v0.6.2

[stolarczyk]

fix #32

[codecov]

Codecov Report

Merging #33 (abe2ef3) into master (771ffbf) will increase coverage by 0.78%.
The diff coverage is 72.54%.

@@            Coverage Diff             @@
##           master      #33      +/-   ##
==========================================
+ Coverage   85.58%   86.36%   +0.78%     
==========================================
  Files          18       19       +1     
  Lines         853      851       -2     
==========================================
+ Hits          730      735       +5     
+ Misses        123      116       -7     

Impacted Files	Coverage Δ
setup.py	0.00% <0.00%> (ø)
ubiquerg/collection.py	56.89% <50.00%> (ø)
ubiquerg/cli_tools.py	82.73% <68.75%> (ø)
ubiquerg/files.py	75.40% <80.00%> (ø)
tests/test_web.py	100.00% <100.00%> (ø)
ubiquerg/_version.py	100.00% <100.00%> (ø)
ubiquerg/environment.py	100.00% <100.00%> (ø)
ubiquerg/paths.py	69.56% <100.00%> (-1.27%)	⬇️
ubiquerg/web.py	100.00% <100.00%> (+22.22%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 771ffbf...abe2ef3. Read the comment docs.

[nsheff]

this switches to a regex-based approach. this will be much slower, but will probably be more what we're looking for... @vreuter any objections?

[stolarczyk]

That's an overkill for such insignificant problem... but ran couple of tests and you're right, regex is slower since it depends on the input length (O(n) time). The urlparse approach runs in O(1) time

import time
from urllib.parse import urlparse 
from ubiquerg import is_url

def check_time(fun, n=10000):
    results = []
    for _ in range(1, n):
        start = time.time()
        fun(string)
        end = time.time() 
        results.append((end - start))
    return (sum(results) / n) * 1000


def is_url_old(string): 
    try:
        result = urlparse(string)
        return all([result.scheme, result.netloc, result.path])
    except:
        return False

strings = [f"http://www.g{'o' * 10000}gle.com'", f"http://www.g{'o' * 5000}gle.com'", "http://www.google.com", ""]
for string in strings:
    print(f"new time: {check_time(fun=is_url)}ms")
    print(f"old time: {check_time(fun=is_url_old)}ms\n")

~ python3 ~/Desktop/check_time.py 
new time: 1.1539672136306762ms
old time: 0.003525686264038086ms

new time: 0.6307248592376709ms
old time: 0.003618669509887695ms

new time: 0.004984068870544433ms
old time: 0.0035085678100585938ms

new time: 0.0031899452209472657ms
old time: 0.0036333560943603515ms

[vreuter]

🤩

[vreuter]

I gave an optional suggestion as a comment but looks good, w/e you guys think as far as tradeoff b/w runtime vs. behavior of the implementation.