Closed ttung closed 5 years ago
Merging #112 into master will decrease coverage by
0.08%
. The diff coverage is66.66%
.
@@ Coverage Diff @@
## master #112 +/- ##
==========================================
- Coverage 75.49% 75.41% -0.09%
==========================================
Files 29 29
Lines 906 907 +1
==========================================
Hits 684 684
- Misses 222 223 +1
Impacted Files | Coverage Δ | |
---|---|---|
slicedimage/url/resolve.py | 72.09% <66.66%> (-1.72%) |
:arrow_down: |
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 ba15786...c37a914. Read the comment docs.
When extracting the path component from urllib.parse.urlparse, the path always starts with '/'. When tokenizing the url into its components, it becomes '/', 'c:', 'windows', 'system32'. Unfortunately, when combined into a WindowsPath, it becomes 'c:windows/system32', which honestly makes no sense to me.
The approach here is to strip the prefix '/', and when we do that, WindowsPath can parse the posix path correctly.
Thanks to @dany-fu for reporting this.
Test plan: loaded up a very empty experiment.json on a Windows VM.