android Archives - Page 2 of 3 - "Where to now?"

Categoryandroid

copy assets script for multiple resolutions for android

August 14, 2013 / Jan Rabe / 0 Comments

Assuming you want to build multiple apks for multiple resolutions on android or just have faster apk upload time to your android device during development.

Considering you want to copy from /myapp/resources/ to /myapp/android/assets/ for several resolutions and you want only to click once for each resolution build.
To accomplish this i recommend using rsycn/ and cygwin.

You might have these script files in:

/myapp/scripts/320x_CopyAssets.bat
/myapp/scripts/480x_CopyAssets.bat
/myapp/scripts/540x_CopyAssets.bat
/myapp/scripts/600x_CopyAssets.bat
/myapp/scripts/720x_CopyAssets.bat
/myapp/scripts/768x_CopyAssets.bat
/myapp/scripts/800x_CopyAssets.bat
/myapp/scripts/1080x_CopyAssets.bat
/myapp/scripts/copy_assets.sh
/myapp/IgnoreCopyAssets/IgnoreCopyAssets.txt
/myapp/IgnoreCopyAssets/320x_IgnoreCopyAssets.txt
/myapp/IgnoreCopyAssets/480x_IgnoreCopyAssets.txt
/myapp/IgnoreCopyAssets/540x_IgnoreCopyAssets.txt
/myapp/IgnoreCopyAssets/600x_IgnoreCopyAssets.txt
/myapp/IgnoreCopyAssets/720x_IgnoreCopyAssets.txt
/myapp/IgnoreCopyAssets/768x_IgnoreCopyAssets.txt
/myapp/IgnoreCopyAssets/800x_IgnoreCopyAssets.txt
/myapp/IgnoreCopyAssets/1080x_IgnoreCopyAssets.txt

Where you merely send the desired resolution at to the actually copy_assets shell script. (e.g the 768x_CopyAssets.bat)

@echo off
set PATH=%~dp0msys\bin;%~dp0win-rsync;%PATH%
sh copy_assets.sh WIN 768
PAUSE

@echo off

set PATH=%~dp0msys\bin;%~dp0win-rsync;%PATH%

sh copy_assets.sh WIN 768

PAUSE

The copy_assets.sh would look like:

SCRIPT_PATH=$( (cd -P $(dirname $0) && pwd) )
RSYNC_CMD="rsync -aCx --progress --exclude-from 'IgnoreCopyAssets/IgnoreCopyAssets.txt'"
if [ $# -ge 1 ];then 

	if [ $# -ge 2 ];then 
		case "$2" in
		320)  RSYNC_CMD="$RSYNC_CMD --exclude-from 'IgnoreCopyAssets/320x_IgnoreCopyAssets.txt'"
		;;
		480)  RSYNC_CMD="$RSYNC_CMD --exclude-from 'IgnoreCopyAssets/480x_IgnoreCopyAssets.txt'"
		;;
		540)  RSYNC_CMD="$RSYNC_CMD --exclude-from 'IgnoreCopyAssets/540x_IgnoreCopyAssets.txt'"
		;;
		600)  RSYNC_CMD="$RSYNC_CMD --exclude-from 'IgnoreCopyAssets/600x_IgnoreCopyAssets.txt'"
		;;
		720)  RSYNC_CMD="$RSYNC_CMD --exclude-from 'IgnoreCopyAssets/720x_IgnoreCopyAssets.txt'"
		;;
		768)  RSYNC_CMD="$RSYNC_CMD --exclude-from 'IgnoreCopyAssets/768x_IgnoreCopyAssets.txt'"
		;;
		800)  RSYNC_CMD="$RSYNC_CMD --exclude-from 'IgnoreCopyAssets/800x_IgnoreCopyAssets.txt'"
		;;
		1080)  RSYNC_CMD="$RSYNC_CMD --exclude-from 'IgnoreCopyAssets/1080x_IgnoreCopyAssets.txt'"
		;;
		esac
	fi
	
	if [ $1 = "WIN" ];then
		SCRIPT_PATH=$(dirname $0) # Otherwise rsync will think it's remote
		RSYNC_CMD="$RSYNC_CMD --chmod=ugo=rwX" # Otherwise permissions are screwed up
	fi
fi

RSYNC_CMD_W_DELETE="$RSYNC_CMD --delete"

###############################################################################

echo RUN ASSETS SYNCING SHELL SCRIPT

asset_source_path=$SCRIPT_PATH/../resources
asset_destination_path=$SCRIPT_PATH/../android/assets

$RSYNC_CMD_W_DELETE $asset_source_path/* $asset_destination_path

# The wildcard in rsync leads missing deletion in the top-level directory so let's fix that
diff -arq $asset_source_path $asset_destination_path | grep "Only in $asset_destination_path" | cut -d" " -f4 | xargs -I{} rm -r -f $asset_destination_path/{}

SCRIPT_PATH=$( (cd -P $(dirname $0) && pwd) )

RSYNC_CMD="rsync -aCx --progress --exclude-from 'IgnoreCopyAssets/IgnoreCopyAssets.txt'"

if [ $# -ge 1 ];then

if [ $# -ge 2 ];then

case "$2" in

320) RSYNC_CMD="$RSYNC_CMD --exclude-from 'IgnoreCopyAssets/320x_IgnoreCopyAssets.txt'"

;;

480) RSYNC_CMD="$RSYNC_CMD --exclude-from 'IgnoreCopyAssets/480x_IgnoreCopyAssets.txt'"

;;

540) RSYNC_CMD="$RSYNC_CMD --exclude-from 'IgnoreCopyAssets/540x_IgnoreCopyAssets.txt'"

;;

600) RSYNC_CMD="$RSYNC_CMD --exclude-from 'IgnoreCopyAssets/600x_IgnoreCopyAssets.txt'"

;;

720) RSYNC_CMD="$RSYNC_CMD --exclude-from 'IgnoreCopyAssets/720x_IgnoreCopyAssets.txt'"

;;

768) RSYNC_CMD="$RSYNC_CMD --exclude-from 'IgnoreCopyAssets/768x_IgnoreCopyAssets.txt'"

;;

800) RSYNC_CMD="$RSYNC_CMD --exclude-from 'IgnoreCopyAssets/800x_IgnoreCopyAssets.txt'"

;;

1080) RSYNC_CMD="$RSYNC_CMD --exclude-from 'IgnoreCopyAssets/1080x_IgnoreCopyAssets.txt'"

;;

esac

if [ $1 = "WIN" ];then

SCRIPT_PATH=$(dirname $0) # Otherwise rsync will think it's remote

RSYNC_CMD="$RSYNC_CMD --chmod=ugo=rwX" # Otherwise permissions are screwed up

RSYNC_CMD_W_DELETE="$RSYNC_CMD --delete"

###############################################################################

echo RUN ASSETS SYNCING SHELL SCRIPT

asset_source_path=$SCRIPT_PATH/../resources

asset_destination_path=$SCRIPT_PATH/../android/assets

$RSYNC_CMD_W_DELETE $asset_source_path/* $asset_destination_path

# The wildcard in rsync leads missing deletion in the top-level directory so let's fix that

diff -arq $asset_source_path $asset_destination_path | grep "Only in $asset_destination_path" | cut -d" " -f4 | xargs -I{} rm -r -f $asset_destination_path/{}

And in case you want to ignore files or dirs you can use IgnoreCopyAssets.txt

.svn
.git/
HelloWorld.png

.svn

.git/

HelloWorld.png

Or the other resolutions (e.g the 768x_IgnoreCopyAssets.txt) Note: you can use # to comment files or directories out. ;)

graphics/320x
graphics/480x
graphics/540x
graphics/600x
graphics/720x
#graphics/768x
graphics/800x
graphics/1080x

graphics/320x

graphics/480x

graphics/540x

graphics/600x

graphics/720x

#graphics/768x

graphics/800x

graphics/1080x

Conclusion

Easy to use scripts to copy assets and remove all unwanted ones from your assets folder plus the posibility to ignore files and dirs on demand. Enjoy :)

c++ arc4random on android, windows and linux

August 14, 2013 / Jan Rabe / 3 Comments

Turns out native android development doesn’t supply you with a real ar4random method. However adapting the OpenBSD’s arc4random.c does the trick. And even works on windows without using CryptoAPI (thanks to Chris K. Jester-Young for these suggestions). Additionally with boost *ducks* it should be more or less multithreadsafe.

Mixed (red dots are generated by arc4; Math.Random() Arc4random)

/*	$OpenBSD: arc4random.c,v 1.24 2013/06/11 16:59:50 deraadt Exp $	*/

/*
 * Copyright (c) 1996, David Mazieres <dm@uun.org>
 * Copyright (c) 2008, Damien Miller <djm@openbsd.org>
 *
 * Permission to use, copy, modify, and distribute this software for any
 * purpose with or without fee is hereby granted, provided that the above
 * copyright notice and this permission notice appear in all copies.
 *
 * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
 * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
 * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
 * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
 * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
 * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
 * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
 */

/*
 * Arc4 random number generator for OpenBSD.
 *
 * This code is derived from section 17.1 of Applied Cryptography,
 * second edition, which describes a stream cipher allegedly
 * compatible with RSA Labs "RC4" cipher (the actual description of
 * which is a trade secret).  The same algorithm is used as a stream
 * cipher called "arcfour" in Tatu Ylonen's ssh package.
 *
 * RC4 is a registered trademark of RSA Laboratories.
 */

#include <boost\thread.hpp>

#if (CC_TARGET_PLATFORM == CC_PLATFORM_WIN32)
#include <Windows.h>
#endif

typedef uint8_t  u_int8_t;
typedef uint16_t u_int16_t;
typedef uint32_t u_int32_t;
typedef uint64_t u_int64_t;

typedef unsigned char u_char;

struct arc4_stream {
	u_int8_t i;
	u_int8_t j;
	u_int8_t s[256];
};

#define pid_t boost::thread::id

static int rs_initialized;
static struct arc4_stream rs;
static pid_t arc4_stir_pid;
static int arc4_count;

static inline u_int8_t arc4_getbyte(void);

static inline void arc4_init(void)
{
	int     n;

	for (n = 0; n < 256; n++)
		rs.s[n] = n;
	rs.i = 0;
	rs.j = 0;
}

static inline void arc4_addrandom(u_char *dat, int datlen)
{
	int     n;
	u_int8_t si;

	rs.i--;
	for (n = 0; n < 256; n++) {
		rs.i = (rs.i + 1);
		si = rs.s[rs.i];
		rs.j = (rs.j + si + dat[n % datlen]);
		rs.s[rs.i] = rs.s[rs.j];
		rs.s[rs.j] = si;
	}
	rs.j = rs.i;
}

static const char * urandom = "/dev/urandom";

static void fillRandom(u_char *rnd, size_t	len)
{
	for (int i = 0; i < (int)len; ++i)
		rnd[i] = i & 255;

	for (int k = 0; k < (int)len*8; ++k)
	{
		int v1 = rand() % len;
		int v2 = rand() % len;
		u_char temp = rnd[v1];
		rnd[v1] = rnd[v2];
		rnd[v2] = temp;
	}
}

static void arc4_stir(void)
{
	int     i; 
	u_char rnd[128];

	if (!rs_initialized) {
		arc4_init();
		rs_initialized = 1;
	}

	#if (CC_TARGET_PLATFORM == CC_PLATFORM_WIN32)

	HMODULE hLib=LoadLibraryA("ADVAPI32.DLL");
	if (hLib) {
		BOOLEAN (APIENTRY *pfn)(void*, ULONG) = (BOOLEAN (APIENTRY *)(void*,ULONG))GetProcAddress(hLib,"SystemFunction036");
		if (pfn) {
			if(pfn(rnd,sizeof(rnd))) {
				// use buff full of random goop
			}
		}
		FreeLibrary(hLib);
	}

	#elif (CC_TARGET_PLATFORM == CC_PLATFORM_ANDROID)
	
		size_t	len = sizeof(rnd);
		// open /dev/urandom and read 128 bytes
		FILE *in = fopen(urandom,"r");
		if (in == NULL)
		{
			CCLog("Couldn't open file %s", urandom);
			fillRandom(rnd,len);
		} else 
		{
			size_t ret = fread(rnd,1,len,in);
			if(ret != len) 
			{
				CCLog("%d != %d", ret, len);
				fillRandom(rnd,len);
			}
			fclose(in);
		}

	#elif (CC_TARGET_PLATFORM == CC_PLATFORM_LINUX)

		int mib[2]; 
		// unix access to /dev/urandom even when urandom is accessable for some weird reason
		mib[0] = CTL_KERN;	
		mib[1] = KERN_ARND;	
		sysctl(mib, 2, rnd, &len, NULL, 0);

	#else

		fillRandom(rnd,len);

	#endif

	arc4_addrandom(rnd, sizeof(rnd));

	/*
	 * Discard early keystream, as per recommendations in:
	 * http://www.wisdom.weizmann.ac.il/~itsik/RC4/Papers/Rc4_ksa.ps
	 */
	for (i = 0; i < 256; i++)
		(void)arc4_getbyte();
	arc4_count = 1600000;
}

static void arc4_stir_if_needed(void)
{
	pid_t pid = boost::this_thread::get_id();

	if (arc4_count <= 0 || !rs_initialized /*|| arc4_stir_pid != pid*/) { // other threads use the same random generator for us
		arc4_stir_pid = pid;
		arc4_stir();
	}
}

static inline u_int8_t arc4_getbyte(void)
{
	u_int8_t si, sj;

	rs.i = (rs.i + 1);
	si = rs.s[rs.i];
	rs.j = (rs.j + si);
	sj = rs.s[rs.j];
	rs.s[rs.i] = sj;
	rs.s[rs.j] = si;
	return (rs.s[(si + sj) & 0xff]);
}

static inline u_int32_t arc4_getword(void)
{
	u_int32_t val;
	val = arc4_getbyte() << 24;
	val |= arc4_getbyte() << 16;
	val |= arc4_getbyte() << 8;
	val |= arc4_getbyte();
	return val;
}

void arc4random_stir(void)
{
	arc4_stir();
}

void arc4random_addrandom(u_char *dat, int datlen)
{
	if (!rs_initialized)
		arc4_stir();
	arc4_addrandom(dat, datlen);
}

u_int32_t arc4randomT(void)
{
	u_int32_t val;
	arc4_count -= 4;
	arc4_stir_if_needed();
	val = arc4_getword();
	return val;
}

void arc4random_buf(void *_buf, size_t n)
{
	u_char *buf = (u_char *)_buf;
	arc4_stir_if_needed();
	while (n--) {
		if (--arc4_count <= 0)
			arc4_stir();
		buf[n] = arc4_getbyte();
	}
}

/*
 * Calculate a uniformly distributed random number less than upper_bound
 * avoiding "modulo bias".
 *
 * Uniformity is achieved by generating new random numbers until the one
 * returned is outside the range [0, 2**32 % upper_bound).  This
 * guarantees the selected random number will be inside
 * [2**32 % upper_bound, 2**32) which maps back to [0, upper_bound)
 * after reduction modulo upper_bound.
 */
u_int32_t arc4random_uniformT(u_int32_t upper_bound)
{
	u_int32_t r, min;

	if (upper_bound < 2)
		return 0;

	/* 2**32 % x == (2**32 - x) % x */
	min = -upper_bound % upper_bound;

	/*
	 * This could theoretically loop forever but each retry has
	 * p > 0.5 (worst case, usually far better) of selecting a
	 * number inside the range we need, so it should rarely need
	 * to re-roll.
	 */
	for (;;) {
		r = arc4randomT();
		if (r >= min)
			break;
	}

	return r % upper_bound;
}

static boost::mutex arc4mutex;

unsigned int arc4random_uniform(unsigned int upper_bound)
{
	arc4mutex.lock();
	unsigned int ret = arc4random_uniformT(upper_bound);
	arc4mutex.unlock();
	return ret;
}

unsigned int arc4random() 
{
	arc4mutex.lock();
	unsigned int ret = arc4randomT();
	arc4mutex.unlock();
	return ret;
}

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168

169

170

171

172

173

174

175

176

177

178

179

180

181

182

183

184

185

186

187

188

189

190

191

192

193

194

195

196

197

198

199

200

201

202

203

204

205

206

207

208

209

210

211

212

213

214

215

216

217

218

219

220

221

222

223

224

225

226

227

228

229

230

231

232

233

234

235

236

237

238

239

240

241

242

243

244

245

246

247

248

249

250

251

252

253

254

255

256

257

258

259

260

261

262

263

264

265

266

267

268

269

270

271

272

273

274

275

276

277

278

279

280

281

282

283

284

285

286

287

/* $OpenBSD: arc4random.c,v 1.24 2013/06/11 16:59:50 deraadt Exp $ */

* Permission to use, copy, modify, and distribute this software for any

* purpose with or without fee is hereby granted, provided that the above

* copyright notice and this permission notice appear in all copies.

* THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES

* WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF

* MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR

* ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES

* WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN

* ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF

* OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.

* Arc4 random number generator for OpenBSD.

* This code is derived from section 17.1 of Applied Cryptography,

* second edition, which describes a stream cipher allegedly

* compatible with RSA Labs "RC4" cipher (the actual description of

* which is a trade secret). The same algorithm is used as a stream

* cipher called "arcfour" in Tatu Ylonen's ssh package.

* RC4 is a registered trademark of RSA Laboratories.

#include <boost\thread.hpp>

#if (CC_TARGET_PLATFORM == CC_PLATFORM_WIN32)

#include <Windows.h>

#endif

typedef uint8_t u_int8_t;

typedef uint16_t u_int16_t;

typedef uint32_t u_int32_t;

typedef uint64_t u_int64_t;

typedef unsigned char u_char;

struct arc4_stream {

u_int8_t i;

u_int8_t j;

u_int8_t s[256];

};

#define pid_t boost::thread::id

static int rs_initialized;

static struct arc4_stream rs;

static pid_t arc4_stir_pid;

static int arc4_count;

static inline u_int8_t arc4_getbyte(void);

static inline void arc4_init(void)

{

int n;

for (n = 0; n < 256; n++)

rs.s[n] = n;

rs.i = 0;

rs.j = 0;

}

static inline void arc4_addrandom(u_char *dat, int datlen)

{

int n;

u_int8_t si;

rs.i--;

for (n = 0; n < 256; n++) {

rs.i = (rs.i + 1);

si = rs.s[rs.i];

rs.j = (rs.j + si + dat[n % datlen]);

rs.s[rs.i] = rs.s[rs.j];

rs.s[rs.j] = si;

}

rs.j = rs.i;

}

static const char * urandom = "/dev/urandom";

static void fillRandom(u_char *rnd, size_t len)

{

for (int i = 0; i < (int)len; ++i)

rnd[i] = i & 255;

for (int k = 0; k < (int)len*8; ++k)

{

int v1 = rand() % len;

int v2 = rand() % len;

u_char temp = rnd[v1];

rnd[v1] = rnd[v2];

rnd[v2] = temp;

}

static void arc4_stir(void)

{

int i;

u_char rnd[128];

if (!rs_initialized) {

arc4_init();

rs_initialized = 1;

}

#if (CC_TARGET_PLATFORM == CC_PLATFORM_WIN32)

HMODULE hLib=LoadLibraryA("ADVAPI32.DLL");

if (hLib) {

BOOLEAN (APIENTRY *pfn)(void*, ULONG) = (BOOLEAN (APIENTRY *)(void*,ULONG))GetProcAddress(hLib,"SystemFunction036");

if (pfn) {

if(pfn(rnd,sizeof(rnd))) {

// use buff full of random goop

}

FreeLibrary(hLib);

}

#elif (CC_TARGET_PLATFORM == CC_PLATFORM_ANDROID)

size_t len = sizeof(rnd);

// open /dev/urandom and read 128 bytes

FILE *in = fopen(urandom,"r");

if (in == NULL)

{

CCLog("Couldn't open file %s", urandom);

fillRandom(rnd,len);

} else

{

size_t ret = fread(rnd,1,len,in);

if(ret != len)

{

CCLog("%d != %d", ret, len);

fillRandom(rnd,len);

}

fclose(in);

}

#elif (CC_TARGET_PLATFORM == CC_PLATFORM_LINUX)

int mib[2];

// unix access to /dev/urandom even when urandom is accessable for some weird reason

mib[0] = CTL_KERN;

mib[1] = KERN_ARND;

sysctl(mib, 2, rnd, &len, NULL, 0);

#else

fillRandom(rnd,len);

#endif

arc4_addrandom(rnd, sizeof(rnd));

* Discard early keystream, as per recommendations in:

* http://www.wisdom.weizmann.ac.il/~itsik/RC4/Papers/Rc4_ksa.ps

for (i = 0; i < 256; i++)

(void)arc4_getbyte();

arc4_count = 1600000;

}

static void arc4_stir_if_needed(void)

{

pid_t pid = boost::this_thread::get_id();

if (arc4_count <= 0 || !rs_initialized /*|| arc4_stir_pid != pid*/) { // other threads use the same random generator for us

arc4_stir_pid = pid;

arc4_stir();

}

static inline u_int8_t arc4_getbyte(void)

{

u_int8_t si, sj;

rs.i = (rs.i + 1);

si = rs.s[rs.i];

rs.j = (rs.j + si);

sj = rs.s[rs.j];

rs.s[rs.i] = sj;

rs.s[rs.j] = si;

return (rs.s[(si + sj) & 0xff]);

}

static inline u_int32_t arc4_getword(void)

{

u_int32_t val;

val = arc4_getbyte() << 24;

val |= arc4_getbyte() << 16;

val |= arc4_getbyte() << 8;

val |= arc4_getbyte();

return val;

}

void arc4random_stir(void)

{

arc4_stir();

}

void arc4random_addrandom(u_char *dat, int datlen)

{

if (!rs_initialized)

arc4_stir();

arc4_addrandom(dat, datlen);

}

u_int32_t arc4randomT(void)

{

u_int32_t val;

arc4_count -= 4;

arc4_stir_if_needed();

val = arc4_getword();

return val;

}

void arc4random_buf(void *_buf, size_t n)

{

u_char *buf = (u_char *)_buf;

arc4_stir_if_needed();

while (n--) {

if (--arc4_count <= 0)

arc4_stir();

buf[n] = arc4_getbyte();

}

* Calculate a uniformly distributed random number less than upper_bound

* avoiding "modulo bias".

* Uniformity is achieved by generating new random numbers until the one

* returned is outside the range [0, 2**32 % upper_bound). This

* guarantees the selected random number will be inside

* [2**32 % upper_bound, 2**32) which maps back to [0, upper_bound)

* after reduction modulo upper_bound.

u_int32_t arc4random_uniformT(u_int32_t upper_bound)

{

u_int32_t r, min;

if (upper_bound < 2)

return 0;

/* 2**32 % x == (2**32 - x) % x */

min = -upper_bound % upper_bound;

* This could theoretically loop forever but each retry has

* p > 0.5 (worst case, usually far better) of selecting a

* number inside the range we need, so it should rarely need

* to re-roll.

for (;;) {

r = arc4randomT();

if (r >= min)

break;

}

return r % upper_bound;

}

static boost::mutex arc4mutex;

unsigned int arc4random_uniform(unsigned int upper_bound)

{

arc4mutex.lock();

unsigned int ret = arc4random_uniformT(upper_bound);

arc4mutex.unlock();

return ret;

}

unsigned int arc4random()

{

arc4mutex.lock();

unsigned int ret = arc4randomT();

arc4mutex.unlock();

return ret;

}

And here are the convinience methods for it:

unsigned int arc4random();
unsigned int arc4random_uniform(unsigned int upper_bound);

int earc4random()
{
	return arc4random();
}

int earc4random(int maxValue)
{
	return arc4random_uniform(maxValue);
}

int earc4random(int minValue, int maxValue)
{
	return arc4random_uniform(maxValue-minValue) + minValue;
}

unsigned int arc4random();

unsigned int arc4random_uniform(unsigned int upper_bound);

int earc4random()

{

return arc4random();

}

int earc4random(int maxValue)

{

return arc4random_uniform(maxValue);

}

int earc4random(int minValue, int maxValue)

{

return arc4random_uniform(maxValue-minValue) + minValue;

}

So you can use it like:

unsigned int random[100];
for(int i = 0 ; i < 100; ++i)
{
  random[i] = earc4random(10,30);
}

unsigned int random[100];

for(int i = 0 ; i < 100; ++i)

{

random[i] = earc4random(10,30);

}

Enjoy. (:

How to get the MAC address on Android

May 24, 2013 / Jan Rabe / 2 Comments

(A test project can be found on github)

Since you can easily get confused with the many permissions provided by google, let me show you two ways to get the mac address on android. The first approach would be to use the WifiManager.getMacAdress().

/**
 * Returns MAC Address.
 * 
 * Note: requires  in AndroidManifest.xml
 * (credits: http://stackoverflow.com/a/15203979)
 *
 * @return MAC Address.
 */
public static String getMACAddress(@NotNull Activity activity) {
	return ((WifiManager) activity.getSystemService(Context.WIFI_SERVICE)).getConnectionInfo().getMacAddress();
}

/**

* Returns MAC Address.

* Note: requires in AndroidManifest.xml

* (credits: http://stackoverflow.com/a/15203979)

* @return MAC Address.

public static String getMACAddress(@NotNull Activity activity) {

return ((WifiManager) activity.getSystemService(Context.WIFI_SERVICE)).getConnectionInfo().getMacAddress();

}

However there are Disadvantages:

Device should have Wi-Fi (where not all devices have Wi-Fi)
If Wi-Fi present in device should be turned on otherwise does not report the MAC address
And the suspicious permission ACCESS_WIFI_STATE

Another approach is the use the NetworkInterface.getHardwareAddress()

/**
 * Returns MAC address of the given interface name.
 * 
 * credits to http://stackoverflow.com/questions/6064510/how-to-get-ip-address-of-the-device/13007325#13007325
 * 
 * Note: requires   and 
 *                  in AndroidManifest.xml
 * 
 * @param interfaceName eth0, wlan0 or NULL=use first interface 
 * @return  mac address or empty string
 */
  @TargetApi(Build.VERSION_CODES.GINGERBREAD)
 public static String getMACAddress(@NotNull final String interfaceName) {
     String macaddress = "";
     try {
         final List&lt;NetworkInterface&gt; interfaces = Collections.list(NetworkInterface.getNetworkInterfaces());
         for (final NetworkInterface intf : interfaces) {
             if (interfaceName != null) {
                 if (!intf.getName().equalsIgnoreCase(interfaceName)) continue;
             }
             final byte[] mac = intf.getHardwareAddress();
             if (mac==null) continue;
             final StringBuilder buf = new StringBuilder();
             for (int idx=0; idx &lt;= mac.length; idx++)
                 buf.append(String.format("%02X:", mac[idx]));
             if (buf.length()&gt;=0) buf.deleteCharAt(buf.length()-1);
                 macaddress = buf.toString();
             if(!macaddress.equals("")) break;
         }
     } catch (final Exception ex) { } // for now eat exceptions
     return macaddress ;
 }

/**

* Returns MAC address of the given interface name.

* credits to http://stackoverflow.com/questions/6064510/how-to-get-ip-address-of-the-device/13007325#13007325

* Note: requires and

* in AndroidManifest.xml

* @param interfaceName eth0, wlan0 or NULL=use first interface

* @return mac address or empty string

@TargetApi(Build.VERSION_CODES.GINGERBREAD)

public static String getMACAddress(@NotNull final String interfaceName) {

String macaddress = "";

try {

final List<NetworkInterface> interfaces = Collections.list(NetworkInterface.getNetworkInterfaces());

for (final NetworkInterface intf : interfaces) {

if (interfaceName != null) {

if (!intf.getName().equalsIgnoreCase(interfaceName)) continue;

}

final byte[] mac = intf.getHardwareAddress();

if (mac==null) continue;

final StringBuilder buf = new StringBuilder();

for (int idx=0; idx <= mac.length; idx++)

buf.append(String.format("%02X:", mac[idx]));

if (buf.length()>=0) buf.deleteCharAt(buf.length()-1);

macaddress = buf.toString();

if(!macaddress.equals("")) break;

}

} catch (final Exception ex) { } // for now eat exceptions

return macaddress ;

}

Disadvantage: uses permissions too, although they are required for your app to access the internet anyway and more likely commonly known to the end-user.

Conclusion:

Carring about permissions can be a pain in the butt there are even other ways if you just want identifying app installations, too.

Android libgdx ShaderAssetLoader & StillModelLoader

May 11, 2013 / Jan Rabe / 0 Comments

I totally love how libgdx brings a lot of interfaces for your application to connect to the graphics library. For instance the asset manager is really thought through well in particular. It has some really useful methods like Assets.manager.getProgress(). Especially since you can easily write your own asset loader and be able to utilize the loading progress in your app.

1) Simply add your loader to the asset manager at some point.

AssetManager manager = new AssetManager();
manager.setLoader( StillModel.class, new StillModelAssetLoader( new InternalFileHandleResolver() ) );
manager.setLoader( ShaderProgram.class, new ShaderAssetLoader( new InternalFileHandleResolver() ) );

AssetManager manager = new AssetManager();

manager.setLoader( StillModel.class, new StillModelAssetLoader( new InternalFileHandleResolver() ) );

manager.setLoader( ShaderProgram.class, new ShaderAssetLoader( new InternalFileHandleResolver() ) );

2) load your assets at required places, e.g. loading screen
(Note: SHADER_PHONG = “data/Graphics/Shader/Phong”; and MODEL_TEAPOT = “data/Graphics/Shader/teapot.obj”)

manager.load( SHADER_PHONG, ShaderProgram.class );
manager.load( MODEL_TEAPOT, StillModel.class );

1 2	manager.load( SHADER_PHONG, ShaderProgram.class ); manager.load( MODEL_TEAPOT, StillModel.class );

3) get your shader or model wherever you need them

ShaderProgram phong = manager.get( SHADER_PHONG, ShaderProgram.class );
StillModel teapotMesh = manager.get( MODEL_TEAPOT, StillModel.class );

1 2	ShaderProgram phong = manager.get( SHADER_PHONG, ShaderProgram.class ); StillModel teapotMesh = manager.get( MODEL_TEAPOT, StillModel.class );

4) free the resource by unloading when you’re done with them (e.g. disposing your app; changing screens)

manager.unload( SHADER_PHONG );
manager.unload( MODEL_TEAPOT);

1 2	manager.unload( SHADER_PHONG ); manager.unload( MODEL_TEAPOT);

That’s the shader asset loader for loading custom shaders in libgdx. (Note: adapt your extensions)

public class ShaderAssetLoader extends SynchronousAssetLoader&lt;ShaderProgram, AssetLoaderParameters&gt; {

    private static final String VERTEX_SHADER_EXTENSION = ".vsh";
    private static final String FRAGMENT_SHADER_EXTENSION = ".fsh";

    public ShaderAssetLoader ( final FileHandleResolver resolver ) {
        super( resolver );
    }

    @Override
    public ShaderProgram load ( final AssetManager assetManager, final String fileName, final AssetLoaderParameters parameter ) {
        final ShaderProgram shader = new ShaderProgram( resolve( fileName + VERTEX_SHADER_EXTENSION ), resolve( fileName + FRAGMENT_SHADER_EXTENSION ) );
        if ( shader.isCompiled() == false ) {
            throw new IllegalStateException( shader.getLog() );
        }
        return shader;
    }

    @Override
    public Array getDependencies ( final String fileName, final AssetLoaderParameters parameter ) {
        return null;
    }
}

public class ShaderAssetLoader extends SynchronousAssetLoader<ShaderProgram, AssetLoaderParameters> {

private static final String VERTEX_SHADER_EXTENSION = ".vsh";

private static final String FRAGMENT_SHADER_EXTENSION = ".fsh";

public ShaderAssetLoader ( final FileHandleResolver resolver ) {

super( resolver );

}

@Override

public ShaderProgram load ( final AssetManager assetManager, final String fileName, final AssetLoaderParameters parameter ) {

final ShaderProgram shader = new ShaderProgram( resolve( fileName + VERTEX_SHADER_EXTENSION ), resolve( fileName + FRAGMENT_SHADER_EXTENSION ) );

if ( shader.isCompiled() == false ) {

throw new IllegalStateException( shader.getLog() );

}

return shader;

}

@Override

public Array getDependencies ( final String fileName, final AssetLoaderParameters parameter ) {

return null;

}

And here is the StillModel loader for loading 3D meshes:

public class StillModelAssetLoader extends SynchronousAssetLoader&lt;StillModel, AssetLoaderParameters&gt; {

    public StillModelAssetLoader ( FileHandleResolver resolver ) {
        super( resolver );
    }

    @Override
    public StillModel load ( AssetManager assetManager, String fileName, AssetLoaderParameters parameter ) {
        return ModelLoaderRegistry.loadStillModel( resolve( fileName ) );
    }

    @Override
    public Array getDependencies ( String fileName, AssetLoaderParameters parameter ) {
        return null;
    }
}

public class StillModelAssetLoader extends SynchronousAssetLoader<StillModel, AssetLoaderParameters> {

public StillModelAssetLoader ( FileHandleResolver resolver ) {

super( resolver );

}

@Override

public StillModel load ( AssetManager assetManager, String fileName, AssetLoaderParameters parameter ) {

return ModelLoaderRegistry.loadStillModel( resolve( fileName ) );

}

@Override

public Array getDependencies ( String fileName, AssetLoaderParameters parameter ) {

return null;

}

Conclusion: assets are handled well in libgdx :D

My bumpy Samsung Galaxy S2 android upgrade to Lungo (CyanogenMod) for developping purposes

May 10, 2013 / Jan Rabe / 2 Comments

Ok here I was, totally innocent thinking about upgrading my Samsung Galaxy S2. I did that before some time ago and had no problems whatsoever. However when trying to install Codename Lungo Mod (CyanogenMod) as described in things-to-do-before-and-after-installing-a-rom-troubleshooter some problems occured.

lungo

Things to do when installing a new rom:

“wipe data/factory reset”
“wipe cache partition”
Go to advanced and “wipe dalvik cache” and “wipe battery stats”
Go to mounts and storage and ”format system”
The go to “install zip from sdcard> choose zip from sdcard” and select the ROM file you want to install.
In case, the developer has recommended to install GApps too, you should install it just after flashing the ROM (use Step 5).
When it is installed, reboot the device

After wipping everything and trying to “install zip from sdcard> choose zip from sdcard” some weird “Status 7” assertion errors appeared. Which basically denied the installation because clockworkmod thought the rom was not compatable with my device which was caused by my rather super outdated clockworkmod . Upgrading clockworkmod didn’t work for some weird reason. Please let me know why it didn’t ate the update.zip if you happend to know.

Anyways, so I had to delete the assertions in the rom zip file as described in how-to-fix-android-custom-rom-status-7-installation-abort-assert-failed-error-in-recovery-easily. Well changing the updater-script was no big deal. However how to get the zip on the device which has no rom installed anymore? I could have rolled back the backup, copied the rom.zip on the device, wiped and attempt another install or push the rom via adb onto the device.

android

But that approach probably took just as long in the end. At first my laptop didn’t find the android device anymore although I’m developing on it on a daily basis so the USB driver were installed. After some research I’ve found a website describing that very problem how-to-solve-adb-device-not-found-error. Which basically let me install PdaNet. My Samsung mobile phone could be found via “adb devices” after I went into clockworkmod start screen, then plugged in the USB cable and then went into the mounting screen. The article stated to use adbfastboot but I have android platform tools installed and just used the officially shipped by google adb tool. Now I could simply “adb push rom.zip /sdcard/rom.zip” and “install zip from sdcard> choose zip from sdcard” , select the lungo and install the rom.

To my surprise the “Enable USB Debugging” menu item was missing which had to be enabled by tapping 7 times on the about information for some weird reason. Is Google hidding stuff like Mac now? Also to note that if you want to get your device “online” when developing look into how-to-fix-adb-device-is-listed-as-offline-on-android-4-2-2.

TL;DR: upgrading to a new rom can be quiet more troublesome than advertised. ^^’

Bonus: on this tiny journey I stumbled on a collection of android rom installation errors and also collection of useful Tools, Drivers, utilities, downloaders and various fixes for Androids and PC. .

Android: Saving Screenshots in OpenGLES

May 9, 2013 / Jan Rabe / 2 Comments

1) Get your helper methods:

public static void saveBitmap(@NotNull Bitmap screenshot, @NotNull String filePath, @NotNull String fileName) {
	OutputStream outStream = null;
	filePath = Environment.getExternalStorageDirectory().toString() + "/" + filePath;
	File dir = new File( filePath );
	dir.mkdirs();
	File output = new File( filePath, fileName );
	try {
		outStream = new FileOutputStream( output );
		screenshot.compress( Bitmap.CompressFormat.PNG, 100, outStream );
		outStream.flush();
		outStream.close();
		Logger.v( TAG, "Saving Screenshot [" + filePath + fileName + "]" );
	} catch ( FileNotFoundException e ) {
		Logger.e( TAG, "" + e.getMessage() );
	} catch ( IOException e ) {
		Logger.e( TAG, "" + e.getMessage() );
	}
}

@NotNull
public Bitmap captureScreenShot() {
	int width = Gdx.graphics.getWidth(); // use your favorite width
	int height = Gdx.graphics.getHeight(); // use your favorite height
	int screenshotSize = width * height;
	ByteBuffer bb = ByteBuffer.allocateDirect( screenshotSize * 4 );
	bb.order( ByteOrder.nativeOrder() );
	// any opengl context will do
	Gdx.graphics.getGL20().glReadPixels( 0, 0, width, height, Gdx.graphics.getGL20().GL_RGBA, Gdx.graphics.getGL20().GL_UNSIGNED_BYTE, bb );
	int pixelsBuffer[] = new int[screenshotSize];
	bb.asIntBuffer().get( pixelsBuffer );
	bb = null;
	Bitmap bitmap = Bitmap.createBitmap( width, height, Bitmap.Config.RGB_565 );
	bitmap.setPixels( pixelsBuffer, screenshotSize - width, -width, 0, 0, width, height );
	pixelsBuffer = null;

	short sBuffer[] = new short[screenshotSize];
	ShortBuffer sb = ShortBuffer.wrap( sBuffer );
	bitmap.copyPixelsToBuffer( sb );

	//Making created bitmap (from OpenGL points) compatible with Android bitmap
	for ( int i = 0; i &lt; screenshotSize; ++i ) {
		short v = sBuffer[i];
		sBuffer[i] = ( short ) ( ( ( v &amp; 0x1f ) &lt;&lt; 11 ) | ( v &amp; 0x7e0 ) | ( ( v &amp; 0xf800 ) &gt;&gt; 11 ) );
	}
	sb.rewind();
	bitmap.copyPixelsFromBuffer( sb );
	return bitmap;
}

@Override
public void captureScreenShot(@NotNull final String path, @NotNull final String fileName) {
	Bitmap screenshot = captureScreenShot();
	saveBitmap( screenshot, path, fileName );
	screenshot.recycle();
}

public static void saveBitmap(@NotNull Bitmap screenshot, @NotNull String filePath, @NotNull String fileName) {

OutputStream outStream = null;

filePath = Environment.getExternalStorageDirectory().toString() + "/" + filePath;

File dir = new File( filePath );

dir.mkdirs();

File output = new File( filePath, fileName );

try {

outStream = new FileOutputStream( output );

screenshot.compress( Bitmap.CompressFormat.PNG, 100, outStream );

outStream.flush();

outStream.close();

Logger.v( TAG, "Saving Screenshot [" + filePath + fileName + "]" );

} catch ( FileNotFoundException e ) {

Logger.e( TAG, "" + e.getMessage() );

} catch ( IOException e ) {

Logger.e( TAG, "" + e.getMessage() );

}

@NotNull

public Bitmap captureScreenShot() {

int width = Gdx.graphics.getWidth(); // use your favorite width

int height = Gdx.graphics.getHeight(); // use your favorite height

int screenshotSize = width * height;

ByteBuffer bb = ByteBuffer.allocateDirect( screenshotSize * 4 );

bb.order( ByteOrder.nativeOrder() );

// any opengl context will do

Gdx.graphics.getGL20().glReadPixels( 0, 0, width, height, Gdx.graphics.getGL20().GL_RGBA, Gdx.graphics.getGL20().GL_UNSIGNED_BYTE, bb );

int pixelsBuffer[] = new int[screenshotSize];

bb.asIntBuffer().get( pixelsBuffer );

bb = null;

Bitmap bitmap = Bitmap.createBitmap( width, height, Bitmap.Config.RGB_565 );

bitmap.setPixels( pixelsBuffer, screenshotSize - width, -width, 0, 0, width, height );

pixelsBuffer = null;

short sBuffer[] = new short[screenshotSize];

ShortBuffer sb = ShortBuffer.wrap( sBuffer );

bitmap.copyPixelsToBuffer( sb );

//Making created bitmap (from OpenGL points) compatible with Android bitmap

for ( int i = 0; i < screenshotSize; ++i ) {

short v = sBuffer[i];

sBuffer[i] = ( short ) ( ( ( v & 0x1f ) << 11 ) | ( v & 0x7e0 ) | ( ( v & 0xf800 ) >> 11 ) );

}

sb.rewind();

bitmap.copyPixelsFromBuffer( sb );

return bitmap;

}

@Override

public void captureScreenShot(@NotNull final String path, @NotNull final String fileName) {

Bitmap screenshot = captureScreenShot();

saveBitmap( screenshot, path, fileName );

screenshot.recycle();

}

2) get a queue ready

protected ConcurrentLinkedQueue screenshotRequestedQueue = new ConcurrentLinkedQueue&lt;&gt;();

1	protected ConcurrentLinkedQueue screenshotRequestedQueue = new ConcurrentLinkedQueue<>();

3) add new filepath the the queue via e.g. events:

@Override
public boolean tap(float x, float y, int count, int button) {
	screenshotRequestedQueue.add( Math.random() * Integer.MAX_VALUE + ".png" );
	return false;
}

@Override

public boolean tap(float x, float y, int count, int button) {

screenshotRequestedQueue.add( Math.random() * Integer.MAX_VALUE + ".png" );

return false;

}

4) IMPORTANT! save the screen at the end of the opengl render process. (Note: If try to capture somewhere in the middle you might end up with a black screen or any other funny effects.)

@Override
public void render(float delta) {

	...
	// render your stuff
	...

	// capture screen on request
	if ( !screenshotRequestedQueue.isEmpty() )
		captureScreenShot(SCREENSHOT_FOLDER, screenshotRequestedQueue.poll() );
}

@Override

public void render(float delta) {

...

// render your stuff

...

// capture screen on request

if ( !screenshotRequestedQueue.isEmpty() )

captureScreenShot(SCREENSHOT_FOLDER, screenshotRequestedQueue.poll() );

}

DroidCon Wednesday 10-04-2013 at Kosmos Berlin

April 10, 2013 / Jan Rabe / 0 Comments

11:15 Dos and Dont’s of Android UI by Juhari Lebtimäki (SnappT)

He talked about a lot of generic and common sense design guidelines. Which you probably know all along by books like “Don’t make me think” by Steve Krug just applied to on android. He gave a couple of recommended principles like

designing for mdpi resolutions (1dp = 1px = mdpi) and scale from there (android dimensions)
use conventional icons when appropriate
use 9-patch
action bar on top and not on bottom
naturally swipe between slides (instead of teleporting by tabs)
errors on top like google play app (leats surprising)
be careful with tutorials, instructions are bad, try to be self-explanatory
popups are bad, because it interrupts use flow for no real reason and
use notifications instead, android did a good job at that
first time start like evernote for complicated usages
be conistent with the navigation as much as possible
let the user use the app without login for the first start even with restricted functionionality, avoid invoking the user to feel like doing a blind bargain
don’t hide the status bar, your app is not the reason the user has his phone!
customization is not an alternative to design
user tests for obvious reasons.

Well nothing really new but it was a well hold presentation nonetheless.

11:40 ProGuard, Optimizer and Obfuscator in the Android SDK by Eric Lafortune (Saikoa)

He basically gave an overview about what proguard does. That it is shrink, optimizes and obfuscates the final release versions of your android apk. Turns out it is a rather powerful tool. I’m using it for protecting code and removing logging so far. There is also a commercial version called dexguard.

12:20 Tracking User Behaviour Creatively by Kiana Tennyson (Manheim)

Google Analytics how-to tutorial kind of talk. Let’s hope people with bad intentions don’t abuse it.

12:20 Build your own app analytics tools by Matthais Huber and Monica Tudora (1&1 Internet)

Monica showed some simple code examples how to track user input and publishing it to private servers in order to improve the app with ACRA and heatmaps. Probably bigget advantage of that approach is that you don’t need to publish anylitic statistics on public and / or 3rd person services. Also she made me aware of how easy it is to create own analytic tools. It can be simple and still can be useful.

13:00 LUNCH

14:00 Firefox OS by Andreas Gal (Modzilla)

It’s quiet interesting concept of having an entire OS being programmable by html, js and css. Sadly very presented too quiet.

14:45 Re-Engineering the Facebook SDK by Connor Treacy (Facebook)

Connor Treacy presented the latest Facebook SDK 3 changes compared to 2. The sdk is easier to use now, and probably easier as well.

15:25 Real Time High Performance Multimedia Systems on Android by Prashant Dixit (Ittiam Systems)

slides will follow…

16:35 Biitstrap Continues Testing for Android in 5 Minutes by Andreas Lüdeke (TestObject)

http://www.cloudbees.com/#slide-2 and http://testobject.com/ in toolchain

DroidCon Tuesday 9-04-2013 at Kosmos Berlin

April 9, 2013 / Jan Rabe / 0 Comments

11:15 Efficient Android Threading by Anders Göransson (Jayway)

Anders gave a nice overview with code examples about multiple ways to do threading on android. Especially the use cases were distinguished rather well.

11:55 Incorporating touch feedback in your apps by Peter van der Linden (Immerson)

Very good and entertaining talk about the haptic feedback sdk. (demo app in android market) I’m definitely going to test in the near future. Because it looks easy to use but still powerful to give vibration feedback with your android device. And a tweet about it got me a stylish android hat :D

12:35 Libgdx game development framework by Alexander Fröhlich and Thomas Madsen (Andlabs)

Well it was kinda like the same talk as yesterday. You had the feeling they only used it for like one day and not went any further. Maybe it’s the restricted time limit of half an hour for the talk. At any rate. I didn’t learn anything new that I couldn’t have learned within a few tutorials myself. Except maybe the recommendation about tile map editors – like tile map editor – and the Physics Body Editor, especially the last one looked really promising for future boxed2d projects. At least the presentation has been done with prezi so the slide transitions weren’t boring. xD

13:15 LUNCH ^_^ <3

14:15 Designing Accessible Android Applications by Sonia Sharman (Google)

Sonia basically advertised to use android build-in functionality like AccessabilityEvents and AccessabilityServices in order to support the use for sight or else handicapped people. It reminded of accessibility guide lines for websites. For instance, minimize chatter, provide feedback, label UI elements / tagging and grouping view elements properly in order to help text to speech programs to be actually useful. I think it’s very important but I can also imagine to most of the developer it will be last on their checklist since their android apps are not needed to be tagged like websites do for SEO reasons. She also mentioned talkback and tripple-tap-zoom. Further informations can be found in the official android documentation.

Also quiet interesting is the Google I/O 2012 talk about it:

15:00 Improving Android Experience by Pavel Lahoda (Actiwerks)

Pavel talked about how to improve the UI. He suggested that android developer can learn from web developers in order to support multiple screen resolutions and layouts for instance using responsive design. Some mentioned solutions were standout – floating windows, fly-out menu, actionbarsherlock. He also pointed out that the native android sdk doesn’t provide enough functionality for that purpose. Not for all devices and all screen resolutions anyway.

15:40 Location and vision based AR by Simon Heinen and Mostafa Akbari (Bitstarts)

Well regrettably this was merely an advertising show off bitstars AR SDK. They basically use a JMonkey Engine to render, GPS (>10-30 meter precision), WIFI, Step detection (> 5 meter precision) and SLAM (>1 centimeter precision) for getting precise locations. At any rate augmented reality is cool especially without markers and I’m looking forward for awesome applications in the future.

16:50 Unifying Callbacks in Android by Matt Brenner (UnME2)

Matt introduced his cooked up EventConsumer Pattern. Which seems to be a way to minimize the boilerplate for events. Although based on the reaction by the audience it didn’t quiet convince everyone. I think it’s a good idea to minimize doubled code though.

17:30 Understanding the GPU to write better code by Guillem Vinals Gangolells (Imagination Technologies)

The last talk today was quiet interesting. Rather low level but a lot of generalised recommendation. Mostly Guillem talked the graphic pipeline for OpenGL on android. I’m going to put his presentation online as soon as they publish it.

Conclusion

Today were far better and mostly way more organised talks. I learned quiet a few new things for which I’m grateful. For instance the haptic sdk, the physics body editor, standalone, fly-out actionbarsherlock are worth looking into. I would have wished for deeper information (not something i can read up in first timer tutorials) at the libgdx talk and less vacuum-cleaner salesman feeling at the android augmented reality talk. Other than that it was well worth attending.

DroidCon Barcamp 8-04-2013 at Kosmos Berlin

April 8, 2013 / Jan Rabe / 0 Comments

This is going to be a short recap for myself about the DroidCon 2013 in the Kosmos Berlin. In deepest respect I thank every presenter for their hard work who took the time to share their knowledge and insights. The following is purely my observation and might be very subjective. If anything appears to be offending please read it as my personal opinion.

11:45 – Interaction and beyond – Wolter

He basically showed of a couple of different devices to develop for and encouraged everyone to think out of the box to get your application as appropriate as possible to increase the end user experience. For instance:

Nothing new in particular though.

12:45 – Libgdx Thomas

He gave a very brief overview about the high level layer of the framework and some components like the SpriteBatch, InputProcessors, how to get the deltaT value, and how to swap some screens. I was a bit disappointed, since I was – work related – looking forward for this talk in particular.

13:15 FOOD! ^_^ Well there was a very delicious all you can eat lunch buffet, which basically amortised the 10€ entry fee* already. (*which got sponsored by my awesome employer, thank you very much Wooga <3)

14:15 – Push Notifications with Netmera / Android Software Craftsmanship

Well this talk was kinda boring, since the presenter only walked you through the process of using their service layer for using google cloud notifications. He didn’t give much insight or reasoning behind the steps he clicked through. Felt like something I could read up later in a tutorial. So I changed to: Android Software Craftsmanship. Which basically was a free highly unprepared discussion about the common ways to develop software. (not a bad thing in this case) It appeared to be consens for all of he participants to use the test-driven approach. Using version controle like git and svn and even use a continues integration development layer to not mess up things in the long run. If something is not automatable: use conventions and iterate over them so they don’t force outdated rules. Also to abstract layers like DAO-Pattern in order to test them more separately has been recommended. Another recommendation was to use integration and behavioural tests for obvious reasons.

The very most and interesting part for myself was to get to know code retreat. Which basically makes you start a project (possibly pair programming) and work on it for 45 min and then throw every code base away in order to start over the same project again. The goal is not to finish the project but to approach the same problem 6-7 times. Usually differently and most likely better. (e.g. Coneway’s Game of Life) Personal I think it’s a pretty cool and fun idea to improve your own skills.

15:00 Mobility Android

Basically they gave an introduction about their flat search app and some thought process behind certain layout questions. They also opened a discussion on how to improve it even further. Turns out there is always someone with a bright idea to about everything. I think these kind of discussions are very worthwhile attending since you won’t forget the reasons for solving problems in a certain kind of way.

15:45 Reverse Engineering Ingris

Dirk Jäckel put his presentation online. Well – as the title states. If you’re into these kind of stuff it’s worth reading. For me however it was a bit too unrelated to everything I’m doing. So I’m not going into details about that one.

15:45 NFC in Games by Hannes Breul

Since the reverse engineering talk was shorter than expected I joined that talk at the middle. He was giving a good summery about where NFC could be used and about it’s strengths and weaknesses. For instance it can be used for location based games, crowd games with single/multiple locations, since it can be used for transmitting content. I didn’t quiet get all the details, but it sounded quiet interesting. I’m going to link his presentation, as soon as he puts it online.

16:15 Sascha Pallenberg from MobileGeeks opened a discussion about the awareness of the public imagine from younger generations on the internet. And that there should be some sort of education for it. So young people don’t put content online which could ruin their job carrer years ahead before actually applying. Which is troublesome since the knowledge about teachers and younger generations seems to be reverse. Pupils might know more about key technologies than teachers but are not entirely aware of the consequences of their public and transparent internet image and how hard it is to erase unfavourable content. It turned out to be some quiet interesting discussion to follow for one because of Sascha’s well communication skills and for the other because the very different opinion about how transparency could be interpreted in the near future by HR people.

17:15 What to consider when teaming up with a co-founder.

Was quiet interesting as well. Especially the consens to write everything down before starting anything. Better be safe than sorry I suppose.

17:45 Demo/Judgment

Basically a couple of startups showed off their latest android app and were looking for feedback, new designers, developer, business people to join their team or to find new investors. I’m gonna talk about this later when the judgement people are done judging and put their stuff online.

Conclusion

Monday the Barcamp was probably designed for people who haven’t done much with android yet I suppose, since it was rather surficial for my taste. Anyway I got to know some new people which was totally worth participating. It’s all about the people there after all. And the food was quiet enjoyable, which usually convince me to recommend it. :D

How to boost android canvas performance with a buffered Image

January 7, 2013 / Jan Rabe / 2 Comments

When it comes to drawing to canvas on android one very big performance boost can be gained by using a buffered bitmap image. The idea is to draw lines to a bitmap. And then draw the bitmap to the screen.
This approach avoids unnecessarily redrawing elements that has been drawn already multiple times.

1) Create a bitmap in your view.

public View(Activity activity) {
    super(activity);
    DisplayMetrics metrics = new DisplayMetrics();
    activity.getWindowManager().getDefaultDisplay().getMetrics(metrics);
    bufferedImage = Bitmap.createBitmap(metrics.widthPixels, metrics.heightPixels, Bitmap.Config.ARGB_8888);
    .
    .
    .
}

public View(Activity activity) {

super(activity);

DisplayMetrics metrics = new DisplayMetrics();

activity.getWindowManager().getDefaultDisplay().getMetrics(metrics);

bufferedImage = Bitmap.createBitmap(metrics.widthPixels, metrics.heightPixels, Bitmap.Config.ARGB_8888);

}

2) Remember all touch events.

@Override
public boolean onTouch(View view, MotionEvent motionEvent) {
    int action = motionEvent.getAction();
    Point point = new Point(motionEvent.getX(), motionEvent.getY());
    switch (action) {
        case MotionEvent.ACTION_DOWN: // touch down
            linePathManager.startLine(point);
            break;
        case MotionEvent.ACTION_MOVE:  // touch drag
            linePathManager.addPoint(point);
            break;
        case MotionEvent.ACTION_UP: // stop touching
            linePathManager.endLine();
            break;
    }
    invalidate();
    return super.onTouchEvent(motionEvent);
}

@Override

public boolean onTouch(View view, MotionEvent motionEvent) {

int action = motionEvent.getAction();

Point point = new Point(motionEvent.getX(), motionEvent.getY());

switch (action) {

case MotionEvent.ACTION_DOWN: // touch down

linePathManager.startLine(point);

break;

case MotionEvent.ACTION_MOVE: // touch drag

linePathManager.addPoint(point);

break;

case MotionEvent.ACTION_UP: // stop touching

linePathManager.endLine();

break;

}

invalidate();

return super.onTouchEvent(motionEvent);

}

3) Process all touch events. (Draw lines or something)

private Path path = new Path();

public void startLine(@NotNull Point point) {
    path.moveTo(point.x, point.y);
}
 
public void add(@NotNull Point p) {
    path.lineTo(p.x, p.y);
    path.moveTo(p.x, p.y);
}

public void close() {
    path.close();
}

private Path path = new Path();

public void startLine(@NotNull Point point) {

path.moveTo(point.x, point.y);

}

public void add(@NotNull Point p) {

path.lineTo(p.x, p.y);

path.moveTo(p.x, p.y);

}

public void close() {

path.close();

}

4) Draw to buffered bitmap image.

lineManager.draw():

private Canvas canvas = new Canvas();

public void draw(Bitmap bitmap, Paint paint) {
    canvas.setBitmap(bitmap);
    canvas.drawPath(path, paint);
}

private Canvas canvas = new Canvas();

public void draw(Bitmap bitmap, Paint paint) {

canvas.setBitmap(bitmap);

canvas.drawPath(path, paint);

}

5) And finally draw the buffered bitmap image.

@Override
public void onDraw(Canvas canvas) {
    super.onDraw(canvas);
    linePathManager.draw(bufferedImage, paint);
    canvas.drawBitmap(bufferedImage, 0, 0, paint);
}

@Override

public void onDraw(Canvas canvas) {

super.onDraw(canvas);

linePathManager.draw(bufferedImage, paint);

canvas.drawBitmap(bufferedImage, 0, 0, paint);

}

Bonus: invalidate canvas only around finger area and only every 90 milliseconds.

@Override
public boolean onTouch(View view, MotionEvent motionEvent) {
    int action = motionEvent.getAction();

    Point point = new Point(motionEvent.getX(), motionEvent.getY());
    switch (action) {
        case MotionEvent.ACTION_DOWN: // touch down
            linePathManager.startLine(point);
            break;
        case MotionEvent.ACTION_MOVE:  // touch drag
            linePathManager.addPoint(point);
            break;
        case MotionEvent.ACTION_UP:
            linePathManager.endLine();
            break;
    }

    // invalidate only finger area
    if (Calendar.getInstance().getTimeInMillis() - startTime > 90) {
        int rectangleSize = (int) (100 * metric.density);
        invalidate((int) (point.x - rectangleSize / 1.5), (int) (point.y - rectangleSize / 1.5), (int) point.x + rectangleSize, (int) point.y + rectangleSize);
        startTime = Calendar.getInstance().getTimeInMillis();
    }
    return true;
}

@Override

public boolean onTouch(View view, MotionEvent motionEvent) {

int action = motionEvent.getAction();

Point point = new Point(motionEvent.getX(), motionEvent.getY());

switch (action) {

case MotionEvent.ACTION_DOWN: // touch down

linePathManager.startLine(point);

break;

case MotionEvent.ACTION_MOVE: // touch drag

linePathManager.addPoint(point);

break;

case MotionEvent.ACTION_UP:

linePathManager.endLine();

break;

}

// invalidate only finger area

if (Calendar.getInstance().getTimeInMillis() - startTime > 90) {

int rectangleSize = (int) (100 * metric.density);

invalidate((int) (point.x - rectangleSize / 1.5), (int) (point.y - rectangleSize / 1.5), (int) point.x + rectangleSize, (int) point.y + rectangleSize);

startTime = Calendar.getInstance().getTimeInMillis();

}

return true;

}

My Tweets

Frequently visiting

My Cloud

Amazing people

Favorite Tools

APIs

Online Tools

Tutorials

My Latest Animes

Categoryandroid

copy assets script for multiple resolutions for android

c++ arc4random on android, windows and linux

How to get the MAC address on Android

Android libgdx ShaderAssetLoader & StillModelLoader

My bumpy Samsung Galaxy S2 android upgrade to Lungo (CyanogenMod) for developping purposes

Android: Saving Screenshots in OpenGLES

DroidCon Wednesday 10-04-2013 at Kosmos Berlin

DroidCon Tuesday 9-04-2013 at Kosmos Berlin

DroidCon Barcamp 8-04-2013 at Kosmos Berlin

How to boost android canvas performance with a buffered Image