-
Notifications
You must be signed in to change notification settings - Fork 1.9k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Publish a new Kivy Launcher for Python 3 #1638
Comments
New launcher is located here and the script for building setup.py. We need to find the key for signing the release. Do we want to have both Python 2 and Python 3 launchers or just Python 3? Also, PyJNIus should have a release before this because there are important things fixed (ref counts, conversion between Python and Java, etc). |
I'd be happy to go with just Python 3, since it keeps the play store listings simple. Does the launcher need any testing I could do? |
I'm not sure there is any testing for the launcher necessary per se. It pretty much just runs a |
I feel like this issue should be tracked in https://github.com/kivy/kivy-launcher/issues but I couldn't transfer it to the kivy-launcher project somehow |
Does this thing replace the It would be neat because then removing |
The missing-license topic apart (that should be fixable if the actual source is tracked down), why would an LGPL file in the source be dangerous as long as it's properly declared? Python does not link, so even GPL requirements would only pertain to that file and not the "whole program", and with LGPL, it's even made explicit that this can be mixed with any license as long as the file stays LGPL. |
@chrysn I am not a lawyer, did you ask yours or did you just make that up? Again I am not a lawyer so this is probably all nonsense, BUT:
Says who? the LGPL is written with C/C++ linking in mind, and while I get the argument python importing is similar to dynamic linking, did you actually ask someone competent about that? because just relying on a hunch here is a horrible idea In overall the LGPL is just quite complicated to comply with, and on top of things the files are NOT properly declared right now. So I stick with my opinion that the only really sane option is to just rip them out. |
Neither am I, nor any of the options; my comment came from established practice in other projects, eg. what Debian accepts to be present in its binary packages. I'd be happy to run through all those arguments, but before we waste both your and my time, is LGPL compatibility really the showstopper? And if so, do you have pointers to the possibly-offending files (for I didn't find them on a quick search but probably lacked the right parameters) to clarify their license? For if the renpy code should go out for technical reasons, or it turns out it's not as LGPL licensed as we imagine, continuing this would be interesting but pointless. |
I did dig deeper into this and forgot to update: it's not actually LGPL but the same license as p4a, so forget all I said (wiki just mentioned LGPL for renpy because of other things)
But yes, LGPL for apk has the risk of being a huge showstopper in general, or not what do I know (not a lawyer) but just see above. LGPL is not that easy to fulfill for combined binary packages
…On April 8, 2019 3:32:43 PM GMT+02:00, chrysn ***@***.***> wrote:
> I am not a lawyer, did you ask yours or did you just make that up?
Neither am I, nor any of the options; my comment came from established
practice in other projects, eg. what Debian accepts to be present in
its binary packages.
I'd be happy to run through all those arguments, but before we waste
both your and my time, is LGPL compatibility really the showstopper?
And if so, do you have pointers to the possibly-offending files (for I
didn't find them on a quick search but probably lacked the right
parameters) to clarify their license? For if the renpy code should go
out for technical reasons, or it turns out it's not as LGPL licensed as
we imagine, continuing this would be interesting but pointless.
--
You are receiving this because you commented.
Reply to this email directly or view it on GitHub:
#1638 (comment)
|
I did dig deeper into this and forgot to update: it's not actually
LGPL but the same license as p4a, so forget all I said (wiki just
mentioned LGPL for renpy because of other things)
Thanks for the update, good to have that cleared.
But yes, LGPL for apk has the risk of being a huge showstopper in
general, or not what do I know (not a lawyer) but just see above. LGPL
is not that easy to fulfill for combined binary packages
I can see it is for nonfree apps. For apps that have source (like the
launcher), there's no need to have the library swappable as a dynamic
library (source is OK too per 4.d.0), and the referencing the copyright
statement (4.c) is only necessary if there is a copyright section in the
combined work. (But even my car does that :-) )
|
The launcher doesn't link its own source inside itself, does it? So yes it WOULD be (if it was LGPL) a problem even for that, because people will distribute it without realizing that. The LGPL is really more complicated than people make it, it's only not a problem if you ALWAYS distribute with full source or instructions to get it, which almost nobody other than linux distributions will do per default in all cases. Unless I am wrong of course, since, not a lawyer
I'm just saying, maybe be careful suggesting LGPL compliance is trivial, there are many potential pitfalls and this can hurt people later if they're not careful
|
Closing as stale |
Issue created to mark this goal for the 2019 Belgium hackathon.
The text was updated successfully, but these errors were encountered: