Let's assume that I have a property padding that defines the size of an inner border of my UIView subclass. It's merely a decorative element and has a default value of 2 pixels on each side.

-(void) initWithCoder:(NSCoder *) coder
{
   self = [super initWithCoder:coder];
   if( self)
      self->_padding = UIEdgeInsetsMake( 2, 2, 2, 2);
    return( self);
}
// and the same for initWithFrame:

I want this to be skinnable so I make it an UI_APPEARANCE_SELECTOR.

@property(nonatomic) UIEdgeInsets padding   UI_APPEARANCE_SELECTOR;

In my UIView subclass I want to host a contentView of some random size (code not shown). How do I calculate the frame.size of my view ? I get the frame of the contentView and add the padding. Simple.

   UIEdgeInsetsInsetRect   insets;
   
   insets        = [self padding];
   insets.top    = - insets.top;
   insets.left   = - insets.left;
   insets.bottom = - insets.bottom;
   insets.right  = - insets.right;
   frame         = UIEdgeInsetsInsetRect( [_contentView frame], 
                                          insets);

Will this work ? It depends. It depends on WHEN I call it.

Here's where the research comes in. This is the order in which a typical iOS program calls some of the key classes instance methods. This is not a stack trace, just the chronological order. The -[UIView setPadding:] call is the UIAppearance setting the value in the UIView:

-[UIWindow initWithFrame:]
-[UIViewController initWithNibName:]
-[UIWindow setRootViewController:]
-[UIWindow makeKeyAndVisible]
-[UIView initWithCoder:]
-[UIView awakeFromNib]
-[UIViewController viewDidLoad]
-[UIViewController viewWillAppear:]
-[UIView willMoveToWindow:]
+[UIView requiresConstraintBasedLayout]
-[UIView setPadding:]
-[UIView didMoveToWindow]
-[UIView layoutSubviews]
-[UIView drawRect:]
-[UIViewController viewDidAppear:]

Rule #1: Do not access UI_APPEARANCE_SELECTOR properties if -[UIView window] returns nil

Now you may think: "hey, do the code in -[UIView layoutSubviews]". But as the name indicates, that method should layout the subviews of my UIView and not change its own frame. That should have been set during the superview's layoutSubviews method. Otherwise the superview could be surprised and misconfigured. In the end, it's really the job of the UIViewController to set my UIView's frame.

But the UIViewController doesn't have a chance anymore to do this.

The clever 90% Solution

How can this be fixed ? Well it would be nice, if the UIView subclass offered a method like +frameForContentRect:, so the view's frame can be calculated in advance, during -[UIViewController viewDidLoad] given the frame of the future contentView:

+ (CGRect) frameForContentRect:(CGRect) rect
{
   id <UIAppearance>       appearance;
   UIEdgeInsetsInsetRect   insets;

   appearance    = [self appearance];
   insets        = [appearance padding];
   insets.top    = - insets.top;
   insets.left   = - insets.left;
   insets.bottom = - insets.bottom;
   insets.right  = - insets.right;
   return( UIEdgeInsetsInsetRect( rect, insets));
}

Solution ? Don't set the default value in the instance but instead preset the UIAppearance with it at +load time and delete the instance default value code from initWithCoder: (and initWithFrame:)

+ (void) load
{
  [[self appearance] setPadding:UIEdgeInsetsMake( 2.0, 2.0, 2.0, 2.0)]);
}

Not always, because the value of padding could also be set by user code on an individual instance. And then the UIAppearance value won't be used. Obviously there is no chance for knowing that in the custom +frameForContentRect: method, since it's a class method.

Rule #2: Set UI_APPEARANCE_SELECTORs default values in +load

I think using +load to set default UIAppearance values is still a good and rather clean method, but for the problem at hand it's not the whole solution. It's pretty clean, because you can be sure that the default is set earlier than -[id <UIApplicationDelegate> application:didFinishLaunchingWithOptions:], which one would expect the application coder to set his UIAppearance values.

The painful 100% Solution

Rule #3: Geometry UI_APPEARANCE_SELECTORs need special attention

]]> http://www.mulle-kybernetik.com/weblog/2013/04/uiappearance_magic_pain.html http://www.mulle-kybernetik.com/weblog/2013/04/uiappearance_magic_pain.html Thu, 25 Apr 2013 22:44:39 +0100 UIAppearance for my code, so that I can easily "skin" my iOS applications. What I wasn't really clear about is how to properly handle default values, because like most recent Apple code, it's just magic. I hate magic.

To test the best way to handle default values for my UIViews, I wanted to test UIAppearance first with the most simple and minimal code possible and go from there. I didn't even make it...

Step 1: Create a new iOS application with a single view

Step 2: Enter the following lines in the app delegate:

- (BOOL) application:(UIApplication *) application 
   didFinishLaunchingWithOptions:(NSDictionary *) launchOptions
{
   UIButton   *button;
   
   [[UIButton appearance] setTintColor:[UIColor orangeColor]];
   button = [UIButton buttonWithType:UIButtonTypeRoundedRect];
   NSLog( @"%@", [button tintColor]);
 }

Step 3: Run and observe console. Notice that it doesn't work, tintColor returns nil. Which is surprising.

The documentation says:

tintColor
The tint color for the button.

@property(nonatomic, retain) UIColor *tintColor
Discussion
The default value is nil.

This property is not valid for all button types.

   UIButton   *button;
   NSUInteger   i;
   
   for( i = 0; i < 10; i++)
   {
      button = [UIButton buttonWithType:i];
      NSParameterAssert( button);
      [button setTintColor:[UIColor greenColor]];
      NSLog( @"%ld: %@", (long) i, [button tintColor]);
   }

2013-04-24 15:57:11.496 ApperanceTest[2732:c07] 0: (null)
2013-04-24 15:57:11.498 ApperanceTest[2732:c07] 1: UIDeviceRGBColorSpace 0 1 0 1
2013-04-24 15:57:11.499 ApperanceTest[2732:c07] 2: (null)
2013-04-24 15:57:11.499 ApperanceTest[2732:c07] 3: (null)
2013-04-24 15:57:11.500 ApperanceTest[2732:c07] 4: (null)
2013-04-24 15:57:11.500 ApperanceTest[2732:c07] 5: (null)
2013-04-24 15:57:11.500 ApperanceTest[2732:c07] 6: (null)
2013-04-24 15:57:11.501 ApperanceTest[2732:c07] 7: (null)
2013-04-24 15:57:11.501 ApperanceTest[2732:c07] 8: (null)
2013-04-24 15:57:11.502 ApperanceTest[2732:c07] 9: (null)

So in the end although tintColor is advertised as being modifiable through UIAppearance in the header @property(nonatomic,retain) UIColor *tintColor NS_AVAILABLE_IOS(5_0) UI_APPEARANCE_SELECTOR; // default is nil. only valid for some button types, it really seems not to be the case.

I would on a hunch assume, that's because UIButton could be a class cluster and the subclasses implement their own tintColor accessors and the whole magic breaks down.

As the button is indeed an instance of a class cluster subclass called UIRoundedRectButton, I tried [[NSClassFromString(@"UIRoundedRectButton") appearance] setTintColor:[UIColor greenColor]]; and then it worked as I'd expected it in the first place.

Create an empty single view IOS project and create a UIView subclass:

"MulleView.h"

#import <UIKit/UIKit.h>

@interface _MulleView : UIView < UIAppearanceContainer>:
@end

@interface MulleView : _MulleView < UIAppearanceContainer>
@end

"MulleView.m"

#import "MulleView.h"


@implementation _MulleView

+ (void) load
{
   [[self appearance] setXColor:[UIColor blueColor]];
}

- (void) setXColor:(UIColor *) color
{
   NSLog( @"%s", __PRETTY_FUNCTION__);
}

@end


@implementation MulleView

- (void) willMoveToWindow:(UIWindow *)newWindow
{
   NSLog( @"%s", __PRETTY_FUNCTION__);
}


- (void) setXColor:(UIColor *) color
{
   NSLog( @"%s", __PRETTY_FUNCTION__);
}

@end

-[MulleView willMoveToWindow:]
-[_MulleView setXColor:]

Das neue Wahlkampfmotto der SPD "Das Wir entscheidet" steht ja unter keinem guten Stern, daher habe ich hier mal schnell einen schönen Ersatz geschaffen. Viel ehrlicher und nicht so verdruckst, als ob man keine Ahnung von nix hat.

1. Download the old Navigator EX 1.0 mpnexosx105ej7.dmg

And that's what I did. The code fairly quickly was able to use a template like this:

<html>
<!-- rendered by {{ [[NSProcessInfo processInfo] processName] }} on {{ [NSDate date]] }} -->
<body>
{# render an old skool HTML table #}
{% MulleScionEven = "#DDDDDD" %}
{% MulleScionOdd = "#FFFFFF" %}
{% for item in [NSTimeZone knownTimeZoneNames] %}
{% if item#.n %}
{% if item#.isFirst %}
<table>
<tr><th>TimeZone</th></tr>
{% endif %}
<tr bgcolor="{{ item#.evenOdd }}"><td>{{ item }}</td></tr>
{% if item#.isLast %}
</table>
{% endif %}
{% else %}
Sorry, no timezone info available.
{% endif %}
{% endfor %}
</body>
</html>

<html>
<!-- rendered by mulle-scion on 2013-02-27 16:27:05 +0000 -->
<body>
<table>
<tr><th>TimeZone</th></tr>
<tr bgcolor="#DDDDDD"><td>Africa/Abidjan</td></tr>
<tr bgcolor="#FFFFFF"><td>Africa/Accra</td></tr>
<tr bgcolor="#DDDDDD"><td>Africa/Addis_Ababa</td></tr>
<tr bgcolor="#FFFFFF"><td>Africa/Algiers</td></tr>
<tr bgcolor="#DDDDDD"><td>Africa/Asmara</td></tr>
<tr bgcolor="#FFFFFF"><td>Africa/Bamako</td></tr>

{% extends "base.scion" %}

This text is after extends, so it won't be printed.
It's a confusing way to comment...:)
All the blocks following are slurped up and remembered till the end of the
enveloping template. Then blocks are substituted.

{% block header %}
this will be overridden shortly by the following block
{% endblock %}

{% block header %}
replaced the whole header
{% endblock %}

{% block body_extension %}
extended the body
{% endblock %}

{% block footer_extension %}
{% block footer_extension2 %}
default footer_extension2 text
{% endblock %}
{% endblock %}

{% block hackish_extension %}A bit of a mind bender{% endblock %}

{% extends "hackish.scion" %}

 Pretty nifty :)

The people's guess is, that it's an ARP request that wakes up the Synology. That sounds sensible and somewhat unavoidable. By my own observations, I'd say that ANY ARP request wakes up the Synology from deep sleep, not just the ones looking for the NAS ethernet address. ARPs are broadcast, but other broadcast traffic I observed didn't affect it.

Case in point, when my "workstation" turns off or goes to sleep, my NAS magically awakes, presumably because my gateway is looking for the workstation again. (Speculation). Also when my "gateway" periodically awakens at night because of Wake on Demand, then it will apparently also send out ARP packets.

The main and fixable problem seems to be, that, if the NAS has "accidentally" risen, it won't find sleep again soon. Instead it waits again for the configured period. (30 minutes in my case).

Another speculation is, that periodic awakenings from sleep mode have something to do with the ntp daemon, who's trying to contact a time server about every hour. I have now disabled ntp for testing, let's see how that works. On a hunch I would say, that this isn't the problem, as the ntp daemon shouldn't be running in sleep mode in the first place.

I have to say, that I was never happy with the MyBookLive. That NAS has only 256 MB of RAM, but with all services running, it is swapping very often, resulting in really bad performance. There are a lot of indexing scripts Western Digital uses, that are total crap and they are being run much too often (like for every file change). While it got a good review in c't magazine, I can't recommend it at all. Opening my MP3 collection in the Finder takes like 20s and sometimes much more. It will be relegated to it's only proper use, a Time Machine storage.

The Synology DS213+ main benefit is having a good sleep mode and a really, really excellent user interface. Like two generations ahead of what WD had to offer. From the way I read it, their partitioned RAID-1 system also sounds like a big advantage over plain RAID-1.

Another plus, it's very quiet, if it's not doing anything. (Which it usually doesn't in my case).

Everybody and their mother are putting Western Digital RED disks in their NAS. But wait a minute, why would I care so much, when I am using RAID-1 ? I  saved 20% of my money and got a potentially much faster harddisk instead, a Seagate ST3000DM001  with 7200 RPM. I like to save energy, when I don't need the system. But otherwise I prefer fast over reliable :)

But it isn't quite as powerful as I would like. Copying files with Samba, is not a problem at all for the DS213+. But if you use rsync, which to my surprise, is quite the CPU hog, you will probably not be able to saturate your harddisks. Opening my MP3 collection in the finder now only takes 3 seconds, which isn't great, but manageable.

I wouldn't buy anything less than a DS213+ for a home NAS now.

Below in the drawing is my current home setup. My DSL router 192.168.0.1 ("fritz.box") creates a WIFI network 192.168.0.0/24. I then have another ethernet based network 192.168.1.0/24, with machines, that have no wireless capabilities.

This used to work fine, if one machine with an ethernet card (192.168.1.2) and a wireless card (192.168.0.2) running OS X, provided internet sharing for the wired net. But when I recently added a wireless only machine 192.168.0.3 it became apparent, that this machine was not able to talk to the 192.168.1.0/24 network.

It took me quite a while to figure out how to do it, although the information is available, if you know which keywords to google for.

First things first. Apple's "Internet Sharing" is doing NAT, which in this case is harmful. NAT and bridging can not be used reliably together (says the OpenBSD pf documentation). But I need to use "real" ethernet bridging, which can be done in OS X since 10.8 apparently.

Create a permanent ethernet bridge on OS X

I put this shell script into /usr/local/sbin as mulle-ethernet-bridge.sh

#! /bin/sh
# ######################################
#  coded by Nat!
#  2013 Mulle kybernetiK
#  GPL

command=${1:-start}
shift
proxyarp=${1:-no}
shift

start()
{
	sysctl -w net.inet.ip.forwarding=1
	sysctl -w net.inet.ip.fw.enable=1
	if [ "$proxyarp" != "no" ]
	then
		sysctl -w net.link.ether.inet.proxyall=1
	fi

	ifconfig bridge0 create
	ifconfig bridge0 addm en0
	ifconfig bridge0 addm en1
	ifconfig bridge0 up
	if [ $? -eq 0 ]
	then
		syslog -s "Mulle Ethernet Bridge is up"
	else
		syslog -s "Mulle Ethernet Bridge failure"
	fi
}


stop()
{
	ifconfig bridge0 destroy

	sysctl -w net.inet.ip.forwarding=0
	sysctl -w net.inet.ip.fw.enable=0
	sysctl -w net.link.ether.inet.proxyall=0

	syslog -s "Mulle Ethernet Bridge is down"
}



case "$command" in
	start*)	start
		;;
	
	stop*)	stop
		;;
esac

chmod 755 /usr/local/sbin/mulle-ethernet-bridge.sh

<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">

<plist version="1.0">
    <dict>
        <key>Label</key>
        <string>com.mulle-kybernetik.admin.ethernet-bridge</string>
        <key>ProgramArguments</key>
        <array>
            <string>/usr/local/sbin/mulle-ethernet-bridge.sh</string>
        </array>
        <key>UserName</key>
        <string>root</string>
        <key>GroupName</key>
        <string>wheel</string>
        <key>RunAtLoad</key>
        <true/>
        <key>LaunchOnlyOnce</key>
        <true/>   
    </dict>
</plist>

sudo chmod 644 /Library/LaunchDaemons/com.mulle-kybernetik.admin.ethernet-bridge.plist
sudo chown root:wheel /Library/LaunchDaemons/com.mulle-kybernetik.admin.ethernet-bridge.plist
sudo launchctl load /Library/LaunchDaemons/com.mulle-kybernetik.admin.ethernet-bridge.plist

bridge0: flags=8863 mtu 1500
	ether ac:dc:18:48:20:13 
	Configuration:
		priority 0 hellotime 0 fwddelay 0 maxage 0
		ipfilter disabled flags 0x2
	member: en0 flags=3
	         port 4 priority 0 path cost 0
	member: en1 flags=3
	         port 5 priority 0 path cost 0

Adding routes to the wireless boxes

This information needs to be added to the static routing tables of the router. Thankfully the fritz box allows this in its expert settings. So I added:

route -p add 192.168.1.0 mask 255.255.255.0 192.168.0.2 metric 1

WRONG!

Because of the cheapness of android tablets, I can now for the first time create small, simple programs and give them away for christmas to hapless children of my family complete with the hardware. All for EUR 80.

It would be really, really cool if those tablets would be conveniently rootable. They aren't though :( I hope RMS gets onto this.

But, android needs to be programmed in Java. Probably my pet-hated language, C++ being a close second. Human memory being what it is, I thought that it wouldn't be cool, but that it probably would just work.

WRONG!

One of the things always touted about Java is, how convenient the garbage collection is and that one doesn't have to worry about memory management. Which even I remember, isn't true, because in the end most of the time in Java projects is spent hunting down leaks. So  explicit management of the object tree always was necessary. And I hated every second of it. I hated writing it and I hated debugging it. Oh god, the memories are coming back now...

As a little aside, I experimented loading Drawables from inkscape generated SVG with svg-android and svg-android-2. Forget about it. The parsing is javaesque snail paced and most of the time the inkscape SVG can't be parsed by these libraries anyway. I tried simplifying them with scour but that had some side-effects on the gradients, which I didn't like (maybe a precision thing). So I didn't go down that route further.

Anway. So of all the problems, that probably exist in android, I immediately run into this http://code.google.com/p/android/issues/detail?id=8488. I don't know how many other glaring bugs android has, but this is just the typical resource unaware Java lameness.  And this bug is most likely in there from the beginning of android.

I especially love how the first mail is from a guy, who claims to have written parts of the UI graphics code, who is totally oblivious to the fact, that a problem exists and then never shows up in the discussion afterwards. Bug Status ? Declined. There is no bug, because that's normal sucky Java behavior!

What is the folksy workaround ? Use some other crufty Java classes which allow manual management. The other idea, and I suspect a LOT of android is written this way: keep the Java as OS-glue and go native for the real action.

Return-Path: Delivered-To: nat@rechner.firma.de
Received: from localhost (localhost [127.0.0.1])
    by rechner.firma.de (Postfix) with ESMTP id B8B232285BF
    for ; Fri,  7 Dec 2012 21:48:14 +0100 (CET)
X-Virus-Scanned: amavisd-new at firma.de
Received: from rechner.firma.de ([127.0.0.1])
    by localhost (rechner.firma.de [127.0.0.1]) (amavisd-new, port 10024)
    with ESMTP id uzkiaWIfZGHK for ;
    Fri,  7 Dec 2012 21:48:09 +0100 (CET)
Received: from spam.com (spam.com [91.240.71.205])
    by rechner.firma.de (Postfix) with ESMTP id 97E2E2285B9
    for ; Fri,  7 Dec 2012 21:48:08 +0100 (CET)
Received: from localhost (localhost [127.0.0.1])
    by spam.com (Postfix) with ESMTP id 1BCCA31128
    for ; Fri,  7 Dec 2012 21:39:00 +0100 (CET)
Date: Fri, 7 Dec 2012 21:39:00 +0100
To: info@firma.de
From: Fickt-Euch Sell Subject: =?iso-8859-1?Q?Nikolausgeschenk_f=FCr_firma_GmbH?=
Message-ID: X-Priority: 3
Importance: 
X-Mailer:  Microsoft Outlook 6.00.692277
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
Content-Type: text/plain; charset="iso-8859-1"
Guten Tag,
Weihnachten naht - und damit unsere Top-Angebote:
++ Kunst-Weihnachtsbaum: Nordmann-Tanne, NADELT NICHT: statt 199,- Euro* nur 79,- Euro
++ Das perfekte Geschenk für den Hausmann: Set aus Akkuschrauber und Akkubohrer: statt 399,- Euro* nur 89,- Euro
++ Superstarke 3,1 PS-Kettensäge: statt 399,- Euro* nur 139 Euro
Hier sofort bestellen:
http://spam.com/
Denken Sie praktisch - und verschenken Sie eine der nützlichen Produkte aus unserem Online-Shop. Bestellen Sie jetzt und profitieren Sie von den allergrößten Rabatten.
Vorweihnachtliche Grüße
Fickt-Euch Sell
Newsletter Abmelden: http://spam.com/abmelden
* Zu diesen und ähnlichen Preise im Handel gesehen

Then add a shell script to the Service (red). Be sure to check the box (yellow) and change the command to sort (green), Finally save the script. The name you give to your script, will be the name that appears in the Services menu. I chose "Zeilen sortieren".

Now in 10.8.1 everything works fine. But I remember that in former times, it may have been necessary to do some magic to get the service going (login/logut). But maybe I am confusing this with Colorsync or Finder plugins.

Anyway... In Xcode select some lines and run your "Service" on it. If you don't have many services, you might be lucky, and the services will appear directly in the top menu.

Now with these preliminaries out of the way, I have created a small unix tool, that aligns variables (and as a bonus also can align assignment statements).  So this:

   int    foo;
   static char  *x;
   auto volatile int  foo[ 126];

becomes

   int                 foo;
   static char         *x;
   auto volatile int   foo[ 126];

and

   x = foo
    // bla
      y[x] = bar

becomes

   x     = foo
    // bla
   y[x]  = bar

Not so easy this time because Cocoa.framework is missing some symbols.