Fix: Use urllib to combine urls (#5)

[erenes]

No longer does the lack of slash at the end of api- or tus-url cause an issue while using the cli.

[erenes]

At least with this one I was able to trick all the checks into passing ;-)

[LordAro]

Not a huge issue, but the email address used in the commit is not attached to your GH account - you may wish to fix this (either by redoing the commits with a different email address or adding that email address to GH

[Xaroth]

One thing to keep in mind is that urljoin has a different behavior than what was used.

Current:

>>> path = "/bla/something/"
>>> f"https://google.com/test{path}"
'https://google.com/test/bla/something/'

New:

>>> urljoin("https://google.com/test", "/bla/something/")
'https://google.com/bla/something/'
>>>

This effectively breaks use-cases where tus_url/api_url ends on a directory rather than just a domain.

[TrueBrain]

What I did for other places, what might be a way to avoid the weird urllib.parse functions, is that I check if an URL ends with a / when loading the setting, and add it if it is not there. But I haven't really looked into this yet, so this is just a random remark that might completely miss the goal :D

[erenes]

What I did for other places, what might be a way to avoid the weird urllib.parse functions, is that I check if an URL ends with a / when loading the setting, and add it if it is not there. But I haven't really looked into this yet, so this is just a random remark that might completely miss the goal :D

That's plan B, I'll look into urllib.parse pecularities and if I can't figure out a way to support paths with a directory I'll go for that approach. Also I'll squash the commits and fix my e-mail address :)

[erenes]

Based on these tests, I have updated the code to always add a trailing slash to the url, and remove the slash-prefix in all calls to session.x methods:

>>> import urllib.parse
>>> urllib.parse.urljoin('http://openttd.org/xx/yy/zz/', 'blaaaaaatt')
'http://openttd.org/xx/yy/zz/blaaaaaatt'
>>> urllib.parse.urljoin('http://openttd.org/xx/yy/zz', 'blaaaaaatt')
'http://openttd.org/xx/yy/blaaaaaatt'
>>> urllib.parse.urljoin('http://openttd.org/xx/yy/zz', '/blaaaaaatt')
'http://openttd.org/blaaaaaatt'
>>> urllib.parse.urljoin('http://openttd.org/xx/yy/zz///', '/blaaaaaatt')
'http://openttd.org/blaaaaaatt'
>>> urllib.parse.urljoin('http://openttd.org/xx/yy/zz///', 'blaaaaaatt')
'http://openttd.org/xx/yy/zz/blaaaaaatt'

commit coming soon :)

[TrueBrain]

You now basically did both solutions, where either one works too :D That is a bit too much in my opinion.

In general, everyone tries to avoid urllib as it is just horrible in so many ways. One of the oldest Python libraries, and it really shows. If possible, I would be really happy if we can avoid using it.

I think that the change you made with the slashes is sufficient. That means urllib is basically not doing anything anymore, as the links are (after my suggestion) always correct already. Also saves some CPU cycles \o/ (no, that is a joke, that is never a goal here :D)

(and in case you wonder, I consider it horrible because (personal opinion incoming): include-path is weird, you have to include urllib.path, it does magic things with your strings without being really verbose why (which is shown by the fact you had to experiment with it to know what it did :P), and most of all: in 90% of the cases it can be avoided :D )

[TrueBrain]

I would do:

if not api.url.endswith("/"):
  api_url += "/"

or something.