Search Results

Search found 7152 results on 287 pages for 'bug fixing'.

Page 21/287 | < Previous Page | 17 18 19 20 21 22 23 24 25 26 27 28  | Next Page >

  • Nginx: bug using if in location, how do I rectify

    - by Quintin Par
    I am using nginx in reverse proxy mode. In my server section I have this code to set expire and cache control of my static files. location ~* ^.+\.(css|js|png|gif)$ { access_log off; expires max; add_header Cache-Control public; if (!-f $request_filename) { proxy_pass http://localhost:82; } } This is quite obviously creating issues. Can someone help me correct this code to use try_files or rewrite?

    Read the article

  • Strange UIKit bug, table view row stays selected

    - by Can Berk Güder
    I'm facing what appears to be a UIKit bug, and it takes the combination of two less commonly used features to reproduce it, so please bear with me here. I have quite the common view hierarchy: UITabBarController -> UINavigationController -> UITableViewController and the table view controller pushes another table view controller onto the navigation controller's stack when a row is selected. There's absolutely nothing special or fancy in the code here. However, the second UITableViewController, the "detail view controller" if you will, does two things: It sets hidesBottomBarWhenPushed to YES in its init method, so the tab bar is hidden when this controller is pushed: - (id)initWithStyle:(UITableViewStyle)style { if(self = [super initWithStyle:style]) { self.hidesBottomBarWhenPushed = YES; } return self; } It calls setToolbarHidden:NO animated:YES and setToolbarHidden:YES animated:YES on self.navigationController in viewWillAppear: and viewWillDisappear: respectively, causing the UIToolbar provided by UINavigationController to be displayed and hidden with animations: - (void)viewWillAppear:(BOOL)animated { [super viewWillAppear:animated]; [self.navigationController setToolbarHidden:NO animated:YES]; } - (void)viewWillDisappear:(BOOL)animated { [super viewWillDisappear:animated]; [self.navigationController setToolbarHidden:YES animated:YES]; } Now, if the second UITableViewController was pushed by selecting the row at the bottom of the screen (it doesn't have to be the last row) in the first controller, this row does not automatically get deselected when the user immediately or eventually returns to the first controller. Further, the row cannot be deselected by calling deselectRowAtIndexPath:animated: on self.tableView in viewWillAppear: or viewDidAppear: in the first controller. I'm guessing this is a bug in UITableViewController's drawing code which of course only draws visible rows, but unfortunately fails to determine correctly if the bottommost row will be visible in this case. I failed to find anything on this on Google or OpenRadar, and was wondering if anyone else on SO had this problem or knew a solution/workaround.

    Read the article

  • Workaround for GNU Make 3.80 eval bug

    - by bengineerd
    I'm trying to create a generic build template for my Makefiles, kind of like they discuss in the eval documentation. I've run into a known bug with GNU Make 3.80. When $(eval) evaluates a line that is over 193 characters, Make crashes with a "Virtual Memory Exhausted" error. The code I have that causes the issue looks like this. SRC_DIR = ./src/ PROG_NAME = test define PROGRAM_template $(1)_SRC_DIR = $$(SRC_DIR)$(1)/ $(1)_SRC_FILES = $$(wildcard $$($(1)_SRC_DIR)*.c) $(1)_OBJ_FILES = $$($(1)_SRC_FILES):.c=.o) $$($(1)_OBJ_FILES) : $$($(1)_SRC_FILES) # This is the problem line endef $(eval $(call PROGRAM_template,$(PROG_NAME))) When I run this Makefile, I get gmake: *** virtual memory exhausted. Stop. The expected output is that all .c files in ./src/test/ get compiled into .o files (via an implicit rule). The problem is that $$($(1)_SRC_FILES) and $$($(1)_OBJ_FILES) are together over 193 characters long (if there are enough source files). I have tried running the make file on a directory where there is only 2 .c files, and it works fine. It's only when there are many .c files in the SRC directory that I get the error. I know that GNU Make 3.81 fixes this bug. Unfortunately I do not have the authority or ability to install the newer version on the system I'm working on. I'm stuck with 3.80. So, is there some workaround? Maybe split $$($(1)_SRC_FILES) up and declare each dependency individually within the eval?

    Read the article

  • asp.net updatepanel inside hidden panel possible bug

    - by MakkyNZ
    Hi The javascript generated by the asp.net SciptManager control seems to have a bug and cant handle hidden UpdatePanels. A javascript error is thrown when a control within one updated panel tries to make another update panel visible. Is this a bug with ASP.Net ajax? And does anyone have any ideas how to get around this? heres is an example of when im trying to to <script type="text/C#" runat="server"> protected void LinkButton1_Click(object sender, EventArgs e) { Panel1.Visible = true; } </script> <asp:ScriptManager ID="ScriptManager1" runat="server"> </asp:ScriptManager> <asp:UpdatePanel ID="UpdatePanel1" runat="server"> <ContentTemplate> <asp:LinkButton ID="LinkButton1" runat="server" OnClick="LinkButton1_Click" Text="Show Panel"></asp:LinkButton> </ContentTemplate> </asp:UpdatePanel> <asp:Panel ID="Panel1" runat="server" Visible="false"> <asp:UpdatePanel ID="UpdatePanel2" runat="server"> <ContentTemplate> blah bla blah </ContentTemplate> </asp:UpdatePanel> </asp:Panel> this is the javascript error that gets thrown when clicking on the "LinkButton1" link. This error comes from the javascript that is generated by the asp.net ScriptManager control Error: Sys.InvalidOperationException: Could not find UpdatePanel with ID 'ctl00_ContentPlaceHolder1_UpdatePanel2'

    Read the article

  • Possible bug in ASP.net UpdatePanel control?

    - by Ben Robinson
    I have come across what seems to be an annoying bug with asp.net UpdatePanels in 2 seperate projects. If you have some kind of autopostback enabled control that can cause all of the controls in the update panel to have visible=false set, resulting in an empty update panel. When you change the autopostback control back to the postion that would re enable all of the controls in the update panel, it simply does not make a call back to the server and the update panel does not update. If you do anything else that makes a call back on the same page, then the update panel contents magically appear. It is as if asp.net has decided the update panel is empty so there is no point maikng a callback, even though making the call back would fill the updatepanel with content. The only way round this is to add a style of display:none to the controls instead of setting visible=false property. Then it works fine. Has anyone else encountered this problem? Is it a bug as i suspect or is it likely i am doing soemthing wrong? I haven't got time to post example code at the moment as the code i am using is too wrapped up in other unrealted things, if people think it would help i will create a simple example and post it when I get time.

    Read the article

  • Weird bug with C++ lambda expressions in VS2010

    - by Andrei Tita
    In a couple of my projects, the following code: class SmallClass { public: int x1, y1; void TestFunc() { auto BadLambda = [&]() { int g = x1 + 1; //ok int h = y1 + 1; //c2296 int l = static_cast<int>(y1); //c2440 }; int y1_copy = y1; //it works if you create a local copy auto GoodLambda = [&]() { int h = y1_copy + 1; //ok int l = this->y1 + 1; //ok }; } }; generates error C2296: '+' : illegal, left operand has type 'double (__cdecl *)(double)' or alternatively error C2440: 'static_cast' : cannot convert from 'double (__cdecl *)(double)' to 'int' You get the picture. It also happens if catching by value. The error seems to be tied to the member name "y1". It happened in different classes, different projects and with (seemingly) any type for y1; for example, this code: [...] MyClass y1; void TestFunc() { auto BadLambda = [&]()->void { int l = static_cast<int>(y1); //c2440 }; } generates both these errors: error C2440: 'static_cast' : cannot convert from 'MyClass' to 'int' No user-defined-conversion operator available that can perform this conversion, or the operator cannot be called error C2440: 'static_cast' : cannot convert from 'double (__cdecl *)(double)' to 'int' There is no context in which this conversion is possible It didn't, however, happen in a completely new project. I thought maybe it was related to Lua (the projects where I managed to reproduce this bug both used Lua), but I did not manage to reproduce it in a new project linking Lua. It doesn't seem to be a known bug, and I'm at a loss. Any ideas as to why this happens? (I don't need a workaround; there are a few in the code already). Using Visual Studio 2010 Express version 10.0.40219.1 Sp1Rel.

    Read the article

  • CakePHP 1.26: Bug in 'Security' component?

    - by Steve
    Okay, for those of you who may have read this earlier, I've done a little research and completely revamped my question. I've been having a problem where my form requests get blackholed by the Security component, although everything works fine when the Security component is disabled. I've traced it down to a single line in a form: <?php echo $form->create('Audition');?> <fieldset> <legend><?php __('Edit Audition');?></legend> <?php echo $form->input('ensemble'); echo $form->input('position'); echo $form->input('aud_date'); // The following line works fine... echo $form->input('owner'); // ...but the following line blackholes when Security included // and the form is submitted: // echo $form->input('owner', array('disabled'=>'disabled'); ?> </fieldset> <?php echo $form->end('Submit');?> (I've commented out the offending line for clarity) I think I'm following the rules by using the form helper; as far as I can tell, this is a bug in the Security component, but I'm too much of a CakePHP n00b to know for sure. I'd love to get some feedback, and if it's a real bug, I'll submit it to the CakePHP team. I'd also love to know if I'm just being dumb and missing something obvious here.

    Read the article

  • jQuery Accordion + Anchor Tag 'stuck as block' bug?

    - by DA
    Sample page: http://jsbin.com/ohuze/2 This is a simple jQuery UI Accordion. Each accordion panel has an UL (an OL works the same) with this markup: <ol> <li><a href="">Lorep ipsum dolor lorem ipsum dolor lorem ipsum dolor</a>?</li> <li><a href="">Lorep ipsum dolor lorem ipsum dolor lorem ipsum dolor</a>?</li> </ol> In IE6, you'll see that the <a> tag appears to be getting rendered as a block element, so the question mark ends up being pushed outside and not at the end of the line of text. In addition, the bullet and/or list item number is now bottom-aligned with the text rather than top-aligned. I've narrowed it down to the javascript that executes to make the accordion. It's not an issue with jQuery's CSS as disabling that, alone, doesn't resolve the issue. Anyone know what might be going on in IE6 to cause this rendering issue? UPDATE: Apparently, this is also an IE7 issue. UPDATE 2: After some more playing, I've narrowed things down a bit more: the bug has nothing to do with lists. The issue is any anchor tag within a jQuery Accordion will appear as display: block (even though it appears that the CSS still indicates display: inline) the bug has nothing to do with the actual CSS that jQuery UI uses to create the accordion. I created a test page that uses the fully rendered jQuery Accordion post-processed source code and the accompanying CSS. In that situation, the anchor tags remain inline. In conclusion: It appears that the process of rendering the accordion via javascript is messing up the display of the anchor tags. It may be a show/hide issue?

    Read the article

  • Bizarre Bug with our Rails app in IE

    - by Callmeed
    We're experiencing a really bizarre bug in our Rails 2.3.4 app. This bug only happens in Internet Explorer (7 and 8). Here's what happens: A new customer creates an account at https://domain.com/signup/free (notice no subdomain) Their account is identified by a subdomain like "example.domain.com" After signing up, they get a welcome screen with a link to their account's home page They follow the link, then click the "log in" button and attempt to login Even though they provide valid credentials, the app redirects back to their account's root url ... they can never reach their admin area The only way they can login (on IE) is by quitting and re-opening IE ... then it works fine ... Something with their initial session is preventing them from logging in. If it matters, we are using restful_authentication and the ssl_requirement plugin ... I'm not sure if one or both of those has a problem with IE but we are stumped here. Also, I've read IE has an issue with subdomains that contain underscores ... this isn't what's going on.

    Read the article

  • fix needed for bug in TextField/Text

    - by Mark
    Sort of a complicated scenario - just curious if anyone else could come up with something: I have a Text control and when I scroll it and stop the scroll with the cursor over some text that has a url, the cursor doesn't revert to a hand, and also flash player starts acting as if a selection is being made from the last cursor position. So IOW a bonafide bug in flash as far as I can determine. The above probably wasn't completely clear so let me elaborate. If you grab a scrollbar thumb and start moving it up and down, you don't actually have to keep the mouse pointer on the thumb while doing so. When you stop the scroll, the mouse pointer could be outside the browser window, inside your flash application, but not currently on the scroll bar thumb, or wherever. The previously mentioned bug occurs when you stop the scroll with the mouse pointer positioned over text with an html anchor (a hyperlink). At that point the cursor enters into some state of limbo, and doesn't show the url hand pointer, and furthermore acts as if some text selection is taking place from the last cursor position prior to the scroll. So the question would be, what sort of event could I simulate in code to jolt flash out of this erroneous state it is in. And furthermore in what event could I perform this simulated event (given that for example there is no AS3 event to signal the end of a scroll.) To be clear, the Text control in question is on a canvas, and that canvas (call it A) is on another canvas which actually owns the scrollbar, and scrolling takes place by changing the scrollRect of canvas A.

    Read the article

  • TreeView update bug in the VB.NET

    - by CFP
    Consider the following code: Dim Working As Boolean = False Private Sub TreeView1_AfterCheck(ByVal sender As Object, ByVal e As System.Windows.Forms.TreeViewEventArgs) Handles TreeView1.AfterCheck If Working Then Exit Sub Working = True e.Node.Checked = Not e.Node.Checked Working = False End Sub Private Sub TreeView1_MouseClick(ByVal sender As System.Object, ByVal e As System.Windows.Forms.MouseEventArgs) Handles TreeView1.MouseClick If e.Button = Windows.Forms.MouseButtons.Right Then MsgBox("Checked = " & TreeView1.SelectedNode.Checked) End Sub Where TreeView1 is a TreeView added to the form, with CheckBoxes set to true and one node added. The code basically cancel any node checking occuring on the form. Single-clicking the top node to check it works well : your click is immediately canceled. Yet if you double-click the checkbox, it will display a tick. But verifying the check state through a right click will yield a Checked = False dialog. How come? Is it a bug (I'm using the latest .Net Framework 4.0, and he same occurs in 2.0), or am I doing something wrong here? Is there a work around? Thanks! EDIT: Additionally, the MouseDoubleClick event is not raised before you click once again. EDIT 2: Posted a bug report at Microsoft Connect

    Read the article

  • What is your experience of Devtrack?

    - by Luke H
    This question covers bug tracking software in general, but I'm interested to find out more detail specifically about Devtrack. If you have first-hand experience of using it, I'd love to hear about it. How would you compare it to other bug tracking systems you know, what do you feel is good and bad about it, and why?

    Read the article

  • Chinese IME - Shape toggle key won't save

    - by Mikhail Sayapin
    Hello everyone, Whenever I use Google IME or other IMEs like Sogou, I always encounter the same problem: half/full-width toggle. It's easy to set half-width by default in ime prefs, but toggle key is maintained by Windows, and here the fun begins. I use Win7 x64 Ultimate. Open the "Text services and input languages", go to "Advanced key settings". Here I see "Chinese (Simplified) IME - Shape Toggle" is set to "Shift+Space". When I try to change it, it appears as if it has changed, but if you click on "Apply" instead of "OK", or just open the window again, you can see it's still "Shift+Space". The same happens with disabling this toggle - disables and goes back on Apply or reopen. Not sure if it's a bug in Win7, so asking here. Please save me. :) Thanks!

    Read the article

  • Setting the view of the toolbar to Large icons creates a gap in the Windows 7's taskbar

    - by Boris
    If you add a custom toolbar in the Windows 7 taksbar and set the toolbar view to Large Icons (and the icons in the toolbar are not set to Use small icons), the height of the taskbar unexpectedly increases by around 5 pixels, which makes a rather stupid gap at the bottom of the screen. If the option Use small icons is used for the taskbar appearance, the height of the taskbar is normal. It appears that the programmers at Microsoft were not very meticulous while designing the Windows 7 UI; it is obviously a bug. I was wondering if there is a registry hack to fix this or if anyone knows any solution to the problem, except for the obvious one: "use the small icons"? Thanks.

    Read the article

  • "Boot has failed, sleeping forever." Suggest some solution for this.

    - by Lazer
    I downloaded Fedora from here. While trying to install it, I am getting this error: No root device found. Boot has failed, sleeping forever. I tried searching on it (found codehead and fedoraforum) and came to know that it is a known bug, but couldn't find any solution to it. Do you know any solution to this? OR Do you know a distribution of Fedora that does not have these problems? [I am trying to install Fedora on my Windows Vista machine]

    Read the article

  • Firefox and Flash requires 2 clicks after Scrolling?

    - by Kerry
    I've noticed this with me and my partners computers, take a look at this: http://www.uploadify.com/demo/ Those buttons are flash. If you scroll (scroll to the bottom then back up, using your mouse scroll) then hover over a button, it is a normal pointer, not a hand. If you click on the button, then it turns into a hand, then you can click on it to trigger the browse files window. Is there any way to fix this? Is this a Firefox bug (doesn't happen in Chrome or IE8)?

    Read the article

  • iPhone virtual keyboard bug

    - by Chandan Shetty SP
    In iPhone virtual keyboard... 1.Change the alphabetical keypad view to numerical view. 2.Tap on the single quote(') button the view changes to alphabetical. 3.In this view tapping on space twice displays a fullstop. I don't know whether it is apple bug or feature, How to fix this issue through coding? Thanks,

    Read the article

  • restrict customer lookup to contacts bug

    - by Amr H. Abdel Majeed
    I'm using microsoft dynamics crm 4 I inserted the following line to restrict a customer lookup to contacts only (no accounts): crmForm.all.customerid.setAttribute("lookuptypes", "2"); It does restrict the lookup to contacts only, but there is a strange bug: When i select any customer, the icon next to him becomes the account icon, not the contact icon (folder instead of card) Any solutions to this ?

    Read the article

  • Weird bug in header in IE7

    - by Luuk
    Hi all, I have a weird bug in IE7. http://www.luukratief-design.nl/dump/simplefolio/ the navigation has to be centered. same goes for the background. Now every browser does it perfectly. even IE6! (didnt do pngfix yet). The only one who is having problems is IE7 who shifts the whole navbar element to the right. Anyone knows how this is possible?

    Read the article

  • JCarousel wrap set to 'circular' bug

    - by Scott
    I'm trying to set up a JCarousel instance witht the wrap set to circular, however, I noticed an issue where I "inspect element" in Chrome, duplicate elements are added to the html when the slides repeat (it can be viewed on the official example page). http://sorgalla.com/projects/jcarousel/examples/static_circular.html I know this is an open bug on GitHub. I'm just wondering if anyone has found a temporary workaround. Thanks in advance, - Scott

    Read the article

  • Showing bug count or CI build on a TV in the office

    - by Matt Frear
    I recently saw a blog post showing off different displays that some software development shops use for showing their bug count or CI build status on a nice TV in the office. I can't for the life of me find it now, and I can't find it on google either. Anyone know what I'm talking about? (I don't think this belongs on superuser or serverfault either) -Matt

    Read the article

  • Internet Explorer border bug

    - by Skilldrick
    On this page in Internet Explorer 6 and 7 on XP and Vista there is a bug where scrolling the page seems to lose bits of the left and right borders. Has anyone seen this behaviour before, and does anyone have any ideas to avoid it?

    Read the article

  • asp.net jquery datepicker bug

    - by frabiacca
    hi there, i'm using successfully a jquery datepicker with masked input plugin too in a aspx webform. I've noticed a bug: when i insert a date by choosing it on the calendar and then i try to modify it manually, the date suddenly disappears. Have u ever seen this behaviour?

    Read the article

< Previous Page | 17 18 19 20 21 22 23 24 25 26 27 28  | Next Page >