pferd/PFERD/utils.py

99 lines
2.6 KiB
Python
Raw Normal View History

2020-04-20 17:15:47 +02:00
"""
A few utility bobs and bits.
"""
2020-04-24 20:39:30 +02:00
import re
2020-04-20 03:54:47 +02:00
from pathlib import Path, PurePath
from typing import Optional, Tuple, Union
2020-04-20 18:38:18 +02:00
import bs4
2020-04-20 03:54:47 +02:00
import requests
2020-05-08 00:26:33 +02:00
from .progress import ProgressSettings, progress_for, size_from_headers
2020-04-24 20:39:30 +02:00
PathLike = Union[PurePath, str, Tuple[str, ...]]
2020-04-20 03:54:47 +02:00
2020-04-20 17:15:47 +02:00
2020-04-24 20:39:30 +02:00
def to_path(pathlike: PathLike) -> Path:
2020-04-25 19:59:58 +02:00
"""
Convert a given PathLike into a Path.
"""
2020-04-24 20:39:30 +02:00
if isinstance(pathlike, tuple):
return Path(*pathlike)
return Path(pathlike)
2020-04-20 14:29:28 +02:00
2020-04-24 20:39:30 +02:00
Regex = Union[str, re.Pattern]
2020-04-20 17:15:47 +02:00
2020-04-24 20:39:30 +02:00
def to_pattern(regex: Regex) -> re.Pattern:
2020-04-25 19:59:58 +02:00
"""
Convert a regex to a re.Pattern.
"""
2020-04-24 20:39:30 +02:00
if isinstance(regex, re.Pattern):
return regex
return re.compile(regex)
2020-04-20 14:29:28 +02:00
2020-04-20 18:38:18 +02:00
def soupify(response: requests.Response) -> bs4.BeautifulSoup:
2020-04-20 19:27:26 +02:00
"""
Wrap a requests response in a bs4 object.
"""
2020-04-20 18:38:18 +02:00
return bs4.BeautifulSoup(response.text, "html.parser")
2020-05-08 00:26:33 +02:00
def stream_to_path(
response: requests.Response,
target: Path,
progress_name: Optional[str] = None,
chunk_size: int = 1024 ** 2
) -> None:
2020-04-20 17:15:47 +02:00
"""
Download a requests response content to a file by streaming it. This
function avoids excessive memory usage when downloading large files. The
chunk_size is in bytes.
2020-05-08 00:26:33 +02:00
If progress_name is None, no progress bar will be shown. Otherwise a progress
bar will appear, if the download is bigger than an internal threshold.
2020-04-20 17:15:47 +02:00
"""
2020-04-20 19:27:26 +02:00
with response:
2020-05-08 00:26:33 +02:00
length = size_from_headers(response)
if progress_name and length and int(length) > 1024 * 1024 * 10: # 10 MiB
settings: Optional[ProgressSettings] = ProgressSettings(progress_name, length)
else:
settings = None
2020-04-24 20:39:30 +02:00
with open(target, 'wb') as file_descriptor:
2020-05-08 00:26:33 +02:00
with progress_for(settings) as progress:
for chunk in response.iter_content(chunk_size=chunk_size):
file_descriptor.write(chunk)
progress.advance(len(chunk))
2020-04-20 14:29:28 +02:00
def prompt_yes_no(question: str, default: Optional[bool] = None) -> bool:
2020-04-20 17:15:47 +02:00
"""
Prompts the user a yes/no question and returns their choice.
"""
2020-04-20 14:29:28 +02:00
if default is True:
prompt = "[Y/n]"
elif default is False:
prompt = "[y/N]"
else:
prompt = "[y/n]"
text = f"{question} {prompt} "
2020-04-20 17:15:47 +02:00
wrong_reply = "Please reply with 'yes'/'y' or 'no'/'n'."
2020-04-20 14:29:28 +02:00
while True:
response = input(text).strip().lower()
if response in {"yes", "ye", "y"}:
return True
2020-04-20 17:15:47 +02:00
if response in {"no", "n"}:
2020-04-20 14:29:28 +02:00
return False
2020-04-20 17:15:47 +02:00
if response == "" and default is not None:
return default
print(wrong_reply)