Search Results

Search found 1385 results on 56 pages for 'dependent'.

Page 39/56 | < Previous Page | 35 36 37 38 39 40 41 42 43 44 45 46  | Next Page >

  • strange raid5 issue: [closed]

    - by 8steve8
    ok so ive had a 4x2TB(samsung HD204UI w/firmware patch) raid5 array working normally for about a month. It was in a h57 gigabyte motherboard using the intel raid with windows 7 x64. Today I got an intel h67 motherboard, so I upgraded the intel raid drivers to 10.1.0.1008 from 9.6.0.1014, and I'm not sure if i checked after a reboot, but it caused no problems. I swapped in the new dh67 motherboard, and my array status was "failed". 2 of the 4 drives listed themselves as members, while the other two drives listed themselves as non-members. I tried going back to the old h57 mobo, and downgrading the raid drivers, but the issue remains. It's not port dependent, 2 of the drives always come up as non-members regardless of what port or motherboard they are plugged into. the screenshot should show that the SNs match, which begs the question why the software doesn't realize the drive is a member of the array: http://img837.imageshack.us/img837/6145/both.png I'd like to know if anyone has experienced anything similar, and what should i do, can i force the drive to be recognized as a member (without wiping data)?

    Read the article

  • Amazon EC2 - Free memory

    - by Damo
    We have an amazon ec2 small instance running and over the past few days we noticed that the memory is going down and down. On the small instance, we are running apache and tomcat6 Tomcat is started with the following JVM parameters -Xms32m -Xmx128m -XX:PermSize=128m -XX:MaxPermSize=256m We use nagios to monitor stuff like updates to apply, free disk space and memory. Everything else is behaving as expected but our memory is going down all the time. Our app receives approx half a million hits a day When I shutdown apache and tomcat, and ran free -m, we had only 594mb of memory free out out of the 1.7gb of memory. Not much else is running on the small instance and when running the top command I cannot see where the memory is going. The app we run on tomcat is a grails webapp. Could there be a possibility that there is a memory leak within our application? I read online and folks say that a small amazon instance is perfect for running apach and tomcat. I found a few posts online that showed how to setup apache and tomcat to limit the memory usage and I have already performed those steps. The memory is not being used up as quick but the memory is still decreasing over time. We have other amazone ec2 small instances running grails apps and the memory is fairly standard on those nodes. But they would not be receiving as much traffic Just to add, when I run the top command on the problem server, I cannot see where all the memory is being used Any help with this is greatly appreciated The output of free -m when run on my server is as follows total used free shared buffers cached Mem: 1657 1380 277 0 158 773 -/+ buffers/cache: 447 1209 Swap: 895 0 895 In your opinion, does this look ok? At what stage would the OS give back memory, would it wait to the memory reaches 0% or is this OS dependent?

    Read the article

  • What should I use to ping multiple IPs and get notified of time outs?

    - by HumanVirus
    I've been using MultiPing to ping hundreds of IPs (from access points and such) and check their performance (packet loss, latency) and uptime. The program is very easy to use, but I was wondering if someone could recommend me something that would work better and that would also work in Linux. The features I'm looking for are: Notification Types: At least desktop notifications and SMS, but it would be great if it also had e-mail, IM, or other types of notifications. (MultiPing has some of these, but they don't work too well.) Being notified about the root problem only: Since some devices are dependent on others, I'd like to be notified only about the root problem. E.g. Let's say I have A[x.x.x.222]B[x.x.x.33C[x.x.x.44]D[x.x.x.55], and B goes down, therefore C and D will also be down. Is it possible to get a notification only about B being down? Light on resources. Ideally multiplatform or at least available for both Linux and Windows. I've heard about Nagios and Shinken being used for monitoring. Would you recommend that I use something of the sort or would that be too much for my needs? If using Nagios, Shinken, or similar software is recommended, can anyone tell me what sites I should go to or what books I should get that would be good for someone who is totally new at this? I'd appreciate any suggestions.

    Read the article

  • value types in the vm

    - by john.rose
    value types in the vm p.p1 {margin: 0.0px 0.0px 0.0px 0.0px; font: 14.0px Times} p.p2 {margin: 0.0px 0.0px 14.0px 0.0px; font: 14.0px Times} p.p3 {margin: 0.0px 0.0px 12.0px 0.0px; font: 14.0px Times} p.p4 {margin: 0.0px 0.0px 15.0px 0.0px; font: 14.0px Times} p.p5 {margin: 0.0px 0.0px 0.0px 0.0px; font: 14.0px Courier} p.p6 {margin: 0.0px 0.0px 0.0px 0.0px; font: 14.0px Courier; min-height: 17.0px} p.p7 {margin: 0.0px 0.0px 0.0px 0.0px; font: 14.0px Times; min-height: 18.0px} p.p8 {margin: 0.0px 0.0px 0.0px 36.0px; text-indent: -36.0px; font: 14.0px Times; min-height: 18.0px} p.p9 {margin: 0.0px 0.0px 12.0px 0.0px; font: 14.0px Times; min-height: 18.0px} p.p10 {margin: 0.0px 0.0px 12.0px 0.0px; font: 14.0px Times; color: #000000} li.li1 {margin: 0.0px 0.0px 0.0px 0.0px; font: 14.0px Times} li.li7 {margin: 0.0px 0.0px 0.0px 0.0px; font: 14.0px Times; min-height: 18.0px} span.s1 {font: 14.0px Courier} span.s2 {color: #000000} span.s3 {font: 14.0px Courier; color: #000000} ol.ol1 {list-style-type: decimal} Or, enduring values for a changing world. Introduction A value type is a data type which, generally speaking, is designed for being passed by value in and out of methods, and stored by value in data structures. The only value types which the Java language directly supports are the eight primitive types. Java indirectly and approximately supports value types, if they are implemented in terms of classes. For example, both Integer and String may be viewed as value types, especially if their usage is restricted to avoid operations appropriate to Object. In this note, we propose a definition of value types in terms of a design pattern for Java classes, accompanied by a set of usage restrictions. We also sketch the relation of such value types to tuple types (which are a JVM-level notion), and point out JVM optimizations that can apply to value types. This note is a thought experiment to extend the JVM’s performance model in support of value types. The demonstration has two phases.  Initially the extension can simply use design patterns, within the current bytecode architecture, and in today’s Java language. But if the performance model is to be realized in practice, it will probably require new JVM bytecode features, changes to the Java language, or both.  We will look at a few possibilities for these new features. An Axiom of Value In the context of the JVM, a value type is a data type equipped with construction, assignment, and equality operations, and a set of typed components, such that, whenever two variables of the value type produce equal corresponding values for their components, the values of the two variables cannot be distinguished by any JVM operation. Here are some corollaries: A value type is immutable, since otherwise a copy could be constructed and the original could be modified in one of its components, allowing the copies to be distinguished. Changing the component of a value type requires construction of a new value. The equals and hashCode operations are strictly component-wise. If a value type is represented by a JVM reference, that reference cannot be successfully synchronized on, and cannot be usefully compared for reference equality. A value type can be viewed in terms of what it doesn’t do. We can say that a value type omits all value-unsafe operations, which could violate the constraints on value types.  These operations, which are ordinarily allowed for Java object types, are pointer equality comparison (the acmp instruction), synchronization (the monitor instructions), all the wait and notify methods of class Object, and non-trivial finalize methods. The clone method is also value-unsafe, although for value types it could be treated as the identity function. Finally, and most importantly, any side effect on an object (however visible) also counts as an value-unsafe operation. A value type may have methods, but such methods must not change the components of the value. It is reasonable and useful to define methods like toString, equals, and hashCode on value types, and also methods which are specifically valuable to users of the value type. Representations of Value Value types have two natural representations in the JVM, unboxed and boxed. An unboxed value consists of the components, as simple variables. For example, the complex number x=(1+2i), in rectangular coordinate form, may be represented in unboxed form by the following pair of variables: /*Complex x = Complex.valueOf(1.0, 2.0):*/ double x_re = 1.0, x_im = 2.0; These variables might be locals, parameters, or fields. Their association as components of a single value is not defined to the JVM. Here is a sample computation which computes the norm of the difference between two complex numbers: double distance(/*Complex x:*/ double x_re, double x_im,         /*Complex y:*/ double y_re, double y_im) {     /*Complex z = x.minus(y):*/     double z_re = x_re - y_re, z_im = x_im - y_im;     /*return z.abs():*/     return Math.sqrt(z_re*z_re + z_im*z_im); } A boxed representation groups component values under a single object reference. The reference is to a ‘wrapper class’ that carries the component values in its fields. (A primitive type can naturally be equated with a trivial value type with just one component of that type. In that view, the wrapper class Integer can serve as a boxed representation of value type int.) The unboxed representation of complex numbers is practical for many uses, but it fails to cover several major use cases: return values, array elements, and generic APIs. The two components of a complex number cannot be directly returned from a Java function, since Java does not support multiple return values. The same story applies to array elements: Java has no ’array of structs’ feature. (Double-length arrays are a possible workaround for complex numbers, but not for value types with heterogeneous components.) By generic APIs I mean both those which use generic types, like Arrays.asList and those which have special case support for primitive types, like String.valueOf and PrintStream.println. Those APIs do not support unboxed values, and offer some problems to boxed values. Any ’real’ JVM type should have a story for returns, arrays, and API interoperability. The basic problem here is that value types fall between primitive types and object types. Value types are clearly more complex than primitive types, and object types are slightly too complicated. Objects are a little bit dangerous to use as value carriers, since object references can be compared for pointer equality, and can be synchronized on. Also, as many Java programmers have observed, there is often a performance cost to using wrapper objects, even on modern JVMs. Even so, wrapper classes are a good starting point for talking about value types. If there were a set of structural rules and restrictions which would prevent value-unsafe operations on value types, wrapper classes would provide a good notation for defining value types. This note attempts to define such rules and restrictions. Let’s Start Coding Now it is time to look at some real code. Here is a definition, written in Java, of a complex number value type. @ValueSafe public final class Complex implements java.io.Serializable {     // immutable component structure:     public final double re, im;     private Complex(double re, double im) {         this.re = re; this.im = im;     }     // interoperability methods:     public String toString() { return "Complex("+re+","+im+")"; }     public List<Double> asList() { return Arrays.asList(re, im); }     public boolean equals(Complex c) {         return re == c.re && im == c.im;     }     public boolean equals(@ValueSafe Object x) {         return x instanceof Complex && equals((Complex) x);     }     public int hashCode() {         return 31*Double.valueOf(re).hashCode()                 + Double.valueOf(im).hashCode();     }     // factory methods:     public static Complex valueOf(double re, double im) {         return new Complex(re, im);     }     public Complex changeRe(double re2) { return valueOf(re2, im); }     public Complex changeIm(double im2) { return valueOf(re, im2); }     public static Complex cast(@ValueSafe Object x) {         return x == null ? ZERO : (Complex) x;     }     // utility methods and constants:     public Complex plus(Complex c)  { return new Complex(re+c.re, im+c.im); }     public Complex minus(Complex c) { return new Complex(re-c.re, im-c.im); }     public double abs() { return Math.sqrt(re*re + im*im); }     public static final Complex PI = valueOf(Math.PI, 0.0);     public static final Complex ZERO = valueOf(0.0, 0.0); } This is not a minimal definition, because it includes some utility methods and other optional parts.  The essential elements are as follows: The class is marked as a value type with an annotation. The class is final, because it does not make sense to create subclasses of value types. The fields of the class are all non-private and final.  (I.e., the type is immutable and structurally transparent.) From the supertype Object, all public non-final methods are overridden. The constructor is private. Beyond these bare essentials, we can observe the following features in this example, which are likely to be typical of all value types: One or more factory methods are responsible for value creation, including a component-wise valueOf method. There are utility methods for complex arithmetic and instance creation, such as plus and changeIm. There are static utility constants, such as PI. The type is serializable, using the default mechanisms. There are methods for converting to and from dynamically typed references, such as asList and cast. The Rules In order to use value types properly, the programmer must avoid value-unsafe operations.  A helpful Java compiler should issue errors (or at least warnings) for code which provably applies value-unsafe operations, and should issue warnings for code which might be correct but does not provably avoid value-unsafe operations.  No such compilers exist today, but to simplify our account here, we will pretend that they do exist. A value-safe type is any class, interface, or type parameter marked with the @ValueSafe annotation, or any subtype of a value-safe type.  If a value-safe class is marked final, it is in fact a value type.  All other value-safe classes must be abstract.  The non-static fields of a value class must be non-public and final, and all its constructors must be private. Under the above rules, a standard interface could be helpful to define value types like Complex.  Here is an example: @ValueSafe public interface ValueType extends java.io.Serializable {     // All methods listed here must get redefined.     // Definitions must be value-safe, which means     // they may depend on component values only.     List<? extends Object> asList();     int hashCode();     boolean equals(@ValueSafe Object c);     String toString(); } //@ValueSafe inherited from supertype: public final class Complex implements ValueType { … The main advantage of such a conventional interface is that (unlike an annotation) it is reified in the runtime type system.  It could appear as an element type or parameter bound, for facilities which are designed to work on value types only.  More broadly, it might assist the JVM to perform dynamic enforcement of the rules for value types. Besides types, the annotation @ValueSafe can mark fields, parameters, local variables, and methods.  (This is redundant when the type is also value-safe, but may be useful when the type is Object or another supertype of a value type.)  Working forward from these annotations, an expression E is defined as value-safe if it satisfies one or more of the following: The type of E is a value-safe type. E names a field, parameter, or local variable whose declaration is marked @ValueSafe. E is a call to a method whose declaration is marked @ValueSafe. E is an assignment to a value-safe variable, field reference, or array reference. E is a cast to a value-safe type from a value-safe expression. E is a conditional expression E0 ? E1 : E2, and both E1 and E2 are value-safe. Assignments to value-safe expressions and initializations of value-safe names must take their values from value-safe expressions. A value-safe expression may not be the subject of a value-unsafe operation.  In particular, it cannot be synchronized on, nor can it be compared with the “==” operator, not even with a null or with another value-safe type. In a program where all of these rules are followed, no value-type value will be subject to a value-unsafe operation.  Thus, the prime axiom of value types will be satisfied, that no two value type will be distinguishable as long as their component values are equal. More Code To illustrate these rules, here are some usage examples for Complex: Complex pi = Complex.valueOf(Math.PI, 0); Complex zero = pi.changeRe(0);  //zero = pi; zero.re = 0; ValueType vtype = pi; @SuppressWarnings("value-unsafe")   Object obj = pi; @ValueSafe Object obj2 = pi; obj2 = new Object();  // ok List<Complex> clist = new ArrayList<Complex>(); clist.add(pi);  // (ok assuming List.add param is @ValueSafe) List<ValueType> vlist = new ArrayList<ValueType>(); vlist.add(pi);  // (ok) List<Object> olist = new ArrayList<Object>(); olist.add(pi);  // warning: "value-unsafe" boolean z = pi.equals(zero); boolean z1 = (pi == zero);  // error: reference comparison on value type boolean z2 = (pi == null);  // error: reference comparison on value type boolean z3 = (pi == obj2);  // error: reference comparison on value type synchronized (pi) { }  // error: synch of value, unpredictable result synchronized (obj2) { }  // unpredictable result Complex qq = pi; qq = null;  // possible NPE; warning: “null-unsafe" qq = (Complex) obj;  // warning: “null-unsafe" qq = Complex.cast(obj);  // OK @SuppressWarnings("null-unsafe")   Complex empty = null;  // possible NPE qq = empty;  // possible NPE (null pollution) The Payoffs It follows from this that either the JVM or the java compiler can replace boxed value-type values with unboxed ones, without affecting normal computations.  Fields and variables of value types can be split into their unboxed components.  Non-static methods on value types can be transformed into static methods which take the components as value parameters. Some common questions arise around this point in any discussion of value types. Why burden the programmer with all these extra rules?  Why not detect programs automagically and perform unboxing transparently?  The answer is that it is easy to break the rules accidently unless they are agreed to by the programmer and enforced.  Automatic unboxing optimizations are tantalizing but (so far) unreachable ideal.  In the current state of the art, it is possible exhibit benchmarks in which automatic unboxing provides the desired effects, but it is not possible to provide a JVM with a performance model that assures the programmer when unboxing will occur.  This is why I’m writing this note, to enlist help from, and provide assurances to, the programmer.  Basically, I’m shooting for a good set of user-supplied “pragmas” to frame the desired optimization. Again, the important thing is that the unboxing must be done reliably, or else programmers will have no reason to work with the extra complexity of the value-safety rules.  There must be a reasonably stable performance model, wherein using a value type has approximately the same performance characteristics as writing the unboxed components as separate Java variables. There are some rough corners to the present scheme.  Since Java fields and array elements are initialized to null, value-type computations which incorporate uninitialized variables can produce null pointer exceptions.  One workaround for this is to require such variables to be null-tested, and the result replaced with a suitable all-zero value of the value type.  That is what the “cast” method does above. Generically typed APIs like List<T> will continue to manipulate boxed values always, at least until we figure out how to do reification of generic type instances.  Use of such APIs will elicit warnings until their type parameters (and/or relevant members) are annotated or typed as value-safe.  Retrofitting List<T> is likely to expose flaws in the present scheme, which we will need to engineer around.  Here are a couple of first approaches: public interface java.util.List<@ValueSafe T> extends Collection<T> { … public interface java.util.List<T extends Object|ValueType> extends Collection<T> { … (The second approach would require disjunctive types, in which value-safety is “contagious” from the constituent types.) With more transformations, the return value types of methods can also be unboxed.  This may require significant bytecode-level transformations, and would work best in the presence of a bytecode representation for multiple value groups, which I have proposed elsewhere under the title “Tuples in the VM”. But for starters, the JVM can apply this transformation under the covers, to internally compiled methods.  This would give a way to express multiple return values and structured return values, which is a significant pain-point for Java programmers, especially those who work with low-level structure types favored by modern vector and graphics processors.  The lack of multiple return values has a strong distorting effect on many Java APIs. Even if the JVM fails to unbox a value, there is still potential benefit to the value type.  Clustered computing systems something have copy operations (serialization or something similar) which apply implicitly to command operands.  When copying JVM objects, it is extremely helpful to know when an object’s identity is important or not.  If an object reference is a copied operand, the system may have to create a proxy handle which points back to the original object, so that side effects are visible.  Proxies must be managed carefully, and this can be expensive.  On the other hand, value types are exactly those types which a JVM can “copy and forget” with no downside. Array types are crucial to bulk data interfaces.  (As data sizes and rates increase, bulk data becomes more important than scalar data, so arrays are definitely accompanying us into the future of computing.)  Value types are very helpful for adding structure to bulk data, so a successful value type mechanism will make it easier for us to express richer forms of bulk data. Unboxing arrays (i.e., arrays containing unboxed values) will provide better cache and memory density, and more direct data movement within clustered or heterogeneous computing systems.  They require the deepest transformations, relative to today’s JVM.  There is an impedance mismatch between value-type arrays and Java’s covariant array typing, so compromises will need to be struck with existing Java semantics.  It is probably worth the effort, since arrays of unboxed value types are inherently more memory-efficient than standard Java arrays, which rely on dependent pointer chains. It may be sufficient to extend the “value-safe” concept to array declarations, and allow low-level transformations to change value-safe array declarations from the standard boxed form into an unboxed tuple-based form.  Such value-safe arrays would not be convertible to Object[] arrays.  Certain connection points, such as Arrays.copyOf and System.arraycopy might need additional input/output combinations, to allow smooth conversion between arrays with boxed and unboxed elements. Alternatively, the correct solution may have to wait until we have enough reification of generic types, and enough operator overloading, to enable an overhaul of Java arrays. Implicit Method Definitions The example of class Complex above may be unattractively complex.  I believe most or all of the elements of the example class are required by the logic of value types. If this is true, a programmer who writes a value type will have to write lots of error-prone boilerplate code.  On the other hand, I think nearly all of the code (except for the domain-specific parts like plus and minus) can be implicitly generated. Java has a rule for implicitly defining a class’s constructor, if no it defines no constructors explicitly.  Likewise, there are rules for providing default access modifiers for interface members.  Because of the highly regular structure of value types, it might be reasonable to perform similar implicit transformations on value types.  Here’s an example of a “highly implicit” definition of a complex number type: public class Complex implements ValueType {  // implicitly final     public double re, im;  // implicitly public final     //implicit methods are defined elementwise from te fields:     //  toString, asList, equals(2), hashCode, valueOf, cast     //optionally, explicit methods (plus, abs, etc.) would go here } In other words, with the right defaults, a simple value type definition can be a one-liner.  The observant reader will have noticed the similarities (and suitable differences) between the explicit methods above and the corresponding methods for List<T>. Another way to abbreviate such a class would be to make an annotation the primary trigger of the functionality, and to add the interface(s) implicitly: public @ValueType class Complex { … // implicitly final, implements ValueType (But to me it seems better to communicate the “magic” via an interface, even if it is rooted in an annotation.) Implicitly Defined Value Types So far we have been working with nominal value types, which is to say that the sequence of typed components is associated with a name and additional methods that convey the intention of the programmer.  A simple ordered pair of floating point numbers can be variously interpreted as (to name a few possibilities) a rectangular or polar complex number or Cartesian point.  The name and the methods convey the intended meaning. But what if we need a truly simple ordered pair of floating point numbers, without any further conceptual baggage?  Perhaps we are writing a method (like “divideAndRemainder”) which naturally returns a pair of numbers instead of a single number.  Wrapping the pair of numbers in a nominal type (like “QuotientAndRemainder”) makes as little sense as wrapping a single return value in a nominal type (like “Quotient”).  What we need here are structural value types commonly known as tuples. For the present discussion, let us assign a conventional, JVM-friendly name to tuples, roughly as follows: public class java.lang.tuple.$DD extends java.lang.tuple.Tuple {      double $1, $2; } Here the component names are fixed and all the required methods are defined implicitly.  The supertype is an abstract class which has suitable shared declarations.  The name itself mentions a JVM-style method parameter descriptor, which may be “cracked” to determine the number and types of the component fields. The odd thing about such a tuple type (and structural types in general) is it must be instantiated lazily, in response to linkage requests from one or more classes that need it.  The JVM and/or its class loaders must be prepared to spin a tuple type on demand, given a simple name reference, $xyz, where the xyz is cracked into a series of component types.  (Specifics of naming and name mangling need some tasteful engineering.) Tuples also seem to demand, even more than nominal types, some support from the language.  (This is probably because notations for non-nominal types work best as combinations of punctuation and type names, rather than named constructors like Function3 or Tuple2.)  At a minimum, languages with tuples usually (I think) have some sort of simple bracket notation for creating tuples, and a corresponding pattern-matching syntax (or “destructuring bind”) for taking tuples apart, at least when they are parameter lists.  Designing such a syntax is no simple thing, because it ought to play well with nominal value types, and also with pre-existing Java features, such as method parameter lists, implicit conversions, generic types, and reflection.  That is a task for another day. Other Use Cases Besides complex numbers and simple tuples there are many use cases for value types.  Many tuple-like types have natural value-type representations. These include rational numbers, point locations and pixel colors, and various kinds of dates and addresses. Other types have a variable-length ‘tail’ of internal values. The most common example of this is String, which is (mathematically) a sequence of UTF-16 character values. Similarly, bit vectors, multiple-precision numbers, and polynomials are composed of sequences of values. Such types include, in their representation, a reference to a variable-sized data structure (often an array) which (somehow) represents the sequence of values. The value type may also include ’header’ information. Variable-sized values often have a length distribution which favors short lengths. In that case, the design of the value type can make the first few values in the sequence be direct ’header’ fields of the value type. In the common case where the header is enough to represent the whole value, the tail can be a shared null value, or even just a null reference. Note that the tail need not be an immutable object, as long as the header type encapsulates it well enough. This is the case with String, where the tail is a mutable (but never mutated) character array. Field types and their order must be a globally visible part of the API.  The structure of the value type must be transparent enough to have a globally consistent unboxed representation, so that all callers and callees agree about the type and order of components  that appear as parameters, return types, and array elements.  This is a trade-off between efficiency and encapsulation, which is forced on us when we remove an indirection enjoyed by boxed representations.  A JVM-only transformation would not care about such visibility, but a bytecode transformation would need to take care that (say) the components of complex numbers would not get swapped after a redefinition of Complex and a partial recompile.  Perhaps constant pool references to value types need to declare the field order as assumed by each API user. This brings up the delicate status of private fields in a value type.  It must always be possible to load, store, and copy value types as coordinated groups, and the JVM performs those movements by moving individual scalar values between locals and stack.  If a component field is not public, what is to prevent hostile code from plucking it out of the tuple using a rogue aload or astore instruction?  Nothing but the verifier, so we may need to give it more smarts, so that it treats value types as inseparable groups of stack slots or locals (something like long or double). My initial thought was to make the fields always public, which would make the security problem moot.  But public is not always the right answer; consider the case of String, where the underlying mutable character array must be encapsulated to prevent security holes.  I believe we can win back both sides of the tradeoff, by training the verifier never to split up the components in an unboxed value.  Just as the verifier encapsulates the two halves of a 64-bit primitive, it can encapsulate the the header and body of an unboxed String, so that no code other than that of class String itself can take apart the values. Similar to String, we could build an efficient multi-precision decimal type along these lines: public final class DecimalValue extends ValueType {     protected final long header;     protected private final BigInteger digits;     public DecimalValue valueOf(int value, int scale) {         assert(scale >= 0);         return new DecimalValue(((long)value << 32) + scale, null);     }     public DecimalValue valueOf(long value, int scale) {         if (value == (int) value)             return valueOf((int)value, scale);         return new DecimalValue(-scale, new BigInteger(value));     } } Values of this type would be passed between methods as two machine words. Small values (those with a significand which fits into 32 bits) would be represented without any heap data at all, unless the DecimalValue itself were boxed. (Note the tension between encapsulation and unboxing in this case.  It would be better if the header and digits fields were private, but depending on where the unboxing information must “leak”, it is probably safer to make a public revelation of the internal structure.) Note that, although an array of Complex can be faked with a double-length array of double, there is no easy way to fake an array of unboxed DecimalValues.  (Either an array of boxed values or a transposed pair of homogeneous arrays would be reasonable fallbacks, in a current JVM.)  Getting the full benefit of unboxing and arrays will require some new JVM magic. Although the JVM emphasizes portability, system dependent code will benefit from using machine-level types larger than 64 bits.  For example, the back end of a linear algebra package might benefit from value types like Float4 which map to stock vector types.  This is probably only worthwhile if the unboxing arrays can be packed with such values. More Daydreams A more finely-divided design for dynamic enforcement of value safety could feature separate marker interfaces for each invariant.  An empty marker interface Unsynchronizable could cause suitable exceptions for monitor instructions on objects in marked classes.  More radically, a Interchangeable marker interface could cause JVM primitives that are sensitive to object identity to raise exceptions; the strangest result would be that the acmp instruction would have to be specified as raising an exception. @ValueSafe public interface ValueType extends java.io.Serializable,         Unsynchronizable, Interchangeable { … public class Complex implements ValueType {     // inherits Serializable, Unsynchronizable, Interchangeable, @ValueSafe     … It seems possible that Integer and the other wrapper types could be retro-fitted as value-safe types.  This is a major change, since wrapper objects would be unsynchronizable and their references interchangeable.  It is likely that code which violates value-safety for wrapper types exists but is uncommon.  It is less plausible to retro-fit String, since the prominent operation String.intern is often used with value-unsafe code. We should also reconsider the distinction between boxed and unboxed values in code.  The design presented above obscures that distinction.  As another thought experiment, we could imagine making a first class distinction in the type system between boxed and unboxed representations.  Since only primitive types are named with a lower-case initial letter, we could define that the capitalized version of a value type name always refers to the boxed representation, while the initial lower-case variant always refers to boxed.  For example: complex pi = complex.valueOf(Math.PI, 0); Complex boxPi = pi;  // convert to boxed myList.add(boxPi); complex z = myList.get(0);  // unbox Such a convention could perhaps absorb the current difference between int and Integer, double and Double. It might also allow the programmer to express a helpful distinction among array types. As said above, array types are crucial to bulk data interfaces, but are limited in the JVM.  Extending arrays beyond the present limitations is worth thinking about; for example, the Maxine JVM implementation has a hybrid object/array type.  Something like this which can also accommodate value type components seems worthwhile.  On the other hand, does it make sense for value types to contain short arrays?  And why should random-access arrays be the end of our design process, when bulk data is often sequentially accessed, and it might make sense to have heterogeneous streams of data as the natural “jumbo” data structure.  These considerations must wait for another day and another note. More Work It seems to me that a good sequence for introducing such value types would be as follows: Add the value-safety restrictions to an experimental version of javac. Code some sample applications with value types, including Complex and DecimalValue. Create an experimental JVM which internally unboxes value types but does not require new bytecodes to do so.  Ensure the feasibility of the performance model for the sample applications. Add tuple-like bytecodes (with or without generic type reification) to a major revision of the JVM, and teach the Java compiler to switch in the new bytecodes without code changes. A staggered roll-out like this would decouple language changes from bytecode changes, which is always a convenient thing. A similar investigation should be applied (concurrently) to array types.  In this case, it seems to me that the starting point is in the JVM: Add an experimental unboxing array data structure to a production JVM, perhaps along the lines of Maxine hybrids.  No bytecode or language support is required at first; everything can be done with encapsulated unsafe operations and/or method handles. Create an experimental JVM which internally unboxes value types but does not require new bytecodes to do so.  Ensure the feasibility of the performance model for the sample applications. Add tuple-like bytecodes (with or without generic type reification) to a major revision of the JVM, and teach the Java compiler to switch in the new bytecodes without code changes. That’s enough musing me for now.  Back to work!

    Read the article

  • Using SQL Developer to Debug your Anonymous PL/SQL Blocks

    - by JeffS
    Everyone knows that SQL Developer has a PL/SQL debugger – check! Everyone also knows that it’s only setup for debugging standalone PL/SQL objects like Functions, Procedures, and Packages, right? – NO! SQL Developer can also debug your Stored Java Procedures AND it can debug your standalone PLSQL blocks. These bits of PLSQL which do not live in the database are also known as ‘Anonymous Blocks.’ Anonymous PL/SQL blocks can be submitted to interactive tools such as SQL*Plus and Enterprise Manager, or embedded in an Oracle Precompiler or OCI program. At run time, the program sends these blocks to the Oracle database, where they are compiled and executed. Here’s an example of something you might want help debugging: Declare x number := 0; Begin Dbms_Output.Put(Sysdate || ' ' || Systimestamp); For Stuff In 1..100 Loop Dbms_Output.Put_Line('Stuff is equal to ' || Stuff || '.'); x := Stuff; End Loop; End; / With the power of remote debugging and unshared worksheets, we are going to be able to debug this ANON block! The trick – we need to create a dummy stored procedure and call it in our ANON block. Then we’re going to create an unshared worksheet and execute the script from there while the SQL Developer session is listening for remote debug connections. We step through the dummy procedure, and this takes OUT to our calling ANON block. Then we can use watches, breakpoints, and all that fancy debugger stuff! First things first, create this dummy procedure - create or replace procedure do_nothing is begin null; end; Then mouse-right-click on your Connection and select ‘Remote Debug.’ For an in-depth post on how to use the remote debugger, check out Barry’s excellent post on the subject. Open an unshared worksheet using Ctrl+Shift+N. This gives us a dedicated connection for our worksheet and any scripts or commands executed in it. Paste in your ANON block you want to debug. Add in a call to the dummy procedure above to the first line of your BEGIN block like so Begin do_nothing(); ... Then we need to setup the machine for remote debug for the session we have listening – basically we connect to SQL Developer. You can do that via a Environment Variable, or you can just add this line to your script - CALL DBMS_DEBUG_JDWP.CONNECT_TCP( 'localhost', '4000' ); Where ‘localhost’ is the machine where SQL Developer is running and ’4000′ is the port you started the debug listener on. Ok, with that all set, now just RUN the script. Once the PL/SQL call is made, the debugger will be invoked. You’ll end up in the DO_NOTHING() object. Debugging an ANON block from SQL Developer is possible! If you step out to the ANON block, we’ll end up in the script that’s used to call the procedure – which is the script you want to debug. The Anonymous Block is opened in a new SQL Dev page You can now step through the block, using watches and breakpoints as expected. I’m guessing your scripts are going to be a bit more complicated than mine, but this serves as a decent example to get you started. Here’s a screenshot of a watch and breakpoint defined in the anon block being debugged: Breakpoints, watches, and callstacks - oh my! For giggles, I created a breakpoint with a passcount of 90 for the FOR LOOP to see if it works. And of course it does You Might Also EnjoyUsing Pass Counts to Turbo Charge Your PL/SQL BreakpointsSQL Developer Tip: Viewing REFCURSOR OutputThe PL/SQL Debugger Strikes Back: Episode VDebugging PL/SQL with SQL Developer: Episode IVHow to find dependent objects in your PL/SQL Programs using SQL Developer

    Read the article

  • Adding Blog to Your Orchard Website

    - by hajan
    One of the common features in today’s content management systems is to provide you the ability to create your own blog in your website. Also, having a blog is one of the very often needed features for various types of websites. Out of the box, Orchard gives you this, so you can create your own blog in your Orchard website on a pretty easy way. Besides the fact that you can very easily create your own blog, Orchard also gives you some extra features in relation with the support of blogging, such as connecting third-party client applications (e.g. Windows Live Writer) to your blog, so that you can publish blog posts remotely. You can already find all the information provided in this blog post on the http://orchardproject.net website, however I thought it would be nice to make summary in one blog post. I assume you have already installed Orchard and you are already familiar with its environment and administration dashboard. If you haven’t, please read this blog post first.   CREATE YOUR BLOG First of all, go to Orchard Administration Dashboard and click on Blog in the left menu Once you are there, you will see the following screen   Fill the form with all needed data, as in the following example and click Save Right after, you should see the following screen Click New post, and add your first post. After that, go to Homepage (click Your Site in the top-left corner) and you should see the Blog link in your menu After clicking on Blog, you will be directed to the following page Once you click on My First Post, you will see that your blog already supports commenting ability (you can enable/disable this from Administration dashboard in your blog settings) Added comment Adding new comment Submit comment So, with following these steps, you have already setup your blog in your Orchard website.   CONNECT YOUR BLOG WITH WINDOWS LIVE WRITER Since many bloggers prepare their blog posts using third-party client applications, like Windows Live Writer, its very useful if your blog engine has the ability to work with these third-party applications and enable them to make remote posting and publishing. The client applications use XmlRpc interface in order to have the ability to manage and publish the blogs remotely. What is great about Orchard is that it gives you out of the box the XmlRpc and Remote Publishing modules. What you only need to do is to enable these features from the Modules in your Orchard Administration Dashboard. So, lets go through the steps of enabling and making your previously created blog able to work with third-party client applications for blogging. 1. Go to Administration Dashboard and click the Modules After clicking the Modules, you will see the following page: As you can see, you already have Remote Blog Publishing and XmlRpc features for Content Publishing, but both are disabled by default. So, if you click Enable only on Remote Blog Publishing, you will see both of them enabled at once since they are dependent features. After you click Enable, if everything is Ok, the following message should be displayed: So, now we have the featured enabled and ready... The next thing you need to do is to open Windows Live Writer. First, open Windows Live Writer and in your Blog Accounts, click on Add blog account In the next window, chose Other services After that, click on your Blog link in the Orchard website and copy the URL, my URL (on localhost development server) is: http://localhost:8191/blog Then, add your login credentials you use to login in Orchard and click Next. After that, if you have setup everything successfully, the Windows Live Writer will do the rest Once it finishes, you will have window where you can specify the name of your blog you have just connected your Windows Live Writer to... Then... you are done. You can see Windows Live Writer has detected the Orchard theme I am using After you finish with the blog post, click on Publish and refresh the Blog page in your Orchard website You see, we have the blog post directly posted from Windows Live Writer to my Orchard Blog. I hope this was useful blog post. Regards, Hajan Reference and other useful posts: Build incredible content-driven websites using Orchard CMS Create blog on your site with Orchard CMS Blogging using Windows Live Writer in your Orchard CMS Blog Orchard Website

    Read the article

  • Tuxedo 11gR1 Client Server Affinity

    - by todd.little
    One of the major new features in Oracle Tuxedo 11gR1 is the ability to define an affinity between clients and servers. In previous releases of Tuxedo, the only way to ensure that multiple requests from a client went to the same server was to establish a conversation with tpconnect() and then use tpsend() and tprecv(). Although this works it has some drawbacks. First for single-threaded servers, the server is tied up for the entire duration of the conversation and cannot service other clients, an obvious scalability issue. I believe the more significant drawback is that the application programmer has to switch from the simple request/response model provided by tpcall() to the half duplex tpsend() and tprecv() calls used with conversations. Switching between the two typically requires a fair amount of redesign and recoding. The Client Server Affinity feature in Tuxedo 11gR1 allows by way of configuration an application to define affinities that can exist between clients and servers. This is done in the *SERVICES section of the UBBCONFIG file. Using new parameters for services defined in the *SERVICES section, customers can determine when an affinity session is created or deleted, the scope of the affinity, and whether requests can be routed outside the affinity scope. The AFFINITYSCOPE parameter can be MACHINE, GROUP, or SERVER, meaning that while the affinity session is in place, all requests from the client will be routed to the same MACHINE, GROUP, or SERVER. The creation and deletion of affinity is defined by the SESSIONROLE parameter and a service can be defined as either BEGIN, END, or NONE, where BEGIN starts an affinity session, END deletes the affinity session, and NONE does not impact the affinity session. Finally customers can define how strictly they want the affinity scope adhered to using the AFFINITYSTRICT parameter. If set to MANDATORY, all requests made during an affinity session will be routed to a server in the affinity scope. Thus if the affinity scope is SERVER, all subsequent tpcall() requests will be sent to the same server the affinity scope was established with. If the server doesn't offer that service, even though other servers do offer the service, the call will fail with TPNOENT. Setting AFFINITYSTRICT to PRECEDENT tells Tuxedo to try and route the request to a server in the affinity scope, but if that's not possible, then Tuxedo can try to route the request to servers out of scope. All of this begs the question, why? Why have this feature? There many uses for this capability, but the most common is when there is state that is maintained in a server, group of servers, or in a machine and subsequent requests from a client must be routed to where that state is maintained. This might be something as simple as a database cursor maintained by a server on behalf of a client. Alternatively it might be that the server has a connection to an external system and subsequent requests need to go back to the server that has that connection. A more sophisticated case is where a group of servers maintains some sort of cache in shared memory and subsequent requests need to be routed to where the cache is maintained. Although this last case might be able to be handled by data dependent routing, using client server affinity allows the cache to be partitioned dynamically instead of statically.

    Read the article

  • Do NOT Change "Copy Local” project references to false, unless understand subsequences.

    - by Michael Freidgeim
    To optimize performance of visual studio build I've found multiple recommendations to change CopyLocal property for dependent dlls to false,e.g. From http://stackoverflow.com/questions/690033/best-practices-for-large-solutions-in-visual-studio-2008 CopyLocal? For sure turn this offhttp://stackoverflow.com/questions/280751/what-is-the-best-practice-for-copy-local-and-with-project-referencesAlways set the Copy Local property to false and enforce this via a custom msbuild stephttp://codebetter.com/patricksmacchia/2007/06/20/benefit-from-the-c-and-vb-net-compilers-perf/BenefitBenefitMy advice is to always set ‘Copy Local’ to falseSome time ago we've tried to change the setting to false, and found that it causes problem for deployment of top-level projects.Recently I've followed the suggestion and changed the settings for middle-level projects. It didn't cause immediate issues, but I was warned by Readify Consultant Colin Savage about possible errors during deploymentsI haven't undone the changes immediately and we found a few issues during testing.There are many scenarios, when you need to have Copy Local’ left to True.The concerns are highlighted in some stack overflow answers, but they have small number of votes.Top-level projects:  set copy local = true.First of all, it doesn't work correctly for top-level projects, i.e. executables or web sites.As pointed in the answer http://stackoverflow.com/a/6529461/52277for all the references in the one at the top set copy local = true.Alternatively you have to change output directory as it's described in http://www.simple-talk.com/dotnet/.net-framework/partitioning-your-code-base-through-.net-assemblies-and-visual-studio-projects/If you set ‘ Copy Local = false’, VS will, unless you tell it otherwise, place each assembly alone in its own .\bin\Debugdirectory. Because of this, you will need to configure VS to place assemblies together in the same directory. To do so, for each VS project, go to VS > Project Properties > Build tab > Output path, and set the Ouput path to ..\bin\Debugfor debug configuration, and ..\bin\Release for release configuration.Second-level  dependencies:  set copy local = true.Another example when copylocal =false fails on run-time, is when top level assembly doesn't directly referenced one of indirect dependencies.E..g. Top-level assembly A has reference to assembly B with copylocal =true, but assembly B has reference to assembly C with copylocal =false. Most likely assembly C will be missing on runtime and will cause errors E.g. http://stackoverflow.com/questions/602765/when-should-copy-local-be-set-to-true-and-when-should-it-not?lq=1Copy local is important for deployment scenarios and tools. As a general rule you should use CopyLocal=True and http://stackoverflow.com/questions/602765/when-should-copy-local-be-set-to-true-and-when-should-it-not?lq=1 Unfortunately there are some quirks and CopyLocal won't necessary work as expected for assembly references in secondary assemblies structured as shown below.MainApp.exe MyLibrary.dll ThirdPartyLibrary.dll (if in the GAC CopyLocal won't copy to MainApp bin folder)This makes xcopy deployments difficult . .Reflection called DLLs  dependencies:  set copy local = true.E.g user can see error "ISystem.Reflection.ReflectionTypeLoadException: Unable to load one or more of the requested types. Retrieve the LoaderExceptions property for more information."The fix for the issue is recommended in http://stackoverflow.com/a/6200173/52277"I solved this issue by setting the Copy Local attribute of my project's references to true."In general, the problems with investigation of deployment issues may overweight the benefits of reduced build time. Setting the Copy Local to false without considering deployment issues is not a good idea.

    Read the article

  • Another sound not working post

    - by Thomas Smart
    Tried all the other "sound not working" posts i think, lost count. purge/reinstall alsa and pulse, reboot, add user to audio group, various lines in the alsa config file such as "options snd-hda-intel model=" then tried different options like generic, auto, basic, default, etc. tried pulseaudio -k && sudo alsa force-reload a few times, with and without rebooting. Hardware: 16gb ram, core I7-4790, Intel Haswell mboard with onboard sound and graphics Multimedia: Audio Adapter: HDA-Intel-HDA Intel HDMI OS: Ubuntu server 14.04 with ubuntu-desktop installed. GUI sound settings lists only the dummy sound card alsamixer -c 0 ¦ Card: HDA Intel HDMI F1: Help ¦ ¦ Chip: Intel Haswell HDMI F2: System information ¦ ¦ View: F3:[Playback] F4: Capture F5: All F6: Select sound card ¦ ¦ Item: S/PDIF ¦ ¦ +--+ ¦ ¦ ¦OO¦ ¦ ¦ +--+ ¦ ¦ < S/PDIF > ¦ aplay -l **** List of PLAYBACK Hardware Devices **** card 0: HDMI [HDA Intel HDMI], device 3: HDMI 0 [HDMI 0] Subdevices: 1/1 Subdevice #0: subdevice #0 aplay -L default Playback/recording through the PulseAudio sound server null Discard all samples (playback) or generate zero samples (capture) pulse PulseAudio Sound Server hdmi:CARD=HDMI,DEV=0 HDA Intel HDMI, HDMI 0 HDMI Audio Output dmix:CARD=HDMI,DEV=3 HDA Intel HDMI, HDMI 0 Direct sample mixing device dsnoop:CARD=HDMI,DEV=3 HDA Intel HDMI, HDMI 0 Direct sample snooping device hw:CARD=HDMI,DEV=3 HDA Intel HDMI, HDMI 0 Direct hardware device without any conversions plughw:CARD=HDMI,DEV=3 HDA Intel HDMI, HDMI 0 Hardware device with all software conversions cat /proc/asound/cards 0 [HDMI ]: HDA-Intel - HDA Intel HDMI HDA Intel HDMI at 0xf7d14000 irq 46 cat /proc/asound/devices 1: : sequencer 2: [ 0- 3]: digital audio playback 3: [ 0- 0]: hardware dependent 4: [ 0] : control 33: : timer mplayer -ao alsa:device=hdmi /usr/share/sounds/ubuntu/stereo/system-ready.ogg MPlayer 1.1-4.8 (C) 2000-2012 MPlayer Team mplayer: could not connect to socket mplayer: No such file or directory Failed to open LIRC support. You will not be able to use your remote control. Playing /usr/share/sounds/ubuntu/stereo/system-ready.ogg. libavformat version 54.20.4 (external) Mismatching header version 54.20.3 libavformat file format detected. [lavf] stream 0: audio (vorbis), -aid 0 Load subtitles in /usr/share/sounds/ubuntu/stereo/ ========================================================================== Opening audio decoder: [ffmpeg] FFmpeg/libavcodec audio decoders libavcodec version 54.35.0 (external) AUDIO: 44100 Hz, 1 ch, floatle, 80.0 kbit/5.67% (ratio: 10000->176400) Selected audio codec: [ffvorbis] afm: ffmpeg (FFmpeg Vorbis) ========================================================================== [AO_ALSA] alsa-lib: confmisc.c:768:(parse_card) cannot find card '1' [AO_ALSA] alsa-lib: conf.c:4248:(_snd_config_evaluate) function snd_func_card_driver returned error: No such file or directory [AO_ALSA] alsa-lib: confmisc.c:392:(snd_func_concat) error evaluating strings [AO_ALSA] alsa-lib: conf.c:4248:(_snd_config_evaluate) function snd_func_concat returned error: No such file or directory [AO_ALSA] alsa-lib: confmisc.c:1251:(snd_func_refer) error evaluating name [AO_ALSA] alsa-lib: conf.c:4248:(_snd_config_evaluate) function snd_func_refer returned error: No such file or directory [AO_ALSA] alsa-lib: conf.c:4727:(snd_config_expand) Evaluate error: No such file or directory [AO_ALSA] alsa-lib: pcm.c:2239:(snd_pcm_open_noupdate) Unknown PCM hdmi [AO_ALSA] Playback open error: No such file or directory Failed to initialize audio driver 'alsa:device=hdmi' Could not open/initialize audio device -> no sound. Audio: no sound Video: no video Exiting... (End of file) mplayer -ao alsa:device=hw=0.3 /usr/share/sounds/ubuntu/stereo/system-ready.ogg MPlayer 1.1-4.8 (C) 2000-2012 MPlayer Team mplayer: could not connect to socket mplayer: No such file or directory Failed to open LIRC support. You will not be able to use your remote control. Playing /usr/share/sounds/ubuntu/stereo/system-ready.ogg. libavformat version 54.20.4 (external) Mismatching header version 54.20.3 libavformat file format detected. [lavf] stream 0: audio (vorbis), -aid 0 Load subtitles in /usr/share/sounds/ubuntu/stereo/ ========================================================================== Opening audio decoder: [ffmpeg] FFmpeg/libavcodec audio decoders libavcodec version 54.35.0 (external) AUDIO: 44100 Hz, 1 ch, floatle, 80.0 kbit/5.67% (ratio: 10000->176400) Selected audio codec: [ffvorbis] afm: ffmpeg (FFmpeg Vorbis) ========================================================================== [AO_ALSA] Format floatle is not supported by hardware, trying default. AO: [alsa] 44100Hz 2ch s16le (2 bytes per sample) Video: no video Starting playback... A: 0.4 (00.4) of 0.8 (00.7) 0.1% Exiting... (End of file) Thank you for your time and help :)

    Read the article

  • Migrating SQL Server Databases – The DBA’s Checklist (Part 3)

    - by Sadequl Hussain
    Continuing from Part 2 of the Database Migration Checklist series: Step 10: Full-text catalogs and full-text indexing This is one area of SQL Server where people do not seem to take notice unless something goes wrong. Full-text functionality is a specialised area in database application development and is not usually implemented in your everyday OLTP systems. Nevertheless, if you are migrating a database that uses full-text indexing on one or more tables, you need to be aware a few points. First of all, SQL Server 2005 now allows full-text catalog files to be restored or attached along with the rest of the database. However, after migration, if you are unable to look at the properties of any full-text catalogs, you are probably better off dropping and recreating it. You may also get the following error messages along the way: Msg 9954, Level 16, State 2, Line 1 The Full-Text Service (msftesql) is disabled. The system administrator must enable this service. This basically means full text service is not running (disabled or stopped) in the destination instance. You will need to start it from the Configuration Manager. Similarly, if you get the following message, you will also need to drop and recreate the catalog and populate it. Msg 7624, Level 16, State 1, Line 1 Full-text catalog ‘catalog_name‘ is in an unusable state. Drop and re-create this full-text catalog. A full population of full-text indexes can be a time and resource intensive operation. Obviously you will want to schedule it for low usage hours if the database is restored in an existing production server. Also, bear in mind that any scheduled job that existed in the source server for populating the full text catalog (e.g. nightly process for incremental update) will need to be re-created in the destination. Step 11: Database collation considerations Another sticky area to consider during a migration is the collation setting. Ideally you would want to restore or attach the database in a SQL Server instance with the same collation. Although not used commonly, SQL Server allows you to change a database’s collation by using the ALTER DATABASE command: ALTER DATABASE database_name COLLATE collation_name You should not be using this command for no reason as it can get really dangerous.  When you change the database collation, it does not change the collation of the existing user table columns.  However the columns of every new table, every new UDT and subsequently created variables or parameters in code will use the new setting. The collation of every char, nchar, varchar, nvarchar, text or ntext field of the system tables will also be changed. Stored procedure and function parameters will be changed to the new collation and finally, every character-based system data type and user defined data types will also be affected. And the change may not be successful either if there are dependent objects involved. You may get one or multiple messages like the following: Cannot ALTER ‘object_name‘ because it is being referenced by object ‘dependent_object_name‘. That is why it is important to test and check for collation related issues. Collation also affects queries that use comparisons of character-based data.  If errors arise due to two sides of a comparison being in different collation orders, the COLLATE keyword can be used to cast one side to the same collation as the other. Continues…

    Read the article

  • Is there a low carbon future for the retail industry?

    - by user801960
    Recently Oracle published a report in conjunction with The Future Laboratory and a global panel of experts to highlight the issue of energy use in modern industry and the serious need to reduce carbon emissions radically by 2050.  Emissions must be cut by 80-95% below the levels in 1990 – but what can the retail industry do to keep up with this? There are three key aspects to the retail industry where carbon emissions can be cut:  manufacturing, transport and IT.  Manufacturing Naturally, manufacturing is going to be a big area where businesses across all industries will be forced to make considerable savings in carbon emissions as well as other forms of pollution.  Many retailers of all sizes will use third party factories and will have little control over specific environmental impacts from the factory, but retailers can reduce environmental impact at the factories by managing orders more efficiently – better planning for stock requirements means economies of scale both in terms of finance and the environment. The John Lewis Partnership has made detailed commitments to reducing manufacturing and packaging waste on both its own-brand products and products it sources from third party suppliers. It aims to divert 95 percent of its operational waste from landfill by 2013, which is a huge logistics challenge.  The John Lewis Partnership’s website provides a large amount of information on its responsibilities towards the environment. Transport Similarly to manufacturing, tightening up on logistical planning for stock distribution will make savings on carbon emissions from haulage.  More accurate supply and demand analysis will mean less stock re-allocation after initial distribution, and better warehouse management will mean more efficient stock distribution.  UK grocery retailer Morrisons has introduced double-decked trailers to its haulage fleet and adjusted distribution logistics accordingly to reduce the number of kilometers travelled by the fleet.  Morrisons measures route planning efficiency in terms of cases moved per kilometre and has, over the last two years, increased the number of cases per kilometre by 12.7%.  See Morrisons Corporate Responsibility report for more information. IT IT infrastructure is often initially overlooked by businesses when considering environmental efficiency.  Datacentres and web servers often need to run 24/7 to handle both consumer orders and internal logistics, and this both requires a lot of energy and puts out a lot of heat.  Many businesses are lowering environmental impact by reducing IT system fragmentation in their offices, while an increasing number of businesses are outsourcing their datacenters to cloud-based services.  Using centralised datacenters reduces the power usage at smaller offices, while using cloud based services means the datacenters can be based in a more environmentally friendly location.  For example, Facebook is opening a massive datacentre in Sweden – close to the Arctic Circle – to reduce the need for artificial cooling methods.  In addition, moving to a cloud-based solution makes IT services more easily scaleable, reducing redundant IT systems that would still use energy.  In store, the UK’s Carbon Trust reports that on average, lighting accounts for 25% of a retailer’s electricity costs, and for grocery retailers, up to 50% of their electricity bill comes from refrigeration units.  On a smaller scale, retailers can invest in greener technologies in store and in their offices.  The report concludes that widely shared objectives of energy security, reduced emissions and continued economic growth are dependent on the development of a smart grid capable of delivering energy efficiency and demand response, as well as integrating renewable and variable sources of energy. The report is available to download from http://emeapressoffice.oracle.com/imagelibrary/detail.aspx?MediaDetailsID=1766I’d be interested to hear your thoughts on the report.   

    Read the article

  • RTL (Arabic and Hebrew) Support for Windows Phone 7

    - by Daniel Moth
    Problem and Background Currently there is no support for Right-To-Left rendering in Windows Phone 7, when developing with Silverlight (itself built on .NET Compact Framework). When I encountered that limitation, I had a flashback to 2005 when I complained about the luck of RTL on NETCF. Unfortunately, the partial solution I proposed back then requires PInvoke and there is no such support on Windows Phone today. Fortunately, my RTL requirements this time were more modest: all I wanted to do was display correctly a translation (of Hebrew or Arabic) in my FREE WP7 translator app. For v1.0 of the app, the code received a string from the service and just put it up on the screen as the translated text. In Arabic and Hebrew, that string (incorrectly) appeared reversed. I knew that, but decided that since it is a platform limitation, I could live with it and so could the users. Yuval P, a colleague at Microsoft, pushed me to offer support for Hebrew (something that I wasn't motivated to pursue if I am honest). After many back and forths, we landed on some code that works. It is certainly not the most efficient code (quite the opposite), but it works and met the bar of minimum effort for v1.1. Thanks Yuval for insisting and contributing most of the code! After Hebrew support was there, I thought the same solution would work for Arabic. Apparently, reversing the Arabic text is not enough: Arabic characters render themselves differently dependent on what preceded/succeeds them(!). So I needed some kind of utility that takes a reversed Arabic string and returns the same string but with the relevant characters "fixed". Luckily, another MS colleague has put out such a library (thanks Bashar): http://arabic4wp7.codeplex.com/. RTL Solution So you have a reversed RTL string and want to make it "right" before displaying on the screen. This is what worked for me (ymmv). Need to split the string into "lines". Not doing this and just reversing the string and sticking it a wrapping text control means that the user not only has to read right to left, they also have to read bottom up. The previous step must take into account a line length that works for both portrait and landscape modes, and of course, not break words in the middle, i.e. find natural breaks. For each line, break it up into words and reverse the order of the words and the order of the letters within each word On the previous step, do not reverse words that should be preserved, e.g. Windows and other such English words that are mixed in with the Arabic or Hebrew words. The same exclusion from reversal applies to numbers. Specifically, for Arabic, once there is a word that is reversed also change its characters. For some code paths, the above has to take into account whether the translation is "from" an RTL language or if it is "to" an RTL language. I packaged the solution in a single code file containing a static class (see the 'Background" section above for… background and credits). Download RTL.cs for your Windows Phone app (to see its usage in action download for FREE "The best translator app") Enjoy, and if you decide to improve on the code, feel free to share back with me… Comments about this post welcome at the original blog.

    Read the article

  • Page output caching for dynamic web applications

    - by Mike Ellis
    I am currently working on a web application where the user steps (forward or back) through a series of pages with "Next" and "Previous" buttons, entering data until they reach a page with the "Finish" button. Until finished, all data is stored in Session state, then sent to the mainframe database via web services at the end of the process. Some of the pages display data from previous pages in order to collect additional information. These pages can never be cached because they are different for every user. For pages that don't display this dynamic data, they can be cached, but only the first time they load. After that, the data that was previously entered needs to be displayed. This requires Page_Load to fire, which means the page can't be cached at that point. A couple of weeks ago, I knew almost nothing about implementing page caching. Now I still don't know much, but I know a little bit, and here is the solution that I developed with the help of others on my team and a lot of reading and trial-and-error. We have a base page class defined from which all pages inherit. In this class I have defined a method that sets the caching settings programmatically. For pages that can be cached, they call this base page method in their Page_Load event within a if(!IsPostBack) block, which ensures that only the page itself gets cached, not the data on the page. if(!IsPostBack) {     ...     SetCacheSettings();     ... } protected void SetCacheSettings() {     Response.Cache.AddValidationCallback(new HttpCacheValidateHandler(Validate), null);     Response.Cache.SetExpires(DateTime.Now.AddHours(1));     Response.Cache.SetSlidingExpiration(true);     Response.Cache.SetValidUntilExpires(true);     Response.Cache.SetCacheability(HttpCacheability.ServerAndNoCache); } The AddValidationCallback sets up an HttpCacheValidateHandler method called Validate which runs logic when a cached page is requested. The Validate method signature is standard for this method type. public static void Validate(HttpContext context, Object data, ref HttpValidationStatus status) {     string visited = context.Request.QueryString["v"];     if (visited != null && "1".Equals(visited))     {         status = HttpValidationStatus.IgnoreThisRequest; //force a page load     }     else     {         status = HttpValidationStatus.Valid; //load from cache     } } I am using the HttpValidationStatus values IgnoreThisRequest or Valid which forces the Page_Load event method to run or allows the page to load from cache, respectively. Which one is set depends on the value in the querystring. The value in the querystring is set up on each page in the "Next" and "Previous" button click event methods based on whether the page that the button click is taking the user to has any data on it or not. bool hasData = HasPageBeenVisited(url); if (hasData) {     url += VISITED; } Response.Redirect(url); The HasPageBeenVisited method determines whether the destination page has any data on it by checking one of its required data fields. (I won't include it here because it is very system-dependent.) VISITED is a string constant containing "?v=1" and gets appended to the url if the destination page has been visited. The reason this logic is within the "Next" and "Previous" button click event methods is because 1) the Validate method is static which doesn't allow it to access non-static data such as the data fields for a particular page, and 2) at the time at which the Validate method runs, either the data has not yet been deserialized from Session state or is not available (different AppDomain?) because anytime I accessed the Session state information from the Validate method, it was always empty.

    Read the article

  • regarding the Windows Phone 7 series, XNA and Visual Basic

    - by Chris Williams
    as long as we're talking about VB... I figured I would share this as well. Hi everyone, I'm about to express a sentiment that might ruffle a few feathers, but I think most of you know me well enough to know I love like accept VB for what it is and that what I'm about to say is with good intentions. (The rest of you, who don't know me, please take my word for it.) The world is full of VB developers, I was one of them for a long time. I think it's safe to assume that none of us are ignorant people who require handholding. We're working professionals, making a living by using our skills as developers. I'm also willing to bet that quite a few of us are fluent in C# as well as VB. It may not be your preferred language, but many of you can do it and you prove that nearly every day. Honestly, I don't know ANY developers or consultants that have only known ONE language ever. So it pains me greatly when I see the word "CAN'T" being tossed around like a crutch... as in "we CAN'T develop for the windows phone or we CAN'T develop XNA games." At MIX, Microsoft hath decreed that C# is the language of choice for developing for the Windows Phone 7. I think it's a safe bet that you won't see VB support if it isn't there already. (Just like XNA... which is up to version 4.0 by now.)  So what? (Yeah... I said it.) I think everyone here can agree that actual coding is only one part of software design and development. There is nothing stopping ANY of you from beginning the process of designing your killer phone app, writing up specs, requirements, doing UI design, workflow, mockups, storyboards, art, etc.... None of these things are language dependent. IF by the time you've got that stuff out of the way, and there's still no VB support, then start doing some rapid prototyping of your app in C# (I know, I know... heresy!)  You still have to spend time learning how the phone does things, what UI tricks do what, what paradigms make sense, how to use to accelerometer and the tilt and the multitouch functionality. I can guarantee you that time spent doing this is a great investment, no matter WHAT extension your code files have. Eventually, you may have a working prototype. IF by this time, there's STILL no VB support... fret not, you've made significant progress on your app. You've designed it, prototyped it, figured out how to use the phone specific features... so you might as well finish it and pat yourself on the back for learning something new... and possibly being first to market with your new app. I'll be happy to argue any and all of these points online or off with anyone who cares to do so, but there is one undeniable point that you simply can't argue:  Your potential customers do not care AT ALL what programming language you used to write the app they are about to purchase. They care that it works. If your biggest concern is being first to market, than stop complaining and get busy because you're running out of time and the 3000+ people who were at MIX certainly aren't waiting for you. They've already started working on their apps.

    Read the article

  • Securing Flexfield Value Sets in EBS 12.2

    - by Sara Woodhull
    Release 12.2 includes a new feature: flexfield value set security. This new feature gives you additional options for ensuring that different administrators have non-overlapping responsibilities, which in turn provides checks and balances for sensitive activities.  Separation of Duties (SoD) is one of the key concepts of internal controls and is a requirement for many regulations including: Sarbanes-Oxley (SOX) Act Health Insurance Portability and Accountability Act (HIPAA) European Union Data Protection Directive. Its primary intent is to put barriers in place to prevent fraud or theft by an individual acting alone. Implementing Separation of Duties requires minimizing the possibility that users could modify data across application functions where the users should not normally have access. For flexfields and report parameters in Oracle E-Business Suite, values in value sets can affect functionality such as the rollup of accounting data, job grades used at a company, and so on. Controlling access to the creation or modification of value set values can be an important piece of implementing Separation of Duties in an organization. New Flexfield Value Set Security feature Flexfield value set security allows system administrators to restrict users from viewing, adding or updating values in specific value sets. Value set security enables role-based separation of duties for key flexfields, descriptive flexfields, and report parameters. For example, you can set up value set security such that certain users can view or insert values for any value set used by the Accounting Flexfield but no other value sets, while other users can view and update values for value sets used for any flexfields in Oracle HRMS. You can also segregate access by Operating Unit as well as by role or responsibility.Value set security uses a combination of data security and role-based access control in Oracle User Management. Flexfield value set security provides a level of security that is different from the previously-existing and similarly-named features in Oracle E-Business Suite: Function security controls whether a user has access to a specific page or form, as well as what operations the user can do in that screen. Flexfield value security controls what values a user can enter into a flexfield segment or report parameter (by responsibility) during routine data entry in many transaction screens across Oracle E-Business Suite. Flexfield value set security (this feature, new in Release 12.2) controls who can view, insert, or update values for a particular value set (by flexfield, report, or value set) in the Segment Values form (FNDFFMSV). The effect of flexfield value set security is that a user of the Segment Values form will only be able to view those value sets for which the user has been granted access. Further, the user will be able to insert or update/disable values in that value set if the user has been granted privileges to do so.  Flexfield value set security affects independent, dependent, and certain table-validated value sets for flexfields and report parameters. Initial State of the Feature upon Upgrade Because this is a new security feature, it is turned on by default.  When you initially install or upgrade to Release 12.2.2, no users are allowed to view, insert or update any value set values (users may even think that their values are missing or invalid because they cannot see the values).  You must explicitly set up access for specific users by enabling appropriate grants and roles for those users.We recommend using flexfield value set security as part of a comprehensive Separation of Duties strategy. However, if you choose not to implement flexfield value set security upon upgrading to or installing Release 12.2, you can enable backwards compatibility--users can access any value sets if they have access to the Values form--after you upgrade. The feature does not affect day-to-day transactions that use flexfields.  However, you must either set up specific grants and roles or enable backwards compatibility before users can create new values or update or disable existing values. For more information, see: Release 12.2 Flexfield Value Set Security Documentation Update for Patch 17305947:R12.FND.C (Document 1589204.1) R12.2 TOI: Implement and Use Application Object Library (AOL) - Flexfields Security and Separation of Duties for Value Sets (recorded training)

    Read the article

  • Adding custom interfaces to your mock instance.

    - by mehfuzh
    Previously, i made a post  showing how you can leverage the dependent interfaces that is implemented by JustMock during the creation of mock instance. It could be a informative post that let you understand how JustMock behaves internally for class or interfaces implement other interfaces into it. But the question remains, how you can add your own custom interface to your target mock. In this post, i am going to show you just that. Today, i will not start with a dummy class as usual rather i will use two most common interfaces in the .NET framework  and create a mock combining those. Before, i start i would like to point out that in the recent release of JustMock we have extended the Mock.Create<T>(..) with support for additional settings though closure. You can add your own custom interfaces , specify directly the real constructor that should be called or even set the behavior of your target. Doing a fast forward directly to the point,  here goes the test code for create a creating a mock that contains the mix for ICloneable and IDisposable using the above mentioned changeset. var myMock = Mock.Create<IDisposable>(x => x.Implements<ICloneable>()); var myMockAsClonable = myMock as ICloneable; bool isCloned = false;   Mock.Arrange(() => myMockAsClonable.Clone()).DoInstead(() => isCloned = true);   myMockAsClonable.Clone();   Assert.True(isCloned);   Here, we are creating the target mock for IDisposable and also implementing ICloneable. Finally, using the “as” for getting the ICloneable reference accordingly arranging it, acting on it and asserting if the expectation is met properly. This is a very rudimentary example, you can do the same for a given class: var realItem = Mock.Create<RealItem>(x => {     x.Implements<IDisposable>();     x.CallConstructor(() => new RealItem(0)); }); var iDispose = realItem as IDisposable;     iDispose.Dispose(); Here, i am also calling the real constructor for RealItem class.  This is to mention that you can implement custom interfaces only for non-sealed classes or less it will end up with a proper exception. Also, this feature don’t require any profiler, if you are agile or running it inside silverlight runtime feel free to try it turning off the JM add-in :-). TIP :  Ability to  specify real constructor could be a useful productivity boost in cases for code change and you can re-factor the usage just by one click with your favorite re-factor tool.   That’s it for now and hope that helps Enjoy!!

    Read the article

  • Threading Overview

    - by ACShorten
    One of the major features of the batch framework is the ability to support multi-threading. The multi-threading support allows a site to increase throughput on an individual batch job by splitting the total workload across multiple individual threads. This means each thread has fine level control over a segment of the total data volume at any time. The idea behind the threading is based upon the notion that "many hands make light work". Each thread takes a segment of data in parallel and operates on that smaller set. The object identifier allocation algorithm built into the product randomly assigns keys to help ensure an even distribution of the numbers of records across the threads and to minimize resource and lock contention. The best way to visualize the concept of threading is to use a "pie" analogy. Imagine the total workset for a batch job is a "pie". If you split that pie into equal sized segments, each segment would represent an individual thread. The concept of threading has advantages and disadvantages: Smaller elapsed runtimes - Jobs that are multi-threaded finish earlier than jobs that are single threaded. With smaller amounts of work to do, jobs with threading will finish earlier. Note: The elapsed runtime of the threads is rarely proportional to the number of threads executed. Even though contention is minimized, some contention does exist for resources which can adversely affect runtime. Threads can be managed individually – Each thread can be started individually and can also be restarted individually in case of failure. If you need to rerun thread X then that is the only thread that needs to be resubmitted. Threading can be somewhat dynamic – The number of threads that are run on any instance can be varied as the thread number and thread limit are parameters passed to the job at runtime. They can also be configured using the configuration files outlined in this document and the relevant manuals.Note: Threading is not dynamic after the job has been submitted Failure risk due to data issues with threading is reduced – As mentioned earlier individual threads can be restarted in case of failure. This limits the risk to the total job if there is a data issue with a particular thread or a group of threads. Number of threads is not infinite – As with any resource there is a theoretical limit. While the thread limit can be up to 1000 threads, the number of threads you can physically execute will be limited by the CPU and IO resources available to the job at execution time. Theoretically with the objects identifiers evenly spread across the threads the elapsed runtime for the threads should all be the same. In other words, when executing in multiple threads theoretically all the threads should finish at the same time. Whilst this is possible, it is also possible that individual threads may take longer than other threads for the following reasons: Workloads within the threads are not always the same - Whilst each thread is operating on the roughly the same amounts of objects, the amount of processing for each object is not always the same. For example, an account may have a more complex rate which requires more processing or a meter has a complex amount of configuration to process. If a thread has a higher proportion of objects with complex processing it will take longer than a thread with simple processing. The amount of processing is dependent on the configuration of the individual data for the job. Data may be skewed – Even though the object identifier generation algorithm attempts to spread the object identifiers across threads there are some jobs that use additional factors to select records for processing. If any of those factors exhibit any data skew then certain threads may finish later. For example, if more accounts are allocated to a particular part of a schedule then threads in that schedule may finish later than other threads executed. Threading is important to the success of individual jobs. For more guidelines and techniques for optimizing threading refer to Multi-Threading Guidelines in the Batch Best Practices for Oracle Utilities Application Framework based products (Doc Id: 836362.1) whitepaper available from My Oracle Support

    Read the article

  • Too Clever for My Own Good

    - by AjarnMark
    Yesterday I caught myself being a little too clever for my own good with some ASP.NET code.  It seems that I have forgotten some of my good old classic HTML and JavaScript skills, and become too dependent on the .NET Framework and WebControls to do the work for me.  Here’s the scenario… In order to improve the User Interface and better communicate to the user when something is happening that they need to wait for, we have started to modify some of our larger (slower) pages to display messages like Processing… or Reloading… while they are cycling through a postback.  (Yes, I understand this could be improved by using AJAX / Callbacks and so on, but even then, you need to let your user know that they need to wait for that section to be re-rendered, so for the moment these pages will continue to use good ol’ Postbacks.)  It’s a very simple trick, really.  All I want to do is when some control triggers a postback, first run a little client-side JavaScript to hide the main contents of the page (such as a GridView) and display the appropriate message.  This lets the user know, “Hey, we’re doing something, don’t click another link or scroll and try to take action right now.” The first places I hooked this up were easy.  Most common cause of a postback:  Buttons.  And when you’re writing the markup or declarative code for an ASP:Button control, there is the handy OnClientClick property which is designed for just this purpose…to run client-side JavaScript before the postback occurs.  This is distinguished from the OnClick property which tells the control what Server-side code to run.  Great!  Done!  Easy! But then there are other controls like DropDownLists and CheckBoxes that we use on our pages with the AutoPostback=True setting which cause postbacks.  And these don’t have OnClientClick or OnClientSelectedIndexChanged events.  So I started getting creative, using an ASP:CustomValidator control in conjunction with setting the CausesValidation and ValidationGroup settings on these controls, which basically caused the action on the control to fire the Custom Validator, which was defined with a Client Side validation function which then did the hide content/show message code (and return a meaningless IsValid setting).  This also caused me to define a different ValidationGroup setting for my real data entry validator controls so that I could control them separately and only have them fire when I really wanted validation, and not just my show/hide trick. For a little while I was pretty proud of myself for coming up with this clever approach to get around what I considered to be a serious oversight on the DropDownList and CheckBox controls declarative syntax.  Then, in the midst of my smugness, just as I was about to commit my changes to the source code repository, it dawned on me that there is a much simpler and much more appropriate way to accomplish this.  All that I really needed to do was to put in my server-side code (I used the Page_Init section) a call to MyControl.Attributes.Add(“onClick”, “myJavaScriptFunctionName()”) for the checkboxes, and for the DropDownLists (which become select tags) use “onChange” instead of “onClick”.  This is exactly the type of thing that the Attributes collection is there for…so you can add attributes to be rendered with the control that you would have otherwise stuck right into the HTML markup if you had been doing this by hand in the first place. Ugh!  A few hours wasted on clever tricks that I ended up completely removing, but I did learn a lot more about custom validators and validation groups in the process.  And got a good reminder that all that stuff (HTML, JavaScript, and CSS) I learned back when I wrote classic ASP pages is still valuable today.  Oh, and one more thing…don’t get lulled into too much reliance on the the whiz-bang tool to do it for you.  After all, WebControls are just another layer of abstraction, and sometimes you need to dig down through the layers and get a little closer to the native language.

    Read the article

  • Question on the implementation of my Entity System

    - by miguel.martin
    I am currently creating an Entity System, in C++, it is almost completed (I have all the code there, I just have to add a few things and test it). The only thing is, I can't figure out how to implement some features. This Entity System is based off a bit from the Artemis framework, however it is different. I'm not sure if I'll be able to type this out the way my head processing it. I'm going to basically ask whether I should do something over something else. Okay, now I'll give a little detail on my Entity System itself. Here are the basic classes that my Entity System uses to actually work: Entity - An Id (and some methods to add/remove/get/etc Components) Component - An empty abstract class ComponentManager - Manages ALL components for ALL entities within a Scene EntitySystem - Processes entities with specific components Aspect - The class that is used to help determine what Components an Entity must contain so a specific EntitySystem can process it EntitySystemManager - Manages all EntitySystems within a Scene EntityManager - Manages entities (i.e. holds all Entities, used to determine whether an Entity has been changed, enables/disables them, etc.) EntityFactory - Creates (and destroys) entities and assigns an ID to them Scene - Contains an EntityManager, EntityFactory, EntitySystemManager and ComponentManager. Has functions to update and initialise the scene. Now in order for an EntitySystem to efficiently know when to check if an Entity is valid for processing (so I can add it to a specific EntitySystem), it must recieve a message from the EntityManager (after a call of activate(Entity& e)). Similarly the EntityManager must know when an Entity is destroyed from the EntityFactory in the Scene, and also the ComponentManager must know when an Entity is created AND destroyed. I do have a Listener/Observer pattern implemented at the moment, but with this pattern I may remove a Listener (which is this case is dependent on the method being called). I mainly have this implemented for specific things related to a game, i.e. Teams, Tagging of entities, etc. So... I was thinking maybe I should call a private method (using friend classes) to send out when an Entity has been activated, deleted, etc. i.e. taken from my EntityFactory void EntityFactory::killEntity(Entity& e) { // if the entity doesn't exsist in the entity manager within the scene if(!getScene()->getEntityManager().doesExsist(e)) { return; // go back to the caller! (should throw an exception or something..) } // tell the ComponentManager and the EntityManager that we killed an Entity getScene()->getComponentManager().doOnEntityWillDie(e); getScene()->getEntityManager().doOnEntityWillDie(e); // notify the listners for(Mouth::iterator i = getMouth().begin(); i != getMouth().end(); ++i) { (*i)->onEntityWillDie(*this, e); } _idPool.addId(e.getId()); // add the ID to the pool delete &e; // delete the entity } As you can see on the lines where I am telling the ComponentManager and the EntityManager that an Entity will die, I am calling a method to make sure it handles it appropriately. Now I realise I could do this without calling it explicitly, with the help of that for loop notifying all listener objects connected to the EntityFactory's Mouth (an object used to tell listeners that there's an event), however is this a good idea (good design, or what)? I've gone over the PROS and CONS, I just can't decide what I want to do. Calling Explicitly: PROS Faster? Since these functions are explicitly called, they can't be "removed" CONS Not flexible Bad design? (friend functions) Calling through Listener objects (i.e. ComponentManager/EntityManager inherits from a EntityFactoryListener) PROS More Flexible? Better Design? CONS Slower? (virtual functions) Listeners can be removed, i.e. may be removed and not get called again during the program, which could cause in a crash. P.S. If you wish to view my current source code, I am hosting it on BitBucket.

    Read the article

  • Controlling server configurations with IPS

    - by barts
    I recently received a customer question regarding how they best could control which packages and which versions were used on their production Solaris 11 servers.  They had considered pointing each server at its own software repository - a common initial approach.  A simpler method leverages one of dependency mechanisms we introduced with Solaris 11, but is not immediately obvious to most people. Typically, most internal IT departments qualify particular versions for production use.  What this customer wanted to do was insure that their operations staff only installed internally qualified versions of Solaris on their servers.  The easiest way of doing this is to leverage the 'incorporate' type of dependency in a small package defined for each server type.  From the reference " Packaging and Delivering Software With the Image Packaging System in Oracle® Solaris 11.1":  The incorporate dependency specifies that if the given package is installed, it must be at the given version, to the given version accuracy. For example, if the dependent FMRI has a version of 1.4.3, then no version less than 1.4.3 or greater than or equal to 1.4.4 satisfies the dependency. Version 1.4.3.7 does satisfy this example dependency. The common way to use incorporate dependencies is to put many of them in the same package to define a surface in the package version space that is compatible. Packages that contain such sets of incorporate dependencies are often called incorporations. Incorporations are typically used to define sets of software packages that are built together and are not separately versioned. The incorporate dependency is heavily used in Oracle Solaris to ensurethat compatible versions of software are installed together. An example incorporate dependency is: depend type=incorporate fmri=pkg:/driver/network/ethernet/[email protected],5.11-0.175.0.0.0.2.1 So, to make sure only qualified versions are installed on a server, create a package that will be installed on the machines to be controlled.  This package will contain an incorporate dependency on the "entire" package, which controls the various components used to be build Solaris.  Every time a new version of Solaris has been qualified for production use, create a new version of this package specifying the new version of "entire" that was qualified.  Once this new control package is available in the repositories configured on the production server, the pkg update command will update that system to the specified version.  Unless a new version of the control package is made available, pkg update will report that no updates are available since no version of the control package can be installed that satisfies the incorporate constraint. Note that if desired, the same package can be used to specify which packages must be present on the system by adding either "require" or "group" dependencies; the latter permits removal of some of the packages, the former does not.  More details on this can be found in either the section 5 pkg man page or the previously mentioned reference document. This technique of using package dependencies to constrain system configuration leverages the SAT solver which is at the heart of IPS, and is basic to how we package Solaris itself.  

    Read the article

  • SQL Server Date Comparison Functions

    - by HighAltitudeCoder
    A few months ago, I found myself working with a repetitive cursor that looped until the data had been manipulated enough times that it was finally correct.  The cursor was heavily dependent upon dates, every time requiring the earlier of two (or several) dates in one stored procedure, while requiring the later of two dates in another stored procedure. In short what I needed was a function that would allow me to perform the following evaluation: WHERE MAX(Date1, Date2) < @SomeDate The problem is, the MAX() function in SQL Server does not perform this functionality.  So, I set out to put these functions together.  They are titled: EarlierOf() and LaterOf(). /**********************************************************                               EarlierOf.sql   **********************************************************/ /**********************************************************   Return the later of two DATETIME variables.   Parameter 1: DATETIME1 Parameter 2: DATETIME2   Works for a variety of DATETIME or NULL values. Even though comparisons with NULL are actually indeterminate, we know conceptually that NULL is not earlier or later than any other date provided.   SYNTAX: SELECT dbo.EarlierOf('1/1/2000','12/1/2009') SELECT dbo.EarlierOf('2009-12-01 00:00:00.000','2009-12-01 00:00:00.521') SELECT dbo.EarlierOf('11/15/2000',NULL) SELECT dbo.EarlierOf(NULL,'1/15/2004') SELECT dbo.EarlierOf(NULL,NULL)   **********************************************************/ USE AdventureWorks GO   IF EXISTS       (SELECT *       FROM sysobjects       WHERE name = 'EarlierOf'       AND xtype = 'FN'       ) BEGIN             DROP FUNCTION EarlierOf END GO   CREATE FUNCTION EarlierOf (       @Date1                              DATETIME,       @Date2                              DATETIME )   RETURNS DATETIME   AS BEGIN       DECLARE @ReturnDate     DATETIME         IF (@Date1 IS NULL AND @Date2 IS NULL)       BEGIN             SET @ReturnDate = NULL             GOTO EndOfFunction       END         ELSE IF (@Date1 IS NULL AND @Date2 IS NOT NULL)       BEGIN             SET @ReturnDate = @Date2             GOTO EndOfFunction       END         ELSE IF (@Date1 IS NOT NULL AND @Date2 IS NULL)       BEGIN             SET @ReturnDate = @Date1             GOTO EndOfFunction       END         ELSE       BEGIN             SET @ReturnDate = @Date1             IF @Date2 < @Date1                   SET @ReturnDate = @Date2             GOTO EndOfFunction       END         EndOfFunction:       RETURN @ReturnDate   END -- End Function GO   ---- Set Permissions --GRANT SELECT ON EarlierOf TO UserRole1 --GRANT SELECT ON EarlierOf TO UserRole2 --GO                                                                                             The inverse of this function is only slightly different. /**********************************************************                               LaterOf.sql   **********************************************************/ /**********************************************************   Return the later of two DATETIME variables.   Parameter 1: DATETIME1 Parameter 2: DATETIME2   Works for a variety of DATETIME or NULL values. Even though comparisons with NULL are actually indeterminate, we know conceptually that NULL is not earlier or later than any other date provided.   SYNTAX: SELECT dbo.LaterOf('1/1/2000','12/1/2009') SELECT dbo.LaterOf('2009-12-01 00:00:00.000','2009-12-01 00:00:00.521') SELECT dbo.LaterOf('11/15/2000',NULL) SELECT dbo.LaterOf(NULL,'1/15/2004') SELECT dbo.LaterOf(NULL,NULL)   **********************************************************/ USE AdventureWorks GO   IF EXISTS       (SELECT *       FROM sysobjects       WHERE name = 'LaterOf'       AND xtype = 'FN'       ) BEGIN             DROP FUNCTION LaterOf END GO   CREATE FUNCTION LaterOf (       @Date1                              DATETIME,       @Date2                              DATETIME )   RETURNS DATETIME   AS BEGIN       DECLARE @ReturnDate     DATETIME         IF (@Date1 IS NULL AND @Date2 IS NULL)       BEGIN             SET @ReturnDate = NULL             GOTO EndOfFunction       END         ELSE IF (@Date1 IS NULL AND @Date2 IS NOT NULL)       BEGIN             SET @ReturnDate = @Date2             GOTO EndOfFunction       END         ELSE IF (@Date1 IS NOT NULL AND @Date2 IS NULL)       BEGIN             SET @ReturnDate = @Date1             GOTO EndOfFunction       END         ELSE       BEGIN             SET @ReturnDate = @Date1             IF @Date2 > @Date1                   SET @ReturnDate = @Date2             GOTO EndOfFunction       END         EndOfFunction:       RETURN @ReturnDate   END -- End Function GO   ---- Set Permissions --GRANT SELECT ON LaterOf TO UserRole1 --GRANT SELECT ON LaterOf TO UserRole2 --GO                                                                                             The interesting thing about this function is its simplicity and the built-in NULL handling functionality.  Its interesting, because it seems like something should already exist in SQL Server that does this.  From a different vantage point, if you create this functionality and it is easy to use (ideally, intuitively self-explanatory), you have made a successful contribution. Interesting is good.  Self-explanatory, or intuitive is FAR better.  Happy coding! Graeme

    Read the article

  • Government Mandates and Programming Languages

    A recent SEC proposal (which, at over 600 pages, I havent read in any detail) includes the following: We are proposing to require the filing of a computer program (the waterfall computer program, as defined in the proposed rule) of the contractual cash flow provisions of the securities in the form of downloadable source code in Python, a commonly used computer programming language that is open source and interpretive. The computer program would be tagged in XML and required to be filed with the Commission as an exhibit. Under our proposal, the filed source code for the computer program, when downloaded and run (by loading it into an open Python session on the investors computer), would be required to allow the user to programmatically input information from the asset data file that we are proposing to require as described above. We believe that, with the waterfall computer program and the asset data file, investors would be better able to conduct their own evaluations of ABS and may be less likely to be dependent on the opinions of credit rating agencies. With respect to any registration statement on Form SF-1 (Section 239.44) or Form SF-3 (Section 239.45) relating to an offering of an asset-backed security that is required to comply with Item 1113(h) of Regulation AB, the Waterfall Computer Program (as defined in Item 1113(h)(1) of Regulation AB) must be written in the Python programming language and able to be downloaded and run on a local computer properly configured with a Python interpreter. The Waterfall Computer Program should be filed in the manner specified in the EDGAR Filer Manual. I dont see how it can be in investors best interests that the SEC demand a particular programming language be used for software related to investment data.  I have a feeling that investors who use computers at all already have software with which they are familiar, and that the vast majority of them are not running an open source scripting language on their machines to do their financial analysis.  In fact, I would wager that most of them are using tools like Excel, and if they really need to script anything, its being done with VBA in Excel. Now, Im not proposing that the SEC should require that the data be provided in Excel format with VBA scripts included so everyone can easily access the data (despite the fact that this would actually be pretty useful generally).  Rather, I think it is ill-advised for a government agency to make recommendations of this nature, period.  If the goal of the recommendation is to ensure that the way things work is codified in a transparent manner, than I can certainly respect that.  It seems to me that this could be accomplished without dictating the technology to use.  To wit: An Excel document could contain all of the data as well as the formulae necessary, and most likely would not require the end-user to install anything on their machine The SEC could simply create a calculator in the cloud such that any/all investors could use a single canonical web-based (or web service based) tool Millions of Java and .NET developers could write their own implementations You can read more about this issue, including the favorable position on it, on Jayanth Varmas blog. Did you know that DotNetSlackers also publishes .net articles written by top known .net Authors? We already have over 80 articles in several categories including Silverlight. Take a look: here.

    Read the article

  • Getting the number of fragments which passed the depth test

    - by Etan
    In "modern" environments, the "NV Occlusion Query" extension provides a method to get the number of fragments which passed the depth test. However, on the iPad / iPhone using OpenGL ES, the extension is not available. What is the most performant approach to implement a similar behaviour in the fragment shader? Some of my ideas: Render the object completely in white, then count all the colors together using a two-pass shader where first a vertical line is rendered and for each fragment the shader computes the sum over the whole row. Then, a single vertex is rendered whose fragment sums all the partial sums of the first pass. Doesn't seem to be very efficient. Render the object completely in white over a black background. Downsample recursively, abusing the hardware linear interpolation between textures until being at a reasonably small resolution. This leads to fragments which have a greyscale level depending on the number of white pixels where in their corresponding region. Is this even accurate enough? Use mipmaps and simply read the pixel on the 1x1 level. Again the question of accuracy and if it is even possible using non-power-of-two textures. The problem wit these approaches is, that the pipeline gets stalled which results in major performance issues. Therefore, I'm looking for a more performant way to accomplish my goal. Using the EXT_OCCLUSION_QUERY_BOOLEAN extension Apple introduced EXT_OCCLUSION_QUERY_BOOLEAN in iOS 5.0 for iPad 2. "4.1.6 Occlusion Queries Occlusion queries use query objects to track the number of fragments or samples that pass the depth test. An occlusion query can be started and finished by calling BeginQueryEXT and EndQueryEXT, respectively, with a target of ANY_SAMPLES_PASSED_EXT or ANY_SAMPLES_PASSED_CONSERVATIVE_EXT. When an occlusion query is started with the target ANY_SAMPLES_PASSED_EXT, the samples-boolean state maintained by the GL is set to FALSE. While that occlusion query is active, the samples-boolean state is set to TRUE if any fragment or sample passes the depth test. When the occlusion query finishes, the samples-boolean state of FALSE or TRUE is written to the corresponding query object as the query result value, and the query result for that object is marked as available. If the target of the query is ANY_SAMPLES_PASSED_CONSERVATIVE_EXT, an implementation may choose to use a less precise version of the test which can additionally set the samples-boolean state to TRUE in some other implementation dependent cases." The first sentence hints on a behavior which is exactly what I'm looking for: getting the number of pixels which passed the depth test in an asynchronous manner without much performance loss. However, the rest of the document describes only how to get boolean results. Is it possible to exploit this extension to get the pixel count? Does the hardware support it so that there may be hidden API to get access to the pixel count? Other extensions which could be exploitable would be debugging features like the number of times the fragment shader was invoked (PSInvocations in DirectX - not sure if something simila is available in OpenGL ES). However, this would also result in a pipeline stall.

    Read the article

  • The long road to bug-free software

    - by Tony Davis
    The past decade has seen a burgeoning interest in functional programming languages such as Haskell or, in the Microsoft world, F#. Though still on the periphery of mainstream programming, functional programming concepts are gradually seeping into the imperative C# language (for example, Lambda expressions have their root in functional programming). One of the more interesting concepts from functional programming languages is the use of formal methods, the lofty ideal behind which is bug-free software. The idea is that we write a specification that describes exactly how our function (say) should behave. We then prove that our function conforms to it, and in doing so have proved beyond any doubt that it is free from bugs. All programmers already use one form of specification, specifically their programming language's type system. If a value has a specific type then, in a type-safe language, the compiler guarantees that value cannot be an instance of a different type. Many extensions to existing type systems, such as generics in Java and .NET, extend the range of programs that can be type-checked. Unfortunately, type systems can only prevent some bugs. To take a classic problem of retrieving an index value from an array, since the type system doesn't specify the length of the array, the compiler has no way of knowing that a request for the "value of index 4" from an array of only two elements is "unsafe". We restore safety via exception handling, but the ideal type system will prevent us from doing anything that is unsafe in the first place and this is where we start to borrow ideas from a language such as Haskell, with its concept of "dependent types". If the type of an array includes its length, we can ensure that any index accesses into the array are valid. The problem is that we now need to carry around the length of arrays and the values of indices throughout our code so that it can be type-checked. In general, writing the specification to prove a positive property, even for a problem very amenable to specification, such as a simple sorting algorithm, turns out to be very hard and the specification will be different for every program. Extend this to writing a specification for, say, Microsoft Word and we can see that the specification would end up being no simpler, and therefore no less buggy, than the implementation. Fortunately, it is easier to write a specification that proves that a program doesn't have certain, specific and undesirable properties, such as infinite loops or accesses to the wrong bit of memory. If we can write the specifications to prove that a program is immune to such problems, we could reuse them in many places. The problem is the lack of specification "provers" that can do this without a lot of manual intervention (i.e. hints from the programmer). All this might feel a very long way off, but computing power and our understanding of the theory of "provers" advances quickly, and Microsoft is doing some of it already. Via their Terminator research project they have started to prove that their device drivers will always terminate, and in so doing have suddenly eliminated a vast range of possible bugs. This is a huge step forward from saying, "we've tested it lots and it seems fine". What do you think? What might be good targets for specification and verification? SQL could be one: the cost of a bug in SQL Server is quite high given how many important systems rely on it, so there's a good incentive to eliminate bugs, even at high initial cost. [Many thanks to Mike Williamson for guidance and useful conversations during the writing of this piece] Cheers, Tony.

    Read the article

  • Application Logging needs work

    Application Logging Application logging is the act of logging events that occur within an application much like how a court report documents what happens in court case. Application logs can be useful for several reasons, but the most common use for logs is to recreate steps to find the root cause of applications errors. Other uses can include the detection of Fraud, verification of user activity, or provide audits on user/data interactions. “Logs can contain different kinds of data. The selection of the data used is normally affected by the motivation leading to the logging. “ (OWASP, 2009) OWASP also stats that logging include applicable debugging information like the event date time, responsible process, and a description of the event. “There are many reasons why a logging system is a necessary part of delivering a distributed application. One of the most important is the ability to track exactly how many users are using the application during different time periods.” (Hatton, 2000) Hatton also states that application logging helps system designers determine whether parts of an application aren't being used as designed. He implies that low usage can be used to identify if users like or do not like aspects of a system based on user usage of the application. This enables application designers to extract why users don't like aspects of an application so that changes can be made to increase its usefulness and effectiveness. “Logging memory usage can also assist you in tuning up the internals of your application. If you're experiencing a randomly occurring problem, being able to match activities performed with the memory status at the time may enable you to discover the cause of the problem. It also gives you a good indication of the health of the distributed server machine at the time any activity is performed. “ (Hatton, 2000) Commonly Logged Application Events (Defined by OWASP) Access of Data Creation of Data Modification of Data in any form Administrative Functions  Configuration Changes Debugging Information(Application Events)  Authorization Attempts  Data Deletion Network Communication  Authentication Events  Errors/Exceptions Application Error Logging The functionality associated with application error logging is actually the combination of proper error handling and applications logging.  If we look back at Figure 4 and Figure 5, these code examples allow developers to handle various types of errors that occur within the life cycle of an application’s execution. Application logging can be applied within the Catch section of the TryCatch statement allowing for the errors to be logged when they occur. By placing the logging within the Catch section specific error details can be accessed that help identify the source of the error, the path to the error, what caused the error and definition of the error that occurred. This can then be logged and reviewed at a later date in order recreate the error that was received based data found in the application log. By allowing applications to log errors developers IT staff can use them to recreate errors that are encountered by end-users or other dependent systems.

    Read the article

< Previous Page | 35 36 37 38 39 40 41 42 43 44 45 46  | Next Page >