Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Python grapples with Apple App Store rejections

lwn.net

Post reply on HN
Print view View original

61–70 of 118 posts

Re: Python grapples with Apple App Store rejections

#61

Post by modeless »

heavyset_go wrote: post #10

It's not just Apple that pulls shenanigans like this. Try building a Python app with PyInstaller while you have Windows Defender live scanning on, which is the default setting. You won't even be able to compile a binary without Defender preventing you from doing so. Similarly, try running the binary produced by PyInstaller with Windows Defender on. Defender will say it's malicious and won't run it. It's a bit dystopi…

I built PyInstaller binaries on Windows a few months ago and had no trouble with Windows Defender at all.

Re: Python grapples with Apple App Store rejections

#62

Post by justinclift »

yashg wrote: post #55
josephcsible wrote: post #11

Earlier quoted context omitted.

It's not just Python apps. It's anything by small-time developers without expensive certificates. I once used MSVC to compile a C program that was little more than a "Hello, World", and Defender called it the Win32/Wacatac Trojan.

The code signing certificates that MS requires are ridiculously expensive. It's a cartel of certificate isseuers. It's a downright robbery. We need something similar to Let's Encrypt for code signing.

SignPath gives free certs to OSS projects:

https://signpath.org/about/

We (sqlitebrowser.org) have recently started using them for signing our Windows builds.

Re: Python grapples with Apple App Store rejections

#63

Post by bastawhiz »

denkmoon wrote: post #57
amelius wrote: post #29

Can we have Separation of Powers on our digital platforms? It is pretty shitty that the one who sells phones also determines what goes on them.

I choose Apple to be the custodian of apps that go on my phone. There is a platform where the user is the one that determines what goes on their phone, it's called Android.

Sure, so why should everyone else with an iOS device have to make the same choice?

Re: Python grapples with Apple App Store rejections

#64

Post by gorgoiler »

The offending string is only there because Python’s urllib has a hard-coded list of schemes which use a hostname component or “netloc”. It’s fine for that list to contain known schemes from RFCs. Anything else — including proprietary third party schemes — should just use a heuristic.

The list is called uses_netloc and is used to help parse the user@host:port part of https, ftp, etc. domains. It’s this list of schemes that includes the forbidden string itms-services, used for Apple’s proprietary iTunes software.

The only code that needs this is urlunsplit and urljoin. If your parsed URL has a netloc then the list isn’t even relevant — if you have a netloc then you are assumed to be in uses_netloc.

This all seems like a much more sensible approach than trying to selectively include or exclude naughty strings from the source code, per some corporation’s passive aggressive demands.

Re: Python grapples with Apple App Store rejections

#65

Post by patrick451 »

heavyset_go wrote: post #10

It's not just Apple that pulls shenanigans like this. Try building a Python app with PyInstaller while you have Windows Defender live scanning on, which is the default setting. You won't even be able to compile a binary without Defender preventing you from doing so. Similarly, try running the binary produced by PyInstaller with Windows Defender on. Defender will say it's malicious and won't run it. It's a bit dystopi…

This is reason enough not to use either windows or macos. Calling these programs malicious is an outright lie. They have not proven this. At most, they can say the program is untrusted or unverified. But calling it malicious is a falsehood and therefore a breach of trust.

Re: Python grapples with Apple App Store rejections

#66

Post by patrick451 »

heavyset_go wrote: post #46
shrimp_emoji wrote: post #37

Earlier quoted context omitted.

Windows isn't a platform for developers. It's a platform for normie consoomers. Isn't that obvious? If you want an engineering OS, use GNU/Linux.

Until you can run a Windows-free build system with WINE (there are a few reported blockers, several others and I have tried) and PyInstaller, cross-platform apps will require developers to compile their Windows ports on Windows itself. Windows is where the users are. Not targeting it is a bad financial decision.

To hell with cross platform. Developing for windows perpetuates a harmful ecosystem. It's really no different than selling ammo to the Sineloa drug cartel.

Re: Python grapples with Apple App Store rejections

#67

Post by denkmoon »

bastawhiz wrote: post #63
denkmoon wrote: post #57

Earlier quoted context omitted.

I choose Apple to be the custodian of apps that go on my phone. There is a platform where the user is the one that determines what goes on their phone, it's called Android.

Sure, so why should everyone else with an iOS device have to make the same choice?

Those with an iOS device can make the exact same choice - to get an Android device. If I don't want Apple to be my custodian, that's the choice I have.

I don't understand why people feel the need to force Apple into a specific strategy when they are not a monopoly. They aren't the only game in town, you don't _need_ an iPhone. Every single person who has an iPhone has chosen to have that device.

Perhaps the reason people don't like the idea that if you want choice you choose the platform that gives you choice, is because Android is a steaming pile of garbage.

Re: Python grapples with Apple App Store rejections

#68

Post by pfraze »

patrick451 wrote: post #66
heavyset_go wrote: post #46

Earlier quoted context omitted.

Until you can run a Windows-free build system with WINE (there are a few reported blockers, several others and I have tried) and PyInstaller, cross-platform apps will require developers to compile their Windows ports on Windows itself. Windows is where the users are. Not targeting it is a bad financial decision.

To hell with cross platform. Developing for windows perpetuates a harmful ecosystem. It's really no different than selling ammo to the Sineloa drug cartel.

It’s a little bit different.

Re: Python grapples with Apple App Store rejections

#69

Post by denkmoon »

areoform wrote: post #59
denkmoon wrote: post #57

Earlier quoted context omitted.

I choose Apple to be the custodian of apps that go on my phone. There is a platform where the user is the one that determines what goes on their phone, it's called Android.

While I am glad that you have the choice to be a digital serf - paying your lord for using land (devices), I would much rather prefer the choice to not be one. That way, you can opt into digital serfdom and people like me can opt out. - Serf here isn’t pejorative, but descriptive. Apple is behaving like a feudal lord. Down to the claims of protecting the land and arbitrating disputes.

Why do you act like Android doesn't exist? You _can_ opt out of "digital serfdom".

Re: Python grapples with Apple App Store rejections

#70

Post by Daishiman »

denkmoon wrote: post #67
bastawhiz wrote: post #63

Earlier quoted context omitted.

Sure, so why should everyone else with an iOS device have to make the same choice?

Those with an iOS device can make the exact same choice - to get an Android device. If I don't want Apple to be my custodian, that's the choice I have. I don't understand why people feel the need to force Apple into a specific strategy when they are not a monopoly. They aren't the only game in town, you don't _need_ an iPhone. Every single person who has an iPhone has chosen to have that device. Perhaps the reason pe…

They're not a monopoly; they're a duopoly. It is a marginally better position but I have no idea how you justify to yourself handing over a consumer's power to the largest tech company in the world.
Post reply on HN