Search Results

Search found 20869 results on 835 pages for 'things i hate'.

Page 776/835 | < Previous Page | 772 773 774 775 776 777 778 779 780 781 782 783  | Next Page >

  • useful customize/enhanced php functions that make thing easy and better

    - by I Like PHP
    Hello All, i like to work in php bcoz it's just amazing language. please share basic, useful, enhanced and customize function that make things better and easy in php and must be used in our all PHP project, i m sharing some of them please share your customize function that may be useful for everyone alternative/ enhanced print_r() and var_dump() function watch( $what ) { echo '<pre>'; if ( is_array( $what ) ) { print_r ( $what ); } else { var_dump ( $what ); } echo '</pre>'; } usage: 1. watch($_POST); // to see all post variable 2. watch($array); // to see any variable may b array, string or a variable enhanced mysql_escape_string() for multidimensional array to prevent sql injection function recursive_escape(&$value) { if (is_array($value)) array_map('recursive_escape', $value); else $value = mysql_escape_string($value); } usage array_map('recursive_escape', $_POST); ---------------------For encoding Get variables-------------------------------------- function nkode($k) { if ( is_array( $k ) ) return array_map("base64_encode",$k); else return base64_encode($k); } ---------------------for decoding varaibles from GET--------------------------------- function dkode($k) { if ( is_array( $k ) ) return array_map("base64_decode",$k); else return base64_decode($k); } Usage <a href="somelink.php?pid=<?php echo nkode($someid)?>"> and on next page(somelink.php) $findID=dkode($_GET[pid]); date convert to mm/dd/yyyy to yyyy-mm-dd( if we use date datatype in mysql) and also change into mm/dd/yyyy to disply on page function dateconvert($date,$func) { if ($func == 1){ //insert conversion list($month, $day, $year) = split('[/.-]', $date); $date = "$year-$month-$day"; return $date; } if ($func == 2){ //output conversion list($year, $month, $day) = split('[-.]', $date); $date = "$month/$day/$year"; return $date; } } usage $firstDate=dateconvert($_POST['firstdate'],1); // for insertion in database $showDate=dateconvert($fetch->date_field,2) // to display on browser to clean(sql injection proof) data before doing some action with that variable function cleandata($data) { $success=0; $data=trim($data); $data=strtolower($data); $data=strip_tags($data); return $data; } usage cleandata($_POST[username]); cleandata($_GET[pid]); please share any basic function that must be used , Thanks

    Read the article

  • Reading Metadata property of GifBitmapDecoder...why is it null?

    - by David
    How can I read the delay, left and top offset data for each frame of a gif? I've gotten this far. Load the Gif var myGif = new GifBitmapDecoder(uri, BitmapCreateOptions.PreservePixelFormat, BitmapCacheOption.OnLoad); Get a frame var frame = myGif.Frames[i]; From MSDN: Native Image Format Metadata Queries read (ushort)Metadata.GetQuery("/grctlext/Delay"), (ushort)Metadata.GetQuery("/imgdesc/Left"), (ushort)Metadata.GetQuery("/imgdesc/Top") But two things don't work. First the Metadata property of both the gif and the frame are always null, even if I try different animated gif files. Second, the Metadata property of the frame doesn't seem to have a GetQuery method. How do I run these queries, what did I miss? Edit: Here is sample code that gives me null metadata. Using a fresh install of VS2010 Premium, on a fresh WPF application. The image file is the one in the comments. using System; using System.Collections.Generic; using System.Linq; using System.Text; using System.Windows; using System.Windows.Controls; using System.Windows.Data; using System.Windows.Documents; using System.Windows.Input; using System.Windows.Media; using System.Windows.Media.Imaging; using System.Windows.Navigation; using System.Windows.Shapes; namespace WpfApplication1 { /// <summary> /// Interaction logic for MainWindow.xaml /// </summary> public partial class MainWindow : Window { public MainWindow() { InitializeComponent(); var uri = new Uri(@"c:\b-414328-animated_gif_.gif"); var myGif = new GifBitmapDecoder(uri, BitmapCreateOptions.PreservePixelFormat, BitmapCacheOption.OnLoad); var frame = myGif.Frames[0]; Title = ""; Title += "Global Metadata is null: " + (myGif.Metadata == null).ToString(); Title += "; Frame Metadata is null: " + (frame.Metadata == null).ToString(); // Crash due to null metadata //var frameData = (BitmapMetadata)frame.Metadata; //var rate = (ushort)frameData.GetQuery("/grctlext/Delay"); } } }

    Read the article

  • iPhone reachability checking

    - by Sneakyness
    I've found several examples of code to do what I want (check for reachability), but none of it seems to be exact enough to be of use to me. I can't figure out why this doesn't want to play nice. I have the reachability.h/m in my project, I'm doing #import <SystemConfiguration/SystemConfiguration.h> And I have the framework added. I also have: #import "Reachability.h" at the top of the .m in which I'm trying to use the reachability. Reachability* reachability = [Reachability sharedReachability]; [reachability setHostName:@"http://www.google.com"]; // set your host name here NetworkStatus remoteHostStatus = [reachability remoteHostStatus]; if(remoteHostStatus == NotReachable) {NSLog(@"no");} else if (remoteHostStatus == ReachableViaWiFiNetwork) {NSLog(@"wifi"); } else if (remoteHostStatus == ReachableViaCarrierDataNetwork) {NSLog(@"cell"); } This is giving me all sorts of problems. What am I doing wrong? I'm an alright coder, I just have a hard time when it comes time to figure out what needs to be put where to enable what I want to do, regardless if I want to know what I want to do or not. (So frustrating) Update: This is what's going on. This is in my viewcontroller, which I have the #import <SystemConfiguration/SystemConfiguration.h> and #import "Reachability.h" set up with. This is my least favorite part of programming by far. FWIW, we never ended up implementing this in our code. The two features that required internet access (entering the sweepstakes, and buying the dvd), were not main features. Nothing else required internet access. Instead of adding more code, we just set the background of both internet views to a notice telling the users they must be connected to the internet to use this feature. It was in theme with the rest of the application's interface, and was done well/tastefully. They said nothing about it during the approval process, however we did get a personal phone call to verify that we were giving away items that actually pertained to the movie. According to their usually vague agreement, you aren't allowed to have sweepstakes otherwise. I would also think this adheres more strictly to their "only use things if you absolutely need them" ideaology as well. Here's the iTunes link to the application, EvoScanner.

    Read the article

  • PHP: How to implement a __get-like method for local function variables

    - by Tom Frost
    I'm no stranger to __get(), and have used it to make some very convenient libraries in the past. However, I'm faced with a new challenge (PHP 5.3, abbreviated and simplified my code for this question): <?php namespace test; class View { function __construct($filename, $varArray) { $this->filename = $filename; $this->varArray = $varArray; } function display() { include($this->filename); } function __get($varName) { if (isset($this->varArray[$varName])) return $this->varArray[$varName]; return "?? $varname ??"; } } ?> Above is a very, very simplified system for loading a View. This code would call the view and display it: <?php require_once("View.php"); use test\View; $view = new View("views/myview.php", array("user" => "Tom")); $view->display(); ?> My goal for this code is to allow the view "myview.php" to contain code like this: <p> Hello <?php echo $user; ?>! Your E-mail is <?php echo $email; ?> </p> And, used with the above code, this would output "Hello Tom! Your E-mail is ?? email ??" However, this won't work. The view is being included within a class method, so when it refers to $user and $email, it's looking for local function variables -- not variables belonging to the View class. For this reason, __get never gets triggered. I could change all my view's variables to things like $this-user and $this-email, but that would be a messy and unintuitive solution. I'd love to find a way where I can reference variables directly WITHOUT having PHP throw an error when an undefined variable is used. Thoughts? Is there a clean way to do this, or am I forced to resort to hacky solutions?

    Read the article

  • How do I apply a servlet filter when serving an HTML page directly?

    - by Philippe Beaudoin
    First off, I'm using Google AppEngine and Guice, but I suspect my problem is not related to these. When the user connect to my (GWT) webapp, the URL is a direct html page. For example, in development mode, it is: http://127.0.0.1:8888/Puzzlebazar.html?gwt.codesvr=127.0.0.1:9997. Now, I setup my web.xml in the following way: <?xml version="1.0" encoding="UTF-8"?> <web-app xmlns="http://java.sun.com/xml/ns/javaee" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://java.sun.com/xml/ns/javaee http://java.sun.com/xml/ns/javaee/web-app_2_5.xsd" version="2.5"> <display-name>PuzzleBazar</display-name> <!-- Default page to serve --> <welcome-file-list> <welcome-file>Puzzlebazar.html</welcome-file> </welcome-file-list> <filter> <filter-name>guiceFilter</filter-name> <filter-class>com.google.inject.servlet.GuiceFilter</filter-class> </filter> <filter-mapping> <filter-name>guiceFilter</filter-name> <url-pattern>/*</url-pattern> </filter-mapping> <!-- This Guice listener hijacks all further filters and servlets. Extra filters and servlets have to be configured in your ServletModule#configureServlets() by calling serve(String).with(Class<? extends HttpServlet>) and filter(String).through(Class<? extends Filter) --> <listener> <listener-class>com.puzzlebazar.server.guice.MyGuiceServletContextListener </listener-class> </listener> </web-app> Since I'm using Guice, I have to configure extra filters in my ServletModule, where I do this: filter("*.html").through( SecurityCookieFilter.class ); But my SecurityCookieFilter.doFilter is never called. I tried things like "*.html*" or <url-pattern>*</url-pattern> but to no avail. Any idea how I should do this?

    Read the article

  • Resize image while rotating image in android

    - by dhams
    Hi every one ,,m working with android project in which i want to rotate image along with touch to some fix pivot point ,,,i have completed all the things but i have facing one problem ,,,while m trying to rotate image the image bitmap is resize....i dont have any idea why it occur ....if somebody have den please give me idea to come over this problem.... my code is below ..... package com.demo.rotation; import android.app.Activity; import android.graphics.Bitmap; import android.graphics.BitmapFactory; import android.graphics.Matrix; import android.os.Bundle; import android.util.Log; import android.view.MotionEvent; import android.view.View; import android.view.View.OnTouchListener; import android.widget.ImageView; import android.widget.ImageView.ScaleType; public class temp extends Activity{ ImageView img1; float startX; float startX2 ; Bitmap source; Bitmap bitmap1 = null; double r; /** Called when the activity is first created. */ @Override public void onCreate(Bundle savedInstanceState) { super.onCreate(savedInstanceState); setContentView(R.layout.main); img1 = (ImageView) findViewById(R.id.img1); img1.setOnTouchListener(img1TouchListener); bitmap1 = BitmapFactory.decodeResource(getResources(), R.drawable.orsl_circle_transparent); } private OnTouchListener img1TouchListener = new OnTouchListener() { @Override public boolean onTouch(View v, MotionEvent event) { switch (event.getAction()) { case MotionEvent.ACTION_MOVE: Log.d("MOVE", "1"); if(source!=null) r = Math.atan2(event.getX() - source.getWidth(), (source.getHeight() / 2) - event.getY()); Log.i("startX" + event.getX(), "startY" + event.getY()); rotate(r,bitmap1, img1); img1.setScaleType(ScaleType.CENTER); break; case MotionEvent.ACTION_DOWN: break; case MotionEvent.ACTION_UP: break; default : break; } return true; } }; private void rotate(double r , Bitmap currentBitmap ,ImageView imageView ) { int rotation = (int) Math.toDegrees(r); Matrix matrix = new Matrix(); matrix.setRotate(rotation, currentBitmap.getWidth()/2, currentBitmap.getHeight()/2); source = Bitmap.createBitmap(currentBitmap, 0, 0, currentBitmap.getWidth(), currentBitmap.getHeight(), matrix, false); imageView.setImageBitmap(source); Log.i("HIGHT OF CURRENT BITMAP", ""+source.getHeight()); } }

    Read the article

  • What to Return? Error String, Bool with Error String Out, or Void with Exception

    - by Ranger Pretzel
    I spend most of my time in C# and am trying to figure out which is the best practice for handling an exception and cleanly return an error message from a called method back to the calling method. For example, here is some ActiveDirectory authentication code. Please imagine this Method as part of a Class (and not just a standalone function.) bool IsUserAuthenticated(string domain, string user, string pass, out errStr) { bool authentic = false; try { // Instantiate Directory Entry object DirectoryEntry entry = new DirectoryEntry("LDAP://" + domain, user, pass); // Force connection over network to authenticate object nativeObject = entry.NativeObject; // No exception thrown? We must be good, then. authentic = true; } catch (Exception e) { errStr = e.Message().ToString(); } return authentic; } The advantages of doing it this way are a clear YES or NO that you can embed right in your If-Then-Else statement. The downside is that it also requires the person using the method to supply a string to get the Error back (if any.) I guess I could overload this method with the same parameters minus the "out errStr", but ignoring the error seems like a bad idea since there can be many reasons for such a failure... Alternatively, I could write a method that returns an Error String (instead of using "out errStr") in which a returned empty string means that the user authenticated fine. string AuthenticateUser(string domain, string user, string pass) { string errStr = ""; try { // Instantiate Directory Entry object DirectoryEntry entry = new DirectoryEntry("LDAP://" + domain, user, pass); // Force connection over network to authenticate object nativeObject = entry.NativeObject; } catch (Exception e) { errStr = e.Message().ToString(); } return errStr; } But this seems like a "weak" way of doing things. Or should I just make my method "void" and just not handle the exception so that it gets passed back to the calling function? void AuthenticateUser(string domain, string user, string pass) { // Instantiate Directory Entry object DirectoryEntry entry = new DirectoryEntry("LDAP://" + domain, user, pass); // Force connection over network to authenticate object nativeObject = entry.NativeObject; } This seems the most sane to me (for some reason). Yet at the same time, the only real advantage of wrapping those 2 lines over just typing those 2 lines everywhere I need to authenticate is that I don't need to include the "LDAP://" string. The downside with this way of doing it is that the user has to put this method in a try-catch block. Thoughts? Is there another way of doing this that I'm not thinking of?

    Read the article

  • Twisted: why is it that passing a deferred callback to a deferred thread makes the thread blocking a

    - by surtyaarthoughts
    I unsuccessfully tried using txredis (the non blocking twisted api for redis) for a persisting message queue I'm trying to set up with a scrapy project I am working on. I found that although the client was not blocking, it became much slower than it could have been because what should have been one event in the reactor loop was split up into thousands of steps. So instead, I tried making use of redis-py (the regular blocking twisted api) and wrapping the call in a deferred thread. It works great, however I want to perform an inner deferred when I make a call to redis as I would like to set up connection pooling in attempts to speed things up further. Below is my interpretation of some sample code taken from the twisted docs for a deferred thread to illustrate my use case: #!/usr/bin/env python from twisted.internet import reactor,threads from twisted.internet.task import LoopingCall import time def main_loop(): print 'doing stuff in main loop.. do not block me!' def aBlockingRedisCall(): print 'doing lookup... this may take a while' time.sleep(10) return 'results from redis' def result(res): print res def main(): lc = LoopingCall(main_loop) lc.start(2) d = threads.deferToThread(aBlockingRedisCall) d.addCallback(result) reactor.run() if __name__=='__main__': main() And here is my alteration for connection pooling that makes the code in the deferred thread blocking : #!/usr/bin/env python from twisted.internet import reactor,defer from twisted.internet.task import LoopingCall import time def main_loop(): print 'doing stuff in main loop.. do not block me!' def aBlockingRedisCall(x): if x<5: #all connections are busy, try later print '%s is less than 5, get a redis client later' % x x+=1 d = defer.Deferred() d.addCallback(aBlockingRedisCall) reactor.callLater(1.0,d.callback,x) return d else: print 'got a redis client; doing lookup.. this may take a while' time.sleep(10) # this is now blocking.. any ideas? d = defer.Deferred() d.addCallback(gotFinalResult) d.callback(x) return d def gotFinalResult(x): return 'final result is %s' % x def result(res): print res def aBlockingMethod(): print 'going to sleep...' time.sleep(10) print 'woke up' def main(): lc = LoopingCall(main_loop) lc.start(2) d = defer.Deferred() d.addCallback(aBlockingRedisCall) d.addCallback(result) reactor.callInThread(d.callback, 1) reactor.run() if __name__=='__main__': main() So my question is, does anyone know why my alteration causes the deferred thread to be blocking and/or can anyone suggest a better solution?

    Read the article

  • Normalization of database for timesheet tool and ensure data integrity

    - by fireeyedboy
    I'm creating a timesheet application. I have the following entities (amongst others): Company Employee = an employee associated with a company Client = a client associated with a company So far I have the following (abbreviated) database setup: Company - id - name Employee - id - companyId (FK to Company.id) - name Client - id - companyId (FK to Company.id) - name Now, I want an employee to be associated with a client, but only if that client is associated with the company the employee works for. How would you guarantee this data integrity on a database level? Or should I just depend on the application to guarantee this data integrity? I thought about creating a many to many table like this: EmployeeClient - employeeId (FK to Employee.id) - companyId \ (combined FK to Client.companyId, Client.id) - clientId / Thus, when I insert a client for an employee along with the employee's company id, the database should prevent this when the client is not associated with the employee's company id. Does this make sense? Because this still doesn't guarantee the employee is associated with the company. How do you deal with these things? UPDATE The scenario is as followed: A company has multiple employees. Employees will only be linked to one company. A company has multiple clients also. Clients will only be linked to one company. (Company is a sandbox, so to speak). An employee of a company can be linked to a client of it's company, but only if the client is part of the company's clientele. In other words: The application will allow a company to create/add employees and create/add clients (hence the companyId FK in the Employee and Client tables). Next, the company will be allowed to assign certain clients to certain of it's employees (EmployeeClient table). Imagine an employee working on projects for a few clients for which s/he can write billable hours, but the employee must not be allowed to write billable hours for clients they are not assigned to by their employer (the company). So, employees will not automatically have access to all their company's clients, but only to those that the company has selected for them. Hopefully this has shed some more light on the matter.

    Read the article

  • do I need to close an audio Clip?

    - by Michael
    have an application that processes real-time data and is supposed to beep when a certain event occurs. The triggering event can occur multiple times per second, and if the beep is already playing when another event triggers the code is just supposed to ignore it (as opposed to interrupting the current beep and starting a new one). Here is the basic code: Clip clickClip public void prepareProcess() { super.prepareProcess(); clickClip = null; try { clipFile = new File("C:/WINDOWS/Media/CHIMES.wav"); ais = AudioSystem.getAudioInputStream(clipFile); clickClip = AudioSystem.getClip(); clickClip.open(ais); fileIsLoaded = true; } catch (Exception ex) { clickClip = null; fileIsLoaded = false; } } public void playSound() { if (fileIsLoaded) { if ((clickClip==null) || (!clickClip.isRunning())) { try { clickClip.setFramePosition(0); clickClip.start(); } catch (Exception ex) { System.out.println("Cannot play click noise"); ex.printStackTrace(); } } } The prepareProcess method gets run once in the beginning, and the playSound method is called every time a triggering event occurs. My question is: do I need to close the clickClip object? I know I could add an actionListener to monitor for a Stop event, but since the event occurs so frequently I'm worried the extra processing is going to slow down the real-time data collection. The code seems to run fine, but my worry is memory leaks. The code above is based on an example I found while searching the net, but the example used an actionListener to close the Clip specifically "to eliminate memory leaks that would occur when the stop method wasn't implemented". My program is intended to run for hours so any memory leaks I have will cause problems. I'll be honest: I have no idea how to verify whether or not I've got a problem. I'm using Netbeans, and running the memory profiler just gave me a huge list of things that I don't know how to read. This is supposed to be the simple part of the program, and I'm spending hours on it. Any help would be greatly appreciated! Michael

    Read the article

  • How can I refactor out needing so many for-loops in rails?

    - by Angela
    I need help refactoring this multi-loop thing. Here is what I have: Campaign has_many Contacts Campaign also has many templates for EVent (Email, Call, and Letter). I need a list of all the Emails, Calls and Letters that are "overdue" for every Contact that belongs to a Campaign. Overdue is determined by a from_today method which looks at the date the Contact was entered in the system and the number of days that needs to pass for any given Event. from_today() outputs the number of days from today that the Event should be done for a given Contact. Here is what I've done, it works for all Emails in a Campaign across all contacts. I was going to try to create another each do loop to change the class names. Wasn't sure where to begin: named_scope, push some things into a method, etcetera, or -- minimum to be able to dynamically change the class names so at least it loops three timees across the different events rather than repeating the code three times: <% @campaigns.each do |campaign| %> <h2><%= link_to campaign.name, campaign %></h2> <% @events.each do |event| %> <%= event %> <% for email in campaign.emails %> <h4><%= link_to email.title, email %> <%= email.days %> days</h4> <% for contact in campaign.contacts.find(:all, :order => "date_entered ASC" ) %> <% if (from_today(contact, email.days) < 0) %> <% if show_status(contact, email) == 'no status'%> <p> <%= full_name(contact) %> is <%= from_today(contact,email.days).abs%> days overdue: <%= do_event(contact, email) %> </p> <% end %> <% end %> <% end %> <% end %> <% end %> <% end %>

    Read the article

  • Zaypay alternatives for payments using call or sms

    - by JohannesH
    We are currently trying to implement a payment provider in zaypay for paying for services using sms or by calling a number. We already have google checkout and paypal working for regular payments but zaypay is rather inflexible, poorly documented and a pain to setup when you have hundreds of products with varying prices. So my question is, do you know of any other european payment providers that take sms and call payments? As a response to Roberts answer/question Hi Robert, I must say that the Zaypay solution is the best and only I've seen thus far regarding phoned payments. However, since its now 2 months ago I finished the implementation of our custom Zaypay UI I can't remember much of the the details of the problems we were having. I'll try to give a brief of them anyways the best I can. First of all I would like to see a redirection type scenario for payalogues. From what I remember you guys are using the JS framework "Prototype" which doesn't play nice with jQuery which we are using so we weren't able to use the popup-type scenario supported by payalogues. Furthermore when implementing our custom interface I remember a lot of missing translations, like words that were codes instead of a word or a phrase. This meant we ended up writing/translating all the messages we needed ourselves. Also, another point of annoyance was the setup of prices and items. I wish we could just send in the order items/prices as a part of the interface like you can in Google Checkout or PayPal (not that they're flawless either), instead of having to define ALL the items you will ever sell through your admin interface beforehand. As far as I can remember it is virtually impossible to use Zaypay for a multi-item order in its current form. Finally there are, as far as I can tell, some security issues that you have to think about when you implement a custom solution... especially a ajax driven one. As I said in my original post you do mention this in the documentation but I believe the documentation wasn't that comprehensive regarding security issues. Again I wish I could give more details but the code & client is long since gone, so I can't look up the comments I wrote. Sorry! Oh yeah, the general API documentation weren't exactly comprehensive and 100% correct either. Again, I don't want to advice people against using Zaypay, I just want to advice that they should try it out first on a realistic prototype and think about their implementation before releasing to production. Maybe its just me who misunderstood a lot of things but I generally had a difficult time using your framework and I was left with a feeling that the API was very new and not thought through from the beginning.

    Read the article

  • ImageMagick on Mac OSX Snow Leopard. Is there any way to get it to compile and run?

    - by ?????
    It seems that I have more trouble getting standard Unix things to run on Snow Leopard than any other platform--including Windows cygwin For the past couple of days, I've been trying to get ImageMagick to run on Snow Leopard. The most obvious way, Mac Ports, fails: tppllc-Mac-Pro:ImageMagick-sl swirsky$ sudo port install imagemagick ---> Computing dependencies for p5-locale-gettext ---> Configuring p5-locale-gettext Error: Target org.macports.configure returned: configure failure: shell command " cd "/opt/local/var/macports/build/_opt_local_var_macports_sources_rsync.macports.org_release_ports_perl_p5-locale-gettext/work/gettext-1.05" && /opt/local/bin/perl Makefile.PL INSTALLDIRS=vendor " returned error 2 Command output: checking for gettext... no checking for gettext in -I/opt/local/include -arch i386 -L/opt/local/lib -lintl...gettext function not found. Please install libintl at Makefile.PL line 18. no Error: Unable to upgrade port: 1 Error: Unable to execute port: upgrade xorg-libXt failed Before reporting a bug, first run the command again with the -d flag to get complete output. tppllc-Mac-Pro:ImageMagick-sl swirsky$ Not wanting to spend another two days figuring out why my libintl doesn't have a "gettext" function, I tried a different route: the script mentioned here: http://github.com/masterkain/ImageMagick-sl This script downloads and installs an ImageMagic independently of MacPorts issues tppllc-Mac-Pro:ImageMagick-sl swirsky$ /usr/local/bin/convert dyld: Library not loaded: /opt/local/lib/libiconv.2.dylib Referenced from: /opt/local/lib/libfontconfig.1.dylib Reason: Incompatible library version: libfontconfig.1.dylib requires version 8.0.0 or later, but libiconv.2.dylib provides version 7.0.0 Trace/BPT trap It downloads everything and compiles fine, but fails when I try to run it, with the message above. So now I'm two steps away from ImageMagick, trying to get a newer libiconv on my machine. I downloaded the latest libiconv, compiled and built it. I put the resulting library in /opt/local/lib, and I still get the same error message: tppllc-Mac-Pro:.libs swirsky$ sudo mv libiconv.2.dylib /opt/local/lib/libiconv.2.dylib tppllc-Mac-Pro:.libs swirsky$ convert dyld: Library not loaded: /opt/local/lib/libiconv.2.dylib Referenced from: /opt/local/lib/libfontconfig.1.dylib Reason: Incompatible library version: libfontconfig.1.dylib requires version 8.0.0 or later, but libiconv.2.dylib provides version 7.0.0 Trace/BPT trap Now here's something interesting. The error message shows it's looking in /opt/local/lib/libiconv.2.dylib. otools -L shows that this does implement 8.0.0: tppllc-Mac-Pro:.libs swirsky$ otool -L /opt/local/lib/libiconv.2.dylib /opt/local/lib/libiconv.2.dylib: /usr/local/lib/libiconv.2.dylib (compatibility version 8.0.0, current version 8.0.0) /usr/lib/libSystem.B.dylib (compatibility version 1.0.0, current version 125.0.0) tppllc-Mac-Pro:.libs swirsky$ And, for good measure, I set the DYLD_LIBRARY_PATH to make sure this directory is the one for dynamic libraries. So even though I do have a library that provides 8.0.0, it's being seen as 7.0.0! Any ideas why this would happen? So here's my question: Is it possible to get ImageMagick to run on OSX Snow Leopard? Are there any binary distributions that have static libraries baked in so I don't have to worry about these issue/

    Read the article

  • Impersonation on Windows 2000 to Windows XP Leaves Connections Open

    - by Tallek
    I'm running on a Windows 2000 Pro SP4 box (off domain) and trying to impersonate a local user on a Windows XP box (on domain). I'm using code very similar to the WindowsImpersonationContextFacade in the question posted here: http://stackoverflow.com/questions/879704/how-can-i-temporarily-impersonate-a-user-to-open-a-file. I am using impersonation to remotely start and stop windows services as well as access network shares (for some automated integration tests). To get this working, i had to use LOGON32_PROVIDER_DEFAULT and LOGON32_LOGON_NEW_CREDENTIALS when calling LogonUser. Everything worked beautifully ( Windows XP on domain to Windows XP on domain, Windows XP on domain to Windows Server 2003 off domain, and even Windows XP on domain to Windows 2000 off domain). The one issue was running on Windows 2000 Pro SP4 off the domain and trying to impersonate a local user on a Windows XP box running on the domain. To get the Windows 2000 piece working, i had to use LOGON32_PROVIDER_WINNT50 and LOGON32_LOGON_NEW_CREDENTIALS when calling LogonUser. This seemed to get me 95% of the way there, i could now impersonate the local user on the XP box and start/stop services as well as access a network share using the impersonated credentials. I'm running in to one problem though, calling Undo impersonation and closing the token handle seems to leave the connection to the remote box open. After about 10 or so impersonation calls, further impersonation attempts will fail with an error saying something about too many connections are currently open. If i look at the Computer Management - System Tools - Shared Folders - Sessions on my remote Windows XP box, i can see about 10 sessions open to the Windows 2000 box. I can manually close these (i think they may eventually close themselves, but not very quickly) and then impersonation begins working again few more times. This open session issue doesn't seem to be a problem in any of my other test scenarios, just when running locally on a Windows 2000 box. Any ideas? Edit 1: After some more testing and trying out many different things, this seems to be an issue with open sessions not being reused. On Windows 2000 only, every call to LogonUser to get a token and then using that token to impersonate seems to result in a new session being created. I'm guessing Windows XP & Windows Server 2003 are reusing open sessions since i don't seem to be having any issues with them. If I call LogonUser once, then cache the token, I seem to be able to make as many calls to impersonate as I need using the cached token without running in to the "too many connections" issue. This seems like an ugly work around though since i can't call CloseHandle() on my token every time i perform impersonation. Anybody have any thoughts or ideas, or am i stuck with this ugly hack? Thanks

    Read the article

  • Basic Ruby On Rails Linking Help

    - by dweebsonduty
    So I am beginning to work with Rails and I get some of the concepts but am stuck on an important one. Let's say I have customers which has many jobs and jobs which belongs to customers. How would I go about creating a new job for a customer? I can create a link that goes to customers/1/jobs/new and I can grab the customer ID but how do I tell it that I am creating a job for customer 1? I know this is the most basic of things but I just need a push in the right direction. This is my form so far: How do I get :customer_id to populate with the customer_id param? <h1>New job</h1> <% form_for(@job) do |f| %> <%= f.error_messages %> <p> <%= f.label :customer_id %><br /> <%= f.text_field :customer_id %> </p> <p> <%= f.label :manufacturer %><br /> <%= f.text_field :manufacturer %> </p> <p> <%= f.label :serial_number %><br /> <%= f.text_field :serial_number %> </p> <p> <%= f.label :problem %><br /> <%= f.text_area :problem %> </p> <p> <%= f.label :notes %><br /> <%= f.text_area :notes %> </p> <p> <%= f.label :status %><br /> <%= f.text_field :status %> </p> <p> <%= f.label :tech_id %><br /> <%= f.text_field :tech_id %> </p> <p> <%= f.submit 'Create' %> </p> <% end %> <%= link_to 'Back', jobs_path %>

    Read the article

  • Django admin fails when using includes in urlpatterns

    - by zenWeasel
    I am trying to refactor out my application a little bit to keep it from getting too unwieldily. So I started to move some of the urlpatterns out to sub files as the documentation proposes. Besides that fact that it just doesn't seem to be working (the items are not being rerouted) but when I go to the admin, it says that 'urlpatterns has not been defined'. The urls.py I have at the root of my application is: if settings.ENABLE_SSL: urlpatterns = patterns('', (r'^checkout/orderform/onepage/(\w*)/$','checkout.views.one_page_orderform',{'SSL':True},'commerce.checkout.views.single_product_orderform'), ) else: urlpatterns = patterns('', (r'^checkout/orderform/onepage/(\w*)/$','commerce.checkout.views.single_product_orderform'), ) urlpatterns+= patterns('', (r'^$', 'alchemysites.views.route_to_home'), (r'^%s/' % settings.DAJAXICE_MEDIA_PREFIX, include('dajaxice.urls')), (r'^/checkout/', include('commerce.urls')), (r'^/offers',include('commerce.urls')), (r'^/order/',include('commerce.urls')), (r'^admin/', include(admin.site.urls)), (r'^accounts/login/$', login), (r'^accounts/logout/$', logout), (r'^(?P<path>.*)/$','alchemysites.views.get_path'), (r'^static/(?P<path>.*)$', 'django.views.static.serve', {'document_root':settings.MEDIA_ROOT}), The urls I have moved out so far are the checkout/offers/order which are all subapps of 'commerce' where the urls.py for the apps are so to be clear. /urls.py in questions (included here) /commerce/urls.py where the urls.py I want to include is: order_info = { 'queryset': Order.objects.all(), } urlpatterns+= patterns('', (r'^offers/$','offers.views.start_offers'), (r'^offers/([a-zA-Z0-9-]*)/order/(\d*)/add/([a-zA-Z0-9-]*)/(\w*)/next/([a-zA-Z0-9-)/$','offers.views.show_offer'), (r'^reports/orders/$', list_detail.object_list,order_info), ) and the applications offers lies under commerce. And so the additional problem is that admin will not work at all, so I'm thinking because I killed it somewhere with my includes. Things I have checked for: Is the urlpatterns variable accidentally getting reset somewhere (i.e. urlpatterns = patterns, instead of urlpatterns+= patterns) Are the patterns in commerce.urls valid (yes, when moved back to root they work). So from there I am stumped. I can move everything back into the root, but was trying to get a little decoupled, not just for theoretical reason but for some short terms ones. Lastly if I enter www.domainname/checkout/orderform/onepage/xxxjsd I get the correct page. However, entering www.domainname/checkout/ gets handled by the alchemysites.views.get_path. If not the answer (because this is pretty darn specific), then is there a good way for troubleshoot urls.py? It seems to just be trial and error. Seems there should be some sort of parser that will tell you what your urlpatterns will do.

    Read the article

  • How do we, as a community, help encourage programming in public schools? (Or state Schools for the U

    - by NoMoreZealots
    PRIMARY MOTIVATION My office gets involved with the "First Robotics" competitions and one thing that lingers year to year is the students typically have no preparation for doing even simple programming as part of the public schools system. While the science classes provide some basic grasp of mechanical and electrical concepts, by in large computer programming gets no coverage from the curriculum. (This my be different in other areas of the country/world.) What makes it worse is there is only a short period of time you have to prepare the student's and help them design the robot. Talking to some professors from local colleges, it's a problem because you can't assume even the most basic understanding for freshman CS majors. Languages like Python, Lua and BASIC are simple enough for at least high school level students, if not younger. SCOPE So how do you get public schools to support a programming, at least to the level of "Try it in BASIC" examples that used to be at the end of a chapter in my Algebra book? At least enough to prepare them for event's such as the FIRST Robotic competitions. Which the primary objectives are to teach problem solving and team work, and to possible foster an interest in Math, Science and Engineering in general. (Not force feed to them, as some people her seem to be implying.) Edit: Why teach kids: (Since 2000 CS enrollment in US colleges has decreased by 70% while college enrollment has increased, this is a PROBLEM.) Saying there is no value in teaching someone programming in Jr./High school because they might think "they know programming." Is like saying there's no value in teaching High school science and physics, because they might decide they "know physics." Leading to abuse like: "I passed a high school physics class, I'm going to develop a Unified Quantum Gravitational Theory." Better Prepared students are better students. Instead it would allows college programs to raise the bar on the entry level courses, allowing students to be weeded out based on their understanding of more advanced material. Plus people who did poorly in that in topic in High school aren't as likely to say "I think there's money in computer's so I'll computer science." Plus if people take it in high school and decide THEN that it's not for them, it's better than them wasting their money to PAY a college to figure that out. The result is that people who take the degree are more likely to succeed and be there for the RIGHT reasons. (i.e. It's what they REALLY want to do. And that's REALLY the key to being good at anything.) Programming is like anything else, the more practice and genuine interest you have the better you get. If you start them later, they get less practice. The earlier give them the opportunity to start, the more practice they will get. All other things equal, the more practice the better the programmer.

    Read the article

  • Direct show video renderers suck?

    - by Daniel
    So I've been looking into the world of media playback for windows and I've started making a C# Media Player using DirectShow. I started off using the VRM-7 windowed video renderer and it was brilliant except it had a couple of small problems (multi monitors, fullscreen). But after some research I found that it's deprecated and I should be using VRM9. So I changed it to use VRM9 windowless then found out that was an old post rofl _< so finally I'm using Vista/Win7 (or XP .net 3) Enhanced Video Renderer (EVR) which is apparently the most up to date Microsoft video renderer and has all the flashy performance/quality things added to it. (tbh I haven't noticed any difference but maybe I need a blue-ray or HQ video to notice it). With using EVR everything is working fine except resizing the video. Its really laggy/choppy/teary and problem something to do with its frame queueing mechanism. To demonstrate my problem open up windows media player classic. View - Options - Playback - output Chose the "EVR" DirectShow Video renderer Now restart wmp class and play a video, while it's playing click and drag a corner to resize it. You'll notice its horribly laggy. This is the exact same problem i am having. But if you chose "EVR Custom Pres. *" or EVR Sync *" resizing works beautifully! So i tried googling around for anything about EVR resizing issues and how to fix it but i couldn't believe how little i could find. I'm guessing "Custom Pres." stands for "Custom Presenter" which sounds like they made their own. Also you'll notice on the right hand size when you swap between EVR and the other EVR's the Resizer drop down on the right greys out. So basically I wan't to know how I can fix this retarded resizing problem and is there any decent documentation out there? There is a fair bit for VMR7/9 but not much for EVR. I downloaded the DirectX SDK which apparently has samples but it was a waste of 500mb of bandwidth as it had nothing relevant. Perhaps there is some way to force it not queueing up frames if that is the problem? If you want code say the word and I'll paste some in. But it's really quite simple and nothing much happens, i'm convinced it's a problem with the EVR renderer. EDIT: Oh and one other thing, what does VLC use? If you go into vlc options and change the renderer to anything but default, they all suck. So is it using VMR7? Or its own?

    Read the article

  • DirectShow EVR resizing window problem

    - by Daniel
    So I've been looking into the world of media playback for windows and I've started making a C# Media Player using DirectShow. I started off using the VRM-7 windowed video renderer and it was brilliant except it had a couple of small problems (multi monitors, fullscreen). But after some research I found that it's deprecated and I should be using VRM9. So I changed it to use VRM9 windowless then found out that was an old post rofl _< so finally I'm using Vista/Win7 (or XP .net 3) Enhanced Video Renderer (EVR) which is apparently the most up to date Microsoft video renderer and has all the flashy performance/quality things added to it. (tbh I haven't noticed any difference but maybe I need a blue-ray or HQ video to notice it). With using EVR everything is working fine except resizing the video. Its really laggy/choppy/teary and probably something to do with its frame queueing mechanism. To demonstrate my problem open up windows media player classic. View - Options - Playback - output Chose the "EVR" DirectShow Video renderer Now restart wmp class and play a video, while it's playing click and drag a corner to resize it. You'll notice its horribly laggy. This is the exact same problem i am having. But if you chose "EVR Custom Pres. *" or EVR Sync *" resizing works beautifully! So i tried googling around for anything about EVR resizing issues and how to fix it but i couldn't believe how little i could find. I'm guessing "Custom Pres." stands for "Custom Presenter" which sounds like they made their own. Also you'll notice on the right hand size when you swap between EVR and the other EVR's the Resizer drop down on the right greys out. So basically I wan't to know how I can fix this retarded resizing problem and is there any decent documentation out there? There is a fair bit for VMR7/9 but not much for EVR. I downloaded the DirectX SDK which apparently has samples but it was a waste of 500mb of bandwidth as it had nothing relevant. Perhaps there is some way to force it not queueing up frames if that is the problem? If you want code say the word and I'll paste some in. But it's really quite simple and nothing much happens, i'm convinced it's a problem with the EVR renderer. EDIT: Oh and one other thing, what does VLC use? If you go into vlc options and change the renderer to anything but default, they all suck. So is it using VMR7? Or its own?

    Read the article

  • compiling a program to run in DOS mode

    - by dygi
    I write a simple program, to run in DOS mode. Everything works under emulated console in Win XP / Vista / Seven, but not in DOS. The error says: this program caonnot be run in DOS mode. I wonder is that a problem with compiler flags or something bigger. For programming i use Code::Blocks v 8.02 with such settings for compilation: -Wall -W -pedantic -pedantic-errors in Project \ Build options \ Compiler settings I've tried a clean DOS mode, booting from cd, and also setting up DOS in Virtual Machine. The same error appears. Should i turn on some more compiler flags ? Some specific 386 / 486 optimizations ? UPDATE Ok, i've downloaded, installed and configured DJGPP. Even resolved some problems with libs and includes. Still have two questions. 1) i can't compile a code, which calls _strdate and _strtime, i've double checked the includes, as MSDN says it needs time.h, but still error says: _strdate was not declared in this scope, i even tried to add std::_strdate, but then i have 4, not 2 errors sazing the same 2) the 2nd code is about gotoxy, it looks like that: #include <windows.h> void gotoxy(int x, int y) { COORD position; position.X = x; position.Y = y; SetConsoleCursorPosition(GetStdHandle(STD_OUTPUT_HANDLE), position); } error says there is no windows.h, so i've put it in place, but then there are many more errors saying some is missing from windows.h, I SUPPOSE it won't work because this functions is strictly for windows right ? is there any way to write similar gotoxy for DOS ? UPDATE2 1) solved using time(); instead of _strdate(); and _strtime(); here's the code time_t rawtime; struct tm * timeinfo; char buffer [20]; time ( &rawtime ); timeinfo = localtime ( &rawtime ); strftime (buffer,80,"%Y.%m.%d %H:%M:%S\0",timeinfo); string myTime(buffer); It now compiles under DJGPP. UPDATE3 Still need to solve a code using gotoxy - replaced it with some other code that compiles (under DJGPP). Thank You all for help. Just learnt some new things about compiling (flags, old IDE's like DJGPP, OpenWatcom) and refreshed memories setting DOS to work :--)

    Read the article

  • Stuck on an ASP.NET/WCF WSDL Parsing Error

    - by Vaccano
    I have a WCF Web Service that my ASP.NET app uses. It has been working fine for quite some time. I just added in a Dev Express Grid (and the Dev Express DLLs) and a new page that uses them and now I am getting parsing errors on the WSDL. But the weird part is that it works fine on my machine but fails on the web server machine. (Both are connecting to the same web services WSDL.) Here is the error message I am getting: Server Error in '/MyWebAppWebDev' Application. -------------------------------------------------------------------------------- Parser Error Description: An error occurred during the parsing of a resource required to service this request. Please review the following specific parse error details and modify your source file appropriately. Parser Error Message: Reference.svcmap: Failed to generate code for the service reference 'MyWebAppService'. Cannot import wsdl:portType Detail: An exception was thrown while running a WSDL import extension: System.ServiceModel.Description.DataContractSerializerMessageContractImporter Error: Referenced type 'WebClientApp.MyWebAppService.ReferenceUpdatesDataContract, WebClientApp, Version=1.0.0.0, Culture=neutral, PublicKeyToken=null' with data contract name 'ReferenceUpdatesDataContract' in namespace 'http://schemas.datacontract.org/2004/07/MyWebAppServiceLibrary.DataContracts' cannot be used since it does not match imported DataContract. Need to exclude this type from referenced types. XPath to Error Source: //wsdl:definitions[@targetNamespace='http://tempuri.org/']/wsdl:portType[@name='IMyWebAppReferenceDataServiceLib'] Cannot import wsdl:binding Detail: There was an error importing a wsdl:portType that the wsdl:binding is dependent on. XPath to wsdl:portType: //wsdl:definitions[@targetNamespace='http://tempuri.org/']/wsdl:portType[@name='IMyWebAppReferenceDataServiceLib'] XPath to Error Source: //wsdl:definitions[@targetNamespace='http://tempuri.org/']/wsdl:binding[@name='MyWebAppServicesDefaultEndpoint'] Cannot import wsdl:port Detail: There was an error importing a wsdl:binding that the wsdl:port is dependent on. XPath to wsdl:binding: //wsdl:definitions[@targetNamespace='http://tempuri.org/']/wsdl:binding[@name='MyWebAppServicesDefaultEndpoint'] XPath to Error Source: //wsdl:definitions[@targetNamespace='http://tempuri.org/']/wsdl:service[@name='MyWebAppReferenceDataServiceLib']/wsdl:port[@name='MyWebAppServicesDefaultEndpoint'] Source Error: [No relevant source lines] Source File: /MyWebAppWebDev/App_WebReferences/MyWebAppService/ Line: 1 I am completely stumped on this. I have checked my web.config endpoint address and it is spot on (and notably is not in the error message above). Any ideas would be welcomed. Things I have tried: Giving permissions to C:\Windows\temp to my Website user name Giving permissions to C:\Windows\temp to my App pool user name Checking to see that none of my data contracts are generic and have IsReference=true in them.

    Read the article

  • jQuery: Counter, Tricky problem with effects for brainy people.

    - by Marius
    Hello there! I made this counter that I think is cool because it only makes visible changes to the numbers being changed between each time it is triggered. This is the code // counter $('a').click(function(){ var u = ''; var newStr = ''; $('.counter').each(function(){ var len = $(this).children('span').length; $(this).children('span').each(function(){ u = u + $(this).text(); }); v = parseInt(u) + 1; v = v + ''; for (i=v.length - 1; i >= 0; i--) { if (v.charAt(i) == u.charAt(i)) { break; } } slce = len - (v.length - (i + 1)) updates = $(this).children('span').slice(slce); $(updates).fadeTo(100,0).delay(100).each(function(index){ f = i + 1 + index; $(this).text(v.charAt(f)); }).fadeTo(100,1); }); }); Markup: <span class="counter"> <span></span><span></span><span></span><span></span><span></span><span></span><span style="margin-right:4px;">9</span><span>9</span><span>9</span><span>9</span> </span> <a href="">Add + 1</a> The problem is that I previously used queue() function to delay() $(this).text(v.charAt(f)); by 100ms, (without queue the text()-function would not be delayed because it isnt in the fx catergory) so that the text would be updated before the object had faded to opacity = 0. That would look stupid. When adding multiple counters, only one of the counters would count. When removing the queue function, both counters would work, but as you can imagine, the delay of the text() would be gone (as it isnt fx-category). It is probably a bit tricky to make out how I can have multiple counters, and still delay the text()-function by 100ms, but I was hoping there is somebody here with spare brain capacity for these things ;) You can see a (NSFW) problem demo here: Just look underneath the sharing icons and you will notice that the text changes WHILE the objects fade out. Looking for some help to sove this problem. I would like to call the text() function once the text has faded to opacity 0, then fade in once the text() has executed. Thank you for your time.

    Read the article

  • SQL efficiency argument, add a column or solvable by query?

    - by theTurk
    I am a recent college graduate and a new hire for software development. Things have been a little slow lately so I was given a db task. My db skills are limited to pet projects with Rails and Django. So, I was a little surprised with my latest task. I have been asked by my manager to subclass Person with a 'Parent' table and add a reference to their custodian in the Person table. This is to facilitate going from Parent to Form when the custodian, not the Parent, is the FormContact. Here is a simplified, mock structure of a sql-db I am working with. I would have drawn the relationship tables if I had access to Visio. We have a table 'Person' and we have a table 'Form'. There is a table, 'FormContact', that relates a Person to a Form, not all Persons are related to a Form. There is a relationship table for Person to Person relationships (Employer, Parent, etc.) I've asked, "Why this couldn't be handled by a query?" Response, Inefficient. (Really!?!) So, I ask, "Why not have a reference to the Form? That would be more efficient since you wouldn't be querying the FormContacts table with the reference from child/custodian." Response, this would essentially make the Parent is a FormContact. (Fair enough.) I went ahead an wrote a query to get from non-FormContact Parent to Form, and tested on the production server. The response time was instantaneous. *SOME_VALUE* is the Parent's fk ID. SELECT FormID FROM FormContact WHERE FormContact.ContactID IN (SELECT SourceContactID FROM ContactRelationship WHERE (ContactRelationship.RelatedContactID = *SOME_VALUE*) AND (ContactRelationship.Relationship = 'Parent')); If I am right, "This is an unnecessary change." What should I do, defend my position or should I concede to the managers request? If I am wrong. What is my error? Is there a better solution than the manager's?

    Read the article

  • JavaScript Resource Management Design Pattern

    - by Adam
    As a web developer, a common problem I find myself tackling is waiting for something to load before doing something else. In particular, I often hide (using either display: none; or visibility: hidden; depending on the situation) elements while waiting for a background image or a CSS file to load. Consider this example from Last.FM. They overlay a semi-transparant PNG over each album art image so that it looks like it's inside a jewel-case. They let it load when it loads, so depending on your internet speed, you may see the art image by itself (without the overlay) temporarily. In this case, the album art looks fine without the jewel-case effect. But in similar situations, I have found that I don't want the user to see the site's design mangled as resources incrementally load. So, in rare cases I have hidden everything from the user until the whole kit and kaboodle has loaded. But this is often a pain to write out, and may force the user to wait for a pretty long time to see anything (besides "loading..." text). I can think of (and have used on occasion) some obvious solutions/compromises: Use some inline CSS so that as certain parts of the DOM load and render, they will immediately have the correct size/position/etc. Immediately render the navigation part of the site, so that if the user wanted to use the current page purely to get somewhere else, they don't have to wait for the rest to load. Load pixelated images first as placeholders for layout while lazy-loading higher quality images as replacements. Something quirky like using a cute animated gif to distract the user during a "loading..." phase. Show useful information as a reference while loading the full UI. (Something akin to Gmail Inbox Preview, etc.) (Sorry if my question was basically just asked and answered...) Despite all of these ideas, I still find myself hoping there are better ways of doing some of these things. So I guess what I'm looking for is some inspiration and/or any creative ways of dealing with this problem that you guys may have seen out in the wild.

    Read the article

  • how to implement a "soft barrier" in multithreaded c++

    - by Jason
    I have some multithreaded c++ code with the following structure: do_thread_specific_work(); update_shared_variables(); //checkpoint A do_thread_specific_work_not_modifying_shared_variables(); //checkpoint B do_thread_specific_work_requiring_all_threads_have_updated_shared_variables(); What follows checkpoint B is work that could have started if all threads have reached only checkpoint A, hence my notion of a "soft barrier". Typically multithreading libraries only provide "hard barriers" in which all threads must reach some point before any can continue. Obviously a hard barrier could be used at checkpoint B. Using a soft barrier can lead to better execution time, especially since the work between checkpoints A and B may not be load-balanced between the threads (i.e. 1 slow thread who has reached checkpoint A but not B could be causing all the others to wait at the barrier just before checkpoint B). I've tried using atomics to synchronize things and I know with 100% certainty that is it NOT guaranteed to work. For example using openmp syntax, before the parallel section start with: shared_thread_counter = num_threads; //known at compile time #pragma omp flush Then at checkpoint A: #pragma omp atomic shared_thread_counter--; Then at checkpoint B (using polling): #pragma omp flush while (shared_thread_counter > 0) { usleep(1); //can be removed, but better to limit memory bandwidth #pragma omp flush } I've designed some experiments in which I use an atomic to indicate that some operation before it is finished. The experiment would work with 2 threads most of the time but consistently fail when I have lots of threads (like 20 or 30). I suspect this is because of the caching structure of modern CPUs. Even if one thread updates some other value before doing the atomic decrement, it is not guaranteed to be read by another thread in that order. Consider the case when the other value is a cache miss and the atomic decrement is a cache hit. So back to my question, how to CORRECTLY implement this "soft barrier"? Is there any built-in feature that guarantees such functionality? I'd prefer openmp but I'm familiar with most of the other common multithreading libraries. As a workaround right now, I'm using a hard barrier at checkpoint B and I've restructured my code to make the work between checkpoint A and B automatically load-balancing between the threads (which has been rather difficult at times). Thanks for any advice/insight :)

    Read the article

< Previous Page | 772 773 774 775 776 777 778 779 780 781 782 783  | Next Page >