eskimo

<http://lists.apple.com/archives/macnetworkprog/2015/Jun/msg00025.html>Share and Enjoy--Quinn The Eskimo!Apple Developer Relations, Developer Technical Support, Core OS/Hardware

I generally do this with a custom build setting. Put the version number in that build setting and then reference it from your various Info.plist's via $(MY_BUILD_SETTING_NAME).Share and Enjoy--Quinn The Eskimo!Apple Developer Relations, Developer Technical Support, Core OS/Hardware

The most likely cause of this is file protection. I posted about this in a thread in the old DevForums.<https://devforums.apple.com/message/807457#807457>Share and Enjoy--Quinn The Eskimo!Apple Developer Relations, Developer Technical Support, Core OS/Hardware

AFAIK Multipeer Connectivity has no new features in iOS 9 / OS X 10.11, which is why there was no session on it this year.Share and Enjoy--Quinn The Eskimo!Apple Developer Relations, Developer Technical Support, Core OS/Hardware

> I'd like to open an issue with the backend provider but there doesn't> seem to be any documentation on the settings they'd need to support> for ATS to work. Has anyone found any documentationProper documentation for this is in the works but you can find a basic outline in WWDC 2015 Session 706 Security and Your Apps.<https://developer.apple.com/videos/>> or know of a way to get more info from the SSL error that comes up> when you fail to connect because of ATS?Right now the error you get back is very generic; that's expected to improve in future seeds.Share and Enjoy--Quinn The Eskimo!Apple Developer Relations, Developer Technical Support, Core OS/Hardware

In which context are you trying to do this? The ability to examine the address space of other processes is getting increasingly restricted over time (for obvious security reasons). If you're creating a shrink wrap product that you intend to ship to end users, you'll need to consider the potential medium-term compatibility issues and write a bunch of complex code to escalate your privileges appropriately. OTOH if you're just hacking away at something locally, it's relatively easy to get this working.Share and Enjoy--Quinn The Eskimo!Apple Developer Relations, Developer Technical Support, Core OS/Hardware

iOS does not provide any public APIs for this. You'll need to write or acquire the necessary code. You can do it with a full-featured security toolkit, like OpenSSL, or just hack something together (this is a relatively simple use of ASN.1).What I recommend is that you restructure your system so that you don't require this. Most folks who have this problem are getting the modulus and exponent from a server that has a full-featured security toolkit (OpenSSL, something Java-ish, and so on). In that case you can have the server build the public key and send you that. Or, to make things even easier, have the server put the public key into a certificate, which is trivial to work with on iOS.Share and Enjoy--Quinn The Eskimo!Apple Developer Relations, Developer Technical Support, Core OS/Hardware

> Could this be related to concurrent calls to> SecKeychainUnlock/SecKeychainCreate on the same keychain?It certainly could be. Normally the only shared keychains on the system (things like the System keychain) are relatively static (they don't come and go, or lock and unlock) so you don't exercise this case.> Any recomendations on how to handle this execenario in wich several> servers use a shared keychain, and any of the could unlock/create it ?I think that once the keychain is created and open you should be OK reading the keychain concurrently. Working on that theory the obvious next step is to move the code that creates and unlocks the keychain to a central service that each of your main servers talks to via XPC.Share and Enjoy--Quinn The Eskimo!Apple Developer Relations, Developer Technical Support, Core OS/Hardware

I didn't know that Directory Utility had a plug-in architecture. Where's the documentation for that?Share and Enjoy--Quinn The Eskimo!Apple Developer Relations, Developer Technical Support, Core OS/Hardware

Honestly, disabling rootless^H^H^H^H^H^H^H^H system integrity protection seems like the wrong answer here. In situations like this, where I need to test on multiple versions of the OS, I set up VMs for each OS version I support and then run my tests on the VMs. That doesn't require disabling SIP /and/ gives you a better representation of the OS as it's deployed on user systems.Share and Enjoy--Quinn The Eskimo!Apple Developer Relations, Developer Technical Support, Core OS/Hardware

This gotcha has bitten folks before. The U+00A0 is a non-breaking space, which makes sense because you don't want the value and the currency symbol to get split across a line break. You should be able to fix your test by putting a non-breakpoint space in your expectedPriceTag string. Use Swift's u{xxxx} notation.btw Your test seems locale sensitive, which could be a problem if someone else runs your test. Even if you're limited to countries that use the Euro, not all countries use 50,00 €. For example, in Ireland this would be €50.00.Share and Enjoy--Quinn The Eskimo!Apple Developer Relations, Developer Technical Support, Core OS/Hardware

In general Wi-Fi should come up during pre-login. There are some situations where it can't--for example, if it requires authentication that can only be done by the user--but in a typical network such user action isn't necessary and Wi-Fi should come up by itself [1].Remember that your pre-login agent will run /very/ early, so you may just have a race condition where your agent is launching before Wi-Fi is up. Your agent should be able to respond to the network coming and going in general, and that includes having the network come up after the agent has started. Reachability can help with this.Share and Enjoy--Quinn The Eskimo!Apple Developer Relations, Developer Technical Support, Core OS/Hardware[1] In a typical home network, using WPA2, the network credentials are store in the system keychain so they are accessible even when no user is logged in.

You're correct that there's no general-purpose API for scanning for Wi-Fi networks.With regards Chromecast, I'm not in a position to reverse engineer other developer's apps on your behalf but there are special-purpose solutions that it may be taking advantage of. For example, MFi accessories can be configured using EAWiFiUnconfiguredAccessoryBrowser.Share and Enjoy--Quinn The Eskimo!Apple Developer Relations, Developer Technical Support, Core OS/Hardware

I think it's safe to say that your choice of language won't have any impact on whatever problems you're having with Multipeer Connectivity.Also, the lack of new features does not imply that no work has been done on the framework; it's common for frameworks to get substantial improvements under the hood without having any impact on their public API [1]. * * *In my experience a lot of folks using Multipeer Connectivity run into problems because their network architecture doesn't match the architecture that Multipeer Connectivity was designed to support. Specifically, if you're building an app that has one server and lots of clients, it's better to use standard networking APIs (Bonjour + TCP/IP) than Multipeer Connectivity because Multipeer Connectivity is really designed to support a set of cooperating peers.It's also important to realise that Multipeer Connectivity is not the only way to get to peer-to-peer Wi-Fi and Bluetooth. The standard networking APIs work with those interfaces just fine.Share and Enjoy--

> 1) I would like to monitor the telephony state of the device in order> to be able to reject incoming VoIP call when the user is already> talking on a telephony call.That should be feasible. When your app resumes in order to process the ring indication (that it got via the old VoIP API or via the silent push you get from PushKit), create and start a new instance of CTCallCenter. If that shows that the user is on a call, don't post a ring indication to the user but rather communicate the busy state back to your VoIP infrastructure.> 2) I would like to reject an incoming telephony call (with a busy> tone) if the user is already talking on a VoIP call.That's not possible on the current system. CT calls always take precedence over VoIP calls. If you'd like to see that change in the future, I encourage you to file a bug describing your requirements.<https://developer.apple.com/bug-reporting/>Please post your bug number, just for the record.Share and Enjoy--Quinn The Eskimo!Apple Deve