Skip to content

Instantly share code, notes, and snippets.

@gilesc
Created May 24, 2021 22:04
Show Gist options
  • Save gilesc/266d6e55197c275fecfc48186a1dbdb0 to your computer and use it in GitHub Desktop.
Save gilesc/266d6e55197c275fecfc48186a1dbdb0 to your computer and use it in GitHub Desktop.
import atexit
import codecs
import functools
import gzip
import os.path
import shutil
import sys
import tempfile
import urllib.request
import diskcache
EXPIRE_TIME = 99999999999
DISK_CACHE_PATH = os.path.join(os.path.dirname(__file__), "lfcache")
cache = diskcache.FanoutCache(DISK_CACHE_PATH, size_limit=int(500e9),
cull_limit=0, timeout=EXPIRE_TIME)
memoize = functools.partial(cache.memoize, tag="memoize", expire=EXPIRE_TIME)
def download(URL):
if not URL in cache:
print(f"Cache miss: {URL}", file=sys.stderr)
with tempfile.TemporaryDirectory() as wd:
def cleanup():
if os.path.exists(wd):
shutil.rmtree(wd)
atexit.register(cleanup)
target = os.path.join(wd, "download.bin")
urllib.request.urlretrieve(URL, target)
with open(target, "rb") as h:
cache.set(URL, h, read=True, tag="download", expire=EXPIRE_TIME)
o = cache.get(URL, read=True)
if URL.endswith(".gz"):
return gzip.GzipFile(fileobj=o, mode="rb")
return o
def download_text(URL, encoding="utf-8"):
handle = download(URL)
reader_class = codecs.getreader(encoding)
return reader_class(handle)
def download_table(uri, **kwargs):
import pandas as pd
h = download(uri)
return pd.read_table(h, **kwargs)
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment