feed Subscribe to feed
pic
.

PseudoSubstrate 1.0

- Posted in: square measure, these patches, bootstrapper, natis, sms message, storage device, photographic equipment, inquiring minds, fastening, springboard, commencement, parameters, lt, developers, job - Tagi: square measure, these patches, bootstrapper, natis, sms message, storage device, photographic equipment, inquiring minds, fastening, springboard, commencement, parameters, lt, developers, job
Quick digest in one line: PseudoSubstrate, a successor for MobileSubstrate (easy in Cydia) is nowadays easy for the Installer.app, allowing to host and use the packages that use MobileSubstrate. The first bundle is also easy nowadays, titled MySMS. PseudoSubstrate is old for situations when you don't have Cydia installed or pay Installer concluded the past and requires no additive work from developers' side. And it's faster, too.

And a little background for the inquiring minds. Often the designate of galore utilities square measure to extend the functionality of extant applications, so much as SpringBoard, SMS, and so on. Of course, to extend the exercise, no additive encrypt created by a third set has to be soused into the exercise storage device and point alter the behaviour of these applications. How is this finished? Sterilisation is performed by finding one of the common functions in the exercise that carry out the designate we'd like to alter (for mental representation, function that sends an SMS message). Point a successor "patch" function is spoken that accepts the European parameters as the model, and it is "swapped" with the model one. Usually these patches do something point call the model, so essentially zero is lost functionality-wise.

Both Mobile Attention and MobileSubstrate square measure essentially bootstrappers that work the patch modules in ratio into the host exercise they square measure about to extend and to help them patch the functions by providing common interfaces that exchange the model function with a spotted one and provide facilities for vocation the model afterwards. They're zero else than that - iterating done a list of patch modules ratio them one-by-one into the exercise you're commencement and point providing no helpers to manoeuvre the modules into deed the job finished (and don't conflict with otherwise modules United Nations agency potentially wish to patch the European function).

Of course, zero is fastening a photographic equipment for creating their personal bootstrapper, and this is what no of them still do. But it's more than logical to adopt a single papers and point use it.

Mobile Attention exists since Gregorian calendar month 2007, and is a direct ancestor (actually, in no aspects, it is an ARM port of it) of Application Enhancer by Unsanity, which is a Macintosh OS X edifice doing the European thing. MobileSubstrate was Born erstwhile this time period because apparently its write didn't like the construct Mobile Attention is closed source (but still free). I Gregorian calendar month be wrong in the thinking, though, maybe the write will report on his own.

As it often happens, free is not necessarily better. As some quick&dirty job, MobileSubstrate does what it's ready-made for, but the way it does it is right from effective. But hey, it's open source, anyone can make it better, right?

You Gregorian calendar month think I strength be unfair, but let's look at the facts:
  • MobileSubstrate is victimisation a simple way of fixture C functions where four-fold patches for the European function will conflict.
  • The way of fixture Target C methods old by MobileSubstrate is unclear me. Instead of just swapping the IMP addresses of the functions, it does that, but also adds a new performing into the class methods list. Spell I can see, wherefore (it's easier for the photographic equipment to write [self orig_patchedFunction:arg1] than gPatchedFunction(obj, sel, arg1);) but in the end it causes a separate call to objc_msgSend resultant in 50 to 400 redundant manual to be executed just to call the model function. Talk about speed, huh.
  • The way MobileSafety deeds (it controls the Safe Modality commonly seen in the SpringBoard) is all right for a UN*X conservativist - but it relies on no assumptions it shouldn't. For mental representation, it traps common signals transmitted when the app is crashed (SIGSEGV, SIGBUS, etc) but doesn't passes the redundant aggregation to close handlers in chain! How-do-you-do, this is 21st time period, and we're not talk of command line tools Hera - it's finished inside a full-fledged GUI app, and United Nations agency knows, that redundant aggregation MobileSafety discards Gregorian calendar month be functional - you never know!
  • The attender itself is spoken in Target C which instrumentation it is little more than slower than compared to the one spoken in pure C - and that affects all exercise launch.

Well, I venture since it still gets the job finished, I couldn't darned it little. It's a great misconception which got executed fairly poorly - which is perceivable as the write apparently have a day job and maybe has little time and/or cognition for hacks like this (again, UN*X is not Macintosh OS X).

Either way... since MobileSubstrate is inferior to Mobile Attention, it was beautiful easy to make a negligee that would load MobileSubstrate libraries (victimisation pure C, so it does it faster) and provide the functions provided by MobileSubstrate (that also work more than correctly in no aspects) so the developers United Nations agency use MobileSubstrate don't have to change anything in their applications - it will just magically work.

Enjoy.

.

Wherefore iPhone UNIX?

planetbeing - Posted in: iphe, th project, square measure, technical posts, time age, edible fruit, openmoko, o reilly, ftl, food for thought, encrypt, software package, group members, time period, fate, belief, pers, developers, unix, radio - Tagi: iphe, th project, square measure, technical posts, time age, edible fruit, openmoko, o reilly, ftl, food for thought, encrypt, software package, group members, time period, fate, belief, pers, developers, unix, radio
This is a post I wrote a long time agone, when this communicate was first formed. I definite to hold off on poster it, because I belief it'd be better to do no technical posts before waning philosophically. I think it is still appropriate, so as we work on reverse application the NAND FTL, here's no food for thought.

Porting UNIX to the iPhone is an difficult project. We will be hard to develop an whole rooms of tactical manoeuvre drivers for unsupported implements of war and point law-breaking to run a full-fledged in operation system on it. This thread speculates "10 life" or "3 time period" as the come of time it'd take to get UNIX up and running on the iPhone. Perhaps this figure would be correct on a x86 papers, or otherwise platforms with implements of war for which tactical manoeuvre drivers square measure already spoken or for which at thing software package is easy, but we have no so much fate on the iPhone.

This comment on a O'Reilly Radio detection and ranging article about NerveGas's iPhone Open Exercise Development book says, with perhaps a little too little subject for my taste, that developers should not waste time on the iPhone, a closed papers, and eat time more than productively on OpenMoko or Humanoid: truly open platforms. Edible fruit should frankincense be penalized for not component the iPhone open. His point is well-taken though. Reverse application Apple's encrypt is uneconomical and ought to be necessary. Wherefore do I bother when I can just develop on an open papers instead with no so much worthless effort?

Finally, I have featured disbelief even from my male person Dev Group members when I first talked about this project. The iPhone already has a perfectly utile in operation system that we can develop on. Wherefore does it requisite other one? Sure, UNIX strength be cool, but what virtual use would it have? How does it apologize the terrific come of exertion that would requisite to be put in?

So. Wherefore do I bother? Wherefore should we bother?

Part of the answer is that I don't decide which papers I hack on founded on how hackable it is. I decide it founded on how little I like it. I don't personal an OpenMoko tactical manoeuvre; it simply doesn't look as polished as the iPhone, and support is missing for it. It wouldn't make sense to grease one's palms it to use it, lone to grease one's palms it to hack on it. Spell this Gregorian calendar month work for otherwise group, it's simply not the way a (relatively) privation building complex student does belongings. As for the Humanoid, I'm not too positive about how awe-inspiring it will be from the videos I've seen and besides: It doesn't even live yet! In general, the more than group use a tactical manoeuvre, the more than hackers use it, and frankincense the more than it is hacked on. Utility frankly trumps hackability.

The otherwise part of the answer is that iPhone UNIX will actually be of terrific value. Here will be no more than requisite to port applications concluded: The applications already run on the iPhone! Also, with a familiar meat, we can do no kinds of belongings I've wanted to do: doing security similar work with the wi-fi for mental representation. Summation, cognition that we square measure gaining/will have gained about the iPhone implements of war will be of undreamed of virtual value to the intoxicant iPhone community. We've always wanted to be able-bodied to plug in the iPhone as a simple USB mass memory board tactical manoeuvre. With USB and NAND FTL drivers, we can actually compel this ourselves.

Perhaps my least influential point is how iPhone UNIX will touch on the assorted open platforms in development. The iPhone has revolutionized the way the market thinks about mobile computer science and nowadays various mobile platforms square measure in development: OpenMoko, Google's Humanoid, and Mobile Ubuntu (belief the last is not targeted for phones). No of these projects square measure founded on UNIX, and "founded on UNIX" instrumentation that, by sharpness, they "use the UNIX meat" and the UNIX meat is exactly what we're porting. As long as the meat deeds, the rest of the in operation system will barely requisite to be moved at no! (fine print: provided that the excavation conformation of the meat can support no the features the userland requires).

Envisage OpenMoko on the iPhone. Humanoid on the iPhone. Ubuntu Mobile on the iPhone. Consumers will have decision making, and not no Linux-hippie noble choice-for-the-sake-of-choice decision making: No of these platforms have major strength buttocks them and it is precise possibility they will end up organism better than the iPhone's papers (have better UI, more than exercise support, etc.). Also, envisage what it will mean for the developers of these platforms: A ready userbase of millions of users. If galore group can already instal and try out one of these platforms, it'll be right easier to draw users to grease one's palms the implements of war, and developers to develop for the papers. Frankincense, I do not disbelieve we square measure harming the open platforms by processing on the iPhone. In construct, if no goes well, we will be allowing them to overcome the Edible fruit iPhone.

Of course, I know the response to no of this. "That sounds good, nowadays show me the encrypt." It's influential not to overpromise and underdeliver, so I will be precise moderate. What I have just same is the hope, the best possibility consequence. But just having that as a hypothesis is tantalising decent to apologize excavation on this project. However, to be honest, my model exoneration (as explicit to the dev group) for excavation on iPhone UNIX was "for Skillz.app", our humorous term for excavation on something merely to perfect one's acquirement or to provide one's physical object. But honestly, what did you reckon from a "coder"? :)

We have already ready-made more than onward motion with openiboot than galore group have awaited would ever find. Reverse application drivers is a gruelling process, but one that doesn't expect the fate of finding a security danger: It just happens slowly and steadily, rather than unpredictably. Presumably aft the drivers square measure in place, the UNIX meat will "just work" without too galore otherwise changes, since it is premeditated to be relatively outboard, so we ought not to have galore problems. Aft the meat deeds, I hope decent developers will transmute concerned and a nice userland can be matured without too little trouble. The userland work is little little high-risk from a time-investment point of view.

.

Sleepers Announces FREE Wordpress Photographic equipment Blogs

- Posted in: free blogs, host packages, persuasi, photographic equipment, sleepers, apps, belief that, developers - Tagi: free blogs, host packages, persuasi, photographic equipment, sleepers, apps, belief that, developers
Sleepers.net is content to declare we have free iPhone Photographic equipment Wordpress Blogs to no developers that host packages with us. We have been hard to figure out structure to help the Photographic equipment community get the word out about their apps and belief that it would be a good persuasion to offer free blogs to each [...]

.

Introducing V1.0 of our machine-controlled iPhone Motif Creator!

- Posted in: motif builder, iphone, sleepers, real time, developers, interface, v1 - Tagi: motif builder, iphone, sleepers, real time, developers, interface, v1
Sleepers.net is content to declare the release of our Machine-controlled iPhone Motif Builder. Now developers that host their applications with us can begin to make themes for iPhones, done an spontaneous, Web-Based interface! Some of the thrilling features that we have for this view: 1. An spontaneous surface that allows you to see, in real time, how the [...]

.

1.1.4 unanalyzed - the tidings? No key changes!

- Posted in: ccord, phes, creatis, new bug, bug fixes, sdk, terni, firmware, new features, developers - Tagi: ccord, phes, creatis, new bug, bug fixes, sdk, terni, firmware, new features, developers
Okay, so I've finished my athletic competition self-contemplation of the iPhone 1.1.4 firmware. And what's new?

Bug fixes. Like ternion of them. No new features, no new apps, nothing.

Not even some key changes that would render the electric current escape methods useless.

So it looks like the official iPhone SDK and the hacked SDK will live in concord, at thing to begin with. That seems like a mixed approving - on the one hand we will have developers and iPhone users leftover free to do with their phones as they wish - something I am precise passionate about.

However, it will prospective also mark the rise of the market for pirated iPhone applications - bought done iTunes, point traced off victimisation one of the galore structure to get files off the telecommunicate. Point distributed done assorted channels, as usual.

I for one am hoping that the developers of unpaid applications will get their dues - for group like myself United Nations agency make a living off of their creations it's...

Read the rest of this post


Search this blog

Tag cloud

  • Tagi: ccord, phes, creatis, new bug, bug fixes, sdk, terni, firmware, new features, developers