Search Results

Search found 25055 results on 1003 pages for 'software bug'.

Page 614/1003 | < Previous Page | 610 611 612 613 614 615 616 617 618 619 620 621  | Next Page >

• What should be done first: Code reviews or Unit tests?

  - by goldenmean

  Hello, If a developer implements code for some module and wants to get it reviewed. What should be the order : *First unit test the module after designing test cases for the module, debugging and fixing the bugs and then give the modified code for peer code review (Pros- Code to be reviewed is 'clean' to a good extent. Reduces some avoidable review comments and rework. Cons- Developer might spend large time debugging/fixing a bug which could have pointed/anticipated in peer code reviews) Or *First do the code review with peers and then go for unit testing. What are your thoughts/experience on this? I believe this approach for unit testing, code reviewing should be programming language agnostic, but it would be interesting to know otherwise(if applicable) with specific examples. -AD

  Read the article

• Cannot set g:checkbox to off for a child on one-to-many relationship

  - by icon911

  Got a weird issue with g:checkbox not being saved when its switched from on to off for a child in one-to-many relationship. For example: class Parent { Boolean enabled static hasMany = [children: Child] static constraints = { enabled(blank: true, nullable: true) } } class Child { Boolean enabled static belongsTo = [parent: Parent] static constraints = { enabled(blank: true, nullable: true) } } Posting to Parent controller true/false values will work for Parent: <g:checkBox name="enabled" value="${parentInstance?.enabled}"/> However, not for Child! When posting to Parent controller I can only go from false to true, trying to change from true to false again will not work: <g:each in="${parentInstance.children}" status="i" var="child"> <g:checkBox name="child[${i}].enabled" value="${child?.enabled}" /> </g:each> That seems to be a bug. Any ideas? Thanks.

  Read the article

• What's the right way to use idlestartup on python 2.6.5?

  - by user210481

  Idlestartup is analogous to pythonstartup variable, but for IDLE, instead of command line. But it seems not to work properly. I'm using python 2.6.5 on Windows. I have the following script assigned to it: from pprint import pprint import sys newPath = 'C:\\Python26\test') sys.path.append(newPath) print "initial config loaded" Both variables Idlestartup and pythonstartup are assigned to the same file (script above). When running IDLE, pprint and sys are NOT available, the final message is NOT printed, but newPath was added to sys.path. Running the command line, pprint and sys are available, the final message is printed and newPath was added to sys.path. Is it a bug? Am I doing something wrong? Thanks

  Read the article

• Axis2 issue with comment in WSDL

  - by Sirs

  I'm using an Axis2 client to access an external Webservice, whose WSDL starts with the following content: <?xml version="1.0" encoding="UTF-8"?><!--Created by TIBCO WSDL--><wsdl:definitions xmlns:wsdl=... My call to sendReceive crashes with the following error: com.ctc.wstx.exc.WstxUnexpectedCharException: Unexpected character 'C' (code 67) in prolog; expected '<' The 'C' is the first character on the comment in the WSDL. Without that comment everything works fine, but as far as my knowledge of basic XML dictates that comment is correct. My question would be: Is this a bug in Axis2 or is the accessed WSDL malformed? Is there any way to prevent Axis2 from crashing under these circumstances?

  Read the article

• .NET Reflector and getters/setters issue

  - by Humberto

  I'm using an up-to-date .NET Reflector to disassemble an internal legacy app whose source code is almost impossible to recover. I need to find the cause of a nasty bug, and then possibly patch it. Reflector did a good job as usual in the re-creation of the project's structure, but soon I discovered that every property call was left "expanded" to its get_() and set_() method signatures, rendering the source code impossible to compile. A quick Visual Studio "Search/Replace" with regex solved these cases, but it's awkward. Is there a way to make Reflector behave correctly?

  Read the article

• Can a DataTemplate be a Page?

  - by dthrasher

  I'm using the MVVM pattern to create a WPF standalone application. My program compiles in Visual Studio 2008, but I frequently get warnings in the editor for my DataTemplates. In my MainWindow.xaml, I've defined the following DataTemplate: <DataTemplate DataType="{x:Type ViewModels:TagViewModel}"> <Views:TagView /> </DataTemplate> Where "TagView" is derived from a Page, rather than an ordinary UserControl. This causes the following message to appear every time I reload the designer in Visual Studio: "Could not create an instance of type 'TagView'. Yet the solution compiles fine and the program seems to work properly. Is this a bug in the Visual Studio 2008 editor? Or am I doing something wrong?

  Read the article

• Simple Windows.Forms binding is failing using framework 4.0.

  - by jyoung

  This works under the .net framework 3.5 client. This fails under the .net framework 4.0 client. Was I doing something that was illegal under 3.5 but just happened to work, or is this a bug? Note that in my project 'PropInt' does not raise change events so using ctx[obj1.PropObj2, "PropInt"] is not an option. public class Obj1 { public Obj2 PropObj2 { get; set; } public Obj1() { PropObj2 = new Obj2(); } } public class Obj2 { public int PropInt { get; set; } } static class Program { [STAThread] static void Main() { var ctx = new BindingContext(); var obj1 = new Obj1(); var x1 = ctx[obj1, "PropObj2.PropInt"]; } }

  Read the article

• Stop 2 identical queries from executing almost simultaneously?

  - by James Simpson

  I have developed an AJAX based game where there is a bug caused (very remote, but in volume it happens at least once per hour) where for some reason two requests get sent to the processing page almost simultaneously (the last one I tracked, the requests were a difference of .0001 ms). There is a check right before the query is executed to make sure that it doesn't get executed twice, but since the difference is so small, the check hasn't finished before the next query gets executed. I'm stumped, how can I prevent this as it is causing serious problems in the game. Just to be more clear, the query is starting a new round in the game, so when it executes twice, it starts 2 rounds at the same time which breaks the game, so I need to be able to stop the script from executing if the previous round isn't over, even if that previous round started .0001 ms ago.

  Read the article

• Not able to pass multiple override parameters using nose-testconfig 0.6 plugin in nosetests

  - by Jaikit

  Hi, I am able to override multiple config parameters using nose-testconfig plugin only if i pass the overriding parameters on commandline. e.g. nosetests -c nose.cfg -s --tc=jack.env1:asl --tc=server2.env2:abc But when I define the same thing inside nose.cfg, than only the value for last parameter is modified. e.g. tc = server2.env2:abc tc = jack.env1:asl I checked the plugin code. It looks fine to me. I am pasting the part of plugin code below: parser.add_option( "--tc", action="append", dest="overrides", default = [], help="Option:Value specific overrides.") configure: if options.overrides: self.overrides = [] overrides = tolist(options.overrides) for override in overrides: keys, val = override.split(":") if options.exact: config[keys] = val else: ns = ''.join(['["%s"]' % i for i in keys.split(".") ]) # BUG: Breaks if the config value you're overriding is not # defined in the configuration file already. TBD exec('config%s = "%s"' % (ns, val)) Let me know if any one has any clue.

  Read the article

• Mongrel with Ruby 1.9.

  - by JussiR

  When starting the server, I get this strange error after updating to Ruby 1.9: " [BUG] cross-thread violation on rb_gc() ruby 1.8.6 (2008-08-11) [i386-mswin32] This application has requested the Runtime to terminate it in an unusual way. Please contact the application's support team for more information. " My mongrel version is 1.1.6, which should be compatible with Ruby 1.9. I also still have the Ruby 1.8.6 installation, so i assume that for some reason mongrel tries to access that and kills ruby? My rails version is 2.3.4, in case it matters.

  Read the article

• Debugging of native code

  - by graham.reeds

  I have a C# Service that is calling a C DLL that was originally written in VC6. There is a bug in the DLL which I am trying to inspect. After having a nightmare trying to get debug to work I eventually added the dll to the VS2005 solution containing the C# Service and added the necessary _CRT_SECURE_NO_WARNINGS. The debug version of the service is registered using 'installutil.exe' tool. I can get the debugger to break just before the line where the dll is entered via a call to System.Diagnostics.Debugger.Break();. I found some instruction on the net regarding stepping into debugging unmanaged code, and enabled the 'Enable unmanaged code debugging' check box, I've also tried turning on the Options-Debugging-Native 'Load DLL exports' and 'Enable RPC Debugging' (even though it's not COM). I've also copied the debug dll and .pdb to the same bin directory as the However the unmanaged code is not being stepped into which is what I really need. UPDATE: I found the Debugging Type in the DLL properties and set it to 'Mixed' as per suggestion on several sites but to no avail.

  Read the article

• How can I tell NetBeans to use the latest available version of a JAR for a library?

  - by Freiheit

  I have a Netbeans project with a library defined which includes several JARs. These JARs are versioned like lib\blah\com.blah.wibble.jar_0.6.1.201004161543 . These are nightly builds from another project so that version changes often. I know I can point NetBeans at the specific JARs with the version name, but this means that every time I get a new version I have to update the NetBeans library. I can also strip the version name from the JARs, but this makes it hard to track down bugs. "What version of the blah JARs?" is usually the second thing we ask when we find a bug. Is it possible to tell Netbeans to use included com.blah.wibble.jar_[??????] where ???? is some sort of automatic pointer to use the latest available version?

  Read the article

• Initiate a PHP class where class name is a variable

  - by ed209

  I need some help with an error I have not encountered before and can't seem to find anywhere. In a PHP mvc framework (just from a tutorial) I have the following: // Initiate the class $className = 'Controller_' . ucfirst($controller); if (class_exists($className)) { $controller = new $className($this->registry); } $className is showing the correct class name (case is also correct). But when I run it I get this in the apache error log (no php error) [Wed Mar 31 10:34:12 2010] [notice] child pid 987 exit signal Segmentation fault (11) Process id is different on every call. I am running PHP 5.3.0 on os x 10.6. This site seems to work on 5.2.11 on another Mac. Not really sure where to go next to debug it. I guess it could be an apache setting as much as a php bug or a problem with the code... any suggestions on where to look next?

  Read the article

• Django's manage.py shell won't indent

  - by hora

  I seem to have run into a strange bug or more likely some setting I am unfamiliar with on my system that is not allowing me to tab when I am in Django's shell (python manage.py shell is how I run it). For obvious reasons this is proving to be annoying since I can't do any loops or conditonals in the shell. If I hit tab it completes all functions that are available to me, like bash does in a terminal. I've tried just using spaces for my indents but I always get an indentation error. Does anyone know why this is happening and what I can do to get tab to work in my shell again? (It may be relevant to know that this is on a Ubuntu 9.04 system). Edit: tab works fine in the regular pythong shell, it's only in django's that it doesn't. Thanks.

  Read the article

• Javascript + Firebug : "cannot access optimized closure" What does it mean?

  - by interstar

  I just got the following error in a piece of javascript (in Firefox 3.5, with Firebug running) cannot access optimized closure I know, superficially, what caused the error. I had a line options.length() instead of options.length Fixing this bug, made the message go away. But I'm curious. What does this mean? What is an optimized closure? Is optimizing an enclosure something that the javascript interpretter does automatically? What does it do?

  Read the article

• How to make sure web services are kept stable from one release to the next?

  - by Tor Hovland

  The company where I work is a software vendor with a suite of applications. There are also a number of web services, and of course they have to be kept stable even if the applications change. We haven't always succeeded with this, and sometimes a customer finds that a service is not behaving as before after upgrading. We now want to handle this better. In general, web services shouldn't change, and if they have to, at least we will know about it and document the change. But how do we ensure this? One idea is to compare the WSDL files with the previous versions at every release. That will make sure the interfaces don't change, but it won't detect that the behavior changes, for example if a bug is introduced in some common library. Another idea is to build up a suite of service tests, for example using soapUI. But then we'll never know if we have covered enough cases. What are some best practices regarding this?

  Read the article

• Error with default argument in Source.getLines (Scala 2.8.0 RC1)

  - by Derek

  assuming I running Scala 2.8.0 RC1, the following scala code should print out the content of the file "c:/hello.txt" for ( line<-Source.fromPath( "c:/hello.txt" ).getLines ) println( line ) However, when I run it, I get the following error <console>:10: error: missing arguments for method getLines in class Source; follow this method with `_' if you want to treat it as a partially applied function Error occured in an application involving default arguments. val it = Source.fromPath("c:/hello.scala").getLines From what I understand, Scala should use the default argument "compat.Platform.EOL" for "getLines". I am wondering if I did wrong or is it a bug in scala 2.8 Thanks

  Read the article

• Support both Standard mode and Quirks mode? Is that possible and necessary?

  - by tshao

  Today I was assigned a bug saying that some page elements don't work on IE8 Quirks mode at all, and I need to fix them. The point is that I believe our pages will always be rendered in Standard mode, because we specify DOCTYPE at the beginning of every page (via master page). I'd think it must be some debugging tools changed that during testing. I managed to convice QA to close it as by design, after a brief explanation to her. Now I start to think the question that whether we should have our page work on both Standard and Quirks mode. Maybe we should try to minimize the problem even if the page is not rendering in a supposed mode? Any standard or best practices on that? Thanks!

  Read the article

• Java multi Generic collection parameters complie error

  - by Geln Yang

  Hi, So strange!Please have a look the code first: public class A { } public class B extends A { } public class C extends A { } public class TestMain { public <T extends A> void test(T a, T b) { } public <T extends A> void test(List<T> a, List<T> b) { } public static void main(String[] args) { new TestMain().test(new B(), new C()); new TestMain().test(new ArrayList<B>(), new ArrayList<C>()); } } The statement "new TestMain().test(new ArrayList(), new ArrayList())" get a "Bound mismatch" compile error, while "new TestMain().test(new B(), new C())" is compiled ok. Bound mismatch: The generic method test(T, T) of type TestMain is not applicable for the arguments (ArrayList, ArrayList). The inferred type ArrayList is not a valid substitute for the bounded parameter It seems the type of the second generic List parameter is limited by the Type of the first.Is it a feature or a bug of the compile program? ps, jdk:1.6,IDE:Eclipse 3.5.1

  Read the article

• Django: problem with merging querysets after annotation

  - by Björn Lilja

  Hi I have a manager for "Dialog" looking like this: class AnnotationManager(models.Manager): def get_query_set(self): return super(AnnotationManager, self).get_query_set().annotate( num_votes=Count('vote', distinct=True), num_comments=Count('comment', distinct=True), num_commentators = Count('comment__user', distinct=True), ) Votes and Comments has a ForeignKey to Dialog. Comments has a ForeignKey to User. When I do this: dialogs_queryset = Dialog.public.filter(organization=organization) dialogs_popularity = dialogs_queryset.exclude(num_comments=0) | dialogs_queryset.exclude(num_votes=0) ...dialogs_popularity will never returned the combination, but only the dialogs with more than 0 comments, or if I change the order of the OR, the dialogs with more than 0 votes! To me, the expected behavior would be to get the dialogs with more than 0 votes AND the dialogs with more than 0 comments. What am I missing? Or is there a bug in the annotation behavior here?

  Read the article

• Placement new in gcc

  - by Roman Prikhodchenko

  I need to find a workaround for a bug with placement new in g++. I now it was fixed in gcc-4.3 but I have to support versions 4.2 and 4.1. For example, following code compiles with an error "error: no matching function for call to 'operator new(long unsigned int, void*&)" template<class T, template<typename> class Alloc> inline void* type_ctor() { Alloc<T> a; void* p = a.allocate(1); new(p) T; return p; } ..... type_ctor<A, NewAllocator >();

  Read the article

• Memory section handling error

  - by Basilevs

  I'm getting a link time error: WARNING: /home/gulevich/development/camac-fedorov/camac/linux/k0607-lsi6/camac-k0607-lsi6.o (.ctors): unexpected non-allocatable section. Did you forget to use "ax"/"aw" in a .S file? Note that for example <linux/init.h> contains section definitions for use in .S files. The code causing the error (assembly in C source): # if defined(__ELF__) # define __SECTION_FLAGS ", \"aw\" , @progbits" /* writable flag needed for ld ".[cd]tors" sections bug workaround) */ # elif defined(__COFF__) # define __SECTION_FLAGS ", \"dr\"" /* untested, may be writable flag needed */ # endif asm ( ".section .ctors" __SECTION_FLAGS "\n" ".globl __ctors_begin__\n" "__ctors_begin__:\n" ".previous\n" ); Is there any way to fix this? The idea is to put a varaible __ctors_begin__ at the beginning of a certain memory section. This code is a legacy that worked fine using a different build system and older compiler. Meaning of this assembly code explained in an answer to my previous question.

  Read the article

• continuous integration web service

  - by Josh Moore

  I am in a position where I could become a team leader of a team distributed over two countries. This team would be the tech. team for a start up company that we plan to bootstrap on limited funds. So I am trying to find out ways to minimize upfront expenses. Right now we are planning to use Java and will have a lot of junit tests. I am planing on using github for VCS and lighthouse for a bug tracker. In addition I want to add a continuous integration server but I do not know of any continuous integration servers that are offered as a web service. Does anybody know if there are continuous integration servers available in a software as a service model? P.S. if anybody knows were I can get these three services at one location that would be great to know to.

  Read the article

• Facebook API returning wrong unread thread count

  - by houbysoft

  I'm trying to query the thread FQL table to get all unread messages, and also the count of unread items in the thread. This is how I query the table: SELECT thread_id,updated_time,snippet,snippet_author,unread FROM thread WHERE folder_id=0 AND unread!=0 From reading the doc to which I linked above, it seems to me that unread should include the count of unread messages in the thread. However, I just tested the above call and Facebook gives me back a value of unread=1, despite the thread in question having 4 unread items. This is how the thread looks on facebook.com (notice the (4), showing that unread should be 4): This is what the API returns to me, which is wrong (notice the "unread":1): { "data":[ { "name":"messages", "fql_result_set":[ { "thread_id":"BLAH BLAH BLAH", "updated_time":1333317140, "snippet":"BLAH BLAH BLAH", "snippet_author":BLAH, "unread":1 } ] } ] } Am I doing something wrong, or is this a bug?

  Read the article

• Replacement for Azure SDK Powershell commandlets for deployments

  - by Frank Rosario

  Hi, We've run into an issue with the New-Deployment Azure Powershell commandlet timing out; we've put in a bug report with MS. While they gave us an explanation for it (the path and timeout threshold used to upload through commandlets is different then what's used by the web portal); they don't have a fix for us. We need to get this running so we can automate our build deployments, so we're looking into developing a custom commandlet to replace New-Deployment using the Azure SDK; hoping this path will not have the timeout issues the commandlet did. But before we go down that route; are there any other scriptable tools I can use to replace the New-Deployment functionality? I looked at Cloudberry for Windows Azure; but that doesn't have a scriptable interface yet. Any constructive input is greatly appreciated.

  Read the article

< Previous Page | 610 611 612 613 614 615 616 617 618 619 620 621  | Next Page >