Search Results

Search found 8224 results on 329 pages for 'sometimes'.

Page 199/329 | < Previous Page | 195 196 197 198 199 200 201 202 203 204 205 206  | Next Page >

• Getting an unexpected "?" at the end of a Registry GetValue in C#

  - by Wilhelm Peraud

  Hi, I use the Registry class to manage values in the Registry on Windows Seven in C#. Registry.GetValue(...); But, I'm facing a curious behavior : Every time, the returned value is the correct one, but sometimes, it is followed by an unexpected "?" When I check the Registry, (regedit), the "?" doesn't exist. I really don't understand from where this question mark come from. Could someone help me please ? Info : - C# - 3.5 framework - windows 7 64 bits (and i want my application to work on both 32 and 64 bits systems) Thank you in advance, Wilhelm

  Read the article

• How reliable are URIs like /index.php/seo_path

  - by Boldewyn

  I noticed, that sometimes (especially where mod_rewrite is not available) this path scheme is used: http://host/path/index.php/clean_url_here --------------------------^ This seems to work, at least in Apache, where index.php is called, and one can query the /clean_url_here part via $_SERVER['PATH_INFO']. PHP even kind of advertises this feature. Also, e.g., the CodeIgniter framework uses this technique as default for their URLs. The question: How reliable is the technique? Are there situations, where Apache doesn't call index.php but tries to resolve the path? What about lighttpd, nginx, IIS, AOLServer? A ServerFault question? I think it's got more to do with using this feature inside PHP code. Therefore I ask here.

  Read the article

• In Android Browser link does not always execute onClick causing focus instead

  - by Artem

  I am trying to program a very standard JS behavior for a link using an HREF onClick handler, and I am facing a strange problem caused by what I believe to be focus/touch mode behavior on Android. Sometimes when I click on the link, instead of executing the action, it simply becomes selected/focused, with either just a focus rectangle or even also with a filled focus rectangle (selected as opposed to just focused?). The pseudo-code right now is <a href="#" onClick="toggleDivBelowToShowHide(); return false;">go</a> I have tried doing something like: <a href="#" onTouchStart="toggleDivBelowToShowHide(); return false;">go</a> But I still get the same pesky problem some of the time.

  Read the article

• How to improve IntelliJ code editor speed?

  - by Hoàng Long

  I am using IntelliJ (Community Edition) for several months, and at first I'm pleased about its speed & simplicity. But now, after upgrading to version 10, it's extremely slow. Sometimes I click a file then it takes 5 - 15 seconds to open that file (it freeze for that time). I don't know if I have done anything which cause that: I have installed 2 plugins(regex, sql), and have 2 versions of IntelliJ on my machine (now the version 9 removed, only version 10 remains). Is there any tips to improve speed of code editor, in general, or specifically IntelliJ? I have some experience when using IntelliJ: 1. Should open IntelliJ a while before working, cause it needs time for indexing. Don't open too many code tabs Open as less other program as possible. I'm using 2 GB RAM WinXP, and it just seems fairly enough for Java, IntelliJ & Chrome at the same time.

  Read the article

• CSS - Force overflowing elements to disappear if partially hidden

  - by Kelso.b

  Let's say we have a box with some short paragraphs: <div style="overflow:hidden"> <p>Some text</p> <p>Some text</p> <p>Some text</p> <p>Some text</p> </div> The height of the box is variable, so sometimes one of the paragraphs' text is partially hidden. Is there a CSS property that would force the paragraph to either display fully or not at all, or would this need to be calculated using javascript?

  Read the article

• firefox: getting access to the list of tabs/windows to restore on startup

  - by robb

  Sometimes ffox fails to restore the previously open tabs/windows. This might be happening when some of the urls to be opened are no longer reachable (e.g. behind a vpn) or after the underlying OS (Windows) has been forcibly restarted (e.g. to complete an automated patch installation). Anyway, after restarting, can this list of urls be recovered somehow? Say for example, I was daft enough to have clicked on "start new session". Can I still get access to the old list of open urls? There is the browser history of course, but it contains a lot of stuff - the urls that were open when ffox last exited are not obvious. It would be neat if they were marked in some way - tagged for example. .robb

  Read the article

• Characters in string changed after downloading HTML from the internet.

  - by Callum Rogers

  Using the following code, I can download the HTML of a file from the internet: WebClient wc = new WebClient(); // .... string downloadedFile = wc.DownloadString("http://www.myurl.com/"); However, sometimes the file contains "interesting" characters like é to é, ? to ↠and ????? to フシギダãƒ. I think it may be something to do with different unicode types or something, as each character gets changed into 2 new ones, perhaps each character being split in half but I have very little knowledge in this area. What do you think is wrong?

  Read the article

• Querying current number of touches on screen without using events on iPhone

  - by nikhil

  I have an application that starts playing a sound when user touches the uiview and changing to different tones as the user slides the finger on the screen. The sound stops when the user lifts the finger. I am using the touchesBegan, Moved and Ended Events for this. My problem is touches Ended (and/or cancelled) is sometimes not fired properly and the sound keeps playing even after the finger is lifted from screen. So as a workaround I would like to implement a timer that would check for the number of touches on the screen and if it is zero it will check and stop the audioplayer if playing. I have been searching for some code that could get me the number of touches like UITouch *touches=[self getAllTouchesonScreen]; or something :)

  Read the article

• How do you get notified of your repos' updates?

  - by furtelwart

  I'm working on several repositories at work and would like to be informed, if anything changes in the SVN repositories. I made a small BAT script (yes, BAT is usefull sometimes) that keeps executing an svn log -r BASE:HEAD on my working copy. It shows all submit comments and revision dates. It works really well but it's not comfortable for different repositories. How do you keep track of changes in your repositories? Do you use a small program you made for yourself? Do you use software someone else made? I'm interested in every approach to this problem. I would like to get notifications and several more information about the commit. The IDE integrated functions are good, but work only if I request the information. I don't want to act to get this information. Platform: Windows, Subversion 1.5 and higher.

  Read the article

• Is it possible to force ignore the :hover pseudoclass for iPhone/iPad users?

  - by christophercamps

  I have some css menus on my site that expand with :hover (without js) This works in a semi-broken way on iDevices, for example a tap will activate the :hover rule and expand the menu, but then tapping elsewhere doesn't remove the :hover. Also if there is a link inside the element that is :hover'ed, you have to tap twice to activate the link (first tap triggers :hover, second tap triggers link). I've been able to make things work nicely on iphone by binding the touchstart event. The problem is that sometimes mobile safari still chooses to trigger the :hover rule from the css instead of my touchstart events! I know this is the problem because when I disable all the :hover rules manually in the css, mobile safari works great (but regular browsers obviously don't anymore). Is there a way to dynamically "cancel" :hover rules for certain elements when the user is on mobile safari? I'm using jQuery.

  Read the article

• Unary NOT/Integersize of the architecture

  - by sid_com

  From "Mastering Perl/Chapter 16/Bit Operators/Unary NOT,~": The unary NOT operator (sometimes called the complement operator), ~, returns the bitwise negation, or 1's complement, of the value, based on integer size of the architecture Why does the following script output two different values? #!/usr/local/bin/perl use warnings; use 5.012; use Config; my $int_size = $Config{intsize} * 8; my $value = 0b1111_1111; my $complement = ~ $value; say length sprintf "%${int_size}b", $value; say length sprintf "%${int_size}b", $complement; Output: 32 64

  Read the article

• Finding the time to program in your spare time?

  - by Omar Kooheji

  I've got about a dozen programming projects bouncing about my head, and I'd love to contribute to some open source projects, the problem I have is that having spent the entire day staring at Visual Studio and or Eclipse (Sometimes both at the same time...) the last thing I feel like doing when I go home is program. How do you build up the motivation/time to work on your own projects after work? I'm not saying that I don't enjoy programming, it's just that I enjoy other things to and it can be hard to even do something you enjoy if you've spent all day already doing it. I think that if I worked at a chocolate factory the last thing I'd want to see when I got home was a Wonka bar.... Related: How do you keep a balance between working, training, health and family?

  Read the article

• Commenting practices?

  - by Tarmon

  Hey Everyone, As a student in computer engineering I have been pressured to type up very detailed comments for everything I do. I can see this being very useful for group projects or in the work place but when you work on your own projects do you spend as much time commenting? As a personal project I am working on grows more and more complicated I sometimes feel as though I should be commenting more but I also feel as though it's a waste of time since I will probably be the only one working on it. Is it worth the time and cluttered code? Thoughts?

  Read the article

• Good practices for intialising properties ?

  - by Rubans

  HI, I have a class property that is a list of strings, List. Sometimes this property is null or if it has been set but the list is empty then count is 0. However elsewhere in my code I need to check whether this property is set, so currently my code check whether it's null and count is 0 which seems messy. if(objectA.folders is null) { if(objectA.folders.count == 0) { // do something } } Any recommendation on how this should be handled? Maybe I should always initialise the property so that it's never null? Appolgies if this is a silly question.

  Read the article

• Best Version control for lone developer

  - by Stephen

  I'm a lone developer at the moment; please share you experiences on what is a good VC setup for a lone developer. My constraints are; I work on multiple machines and need to keep them synced up Sometimes I work offline I'm currently using Subversion(just the client to a remote server), and that is working ok. I'm interested in mecurial and git DVCS, but none of their use-cases make sense to my situation. EDIT: I've migrated my active development to Fossil http://www.fossil-scm.org/ after trialing it with a client. I really like the features to autosync my repositories(reducing accidental forks), the documentation support(both wiki and embedded/versioned) that supports my need to document the code and the project in different spaces, the easy to configure issue tracker, nice access control, skinnable web interface and helpful community.

  Read the article

• Are Sphinx & thinking_sphinx really stable? Not indexing Columns

  - by seb

  I'm encountering strange behaviour from thinking_sphinx/sphinx. My define_index block is about 100 lines, so quite a lot of columns i'm indexing. For full-text searching I only need about 10 attributes, for sorting and filtering I have another approximately 50 columns, mostly floats and integers. By filtering I mean using the "with" or "without" options. Searching does not really work consistently. All of a sudden, one attribute fails to filter. Or if I add a new one, it does not work. Only after a lot of tinkering it suddenly starts working. I cannot really reproduce it. Steps I that sometimes lead me to success where: rm -rf db/sphinx change the attribute definition e.g. has some_attribute = has some_attribute, :sortable = true or = has some_attribute, :sortable = true, :as = "some_attribute" restarting the server assigning a new :as name = has some_attribute, :as = "some_attribute_new" (yes, I did rake ts:rebuild or rake ts:in after every step) Does anybody else encounter similar problems?

  Read the article

• How to implement following with Zend navigation?

  - by simple

  I have a top navigation and a side one Home | Tours | Booking Tour1 ------ Tour2 ------ Tour3 I show the side menu depending on a active top item. But sometimes when the side menu item is clicked I have to show that items children instead of the sidemenu. When no children exist I just show Children of a top menu depending on the active item. I an really having difficulties implementing that kind a logic and any help would be appreciated. //Comment button is not working so I will add comments here After stumbling the zend navigation view helpers, I am coming to the point - Understanding the concept of how in zend V part of MVC is implemented can save someone who is new to the framework many hours. As said in the answer to this question - "Use what is available", thought first we have to know where to find what is already available out there - that is where comes handy to take a look at the concept of helpers and so forth.

  Read the article

• DataTable C# Empty column type

  - by Dested

  I am trying build a DataTable one row at a time using the following code. foreach (var e in Project.ProjectElements[hi.FakeName].Root.Elements()) { index = 0; object[] obj=new object[count]; foreach (var holdingColumn in names) { string d = e.Attribute(holdingColumn.Key).Value; obj[index++] = d; } dt.Rows.Add(obj); } The problem is the DataTable has types tied to the columns. Sometimes im passing null (or an empty string) in that object index and it is telling me that it cant be converted properly to a DateTime (in this case). My question is what should I default this value to, or is there some way to have the DataTable ignore empty values.

  Read the article

• Weird mapping error in linq-to-sql dbml file in VS2010

  - by rwwilden

  Since I switched to VS2010, several times a day I get a compilation error in my dbml file: DBML1005: Mapping between DbType 'bigint' and Type 'MyNamespace.SecurityToken' in Column 'SecurityToken' of Type 'Employee' is not supported When I restart VS2010 the error disappears. I have no problems running my application using this dbml file (specifically, there are no problems getting correct values inside the SecurityToken property of Employee objects). The SecurityToken property is of an enum type defined as follows: [Flags] public enum SecurityToken : long { None = 1, Admin = 2, ...... } The SecurityToken column in the database is of type bigint. Am I missing something? It's especially weird that the error only happens sometimes, when I'm writing code that isn't related at all to the LINQ model.

  Read the article

• Preload Images with javascript

  - by user271619

  Here is my preload script: <script type="text/javascript"> Image_1 = new Image(1,1); Image_1.src = "images/sprites.png"; </script> And it works fine. I can see in the headers the image loading. Here's my question. Even though I have preloaded sprites.png, I notice that when I go to a page on my site that simply displays that entire image, the headers tell me my browser is requesting the server to send the image again. I'm hoping to not sound too naive. But, why would the browser request the image when I preloaded it earlier? Are there other reasons the browser would need to re-request? Also, I put the dimensions as Image(1,1). Sometimes I don't know the dimensions, as that image gets updated from time to time. Maybe that's a reason?

  Read the article

• Need to determine if ELMAH is logging an unhandled exception or one raised by ErrorSignal.Raise()

  - by Ronnie Overby

  I am using the Elmah Logged event in my Global.asax file to transfer users to a feedback form when an unhandled exception occurs. Sometimes I log other handled exceptions. For example: ErrorSignal.FromCurrentContext().Raise(new System.ApplicationException("Program code not found: " + Student.MostRecentApplication.ProgramCode)); // more code that should execute after logging this exception The problem I am having is that the Logged event gets fired for both unhandled and these handled, raised exceptions. Is there a way to determine, in the Logged event handler, whether the exception was raised via ErrorSignal class or was simply unhandled? Are there other Elmah events that I can take advantage of?

  Read the article

• Django admin interface upload failing on request data read error

  - by Jake

  Hi All, This is an updated version of an old question I asked. I've now done a lot more testing, plus the old question got hijacked. I'm getting a request data read error when trying to upload files to the Django admin interface. Files under about 150k work, but bigger files always fail and almost always at around 192k (that's 3 chunks) completed, sometimes at around 160k. The Exception I get is below. File "/usr/lib/python2.4/site-packages/django/http/multipartparser.py", line 405, in read return self._file.read(num_bytes) IOError: request data read error I've tried Chrome and Firefox on Windows and Firefox on Mac - Same results. I can upload to other sites so I don't think it's my connection. I'm running python 2.4, django 1.1, mod_wsgi, on CentOS (a media temple DV server) Locally it's fine (Django development server) Everything I've found on this issue says it's a mod_python issue and that changing to mod_wsgi will fix it, but I am running mod_wsgi. Can anyone help?

  Read the article

• Android ndk build mysteriously failing under cygwin with "Error 126"

  - by Jan Hudec

  I have a JNI application built by ndk-build (using Android NDK r5b and cygwin make 3.81). The build usually works, by occasionally fails with: ... Compile++ thumb : components <= Component.cpp make: *** [/c/.hudson/jobs/Nightly/workspace/application/obj/local/armeabi/objs/components/Component.o] Error 126 make: Leaving directory `/c/.hudson/jobs/Nightly/workspace/application/obj/local/armeabi/objs/components' There is no other error. Make than exits with status 2. It happens in different file each time (the name above is anonymized). It seems to happen more often with parallel builds, but sometimes happens with non-parallel builds too. Does anybody have an idea what it might be or at least how to debug it?

  Read the article

• Send Logcat output of an App to an EmailAdress

  - by Sebi

  Hi Were now testing our application with a few friends. Sometimes there are some errors which dont throw an exception. So I don't really know whats the problem was. So i thought it would be a good idea to implement a menu item which allows to send the logcat file to a email adress, so that we can examine the logcat. Unfortunately I didnt found a hint in the Internet how to extract the Logcat from a phone. How to send a email shouldn't be the problem.

  Read the article

• GWT application throws an exception when run on Google Chrome with compiler output style set to 'OBF

  - by Elifarley

  I'd like to know if you guys have faced the same problem I'm facing, and how you are dealing with it. Sometimes, a small and harmless change in a Java class ensues strange errors at runtime. These errors only happen if BOTH conditions below are true: 2) the application is run on Google Chrome, and 1) the GWT JavaScript compiler output style is set to 'OBF'. So, running the application on Firefox or IE always works. Running with the output style set to 'pretty' or 'detailed' always works, even on Google Chrome. Here's an example of error message that I got: "((TypeError): Property 'top' of object [object DOMWindow] is not a function stack" And here's what I have: - GWT 1.5.3 - GXT 1.2.4 - Google Chrome 4 and 5 - Windows XP In order to get rid of this Heisenbug, I have to either deploy my application without obfuscation or endure a time-consuming trial-and-error process in which I re-implement the change in slightly different ways and re-run the application, until the GWT compiler is happy with my code. Would you have a better idea on how to avoid this?

  Read the article

< Previous Page | 195 196 197 198 199 200 201 202 203 204 205 206  | Next Page >