Sven Gothel (89): Bug 1063 - Utilize own Uri and Uri.Encoded class w/ proper encoding and differentiating encoded/decoded variants by type [Part 1] Bug 1063 - Uri.PATH_LEGAL: Remove reserved 'punct', encoding all reserved chars but '!', allows removal of PATH_MIN_LEGAL ; Add comment for Harmony values for review Bug 1063 - Refine Uri and Uri.Encode ; Fix unit tests (passing on GNU/Linux, OSX and Windows) Bug 1063: Complete Uri impl. and adoption throughout GlueGen Bump 2.3.0 Bug 1063: Further Uri completion - As a result of JOGL Uri adoption Bug 1063: Uri: Refine API doc; Add create(Encoded ..) ; Provide common impl. for getNormalized(), getDirectory(), getParent() and getRelativeOf() Bug 1067 - IOUtil.testDirExec(File) needs to preserve SPACE in exetst.getCanonicalPath() command Bump semver to 0.9.30-SNAPSHOT of jogamp branch, incl. commit 9cdb6bbf6d3248a65aa78ced08211d609c356f91 IOUtil: Fix API doc; getTempDir(..) shall throw IOException instead of RuntimeException for IO failures Bug 1080 - Add read support for memory mapped big file I/O via specialized InputStream impl., incl. mark/reset Adapt unit tests to commit 1350823035597f784f9cf871aa487f896f3d1840: getTempDir(..) throws IOException Bug 1080 - Fix TestByteBufferInputStream: Handle OutOfMemoryError cause in IOException (Add note to FLUSH_NONE); Reduce test load / duration. Bug 1080 - Add write support for memory mapped big file I/O via specialized OutputStream impl. Bug 1080 - Refine MappedByteBuffer*Stream impl. and API [doc], adding stream to stream copy as well as direct memory mapped ByteBuffer access MappedByteBuffer*Stream: Make Scripts OSX: Remove -java7 script, add -java6 script (default uses java8) TestByteBufferCopyStream, TestByteBufferInputStream: Reduce load and test time TestByteBufferInputStream: Reduce FLUSH_SOFT load MappedByteBufferInputStream: Default CacheMode is FLUSH_PRE_HARD now (was FLUSH_PRE_SOFT) SingletonTestCase: Move singleton test impl. from JOGL's UITestCase, inherit from JunitTracer and use it for all tests. gitignore: build*/ and build-temp SingletonTestCase: SINGLE_INSTANCE_LOCK_TO (timeout) 6 -> 12 minutes SingletonTestCase -> SingletonJunitCase: Accomodate ClassLoader lifecycle of static fields; Change name to avoid testing itself. SingletonJunitCase: Refine poll and release behavior SingletonJunitCase: Increase TO 12 -> 15 mins, only sleep for 1 poll period (500 ms) per JVM after test class (cooperative w/ other JVM test instances) Add ExceptionUtils: Exposing dumpStackTrace(..) and dumpThrowable(..) Refine ExceptionUtils: Shorten method name, dumpThrowable uses dumpStack ExceptionUtils.dumpStack(..): Skip self stack entry Bug 1106 - Bitstream: Simplify 'msbFirst' case for bulk operations / Add setting of stream position (optional) Bug 1108 - IOUtil.getTempDir(..): Difficulties to determine executable access permission on Windows via BATCH file execution GlueGen JavaEmitter: Use proper class access modifiers for emitted classes incl. ProcAddressTable test scripts: Bump JRE, add TEMP/TMP settings TempFileCache/TempJarCache: Fix NPE in DEBUG mode Bug 1108 - Fix difficulties to determine executable access permission on Windows via BATCH file execution ProcAddressConfiguration: Allow setting default value for 'GetProcAddressTableExpr' ExceptionUtils.dumpThrowable(..): Also dump all causes of the given Throwable Bug 1124 - Android Build: Use SDK >= 24.0.2 (build-tools >= 21.1.2) and NDK >= 10d Bug 1123 - Android Launcher/ClassLoaderUtil: Fails to work w/ Android 5.* (lollipop/art) Bug 1122 - Add Android API 21 jar file and source zip GlueGen: Make const array-length getter static, if constant. Bug 1122 - Elf Parser: EM_res183 -> EM_AARCH64, detecting AArch64; Updating Shdr and Ehdr via GlueGen Bug 1122: Add AArch64 support (Android, GNU/Linux and in general) GlueGen: Cleanup static generated JNI code (fix method names) Bug 1122: Reflect __LP64__ and _aarch64__ in GlueGen's stdint/stddef and Android compilerflags Refine Native Library Code: Bulk Permissions, Cleanup DynamicLinker impl. - and fix Android AArch64 BionicDynamicLinker (Bug 1122) Bug 1122: Refine AArch64 and OSType/32Bit incl. MachineDescription Bug 1125 - Make ELF Reader 'jogamp.common.os.elf' Stateless Minor Cleanup: Buffers: Remove !JAVA_6 branch; NativeLibrary: Reuse isOSX detection Bug 1126 - Remove static query requirement of MachineDescriptor, find matching StaticConfig at runtime; Fix PPC (Bug 1056) and MIPSLE (Bug 1014) issues. Remove unused armv7 scripts and gluegen piece GlueGen AAPT Task: Android APK native-code files shall be named using 'os.and.arch', instead of android.abi for harmonization GlueGen: Adapt APK install scripts for harmonized os.and.arch names, see commit 3b43a223253176731567c6d8b7a67c9a6110782b Add javadoc arg '-Xdoclint:none' if java >= 1.8 via property 'javadoc.xarg1' avoiding javadoc build failure Fix regression of commit 3caf446e29a3934900b9983dfd72cb8aa0d9e8d7: Win64 is not LP64 build setup script: remove redundant ANT_PATH entry Refine commit 12feaa7d3b1544098f684d851e3caff1ec88cbc8: Add 'throws SecurityException' decl., remove dead code, remove redundant check. Bug 1134 - Enhance GlueGen Compiler: Minimal GL Header Changes _and_ Typesafety Bug 1134 - Fix regression: Static C-Function call must use original API name Bug 1134 - Add ASTLocationTag, locating source of [semantic] errors while parsing / analyzing Bug 1134 - Refine Logging using 'LoggerIf' - Replace System.err w/ Logging where appropriate Bug 1134 - Utilize AliasedSymbol where required in *Configuration; Clarify ProcAddressEmitter criteria Bug 1134 - Fix ProcAddressEmitter.getFunctionPointerTypedefName() ; Fix JavaEmitter's Function/Struct Emission Bug 1134 - GlueGenException.toString(): Produce a 'gcc' like/compatible error message Bug 1134 - Fix regression on OpenSolaris/gcc: Redefine in test1.h (test case) TestByteBuffer*Stream: Reduce mmap load on poor OSX mini test machine to reduce test time (was: > 10 min) Bug 1134 - Pass ASTLocationTag to all types, used for GlueGenException Bug 1134 - Handle Opaque fields in structs; Print struct field name in apidoc of getter/setter Bug 1134 - Fix IntType: Add getCName(..) for proper C-type code; Fix its newCVVariant(..), don't pass given cv-attr as typedef Fix Type.getDebugString() comma separation Bug 1134 - Use ASTLocationTag in Logging (PCPP, Emitter); Refine ASTLocationTag log/exception formatting. Bug 1134 - Logging's formatter now takes care of Type -> getDebugString() when desired, not at 'log' call Bug 1134 - Make ASTLocusTag's text optional Bug 1134 - Fix IntType 'unsigned'; Type: Use copy-ctor instead of Clonable, .. Bug 1134 - Fix aliased typedef struct emission Bug 1134 - Fix CMethodBindingEmitter return type '_res' qualifiers Bug 1134 - ProcAddressCMethodBindingEmitter: Use available 'ProcAddrTypedef' for type validation w/ local generated variant (robustness) Bug 1134 - Fix CMethodBindingEmitter return type '_res' qualifiers (part 2) Bug 1134 - In case of 'undefined type' throw a semantic GlueGenException instead of an NPE Bug 1134 - Add ASTLocusTagProvider for Define and fix newline in c-parser (Expose source location for log/error messages) Bug 1134 - Refine err/log message of 'previous definition is here' Bug 1144 - Add 'DelegateImplementation': Cleanup JavaConfiguration and use manuallyImplement(AliasedSymbol) for FunctionSymbol Bug 1144 - Add 'DelegateImplementation': Sort JavaMethodBindingEmitter argument list; forImplementingMethodCall -> isNativeMethod Bug 1144 - Add 'DelegateImplementation': Cleanup MethodBinding/FunctionBinding Semantics Bug 1144 - Add 'DelegateImplementation', manually impl. may delegate to renamed original Bump for 2.3.0 release Bug 1134 - LOG AliasedSymbol: Don't issue getAliasedString(), logger takes care of that, when it logs. Bug 1144 - Add 'DelegateImplementation': Requires own MethodBinding for delegates impl. / Adding ReturnsOpaque Bug 1144 - Fix commit bbea09816015ecf3596acdcc033553127fcc0ef3 (missing file)