Search Results

Search found 22901 results on 917 pages for 'query bug'.

Page 582/917 | < Previous Page | 578 579 580 581 582 583 584 585 586 587 588 589  | Next Page >

• Using [delphi] MadExcept errorhandling with MS Echange Server 2007

  - by Tony

  I currently use madExcept.MailAsSmtpClient to send my bug reports. However a couple of large clients have upgraded to Exchange Server 2007 and we can't get the SMTP support for our app configured (the app runs on individual workstations so the messages aren't all coming from one IP. We can configure an authenticated account in exchange and access it via SMTP from other clients but it rejects madExcept for some reason). So I have two questions 1) has anyone successfully configured that combo ? or 2) is there an example somewhere of how to use the madExcept.UploadViaHTTP option?

  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

• jQuery TriggerHandler doesn't trigger keyup event set via live

  - by Beck

  Why i can't trigget handler via triggerHandler if it's was bound via live? For example form.find('#code').live('change keyup',function(){ process_code($(this)); }); Startups after window load: /* Triggers */ if(checkCookie('banner_ready_code')) { $('#form').find('#code').triggerHandler('keyup'); } Nothing happens... If i'm binding simply by .change or .keyup, triggerHandler is working properly as intended. If it's bug, where to report it. Too lazy :) If not, please explain. :)

  Read the article

• Tortoisesvn not showing up correct status after commit

  - by Michael

  Hi, I was going to ask this in tigris.org, however they have maintenance in their forum. My environment: Windows 7 x64, Tortoisesvn latest x64, simple repo. What I'm doing: I am adding a new file to repo, then doing SVN Commit. This operation is successful and I can see it in repo from trac or directly. I expect: To see that file's icon as green checkmark. What I have: I see blue PLUS icon, like I haven't done commit. However, if I just create any new file in that folder(without any commit or update), the icon is immediately changing to checkmark. What a magic! I don't expect here anyone to have answer, this might be a bug, but who knows )) Cheers!

  Read the article

• Why can't QFile read from the "~" directory?

  - by hurikhan77

  I've tried the following short example to find out about a bug in a bigger program I am working on. It looks like QFile doesn't support unix (or the shell's) notation for the home directory: #include <QFile> #include <QDebug> int main() { QFile f("~/.vimrc"); if (f.open(QIODevice::ReadOnly)) { qDebug() << f.readAll(); f.close(); } else { qDebug() << f.error(); } } As soon as I replace the "~" with my real home directory path, it works. Is there an easy workaround - some setting to enable? Or do I have to go the "ugly" way and ask QDir for the home directory of the current user and prepend that manually to each path?

  Read the article

• from ggplot2 to OOo workflow?

  - by Andreas

  This is not really a programming question, but I try here none the less. I once used latex for my reports. But the people I work with needs to make small edits and do not have latex skillz. Openoffice is then the way to go. But saving ggplot images with dpi 100 makes for really ugly graphs. dpi = 600 is a no go (e.g. huge legend). So what to do? I currently save (still via ggsave) to eps - which openoffice can import. But performance is not good at all. Googling I found a bug for the poor eps performance in OOo, and also talk about a non-implemented svg feature. But none helps me right now. If you work with ggplot2 and OOo - What do you do? I have been unsuccesfull with pdf conversion for some reason.

  Read the article

• Why doesn't MySQL support millisecond precision?

  - by Byron Whitlock

  So I just found the most frustrating bug ever in MySQL. Apparently the TIMESTAMP field, and supporting functions do not support any greater precision than seconds!? So I am using PHP and Doctrine, and I really need those microseconds (I am using the actAs: [Timestampable] property). I found a that I can use a BIGINT field to store the values. But will doctrine add the milliseconds? I think it just assigns NOW() to the field. I am also worried the date manipulation functions (in SQL) sprinkled through the code will break. I also saw something about compiling a UDF extension. This is not an acceptable workaround because I or a future maintainer will upgrade and poof, change gone. Has anyone found a suitable workaround?

  Read the article

• Is Boost.Tuple compatible with C++0x variadic templates ?

  - by Thomas Petit

  Hi, I was playing around with variadic templates (gcc 4.5) and hit this problem : template <typename... Args> boost::tuple<Args...> my_make_tuple(Args... args) { return boost::tuple<Args...>(args...); } int main (void) { boost::tuple<int, char> t = my_make_tuple(8, 'c'); } GCC error message : sorry, unimplemented: cannot expand 'Arg ...' into a fixed-length argument list In function 'int my_make_tuple(Arg ...)' If I replace every occurrence of boost::tuple by std::tuple, it compiles fine. Is there a problem in boost tuple implementation ? Or is this a gcc bug ? I must stick with Boost.Tuple for now. Do you know any workaround ? Thanks.

  Read the article

• Facebook redirect after login not working when url is dynamic

  - by dythffvrb

  This is my code: $loginUrl = $facebook->getLoginUrl(array( 'scope' => 'publish_actions', 'redirect_uri' => 'http://mysite.com/', )); It works but if I remove redirect_uri it doesn't work anymore. $loginUrl = $facebook->getLoginUrl(array( 'scope' => 'publish_actions' )); According to Facebook domcumentation redirect_uri is optional. https://developers.facebook.com/docs/reference/php/facebook-getLoginUrl/ Im trying to redirect the users to the same url they were on before logging in. Update: This problem occurs when the url is mysite.com/post23 but when url is mysite.com/staticpage or mysite.com there are no problems Any workarounds? EDIT: It looks like a bug, it doesn't work with certain url in the same site I wil try and report it to Facebook.

  Read the article

• PHP fastest method of reading server response

  - by Peter John

  Hi there, im having some real problems with the lag produced by using fgets to grab the server's response to some batch database calls im making. Im sending through a batch of say, 10,000 calls and ive tracked the lag down to fgets causing the hold up in the speed of my application as the response for each call needs to be grabbed. I have found this thread http://bugs.php.net/bug.php?id=32806 which explains the problem quite well, but hes reading a file, not a server response so fread could be a bit tricky as i could get part of the next line, and extra stuff which i dont want. Any help much appreciated!

  Read the article

• MVVM-Light Loaded Evented Executing Twice

  - by user275561

  Let me show the code first, The Control <ItemsControl.ItemsPanel> <ItemsPanelTemplate> <Controls:MatrixGrid x:Name="matrixGrid"> <i:Interaction.Triggers> <i:EventTrigger EventName="Loaded"> <cmd:EventToCommand Command="{Binding MatrixLoaded}" CommandParameter="{Binding ElementName=matrixGrid}" /> </i:EventTrigger> </i:Interaction.Triggers> </Controls:MatrixGrid> <ItemsControl.ItemsPanel> <ItemsPanelTemplate> In The ViewModel Class I have public RelayCommand<MatrixGrid> MatrixLoaded { get; private set; } In The Constructor of the View Model I have MatrixLoaded = new RelayCommand<MatrixGrid>(MatrixGridAction); Now When I put a Breakpoint on the Function MatrixGridAction, The breakpoint is hit twice. Am I doing something wrong or is this a bug?

  Read the article

• Paypal recuring IPN signals

  - by user1548981

  When creating Paypal recurring payments i don't receive any notifications from IPN that the recurring payments were actually made. I do receive the notifications that recurring profiles are created, that regular payments are made, that profile is canceled and so on. But no notifications when Paypal make the recurring payment. The file that process the ipn notification works fine and i think is bug free.It also send me an email every time he got something from Paypal(with row POST data) so i don't think the problem is here. Does Paypal send these notifications.If i set a daily recurring payment i should get a daily notification that payment was made? From what i read in documentation i should get these signals but.. Thanks

  Read the article

• Check whether Excel file is Password protected

  - by Torben Klein

  I am trying to open an Excel (xlsm) file via VBA. It may or may not be protected with a (known) password. I am using this code: On Error Resume Next Workbooks.Open filename, Password:=user_entered_pw opened = (Err.Number=0) On Error Goto 0 Now, this works fine if the workbook has a password. But if it is unprotected, it can NOT be opened. Apparently this is a bug in XL2007 if there is also workbook structure protection active. (http://vbaadventures.blogspot.com/2009/01/possible-error-in-excel-2007.html). On old XL2003, supplying a password would open both unprotected and password protected file. I tried: Workbooks.Open filename, Password:=user_entered_pw If (Err.Number <> 0) Then workbooks.open filename This works for unprotected and protected file. However if the user enters a wrong password it runs into the second line and pops up the "enter password" prompt, which I do not want. How to get around this?

  Read the article

• Has anyone noticed that a WPF file dialog will pass through a click to the UI when double clicking t

  - by Ben

  I have some buttons on my WPF UI and I also need to choose files from time to time. I kept noticing strange problems where when I double-click an item in the file dialog, a button on the main UI would also get clicked. After experimenting, it seems that if you line up an item in the file dialog with a button behind it on the main UI and double click to select the file, it will single-click the button behind it as well. Has anyone else noticed this, or is it just a freak bug with the way I have my UI laid out?

  Read the article

• Is it valid syntax to have ordered and unordered lists in sequence in markdown?

  - by nfm

  I just wrote some markdown and it doesn't seem to render correctly. Is it legal syntax to have an ordered list, followed by newlines, then followed by an unordered list? Or is this a bug in bluecloth? For example: 1. One 2. Two 3. Three * Apple * Banana * Carrot Bluecloth creates a single <ul> and nests apple, banana and carrot as <li>'s under it. Stackoverflow's markdown parser does this too. Am I just doing it wrong? Surely this is a common usage case...

  Read the article

• Resources for looking up differences in rendering behavior between web-browsers and browser bugs

  - by ICodeForCoffee

  I recently encountered a printing issue in Firefox that eventually turned out to be a problem with the fieldset tag we wrapped the entire page in. (Bugzilla: Bug 471015) All browsers have their own rendering quirks and issues, but it can be very hard to know what's causing different behaviors. Sure, you can Google, but that's often taking a shot in the dark about what you think is causing the problem. It also doesn't stop you from having to sort through multiple complains about the same behavior before you find someone who has the issue your looking for. Are there any websites out there that let you search for rendering behavior issues by browser version, browser function, css tag, or html tag? I've seen this SO Question, Wanted: Resource for documented Cross-Browser differences, but I'd like to find something more detailed that includes browser bugs.

  Read the article

• How to create a view of table that contains a timestamp column?

  - by Matt Faus

  This question is an extension of a previous one I have asked. I have a table (2014_05_31_transformed.Video) with a schema that looks like this. I have put up the JSON returned by the BigQuery API describing it's schema in this gist. I am trying to create a view against this table with an API call that looks like this: { 'view': { 'query': u 'SELECT deleted_mod_time FROM [2014_05_31_transformed.Video]' }, 'tableReference': { 'datasetId': 'latest_transformed', 'tableId': u 'Video', 'projectId': 'redacted' } } But, the BigQuery API is returning this error: HttpError: https://www.googleapis.com/bigquery/v2/projects/124072386181/datasets/latest_transformed/tables?alt=json returned "Invalid field name "deleted_mod_time.usec". Fields must contain only letters, numbers, and underscores, start with a letter or underscore, and be at most 128 characters long." The schema that the BigQuery API does not make any distinction between a TIMESTAMP data type and a regular nullable INTEGER data type, so I can't think of a way to programmatically correct this problem. Is there anything I can do, or is this a bug with BigQuery's view implementation?

  Read the article

• Graph call "/me/friends" doesn't contain names if called from a testuser

  - by Marc

  When I log into Facebook with my account and call the Graph API with /me/friends it returns an array like this: [ {id: "6868...", name: "John Smith"}, {id: "7531...", name: "Diane Smith"}, ... ] But if I'm logged in with a Facebook testuser and do the same call I get an array like this: [ {id: "100002927..."}, {id: "100003151..."}, ... ] So I'm missing the name property. Bug? Or did I some mistake? I'm using the JS SDK, but I think that doesn't matter.

  Read the article

• How should the Version field be used in Trac?

  - by Eric

  I use Trac to track bugs, and future changes in my software projects. Tickets in Trac have a "Version" field and I'm trying to figure out the best way to use this field. Say I find a series of bugs in version 1.0 of my software. I create tickets in track for each and assign them to version 1.0. Now say I fix some of the bugs, and add some of the new features and release version 1.1. But some of the old 1.0 bugs are still in 1.1. Should I change their corresponding tickets to version 1.1 because they also now exist in 1.1? Or should I leave them set to version 1.0 as a way of tracking what version the bug was found in, and just assume that any open tickets in older versions still exist in newer versions?

  Read the article

• How can I update the version of an APK which I did not create myself

  - by sylvanaar

  I am new to Android development. I am using a x-platform development tool which builds and signs the APK for me. This tool has a bug and it does not generate APK's with the version number I specify, all APK's generated are version 1.0. I would like to unpack the APK, edit the version number, and then resign and repackage it. It was signed using my keystore originally, so I have the keys needed to sign it again. How can this be done?

  Read the article

• Generating changelogs from commit-comments - bad idea?

  - by Adam Asham

  I work for a small company (<10 developers) where we use Svn and to some extent Git. We have no release system implemented. We don't tag our applications with a release numbers so as you can imagine, we don't keep track of changes between releases very well. With this in mind I've been thinking about generating changelogs from the commit-comments. It doesn't have to be an awful idea if you implement guidelines for how to write comments. Or am I fooling myself thinking that colleagues will follow these guidelines? Most of them don't bother adding comments today but I'm hoping that will change if they become aware of how to formulate themselves and that you don't have to comment on every single commit. Or would we it be more wise to gather information from our bug/project managing software and write the changelogs manually (which unfortunately is something nobody wants to do)?

  Read the article

• Viewing array of pointers to structures in Visual Studio in the debugger

  - by Fozi

  I'm having a problem displaying the contents of a pointer array with its contents in the debugger. If I just add the pointer to the watch only the fist entry is visible. But if I add the length to it the debugger assumes that I have something like array[length][length]. Picture The first entry is the broken one, looks the same with ,2 or with any ,2 x. The second one is what I would like to see, but for more entries. Edit: The two entries displayed as [x][1] are invalid. I'm using VS 2005 but I think I had this problem on 2008 as well. Is this a bug or am I doing something wrong?

  Read the article

• Installing a condition handler in Common Lisp

  - by Paul Nathan

  The HTTP library Drakma on CLISP generates an error USOCKET:UNSUPPORTED due to a bug in Drakma+CLISP. However, it turns out that the CONTINUE restart seems to work fine. Therefore, I spent some time with CLtL and other references trying to determine how to write a restart handler. (defun http-request (url param) (handler-bind ((USOCKET:UNSUPPORTED #'(lambda (x) (invoke-restart 'continue))))) (drakma:http-request url :method :post :parameters param)) According to my best understanding, the above code should trap the error USOCKET:UNSUPPORTED. It doesn't; it seems to ignore the error binder. How do I fix this?

  Read the article

• Negative zero using Crystal Report ToText()

  - by Dan Ward

  Using Crystal Reports 8.5 on Windows Vista or 7, I'm using the ToText function to report a value: totext(Sum ({ap121w7.yrentamt}, {@type1099})*100,"000000000000000000") The result (if yrentamt is 0) is -000000000000000000 The dash (I assume it's a negative sign) is unneccessary and unwanted in my report. Is this a bug, or is there an easy solution? --Note-- I would very much like to avoid the following: if {ap121w7.yrentamt}=0.00 then yrentamt := "000000000000" else yrentamt := totext({ap121w7.yrentamt}*100,"000000000000"); I have about 100 files to fix with multiple formulas per file, and the above solution doesn't seem to work consistently either.

  Read the article

• Wordpress problem

  - by Joann

  Just recently, my client's blog http://robertnogueira.com/ encountered a bug. If you look at the site the footer has been displaced. It's been running fine until I tried to populate the posts for the 6th time with caff content. It got broke and I don't know how to fix it. I chose to ask help here because the folks here usually responds to my question unlike the wordpress forums. Please help me figure this out, I do this for a living. :(

  Read the article

< Previous Page | 578 579 580 581 582 583 584 585 586 587 588 589  | Next Page >