Search Results

Search found 1673 results on 67 pages for 'hack augusto'.

Page 62/67 | < Previous Page | 58 59 60 61 62 63 64 65 66 67  | Next Page >

• Using FFMPEG to reliably convert videos to mp4 for iphone/ipod and flash players

  - by Jake Stevenson

  I need to convert videos for use in both a flash player and the iphone/ipod touch. I'm using the following batch script with ffmpeg: @echo off ffmpeg.exe -i %1 -s qvga -acodec libfaac -ar 22050 -ab 128k -vcodec libx264 -threads 0 -f ipod %2 This always outputs an mp4 file, and I can always play it on my PC. The videos also seem to play fine on my iphone 3GS. But with some input files it won't work for older iphone versions (3G and iPod touch). Here's the ffmpeg output from one such file: D:\ffmpeg>encode.bat d:\temp\recording.flv d:\temp\out.m4v FFmpeg version SVN-r18709, Copyright (c) 2000-2009 Fabrice Bellard, et al. configuration: --enable-memalign-hack --prefix=/mingw --cross-prefix=i686-ming w32- --cc=ccache-i686-mingw32-gcc --target-os=mingw32 --arch=i686 --cpu=i686 --e nable-avisynth --enable-gpl --enable-zlib --enable-bzlib --enable-libgsm --enabl e-libfaac --enable-libfaad --enable-pthreads --enable-libvorbis --enable-libtheo ra --enable-libspeex --enable-libmp3lame --enable-libopenjpeg --enable-libxvid - -enable-libschroedinger --enable-libx264 libavutil 50. 3. 0 / 50. 3. 0 libavcodec 52.27. 0 / 52.27. 0 libavformat 52.32. 0 / 52.32. 0 libavdevice 52. 2. 0 / 52. 2. 0 libswscale 0. 7. 1 / 0. 7. 1 built on Apr 28 2009 04:04:42, gcc: 4.2.4 [flv @ 0x187d650]skipping flv packet: type 18, size 164, flags 0 Input #0, flv, from 'd:\temp\recording.flv': Duration: 00:00:07.17, start: 0.001000, bitrate: N/A Stream #0.0: Video: flv, yuv420p, 320x240, 1k tbr, 1k tbn, 1k tbc Stream #0.1: Audio: nellymoser, 44100 Hz, mono, s16 [libx264 @ 0x13518b0]using cpu capabilities: MMX2 SSE2Fast SSSE3 FastShuffle SSE 4.2 [libx264 @ 0x13518b0]profile Baseline, level 4.2 Output #0, ipod, to 'd:\temp\out.m4v': Stream #0.0: Video: libx264, yuv420p, 320x240, q=2-31, 200 kb/s, 1k tbn, 1k tbc Stream #0.1: Audio: libfaac, 22050 Hz, mono, s16, 128 kb/s Stream mapping: Stream #0.0 -> #0.0 Stream #0.1 -> #0.1 Press [q] to stop encoding frame= 90 fps= 0 q=-1.0 Lsize= 128kB time=6.87 bitrate= 152.4kbits/s video:92kB audio:32kB global headers:1kB muxing overhead 2.620892% [libx264 @ 0x13518b0]slice I:8 Avg QP:29.62 size: 7047 [libx264 @ 0x13518b0]slice P:82 Avg QP:30.83 size: 467 [libx264 @ 0x13518b0]mb I I16..4: 17.9% 0.0% 82.1% [libx264 @ 0x13518b0]mb P I16..4: 0.6% 0.0% 0.0% P16..4: 23.1% 0.0% 0.0% 0.0% 0.0% skip:76.3% [libx264 @ 0x13518b0]final ratefactor: 57.50 [libx264 @ 0x13518b0]SSIM Mean Y:0.9544735 [libx264 @ 0x13518b0]kb/s:8412.6 My suspicion is that it has something to do with the audio encoding. If so, does anyone know how to force it to reencode the audio to the proper format? Any other ideas?

  Read the article

• When -exactly- does the Rails3 application get initialized?

  - by bergyman

  I've been fighting left and right with rails 3 and bundler. There are a few gems out there that don't work properly if the rails application hasn't been loaded yet. factory_girl and shoulda are both examples, even on the rails3 branch. Taking shoulda as an example, when trying to run rake test:units I get the following error: DEPRECATION WARNING: RAILS_ROOT is deprecated! Use Rails.root instead. (called from autoload_macros at c:/code/test_harness/vendor/windows_gems/gems/shoulda-2.10.3/lib/shoulda/autoload_macros.rb:40) c:/code/test_harness/vendor/windows_gems/gems/shoulda-2.10.3/lib/shoulda/autoload_macros.rb:44:in 'join': can't convert #<Class:0x232b7c0> into String (TypeError) from c:/code/test_harness/vendor/windows_gems/gems/shoulda-2.10.3/lib/shoulda/autoload_macros.rb:44:in 'block in autoload_macros' from c:/code/test_harness/vendor/windows_gems/gems/shoulda-2.10.3/lib/shoulda/autoload_macros.rb:44:in 'map' from c:/code/test_harness/vendor/windows_gems/gems/shoulda-2.10.3/lib/shoulda/autoload_macros.rb:44:in 'autoload_macros' from c:/code/test_harness/vendor/windows_gems/gems/shoulda-2.10.3/lib/shoulda/rails.rb:17:in '<top (required)>' Digging a bit deeper into lib/shoulda/rails, I see this: root = if defined?(Rails.root) && Rails.root Rails.root else RAILS_ROOT end # load in the 3rd party macros from vendorized plugins and gems Shoulda.autoload_macros root, File.join("vendor", "{plugins,gems}", "*") So...what's happening here is while Rails.root is defined, Rails.root == nil, so RAILS_ROOT is used, and RAILS_ROOT==nil, which is then being passed on to Shoulda.autoload_macros. Obviously the rails app has yet to be initialized. With Rails3 using Bundler now, there's been some hubub over on the Bundler side about being able to specify an order in which the gems are required, but I'm not sure whether or not this would solve the problem at hand. Ultimately my questions is this: When exactly does the environment.rb file (which actually initializes the application) get pulled in? Is there any harm to bumping up when the app is initialized and have it happen before the Bundler.require line in config/application.rb? I've tried to hack bundler to specify the order myself, and have the rails gem pulled in first, but it doesn't appear to me that requiring the rails gem actually initializes the application. As this line (in config/application.rb) is being called before the app is initialized, any gem in the bundler Gemfile that requires rails to be initialized is going to tank. # Auto-require default libraries and those for the current Rails environment. Bundler.require :default, Rails.env

  Read the article

• jQueryUI autocomplete - when no results are returned

  - by Brian M. Hunt

  I'm wondering how one can catch and add a custom handler when empty results are returned from the server when using jQueryUI autocomplete. There seem to be a few questions on this point related to the various jQuery plugins (e.g. jQuery autocomplete display “No data” error message when results empty), but I am wondering if there's a better/simpler way to achieve the same with the jQueryUI autocomplete. It seems to me this is a common use case, and I thought perhaps that jQueryUI had improved on the jQuery autocomplete by adding the ability to cleanly handle this situation. However I've not been able to find documentation of such functionality, and before I hack away at it I'd like to throw out some feelers in case others have seen this before. While probably not particularly influential, I can have the server return anything - e.g. HTTP 204: No Content to a 200/JSON empty list - whatever makes it easiest to catch the result in jQueryUI's autocomplete. My first thought is to pass a callback with two arguments, namely a request object and a response callback to handle the code, per the documentation: The third variation, the callback, provides the most flexibility, and can be used to connect any data source to Autocomplete. The callback gets two arguments: A request object, with a single property called "term", which refers to the value currently in the text input. For example, when the user entered "new yo" in a city field, the Autocomplete term will equal "new yo". A response callback, which expects a single argument to contain the data to suggest to the user. This data should be filtered based on the provided term, and can be in any of the formats described above for simple local data (String-Array or Object-Array with label/value/both properties). When the response callback receives no data, it inserts returns a special one-line object-array that has a label and an indicator that there's no data (so the select/focus recognize it as the indicator that no-data was returned). This seems overcomplicated. I'd prefer to be able to use a source: "http://...", and just have a callback somewhere indicating that no data was returned. Thank you for reading. Brian

  Read the article

• XmlReader throws on an RSS feed, when the RSS document includes an embedded <script> block.

  - by Cheeso

  The code: using (XmlReader xmlr = XmlReader.Create(new StringReader(allXml))) { var items = from item in SyndicationFeed.Load(xmlr).Items select item; } The exception: Exception: System.Xml.XmlException: Unexpected node type Element. ReadElementString method can only be called on elements with simple or empty content. Line 11, position 25. at System.Xml.XmlReader.ReadElementString() at System.ServiceModel.Syndication.Rss20FeedFormatter.ReadXml(XmlReader reader, SyndicationFeed result) at System.ServiceModel.Syndication.Rss20FeedFormatter.ReadFeed(XmlReader reader) at System.ServiceModel.Syndication.Rss20FeedFormatter.ReadFrom(XmlReader reader) at System.ServiceModel.Syndication.SyndicationFeed.Load[TSyndicationFeed](XmlReader reader) at System.ServiceModel.Syndication.SyndicationFeed.Load(XmlReader reader) at Ionic.ToolsAndTests.ReadRss.Run() in c:\dev\dotnet\ReadRss.cs:line 90 The XML content: <?xml version="1.0" encoding="utf-8"?> <?xml-stylesheet type="text/xsl" href="https://www.ibm.com/developerworks/mydeveloperworks/blogs/roller-ui/styles/rss.xsl" media="screen"?><rss version="2.0" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:atom="http://www.w3.org/2005/Atom" > <channel> <title>Software architecture, software engineering, and Renaissance Jazz</title> <link>https://www.ibm.com/developerworks/mydeveloperworks/blogs/gradybooch</link> <atom:link rel="self" type="application/rss+xml" href="https://www.ibm.com/developerworks/mydeveloperworks/blogs/gradybooch/feed/entries/rss?lang=en" /> <description>Software architecture, software engineering, and Renaissance Jazz</description> <language>en-us</language> <copyright>Copyright <script type='text/javascript'> document.write(blogsDate.date.localize (1273534889181));</script></copyright> <lastBuildDate>Mon, 10 May 2010 19:41:29 -0400</lastBuildDate> As you can see, on line 11, at position 25, there's a script block inside the <copyright> element. Other people have reported similar errors with other XML documents. The way I worked around this was to do a StreamReader.ReadToEnd, then do Regex.Replace on the result of that to yank out the script block, before passing the modified string to XmlReader.Create(). Feels like a hack. Has anyone got a better approach? I don't like this because I have to read in a 125k string into memory. Is it valid rss to include "complex content" like that - a script block inside an element? Thanks.

  Read the article

• Is MVVM pointless?

  - by joebeazelman

  Is orthodox MVVM implementation pointless? I am creating a new application and I considered Windows Forms and WPF. I chose WPF because it's future-proof and offer lots of flexibility. There is less code and easier to make significant changes to your UI using XAML. Since the choice for WPF is obvious, I figured that I may as well go all the way by using MVVM as my application architecture since it offers blendability, separation concerns and unit testability. Theoretically, it seems beautiful like the holy grail of UI programming. This brief adventure; however, has turned into a real headache. As expected in practice, I’m finding that I’ve traded one problem for another. I tend to be an obsessive programmer in that I want to do things the right way so that I can get the right results and possibly become a better programmer. The MVVM pattern just flunked my test on productivity and has just turned into a big yucky hack! The clear case in point is adding support for a Modal dialog box. The correct way is to put up a dialog box and tie it to a view model. Getting this to work is difficult. In order to benefit from the MVVM pattern, you have to distribute code in several places throughout the layers of your application. You also have to use esoteric programming constructs like templates and lamba expressions. Stuff that makes you stare at the screen scratching your head. This makes maintenance and debugging a nightmare waiting to happen as I recently discovered. I had an about box working fine until I got an exception the second time I invoked it, saying that it couldn’t show the dialog box again once it is closed. I had to add an event handler for the close functionality to the dialog window, another one in the IDialogView implementation of it and finally another in the IDialogViewModel. I thought MVVM would save us from such extravagant hackery! There are several folks out there with competing solutions to this problem and they are all hacks and don’t provide a clean, easily reusable, elegant solution. Most of the MVVM toolkits gloss over dialogs and when they do address them, they are just alert boxes that don’t require custom interfaces or view models. I’m planning on giving up on the MVVM view pattern, at least its orthodox implementation of it. What do you think? Has it been worth the trouble for you if you had any? Am I just a incompetent programmer or does MVVM not what it's hyped up to be?

  Read the article

• Semi-complex aggregate select statement confusion

  - by Ian Henry

  Alright, this problem is a little complicated, so bear with me. I have a table full of data. One of the table columns is an EntryDate. There can be multiple entries per day. However, I want to select all rows that are the latest entry on their respective days, and I want to select all the columns of said table. One of the columns is a unique identifier column, but it is not the primary key (I have no idea why it's there; this is a pretty old system). For purposes of demonstration, say the table looks like this: create table ExampleTable ( ID int identity(1,1) not null, PersonID int not null, StoreID int not null, Data1 int not null, Data2 int not null, EntryDate datetime not null ) The primary key is on PersonID and StoreID, which logically defines uniqueness. Now, like I said, I want to select all the rows that are the latest entries on that particular day (for each Person-Store combination). This is pretty easy: --Figure 1 select PersonID, StoreID, max(EntryDate) from ExampleTable group by PersonID, StoreID, dbo.dayof(EntryDate) Where dbo.dayof() is a simple function that strips the time component from a datetime. However, doing this loses the rest of the columns! I can't simply include the other columns, because then I'd have to group by them, which would produce the wrong results (especially since ID is unique). I have found a dirty hack that will do what I want, but there must be a better way -- here's my current solution: select cast(null as int) as ID, PersonID, StoreID, cast(null as int) as Data1, cast(null as int) as Data2, max(EntryDate) as EntryDate into #StagingTable from ExampleTable group by PersonID, StoreID, dbo.dayof(EntryDate) update Target set ID = Source.ID, Data1 = Source.Data1, Data2 = Source.Data2, from #StagingTable as Target inner join ExampleTable as Source on Source.PersonID = Target.PersonID and Source.StoreID = Target.StoreID and Source.EntryDate = Target.EntryDate This gets me the correct data in #StagingTable but, well, look at it! Creating a table with null values, then doing an update to get the values back -- surely there's a better way to do this? A single statement that will get me all the values the first time? It is my belief that the correct join on that original select (Figure 1) would do the trick, like a self-join or something... but how do you do that with the group by clause? I cannot find the right syntax to make the query execute. I am pretty new with SQL, so it's likely that I'm missing something obvious. Any suggestions? (Working in T-SQL, if it makes any difference)

  Read the article

• How to show server errors in Rails 2.3.5 with JSON and jQuery

  - by Fortuity

  I've got in-place editing on a page in my app (using Rails 2.3.5 and jQuery). I want to know how to display an error on the page when the update fails. I'm using ajax (an XMLHttpRequest) to save an update to a Comment object. The controller has an update method like this: def update @comment = Comment.find(params[:id]) respond_to do |format| # if @comment.update_attributes!(params[:comment]) if false #deliberately forcing a fail here to see what happens format.json { render :nothing => true } else format.json { render :json => @comment.errors, :status => :unprocessable_entity } end end end In Firebug, I can see the server returns a "422" (an appropriate validation error status code). But it's a response to an XMLHttpRequest so there is no redirect to an error page. I think I actually want to do this: format.json { render :json => @comment.errors} or maybe this: format.json {render :json => { :status => :error, :message => "Could not be saved" }.to_json, :status => 400 } and trigger some Javascript function that iterates through (and displays) any errors. I'm using a rails plugin http://github.com/janv/rest_in_place/ to implement the in-place editing. It doesn't appear to have any callback function to handle a failure. What are my options? Can I write some Javascript to respond to a failure condition without hacking the plugin? Do I have to hack the rest_in_place plugin to handle a failure condition? Is there a better plugin (for Rails or jQuery) that handles in-place editing, including failure conditions? UPDATE This post from Peter Bui (http://paydrotalks.com/posts/45-standard-json-response-for-rails-and-jquery) was helpful in showing how to handle an error message from the server using XMLHttpRequest.status. I looked at his implementation of a blog using ajax (http://github.com/paydro/talks). I'm surprised at the complexity required to handle a simple error condition. Usually Rails has all the goodness baked in but it seems server errors with JSON are out of scope. Can that be? I also looked at grimen's validatious-on-rails (http://github.com/grimen/validatious-on-rails/) which accommodates models validations when ajax XMLHttpRequest is used. It's not clear to me how I'd use it to handle the general case of a "save" failing when validations succeed. P.S. Please vote me up... so I can use more than one HTML link when I ask my question :-)

  Read the article

• PyGTK: dynamic label wrapping

  - by detly

  It's a known bug/issue that a label in GTK will not dynamically resize when the parent changes. It's one of those really annoying small details, and I want to hack around it if possible. I followed the approach at 16 software, but as per the disclaimer you cannot then resize it smaller. So I attempted a trick mentioned in one of the comments (the set_size_request call in the signal callback), but this results in some sort of infinite loop (try it and see). Does anyone have any other ideas? (You can't block the signal just for the duration of the call, since as the print statements seem to indicate, the problem starts after the function is left.) The code is below. You can see what I mean if you run it and try to resize the window larger and then smaller. (If you want to see the original problem, comment out the line after "Connect to the size-allocate signal", run it, and resize the window bigger.) The Glade file ("example.glade"): <?xml version="1.0"?> <glade-interface> <!-- interface-requires gtk+ 2.16 --> <!-- interface-naming-policy project-wide --> <widget class="GtkWindow" id="window1"> <property name="visible">True</property> <signal name="destroy" handler="on_destroy"/> <child> <widget class="GtkLabel" id="label1"> <property name="visible">True</property> <property name="label" translatable="yes">In publishing and graphic design, lorem ipsum[p][1][2] is the name given to commonly used placeholder text (filler text) to demonstrate the graphic elements of a document or visual presentation, such as font, typography, and layout. The lorem ipsum text, which is typically a nonsensical list of semi-Latin words, is a hacked version of a Latin text by Cicero, with words/letters omitted and others inserted, but not proper Latin[1][2] (see below: History and discovery). The closest English translation would be "pain itself" (dolorem = pain, grief, misery, suffering; ipsum = itself).</property> <property name="wrap">True</property> </widget> </child> </widget> </glade-interface> The Python code: #!/usr/bin/python import pygtk import gobject import gtk.glade def wrapped_label_hack(gtklabel, allocation): print "In wrapped_label_hack" gtklabel.set_size_request(allocation.width, -1) # If you uncomment this, we get INFINITE LOOPING! # gtklabel.set_size_request(-1, -1) print "Leaving wrapped_label_hack" class ExampleGTK: def __init__(self, filename): self.tree = gtk.glade.XML(filename, "window1", "Example") self.id = "window1" self.tree.signal_autoconnect(self) # Connect to the size-allocate signal self.get_widget("label1").connect("size-allocate", wrapped_label_hack) def on_destroy(self, widget): self.close() def get_widget(self, id): return self.tree.get_widget(id) def close(self): window = self.get_widget(self.id) if window is not None: window.destroy() gtk.main_quit() if __name__ == "__main__": window = ExampleGTK("example.glade") gtk.main()

  Read the article

• Enable real fixed positioning on Samsung Android browsers

  - by Mr. Shiny and New ??

  The Android browser, since 2.2, supports fixed positioning, at least under certain circumstances such as when scaling is turned off. I have a simple HTML file with no JS, but the fixed positioning on three Samsung phones I've tried is simply wrong. Instead of true fixed positioning, the header scrolls out of view then pops back into place after the scrolling is done. This doesn't happen on the Android SDK emulator for any configuration I've tested (2.2, 2.3, 2.3 x86, 4.0.4). It also doesn't happen when using the WebView in an app on the Samsung phones: in those cases the positioning works as expected. Is there a way to make the Samsung Android "stock" browser use real fixed positioning? I've tested: 1. Samsung Galaxy 551, Android 2.2 2. Samsung Galaxy S, Android 2.3 3. Samsung Galaxy S II, Android 2.3 Sample code: <html> <head> <meta name="viewport" content="initial-scale=1.0,maximum-scale=1.0,user-scalable=no,width=device-width,height=device-height"> <style> h1 { position: fixed; top: 0; left: 0; height: 32px; background-color: #CDCDCD; color: black; font-size: 32px; line-height: 32px; padding: 2px; width: 100%; margin: 0;} p { margin-top: 36px; } </style> </head> <body> <h1>Header</h1> <p>Long text goes here</p> </body> </html> The expected behaviour is that the grey header fills the top of the screen and stays put no matter how much you scroll. On Samsung Android browsers it seems to scroll out of view then pop back into place once the scrolling is done, as if the fixed-positioning is being simulated using Javascript, which it isn't. Edit Judging by the comments and "answers" it seems that maybe I wasn't clear on what I need. I am looking for a meta tag or css rule/hack or javascript toggle which turns off Samsung's broken fixed-positioning and turns on the Android browser's working fixed-positioning. I am not looking for a Javascript solution that adds broken fixed-positioning to a browser that has no support whatsoever; the Samsung fixed-positioning does that already, it just looks stupid.

  Read the article

• VSTS test deployment and invalid assembly culture

  - by Merlyn Morgan-Graham

  I have a DLL that I'm testing, which links to a DLL that has what I think is an invalid value for AssemblyCulture. The value is "Neutral" (notice the upper-case "N"), whereas the DLL I'm testing, and every other DLL in my project, has a value of "neutral" (because they specify AssemblyCulture("")). When I try to deploy the DLL that links to the problem DLL, I get this error in VSTS: Failed to queue test run '...': Culture is not supported. Parameter name: name Neutral is an invalid culture identifier. <Exception>System.Globalization.CultureNotFoundException: Culture is not supported. Parameter name: name Neutral is an invalid culture identifier. at System.Globalization.CultureInfo..ctor(String name, Boolean useUserOverride) at System.Globalization.CultureInfo..ctor(String name) at System.Reflection.RuntimeAssembly.GetReferencedAssemblies(RuntimeAssembly assembly) at System.Reflection.RuntimeAssembly.GetReferencedAssemblies() at Microsoft.VisualStudio.TestTools.Utility.AssemblyLoadWorker.ProcessChildren(Assembly assembly) at Microsoft.VisualStudio.TestTools.Utility.AssemblyLoadWorker.GetDependentAssemblies(String path) at Microsoft.VisualStudio.TestTools.Utility.AssemblyLoadWorker.GetDependentAssemblies(String path) at Microsoft.VisualStudio.TestTools.Utility.AssemblyLoadStrategy.GetDependentAssemblies(String path) at Microsoft.VisualStudio.TestTools.Utility.AssemblyHelper.GetDependentAssemblies(String path, DependentAssemblyOptions options, String configFile) at Microsoft.VisualStudio.TestTools.TestManagement.DeploymentManager.GetDependencies(String master, String configFile, TestRunConfiguration runConfig, DeploymentItemOrigin dependencyOrigin, List`1 dependencyDeploymentItems, Dictionary`2 missingDependentAssemblies) at Microsoft.VisualStudio.TestTools.TestManagement.DeploymentManager.DoDeployment(TestRun run, FileCopyService fileCopyService) at Microsoft.VisualStudio.TestTools.TestManagement.ControllerProxy.SetupTestRun(TestRun run, Boolean isNewTestRun, FileCopyService fileCopyService, DeploymentManager deploymentManager) at Microsoft.VisualStudio.TestTools.TestManagement.ControllerProxy.SetupRunAndListener(TestRun run, FileCopyService fileCopyService, DeploymentManager deploymentManager) at Microsoft.VisualStudio.TestTools.TestManagement.ControllerProxy.QueueTestRunWorker(Object state)</Exception> Even if I don't link to the DLL (in my VSTS wrapper test, or in the NUnit test), as soon as I add it in my GenericTest file (I'm wrapping NUnit tests), I get that exception. We don't have the source for the problem DLL, and it is also code signed, so I can't solve this by recompiling. Is there a way to skip deploying the dependencies of a DLL DeploymentItem, to fix or disable the culture check, or to work around this by convoluted means (maybe somehow embed the assembly)? Is there a way to override the value for the culture, short of hacking the DLL (and removing code signing so the hack works)? Maybe with an external manifest? Any correct solution must work without weird changes to production code. We can't deploy a hacked DLL, for example. It also must allow the DLL to be instrumented for code coverage. Additional note: I do get a linker warning when compiling the DLL under test that links to the problem DLL, but this hasn't broken anything but VSTS, and multiple versions have shipped.

  Read the article

• Soon to be PhD in Computer Science - Which Path to Follow?

  - by mttr

  I am going to submit my PhD thesis within the next six months. My PhD is on managing the availabiity of large-scale distributed systems, so I have some experience actually building non-trivial systems (+ I have four years experience working as a programmer). I am now trying to figure out what I should do following the PhD. I enjoy research (a quick definition: identify problem, come up with solution, ask interesting questions, find ways to answer them, build system, experiment, contribute some new knowledge and publish). I also like teaching and supervising students. It would seem that a career in academia is the ideal thing to do (can work on non-trivial problems and contribute something of use to some or more people). However, a career in academia has two significant drawbacks. First, it can be difficult to gain access to real systems with real users which then display real problems. This creates the danger that you do work that seems important (to you and maybe to some of your colleagues), but is not really relevant to anything or anyone. Second, the pay is pretty sad. Apparently, you have to sacrifice this for the privilege of doing research. I enjoy programming, but don't just want to hack some web-based system for the rest of my life. That is, working in IT for a bank is not a future I see myself enjoying. I want to work on interesting problms (that's difficult to define clearly): things where you don't know how to start, that take some time to figure out and attack, that require a rigorous approach to demonstrate that the problem has been solved, and problems that need a solution in the real world. Give the experience of people on stackoverflow, what do you think suitable options are and why (or alternatively, what gaps in my thinking does the above reveal)? Is industrial research (aka IBM Research, Microsoft Research) the only alternative avenue to a career in academia? What other areas, companies, occupations, etc. could provide me with stimulating, inspiring work? Which regions, countries am I most likely to find such work? Please share your experience.

  Read the article

• Django : presenting a form very different from the model and with multiple field values in a Django-

  - by sebpiq

  Hi ! I'm currently doing a firewall management application for Django, here's the (simplified) model : class Port(models.Model): number = models.PositiveIntegerField(primary_key=True) application = models.CharField(max_length=16, blank=True) class Rule(models.Model): port = models.ForeignKey(Port) ip_source = models.IPAddressField() ip_mask = models.IntegerField(validators=[MaxValueValidator(32)]) machine = models.ForeignKey("vmm.machine") What I would like to do, however, is to display to the user a form for entering rules, but with a very different organization than the model : Port 80 O Not open O Everywhere O Specific addresses : --------- delete field --------- delete field + add address field Port 443 ... etc Where Not open means that there is no rule for the given port, Everywhere means that there is only ONE rule (0.0.0.0/0) for the given port, and with specific addresses, you can add as many addresses as you want (I did this with JQuery), which will make as many rules. Now I did a version completely "handmade", meaning that I create the forms entirely in my templates, set input names with a prefix, and parse all the POSTed stuff in my view (which is quite painful, and means that there's no point in using a web framework). I also have a class which aggregates the rules together to easily pre-fill the forms with the informations "not open, everywhere, ...". I'm passing a list of those to the template, therefore it acts as an interface between my model and my "handmade" form : class MachinePort(object): def __init__(self, machine, port): self.machine = machine self.port = port @property def fully_open(self): for rule in self.port.rule_set.filter(machine=self.machine): if ipaddr.IPv4Network("%s/%s" % (rule.ip_source, rule.ip_mask)) == ipaddr.IPv4Network("0.0.0.0/0"): return True else : return False @property def partly_open(self): return bool(self.port.rule_set.filter(machine=self.machine)) and not self.fully_open @property def not_open(self): return not self.partly_open and not self.fully_open But all this is rather ugly ! Do anyone of you know if there is a classy way to do this ? In particular with the form... I don't know how to have a form that can have an undefined number of fields, neither how to transform these fields into Rule objects (because all the rule fields would have to be gathered from the form), neither how to save multiple objects... Well I could try to hack into the Form class, but seems like too much work for such a special case. Is there any nice feature I'm missing ?

  Read the article

• Robust LINQ to XML query for sibling key-value pairs

  - by awshepard

  (First post, please be gentle!) I am just learning about LINQ to XML in all its glory and frailty, trying to hack it to do what I want to do: Given an XML file like this - <list> <!-- random data, keys, values, etc.--> <key>FIRST_WANTED_KEY</key> <value>FIRST_WANTED_VALUE</value> <key>SECOND_WANTED_KEY</key> <value>SECOND_WANTED_VALUE</value> <!-- wanted because it's first --> <key>SECOND_WANTED_KEY</key> <value>UNWANTED_VALUE</value> <!-- not wanted because it's second --> <!-- nonexistent <key>THIRD_WANTED_KEY</key> --> <!-- nonexistent <value>THIRD_WANTED_VALUE</value> --> <!-- more stuff--> </list> I want to extract the values of a set of known "wanted keys" in a robust fashion, i.e. if SECOND_WANTED_KEY appears twice, I only want SECOND_WANTED_VALUE, not UNWANTED_VALUE. Additionally, THIRD_WANTED_KEY may or may not appear, so the query should be able to handle that as well. I can assume that FIRST_WANTED_KEY will appear before other keys, but can't assume anything about the order of the other keys - if a key appears twice, its values aren't important, I only want the first one. An anonymous data type consisting of strings is fine. My attempt has centered around something along these lines: var z = from y in x.Descendants() where y.Value == "FIRST_WANTED_KEY" select new { first_wanted_value = ((XElement)y.NextNode).Value, //... } My question is what should that ... be? I've tried, for instance, (ugly, I know) second_wanted_value = ((XElement)y.ElementsAfterSelf() .Where(w => w.Value=="SECOND_WANTED_KEY") .FirstOrDefault().NextNode).Value which should hopefully allow the key to be anywhere, or non-existent, but that hasn't worked out, since .NextNode on a null XElement doesn't seem to work. I've also tried to add in a .Select(t => { if (t==null) return new XElement("SECOND_WANTED_KEY",""); else return t; }) clause in after the where, but that hasn't worked either. I'm open to suggestions, (constructive) criticism, links, references, or suggestions of phrases to Google for, etc. I've done a fair share of Googling and checking around S.O., so any help would be appreciated. Thanks!

  Read the article

• Have problems designing input and output for a calculator app

  - by zoul

  Hello! I am writing a button calculator. I have the code split into model, view and a controller. The model knows nothing about formatting, it is only concerned with numbers. All formatting is done in the view. The model gets its input as keypresses, each keypress is a part of an enum: typedef enum { kButtonUnknown = 0, kButtonMemoryClear = 100, kButtonMemoryPlus = 112, kButtonMemoryMinus = 109, kButtonMemoryRecall = 114, kButtonClear = 99, … }; When the user presses a button (say 1), the model receives a button code (kButtonNum1), adds the corresponding number to a string input buffer ("1") and updates the numeric output value (1.0). The controller then passes the numeric output value to the view that formats it (1). This is all plain, simple and clean, but does not really work. The problem is that when user enters a part of a number (say 0.00, going to enter 0.001), the input does not survive the way through model to view and the display says 0 instead of 0.00. I know why this happens ("0.00"::string parses to 0::double and that gets formatted as 0). What I don’t know is how to design the calculator so that the code stays clean and simple and the numbers will show up on screen exactly as the user types them. I’ve already come with some kind of solution, but that’s essentially a hack and breaks the beautiful and simple flow from the calculator model to the display. Ideas? Current solution keeps track of the calculator state. If the calculator is building a number, I take the calculator input buffer (a string) and directly set the display contents (also a string). Otherwise I take the proper path, ie. take the numeric calculator output, pass it to the view as a double and the view uses its internal formatter to create a string for the display. Example input: input | display | mode ------+---------+------------ 0 | 0 | from string 0. | 0. | from string 0.0 | 0.0 | from string 0.0+ | 0 | from number This is ugly. (1) The calculator has to expose its input buffer and state. (2) The view has to expose its display and allow setting its contents directly using a string. (3) I have to duplicate some of the formatting code to format the string I got from the calculator input buffer. If user enters 12345.000, I have to display 12,345.000 and therefore I’ve got to have a commification code for strings. Yuck.

  Read the article

• Translating itunes affiliate rss via xslt

  - by jd

  I can't get this working for the life of me. Here is a snippet of the xml I get from an RSS feed from itunes affiliate. I want top print the values within tags but I cannot for some reason: <?xml version="1.0" encoding="utf-8"?> <feed xmlns:im="http://itunes.apple.com/rss" xmlns="http://www.w3.org/2005/Atom" xml:lang="en"> <id>http://ax.itunes.apple.com/WebObjects/MZStoreServices.woa/ws/RSS/toppaidapplications/sf=143441/limit=100/genre=6014/xml</id><title>iTunes Store: Top Paid Applications</title><updated>2010-03-24T15:36:42-07:00</updated><link rel="alternate" type="text/html" href="http://itunes.apple.com/WebObjects/MZStore.woa/wa/viewTop?id=25180&amp;popId=30"/><link rel="self" href="http://ax.itunes.apple.com/WebObjects/MZStoreServices.woa/ws/RSS/toppaidapplications/sf=143441/limit=100/genre=6014/xml"/><icon>http://phobos.apple.com/favicon.ico</icon><author><name>iTunes Store</name><uri>http://www.apple.com/itunes/</uri></author><rights>Copyright 2008 Apple Inc.</rights> <entry> <updated>date</updated> <id>someID</id> <title>a</title> <im:name>b</im:name> </entry> <entry> <updated>date2/updated> <id>someID2</id> <title>a2</title> <im:name>b2</im:name> </entry> </feed> If I try <xsl:apply-templates match="entry"/> it spits out the entire contents of file. If I use <xsl:call-template name="entry"> it will show only one entry and I have to use <xsl:value-of select="//*[local-name(.)='name']"/> to get name but that's a hack. I've used xslt before for xml without namespaces and xml that has proper parent child relationships but not like this RSS feed. Notice entry is not wrapped in entries or anything. Any help is appreciated. I want to use xslt because I want to alter the itunes link to go through my affiliate account - so something automated wouldn't work for me.

  Read the article

• Queue ExternalInterface calls to Flash Object in UpdatePanel - Needs Improvement?

  - by Laramie

  A Flash (actually Flex) object is created on an ASP.Net page within an Update Panel using a modified version of the embedCallAC_FL_RunContent.js script so it can be written in dynamically. It is re-created with this script with each partial postback to that panel. There are also other Update Panels on the page. With some postbacks (partial and full), External Interface calls such as $get('FlashObj').ExternalInterfaceFunc('arg1', 0, true); are prepared server-side and added to the page using ScriptManager.RegisterStartupScript. They're embedded in a function and stuffed into Sys.Application's load event, for example Sys.Application.add_load(funcContainingExternalInterfaceCalls). The problem is that because the Flash object's state state may change with each partial postback, the Flash (Flex) object and/or External Interface may not be ready or even exist yet in the DOM when the JavaScript - Flash External Interface call is made. It results in an "Object doesn't support this property or method" exception. I have a working strategy to make the ExternalInterface calls immediately if Flash is ready or else queue them until such time that Flash announces its readiness. //Called when the Flash object is initialized and can accept ExternalInterfaceCalls var flashReady = false; //Called by Flash when object is fully initialized function setFlashReady() { flashReady = true; //Make any queued ExternalInterface calls, then dequeue while (extIntQueue.length > 0) (extIntQueue.shift())(); } var extIntQueue = []; function callExternalInterface(flashObjName, funcName, args) { //reference to the wrapped ExternalInterface Call var wrapped = extWrap(flashObjName, funcName, args); //only procede with ExternalInterface call if the global flashReady variable has been set if (flashReady) { wrapped(); } else { //queue the function so when flashReady() is called next, the function is called and the aruments are passed. extIntQueue.push(wrapped); } } //bundle ExtInt call and hold variables in a closure function extWrap(flashObjName, funcName, args) { //put vars in closure return function() { var funcCall = '$get("' + flashObjName + '").' + funcName; eval(funcCall).apply(this, args); } } I set the flashReady var to dirty whenever I update the Update Panel that contains the Flash (Flex) object. ScriptManager.RegisterClientScriptBlock(parentContainer, parentContainer.GetType(), "flashReady", "flashReady = false;", true); I'm pleased that I got it to work, but it feels like a hack. I am still on the learning curve with respect to concepts like closures why "eval()" is apparently evil, so I'm wondering if I'm violating some best practice or if this code should be improved, if so how? Thanks.

  Read the article

• .NET 2.0 Process Elevation for App Installation

  - by Brian Gillespie

  We have an application written in both C++ and .NET that installs for all users in the Program Files folder. This application downloads new versions of itself (as MSI installers) and spawns the new installer process to replace itself. The install process as it exists today: Copy an install manager app (C#, .NET 2.0) to the temp directory. Call this 'Manager' Manager is executed with elevated privs per this article. The original application exits. Manager spawns the MSI installer (with elevated privs, since the copy is elevated) Manager spawns the new version of the app. The bug: The newly installed app is running in an elevated state. This causes problems I won't enumerate here. Ideally, the launch of the newly installed app would be run with the permissions of the original user. I can't figure out how to demote the app back to being the standard user after elevation. An inelegant hack: (yeah, yeah, this whole process is inelegant anyway) Copy the install manager to the temp directory Run the install manager with standard user privs. Lets call this instance 'LowlyManager'. Original application exits. LowlyManager spawns the app again, this time with elevated privs. Let's name this instance 'UpperManagement' UpperManagement spawns the installer UpperManagement exits gracefully, returning the exit code of the installer. LowlyManager interprets the error code from UpperManagement, and spawns the newly installed application. This time as the original invoker. Is there a better way to do this? (I've left out a bunch of other details before and after these steps that make the process smoother for the user, but this should be enough to understand the core of the problem I'm trying to solve.) Other requirements: We can't install as a per-user app The user shouldn't be presented with an authentication dialog box if UAC would have simply asked "are you sure you want to allow this?". I think this might kill a solution using WindowsImpersonationContext, but I'm not sure. The system needs to work on XP, Vista, and Windows 7 (even if there is a separate process for XP).

  Read the article

• Why don't my scrollbars work properly when programmatically hiding rows in silverlight Datagrid?

  - by Luke Vilnis

  I have a Silverlight datagrid with custom code that allows for +/- buttons on the lefthand side and can display a table with a tree structure. The +/- buttons are bound to a IsExpanded property on my ViewModelRows, as I call them. The visibility of rows is bound to an IsVisible property on the ViewModelRows which is determined based on whether or not all of the parent rows are expanded. Straightforward enough. This code works fine in that if I scroll up and down the grid with PageUp/PageDown or the arrow keys, all the right rows are hidden and everything has the right structure and I can play with the +/- buttons to my hearts content. However, the vertical scroll bar on the right hand side, although it starts off the correct size and it scrolls through the rows smoothly, when I collapse rows and then re-expand them, doesn't go back to its correct size. The scrollbar can still usually be moved around to scroll through the whole collection, but because it is too big, once the bar moves to the bottom, there are still more rows to go through and it sort of jerkily shoots all the way down to the bottom or sometimes fails to scroll at all. This is pretty hard to describe so I included a screenshot with the black lines drawn on to show the difference in scrollbar length even though the two grids have the same number of rows expanded. I think this might be a bug related to the way the Datagrid does virtualization of rows. It seems to me like it isn't properly keeping track of how tall each row is supposed to be when expansion states change. Is there a way to programmatically "poke" (read hack) it to recalculate its scrollbar size on LoadingRow or something ugly like that? I'd include a code sample but there's 2 c# files and 1 xaml file so I wanted to see if anyone else has heard of this sort of issue before I try to make it reproducible in a self-contained way. Once again, scrolling with the arrow keys works fine so I'm pretty sure the underlying logic and binding is working, there's just some issue with the row height not being calculated properly. Since I'm a new user, it won't let me use image tags so here's the link to a picture of the problem: http://img210.imageshack.us/img210/8760/messedupscrollbars.png

  Read the article

• Using XSLT, how do I separate nodes based on their value?

  - by Will Goring

  I have a pretty flat XML structure that I need to reorder into categorised sections and, for the life of me, I can't figure out how to do it in XSLT (not that I'm by any means an expert.) Basically, the original XML looks kinda like: <things> <thing> <value>one</value> <type>a</type> </thing> <thing> <value>two</value> <type>b</type> </thing> <thing> <value>thee</value> <type>b</type> </thing> <thing> <value>four</value> <type>a</type> </thing> <thing> <value>five</value> <type>d</type> </thing> </things> And I need to output something like: <data> <a-things> <a>one</a> <a>four</a> </a-things> <b-things> <b>two</b> <b>three</b> </b-things> <d-things> <d>five</d> </d-things> </data> Note that I can't output <c-things> if there aren't any <c> elements, but I do know ahead of time what the complete list of types is, and it's fairly short so handcoding templates for each type is definitely possible. It feels like I could probably hack something together using <xsl:if> and <xsl:for-each> but it also feels like there must be a more ... 'templatey' way to do it. Can anyone help? Cheers.

  Read the article

• What's the correct way to use Cakephp urls?

  - by Pichan

  Hello all, it's my first post here :) I'm having some difficulties with dealing with urls and parameters. I've gone through the router class api documentation over and over again and found nothing useful. First of all, I'd like to know if there is any 'universal' format in CakePHP(1.3) for handling urls. I'm currently handling all my urls as simple arrays(in the format that Router::url and $html-link accepts) and it's easy as long as I only need to pass them as arguments to cake's own methods. It usually gets tricky if I need something else. Mainly I'm having problems with converting string urls to the basic array format. Let's say I want to convert $arrayUrl to string and than again into url: $arrayUrl=array('controller'=>'SomeController','action'=>'someAction','someValue'); $url=Router::url($arrayUrl); //$url is now '/path/to/site/someController/someAction/someValue' $url=Router::normalize($url); //remove '/path/to/site' $parsed=Router::parse($url); /*$parsed is now Array( [controller] => someController [action] => someAction [named] => Array() [pass] => Array([0] => someValue) [plugin] => ) */ That seems an awful lot of code to do something as simple as to convert between 2 core formats. Also, note that $parsed is still not in the same as $arrayUrl. Of course I could tweak $parsed manually and actually I've done that a few times as a quick patch but I'd like to get to the bottom of this. I also noticed that when using prefix routing, $this-params in controller has the prefix embedded in the action(i.e. [action] = 'admin_edit') and the result of Router::parse() does not. Both of course have the prefix in it's own key. To summarize, how do I convert an url between any of these 3(or 4, if you include the prefix thing) mentioned formats the right way? Of course it would be easy to hack my way through this, but I'd still like to believe that cake is being developed by a bunch of people who have a lot more experience and insight than me, so I'm guessing there's a good reason for this "perceived misbehavior". I've tried to present my problem as good as I can, but due to my rusty english skills, I had to take a few detours :) I'll explain more if needed.

  Read the article

• When does IE7 recompute styles? Doesn't work reliably when a class is added to the body.

  - by Kid A

  I have an interesting problem here. I'm using a class on the element as a switch to drive a fair amount of layout behavior on my site. If the class is applied, certain things happen, and if the class isn't applied, they don't happen. The relevant CSS is roughly like this: .rightSide { display:none; } .showCommentsRight .rightSide { display:block; width:50%; } .showCommentsRight .leftSide { display:block; width:50%; } And the HTML: <body class="showCommentsRight"> <div class="container"></div> <div class="leftSide"></div> <div class="rightSide"></div> </div> <div class="container"></div> <div class="leftSide"></div> <div class="rightSide"></div> </div> <div class="container"></div> <div class="leftSide"></div> <div class="rightSide"></div> </div> </body> I've simplified things but this is essentially the method. The whole page changes layout (hiding the right side in three different areas) when the flag is set on the body. This works in Firefox and IE8. It does not work in IE8 in compatibility mode. What is fascinating is that if you sit there and refresh the page, the results can vary. It will pick a different section's right side to show. Sometimes it will show only the top section's right side, sometimes it will show the middle. I have tried a validator (to look for malformed html), double css formatting, and making sure my IE7 hack sheet wasn't having an effect. So my question is: * Is there a way that this behavior can be made reliable? * When does IE7 decide to re-do styling? Thanks everyone.

  Read the article

• Is this scatter-brained workflow realizable in Git?

  - by Luke Maurer

  This is what I'd like my workflow to look like at a conceptual level: I hack on my new feature for a while I notice a typo in a comment I change it Since the typo is completely unrelated to anything else, I put that change in a pile of comment fixes I keep working on the code I realize I need to flesh out a few utility functions I do so I put that change in its own pile Steps 2, 3, and 4 each repeat throughout the day I finish the new feature and put the changes for that feature in a pile I push nice patches upstream: One with the new feature, a few for the other tweaks, and one with a bunch of comment fixes if enough have accumulated Since I'm both lazy and a perfectionist, I want to be able to do some things out of order: I might correct a typo but forget to put it in the comment fix pile; when I prepare the upstream patches (I'm using git-svn, so I need to be pretty deliberate about these), I'll then pull out the comment fixes at that point. I might forget to separate things altogether until the very end. But I might /also/ have committed some of the piles along the way (sorry, the metaphor is breaking down …). This is all rather like just using Eclipse changesets with SVN, only I can have different changes to the same file in different piles (having to disentangle changes into different commits is what motivated me to move to git-svn, in fact …), and with Git I can have my full discombobulated change history, experimental branches and all, but still make a nice, neat patch. I've just recently started with Git after having wanted to for a good while, and I'm quite happy so far. The biggest way in which the above workflow doesn't really map into Git, though, is that a “bin” can't really be just a local branch, since the working tree only ever reflects the state of a single branch. Or maybe the Git index is a “pile,” and what I want is to have more than one somehow (effectively). I can think of a few ways to approximate what I want (maybe creative use of stash? Intricate stash-checkout-merge dances?), but my grasp on Git isn't solid enough to be sure of how best to put all the pieces together. It's said that Git is more a toolkit than a VCS, so I guess the question comes down to: How do I build this thing with these tools?

  Read the article

• Undesired "flashing" in Jquery dropdown menu?

  - by Tom

  Hi, I've got the following piece of Jquery: $("#collapse-menu > li > a").click(function() { $(this).toggleClass("expanded").toggleClass("collapsed").find("+ ul").slideToggle("medium"); }); What it basically does is expands or collapses a menu of nested "lists" which contain dropdowns (simplified example): <ul id="collapse-menu"> <li><a class="expanded">Basic Details</a> <ul> <li> <select> .... </select> </li> <li> <select> .... </select> </li> The code works absolutely fine EXCEPT when a large value is selected in any of the dropdowns. At that point, when clicked, the menu will still expand/collapse correctly but "flash" quickly while doing so, as if the entire element was being reset somehow. The data itself is fine but it's the flashing that's unwanted. The strange thing is that if a small value is selected in a dropdown, there's no flashing. When a large value is selected (say, above 30 in an age dropdown of 18-99), the flashing starts happening. Can anyone tell me why this is happening? Or whether there's something not right about the Jquery that's causing this. Thanks. UPDATE: Adding a bounty to this. Have tried a few similar plugins/solutions out there on the net but they all seem to suffer from this "flashing" problem when large dropdown values are selected by default. If it helps, here's a typical similar solution: http://www.i-marco.nl/weblog/jquery-accordion-menu/index_collapsed.html# ... but it suffers from the same problem when dropdowns are added inside the accordion. I hope someone has a clean solution, instead of needing to hack this somehow.

  Read the article

• How can I tell groovy/grails not to try to "re-encode" binary data? (Revised title)

  - by ?????

  I have a groovy/grails application that needs to serve images It works fine on my dev box, the image is returned properly. Here's the start of the returned JPEG, as seen by od -cx 0000000 377 330 377 340 \0 020 J F I F \0 001 001 001 001 , d8ff e0ff 1000 464a 4649 0100 0101 2c01 but on the production box, there's some garbage in front, and the d8ff e0ff before the 1000 is missing 0000000 ? ** ** ? ** ** ? ** ** ? ** ** \0 020 J F bfef efbd bdbf bfef efbd bdbf 1000 464a 0000020 I F \0 001 001 001 \0 H \0 H \0 \0 ? ** ** ? 4649 0100 0101 4800 4800 0000 bfef efbd It's the exact same code. I just moved the .war over and run it on a different machine. (Isn't Java supposed to be write once, run everywhere?) Any ideas? An "encoding" problem? The code is sent to the response like this: response.contentType = "image/jpeg"; response.outputStream << out; Here's the code that locates the image on an internal application server and re-serves the image. I've pared down the code a bit to remove the error handling, etc, to make it easier to read. def show = { def address = "http://internal.application.server:9899/img?photoid=${params.id}" def out = new ByteArrayOutputStream() out << new URL(address).openStream() response.contentLength = out.size(); // XXX If you don't do this hack, "head" requests won't work! if (request.method == 'HEAD') { render( text : "", contentType : "image/jpeg" ); } else { response.contentType = "image/jpeg"; response.outputStream << out; } } Update: I tried setting the CharacterEncoding response.setCharacterEncoding("ISO-8859-1"); if (request.method == 'HEAD') { render( text : "", contentType : "image/jpeg" ); } else { response.contentType = "image/jpeg;charset=ISO-8859-1"; response.outputStream << out; } but it made no difference in the output. On my production machine, the binary bytes in the image are re-encoded/escaped as if they were UTF-8 (see Michael's explanation below). It works fine on my development machine.

  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

< Previous Page | 58 59 60 61 62 63 64 65 66 67  | Next Page >