return start_end
all_true = all_true and self.add_item(item)
return all_true
- def __repr__(self, *, width: int = 80, label_formatter: str = None) -> str:
+ def __repr__(self, *, width: int = 80, label_formatter: str = '%d') -> str:
from text_utils import bar_graph
+ txt = ""
max_population: Optional[int] = None
for bucket in self.buckets:
pop = self.buckets[bucket]
last_bucket_start = bucket[0] # beginning of range
if max_population is None or pop > max_population:
max_population = pop # bucket with max items
-
- txt = ""
if max_population is None:
return txt
- max_label_width = None
- lowest_start = None
- highest_end = None
+ max_label_width: Optional[int] = None
+ lowest_start: int = None
+ highest_end: int = None
for bucket in sorted(self.buckets, key=lambda x: x[0]):
start = bucket[0]
if lowest_start is None:
end = bucket[1]
if highest_end is None or end > highest_end:
highest_end = end
- if label_formatter is None:
- if type(start) == int and type(end) == int:
- label_formatter = '%d'
- elif type(start) == float and type(end) == float:
- label_formatter = '%.2f'
- else:
- label_formatter = '%s'
label = f'[{label_formatter}..{label_formatter}): ' % (start, end)
label_width = len(label)
if max_label_width is None or label_width > max_label_width:
max_label_width = label_width
if start == last_bucket_start:
break
+ assert max_label_width
+ assert lowest_start
+ assert highest_end
+
sigma_label = f'[{label_formatter}..{label_formatter}): ' % (
lowest_start,
highest_end,
import logging
import math
import sys
-from typing import List, NamedTuple, Optional
+from typing import Dict, Generator, List, NamedTuple, Optional
from ansi import fg, reset
def get_console_rows_columns() -> RowsColumns:
"""Returns the number of rows/columns on the current console."""
- from exec_utils import cmd_with_timeout
+ from exec_utils import cmd
try:
- rows, columns = cmd_with_timeout(
+ rows, columns = cmd(
"stty size",
timeout_seconds=1.0,
).split()
reset=reset(),
left_end="[",
right_end="]",
-) -> None:
+) -> str:
"""Returns a string containing a bar graph.
>>> bar_graph(0.5, fgcolor='', reset='')
return retval[1:]
-def generate_padded_columns(text: List[str]) -> str:
- max_width = defaultdict(int)
+def generate_padded_columns(text: List[str]) -> Generator:
+ max_width: Dict[int, int] = defaultdict(int)
for line in text:
for pos, word in enumerate(line.split()):
max_width[pos] = max(max_width[pos], len(word))