Search Results

Search found 11972 results on 479 pages for 'writing'.

Page 384/479 | < Previous Page | 380 381 382 383 384 385 386 387 388 389 390 391  | Next Page >

• Usage of Assert.Inconclusive

  - by Johannes Rudolph

  Hi, Im wondering how someone should use Assert.Inconclusive(). I'm using it if my Unit test would be about to fail for a reason other than what it is for. E.g. i have a method on a class that calculates the sum of an array of ints. On the same class there is also a method to calculate the average of the element. It is implemented by calling sum and dividing it by the length of the array. Writing a Unit test for Sum() is simple. However, when i write a test for Average() and Sum() fails, Average() is likely to fail also. The failure of Average is not explicit about the reason it failed, it failed for a reason other than what it should test for. That's why i would check if Sum() returns the correct result, otherwise i Assert.Inconclusive(). Is this to be considered good practice? What is Assert.Inconclusive intended for? Or should i rather solve the previous example by means of an Isolation Framework?

  Read the article

• Tail recursion and memoization with C#

  - by Jay

  I'm writing a function that finds the full path of a directory based on a database table of entries. Each record contains a key, the directory's name, and the key of the parent directory (it's the Directory table in an MSI if you're familiar). I had an iterative solution, but it started looking a little nasty. I thought I could write an elegant tail recursive solution, but I'm not sure anymore. I'll show you my code and then explain the issues I'm facing. Dictionary<string, string> m_directoryKeyToFullPathDictionary = new Dictionary<string, string>(); ... private string ExpandDirectoryKey(Database database, string directoryKey) { // check for terminating condition string fullPath; if (m_directoryKeyToFullPathDictionary.TryGetValue(directoryKey, out fullPath)) { return fullPath; } // inductive step Record record = ExecuteQuery(database, "SELECT DefaultDir, Directory_Parent FROM Directory where Directory.Directory='{0}'", directoryKey); // null check string directoryName = record.GetString("DefaultDir"); string parentDirectoryKey = record.GetString("Directory_Parent"); return Path.Combine(ExpandDirectoryKey(database, parentDirectoryKey), directoryName); } This is how the code looked when I realized I had a problem (with some minor validation/massaging removed). I want to use memoization to short circuit whenever possible, but that requires me to make a function call to the dictionary to store the output of the recursive ExpandDirectoryKey call. I realize that I also have a Path.Combine call there, but I think that can be circumvented with a ... + Path.DirectorySeparatorChar + .... I thought about using a helper method that would memoize the directory and return the value so that I could call it like this at the end of the function above: return MemoizeHelper( m_directoryKeyToFullPathDictionary, Path.Combine(ExpandDirectoryKey(database, parentDirectoryKey)), directoryName); But I feel like that's cheating and not going to be optimized as tail recursion. Any ideas? Should I be using a completely different strategy? This doesn't need to be a super efficient algorithm at all, I'm just really curious. I'm using .NET 4.0, btw. Thanks!

  Read the article

• C Programing - Return libcurl http response to string to the calling function

  - by empty set

  I have a homework and i need somehow to compare two http responses. I am writing it on C (dumb decision) and i use libcurl to make things easier. I am calling the function that uses libcurl to http request and response from another function and i want to return the http response to it. Anyway, the code below doesn't work, any ideas? #include <stdio.h> #include <curl/curl.h> #include <string.h> size_t write_data(void *ptr, size_t size, size_t nmemb, void *stream) { size_t written; written = fwrite(ptr, size, nmemb, stream); return written; } char *handle_url(void) { CURL *curl; char *fp; CURLcode res; char *url = "http://www.yahoo.com"; curl = curl_easy_init(); if (curl) { curl_easy_setopt(curl, CURLOPT_URL, url); curl_easy_setopt(curl, CURLOPT_WRITEFUNCTION, write_data); curl_easy_setopt(curl, CURLOPT_WRITEDATA, fp); res = curl_easy_perform(curl); if(res != CURLE_OK) fprintf(stderr, "curl_easy_perform() failed: %s\n", curl_easy_strerror(res)); curl_easy_cleanup(curl); //printf("\n%s", fp); } return fp; } This solution C libcurl get output into a string works, but not in my case because i want to return this string to the calling function.

  Read the article

• Image appearing in the wrong place.

  - by Luke

  I have a list that I want to precede on each line with a small image. The CSS: .left div.arrow{background-image: url(../images/design/box.png); float:left; height:15px; width:15px; margin-right:2px;} .left a.list:link, a:visited, a:active {color:black; font-size:0.8em; text-decoration:none; display:block; float:left;} The HTML: <div class="panel">My quick links</div> <div class="arrow"></div> <a href="/messages.php?p=new" class="list">Send a new message</a> <br /> <div class="arrow"></div> <a href="/settings.php#password" class="list">Change my password</a> <br /> <div class="arrow"></div> <a href="/settings.php#picture" class="list">Upload a new site image</a> <br /> As you can see, each image should go before the writing. On the first line, the text "Send a new message" has the preceeding image. However, each line afterwards has the image at the end. So "Send a new message" has an image at the start and finish. It is like the images are staying at the end of the previous line. Any ideas?

  Read the article

• How can I sign a Windows Mobile application for internal use?

  - by AR

  I'm developing a Windows Mobile application for internal company use, using the Windows Mobile 6 Professional SDK. Same old story: I've developed and tested on the emulator and all is well, but as soon as I deploy to advice I get an UnauthorizedAccessException when writing files or creating directories. I'm aware that an application installed to a device needs to be signed but I'm running into roadblocks at every turn: Using the project properties 'Devices' window I select 'Sign the project output with this certificate, and choose one of the sample certificates from the SDK. This results in a build error: "The signer's certificate is not valid for signing" when running SignTool. If I try to run SignTool.exe from the commandline, I get an error telling me to run SignTool.exe from a location in the system's PATH. I can't use the 'Signing' tab in the Project Properties to create a test certificate - this is greyed out (presumably for WinMobile projects?). If at all possible, I would like to avoid having to go through Versign or the like to get a Mobile2Market certificate. If I have to go this route for a final version that's fine, but I need to at least be able to test the app on real devices. Any advice would be most welcome!

  Read the article

• Can I detect whether an object has called GC.SuppressFinalize?

  - by Joe White

  Is there a way to detect whether or not an object has called GC.SuppressFinalize? I have an object that looks something like this (full-blown Dispose pattern elided for clarity): public class ResourceWrapper { private readonly bool _ownsResource; private readonly UnmanagedResource _resource; public ResourceWrapper(UnmanagedResource resource, bool ownsResource) { _resource = resource; _ownsResource = ownsResource; if (!ownsResource) GC.SuppressFinalize(this); } ~ResourceWrapper() { if (_ownsResource) // clean up the unmanaged resource } } If the ownsResource constructor parameter is false, then the finalizer will have nothing to do -- so it seems reasonable (if a bit quirky) to call GC.SuppressFinalize right from the constructor. However, because this behavior is quirky, I'm very tempted to note it in an XML doc comment... and if I'm tempted to comment it, then I ought to write a unit test for it. But while System.GC has methods to set an object's finalizability (SuppressFinalize, ReRegisterForFinalize), I don't see any methods to get an object's finalizability. Is there any way to query whether GC.SuppressFinalize has been called on a given instance, short of buying Typemock or writing my own CLR host?

  Read the article

• Help with why my app crashed?

  - by Moshe

  I'm writing an iPad app that is a "kiosk" app. The iPad should be hanging on the wall and the app should just run. I did a test, starting the app last night (Friday, December 31) and letting it run. This morning, when I woke up, it was not running. I just checked the iPad's console and I can't figure out why it crashed. The iPad was plugged in and so the battery is not the issued. I did disable the idleTimer in my application delegate. The app was seen running as late as midnight last night. I would like to note that my app acts as a Bluetooth server through Game Kit and large portion of the console output is occupied by bluetooth status messages. When I opened the iPad, the app was paused and there was a system alert which prompted me to check an "Expiring Provisioning Profile". I tapped "dismiss" and the alert went away. The app crashed about a second after I dismissed the system alert. Any ideas how I can diagnose this problem? Why would my app crash? Here is my iPad's Console log, as copied from Xcode's organizer. Edit: A bit of Googling lead me to this site which says that alert views cause the app to lose focus. Could that be involved? What can I do to fix the problem? EDIT2: My Crash log describes the situation as: Application Specific Information: appname failed to resume in time Elapsed total CPU time (seconds): 10.010 (user 8.070, system 1.940), 100% CPU Elapsed application CPU time (seconds): 9.470, 95% CPU

  Read the article

• Building a ctypes-"based" C library with distutils

  - by Robie Basak

  Following this recommendation, I have written a native C extension library to optimise part of a Python module via ctypes. I chose ctypes over writing a CPython-native library because it was quicker and easier (just a few functions with all tight loops inside). I've now hit a snag. If I want my work to be easily installable using distutils using python setup.py install, then distutils needs to be able to build my shared library and install it (presumably into /usr/lib/myproject). However, this not a Python extension module, and so as far as I can tell, distutils cannot do this. I've found a few references to people other people with this problem: Someone on numpy-discussion with a hack back in 2006. Somebody asking on distutils-sig and not getting an answer. Somebody asking on the main python list and being pointed to the innards of an existing project. I am aware that I can do something native and not use distutils for the shared library, or indeed use my distribution's packaging system. My concern is that this will limit usability as not everyone will be able to install it easily. So my question is: what is the current best way of distributing a shared library with distutils that will be used by ctypes but otherwise is OS-native and not a Python extension module? Feel free to answer with one of the hacks linked to above if you can expand on it and justify why that is the best way. If there is nothing better, at least all the information will be in one place.

  Read the article

• Is there a way to programatically check dependencies of an EXE?

  - by Mason Wheeler

  I've got a certain project that I build and distribute to users. I have two build configurations, Debug and Release. Debug, obviously, is for my use in debugging, but there's an additional wrinkle: the Debug configuration uses a special debugging memory manager, with a dependency on an external DLL. There's been a few times when I've accidentally built and distributed an installer package with the Debug configuration, and it's then failed to run once installed because the users don't have the special DLL. I'd like to be able to keep that from happening in the future. I know I can get the dependencies in a program by running Dependency Walker, but I'm looking for a way to do it programatically. Specifically, I have a way to run scripts while creating the installer, and I want something I can put in the installer script to check the program and see if it has a dependency on this DLL, and if so, cause the installer-creation process to fail with an error. I know how to create a simple CLI program that would take two filenames as parameters, and could run a DependsOn function and create output based on the result of it, but I don't know what to put in the DependsOn function. Does anyone know how I'd go about writing it?

  Read the article

• Advice on whether to use native C++ DLL or not: PINVOKE & Marshaling ?

  - by Bob

  What's the best way to do this....? I have some Native C++ code that uses a lot of Win32 calls together with byte buffers (allocated using HeapAlloc). I'd like to extend the code and make a C# GUI...and maybe later use a basic Win32 GUI (for use where there is no .Net and limited MFC support). (A) I could just re-write the code in C# and use multiple PINVOKEs....but even with the PINVOKES in a separate class, the code looks messy with all the marshaling. I'm also re-writing a lot of code. (B) I could create a native C++ DLL and use PINVOKE to marshal in the native data structures. I'm assuming I can include the native C++ DLL/LIB in a project using C#? (C) Create a mixed mode DLL (Native C++ class plus managed ref class). I'm assuming that this would make it easier to use the managed ref class in C#......but is this the case? Will the managed class handle all the marshaling? Can I use this mixed mode DLL on a platform with no .Net (i.e. still access the native C++ unmanaged component) or do I limit myself to .Net only platforms. One thing that bothers me about each of these options is all the marshalling. Is it better to create a managed data structure (array, string etc.) and pass that to the native C++ class, or, the other way around? Any ideas on what would be considered best practice...?

  Read the article

• Difference in BackgroundWorker thread access in VS2010 / .NET 4.0?

  - by Jonners

  Here's an interesting one - in VS2005 / VS2008 running against .NET 2.0 / .NET 3.0 / .NET 3.5, a BackgroundWorker thread may not directly update controls on a WinForms form that initiated that thread - you'll get a System.InvalidOperationException out of the BackgroundWorker stating "Cross-thread operation not valid: Control 'thecontrol' accessed from a thread other than the thread it was created on". I remember hitting this back in 2004 or so when I first started writing .NET WinForms apps with background threads. There are several ways around the problem - this is not a question asking for that answer. I've been told recently that this kind of operation is now allowed when written in VS2010 / .NET 4.0. This seems unlikely to me, since this kind of object access restriction has always been a fairly fundamental part of thread-safe programming. Allowing a BackgroundWorker thread direct access to objects owned not by itself but by its parent UI form seems contrary to that principle. A trawl through the .NET 4.0 docs hasn't revealed any obvious changes that could account for this behaviour. I don't have VS2010 / .NET 4.0 to test this out - does anyone who has access to that toolset know for sure whether the model has changed to allow that kind of thread interaction? I'd like to either take advantage of it in future, or deploy the cluestick. ;)

  Read the article

• Looking for a web based, embeddable file manager

  - by Kristi H.

  I'm looking for a web based, embeddable file manager. I haven't found anything suitable through Google or the Stack Overflow archives. Does anyone know of a file manager that meets the following criteria? The file manager must… …be embeddable (e.g. Flash or a Java applet) …run from my server (no storing uploads in a remote location) …be licensed for business use (fees are okay) …let me disable uploads. Users shouldn't be able to upload files, only download them …allow me to tag files and/or place them in folders …let me disable authentication or handle it from the backend. My app already has a login system and I don't want to make users log in twice …allow users to select and download multiple files …have an attractive interface The file manager should… …have an external config file …display thumbnail previews for files …be skinnable …be actively developed or at least actively maintained Thanks for your help. I need a sophisticated file manager and I'd like to avoid writing it from scratch.

  Read the article

• Streaming Youtube Videos

  - by Vinay

  Hi All, I am writing an application to play the youtube videos using streaming. First method: I am getting the RTSP URL to the video using GData APIs. Here is the code to play the RTSP url. VideoView mVideoView = new VideoView(this); setContentView(mVideoView); mVideoView.setVideoURI(Uri.parse("rtsp://rtsp2.youtube.com/CiILENy73wIaGQkDwpjrUxOWQBMYESARFEgGUgZ2aWRlb3MM/0/0/0/video.3gp")); mVideoView.start(); But it throws error on both G1 device and emulator (Emulator has some firewall problem as per mailing list) Here is the error message ERROR/PlayerDriver(35): Command PLAYER_INIT completed with an error or info PVMFFailure Second method: A hack way to get the path of 3gp file from http://www.youtube.com/get_video?v=&t=<&<.. After getting the file path and I can call setVideoURI and it plays fine. But it is a hack way to achieve the requirement. I have checked the Youtube App also, it also does the hack way to play the youtube url.(Checked with logcat) I have tried changing from VideoView to MediaPlayer but no change in the error. Is there a "Clean" way to do this? Please let me know your thoughts.

  Read the article

• How to view shell commands used by eclipse "run configurations"

  - by gmale

  Given a "run configuration" in Eclipse, I want to print out the associated shell command that would be used to run it. For example: Right now, in Eclipse, if I click "play" it will run: mvn assembly:directory -Dmaven.test.skip=true I don't see that command, I just know that's what the IDE must run, at some point. However, some of the other run configurations are far more complex with long classpaths and virtual machine options and, frankly, sometimes I have no idea what the equivalent shell command would be (particularly when it comes to Flex). There must be some way to access the shell command that would be associated with a "Run Configuration" in Eclipse/Flex Builder. This information must be available, which leads me to believe someone has written a plugin to display it. Or maybe there's already an option built into Eclipse for accessing this. So is there a way to, essentially, convert an Eclipse run configuration into a shell command? (for context only: I'm asking because I'm writing a bash script that automates everything I do, during development--from populating the Database all the way to opening Firefox and clearing the cache before running the web app. So every command I run from the IDE needs to exist in the script. Some are tricky to figure out.)

  Read the article

• Possible to capture all events in a web browser?

  - by David

  I am working on a pet project and am at the research stage. Quick summary I am trying to intercept all form submits, onclick, and every single keydown. My library of choice is either jquery, or jquery + prototypejs. I figure I can batch up the events into a queue/stack and send it back to the server in time interval batches to keep performance relatively stable. Concerns Form submits and change's would be relatively simple.. Something like $("form :inputs").bind("change", function() { ... record event... }); But how to ensure I get precedence over the applications handlers as I have a habit of putting return false on a lot of my form handlers when there is a validation issue. As I understand it, that effectively stops the event in its tracks. My project For my smaller remote clients I will put their products onto a VPS or run it in my home data center. Currently I use a basic authentication system, given a username/password they see the website and then hopefully send me somewhat sane notes on what is broken or should be tweaked. As a better solution, I've written a simple proxy web server that does the above but allows me to have one DNS entry and then depending on credentials it makes an internal request relaying headers and re-writing URLS as needed. Every single html/text or application/* request is compressed and shoved into a sqlite table so I can partially replay what they've done. Now I am shifting to the frontend and would like to capture clicks, keydown's, and submits on everything on the page.

  Read the article

• C# Breakpoint Weirdness

  - by Dan

  In my program I've got two data files A and B. The data in A is static and the data in B refers back to the data in A. In order to make sure the data in B is invalidated when A is changed, I keep an identifier for each of the links which is a long byte-string identifying the data. I get this string using BitConverter on some of the important properties. My problem is that this scheme isn't working. I save the identifiers initially, and with I reload (with the exact same data in A) the identifiers don't match anymore. It seems the bit converter gives different results when I go to save. The really weird thing about it is, if I place a breakpoint in the save code, I can see the identifier it's writing to the file is fine, and the next load works. If I don't place a breakpoint and say print the identifiers to console instead, they're totally different. It's like when my program is running at full-speed the CPU messes up some instructions. This isn't the first time something like this happens to me. I've seen it in other projects. What gives? Has anyone every experienced this kind of debugging weirdness? I can't explain how stopping the program and not stopping it can change the output. Also, it's not a hardware problem because this happens on my laptop as well.

  Read the article

• C++ operator lookup rules / Koenig lookup

  - by John Bartholomew

  While writing a test suite, I needed to provide an implementation of operator<<(std::ostream&... for Boost unit test to use. This worked: namespace theseus { namespace core { std::ostream& operator<<(std::ostream& ss, const PixelRGB& p) { return (ss << "PixelRGB(" << (int)p.r << "," << (int)p.g << "," << (int)p.b << ")"); } }} This didn't: std::ostream& operator<<(std::ostream& ss, const theseus::core::PixelRGB& p) { return (ss << "PixelRGB(" << (int)p.r << "," << (int)p.g << "," << (int)p.b << ")"); } Apparently, the second wasn't included in the candidate matches when g++ tried to resolve the use of the operator. Why (what rule causes this)? The code calling operator<< is deep within the Boost unit test framework, but here's the test code: BOOST_AUTO_TEST_SUITE(core_image) BOOST_AUTO_TEST_CASE(test_output) { using namespace theseus::core; BOOST_TEST_MESSAGE(PixelRGB(5,5,5)); // only compiles with operator<< definition inside theseus::core std::cout << PixelRGB(5,5,5) << "\n"; // works with either definition BOOST_CHECK(true); // prevent no-assertion error } BOOST_AUTO_TEST_SUITE_END() For reference, I'm using g++ 4.4 (though for the moment I'm assuming this behaviour is standards-conformant).

  Read the article

• GNU Make - Dependencies on non program code

  - by Tim Post

  A requirement for a program I am writing is that it must be able to trust a configuration file. To accomplish this, I am using several kinds of hashing algorithms to generate a hash of the file at compile time, this produces a header with the hashes as constants. Dependencies for this are pretty straight forward, my program depends on config_hash.h, which has a target that produces it. The makefile looks something like this : config_hash.h: $(SH) genhash config/config_file.cfg > $(srcdir)/config_hash.h $(PROGRAM): config_hash.h $(PROGRAM_DEPS) $(CC) ... ... ... I'm using the -M option to gcc, which is great for dealing with dependencies. If my header changes, my program is rebuilt. My problem is, I need to be able to tell if the config file has changed, so that config_hash.h is re-generated. I'm not quite sure how explain that kind of dependency to GNU make. I've tried listing config/config_file.cfg as a dependency for config_hash.h, and providing a .PHONY target for config_file.cfg without success. Obviously, I can't rely on the -M switch to gcc to help me here, since the config file is not a part of any object code. Any suggestions? Unfortunately, I can't post much of the Makefile, or I would have just posted the whole thing.

  Read the article

• Firefox extension: How to inject javascript into page and run it?

  - by el griz

  I'm writing a Firefox extension to allow users to annotate any page with text and/or drawings and then save an image of the page including the annotations. Use cases would be clients reviewing web pages, adding feedback to the page, saving the image of this and emailing it back to the web developer or testers taking annotated screenshots of GUI bugs etc. I wrote the annotation/drawing functionality in javascript before developing the extension. This script adds a <canvas> element to the page to draw upon as well as a toolbar (in a <div>) that contains buttons (each <canvas> elements) for the different draw tools e.g. line, box, ellipse, text, etc. This works fine when manually included in a page. I now need a way for the extension to: Inject this script into any page including pages I don't control. This needs to occur when the user invokes the extension, which can be after the page has loaded. Once injected the init() function in this script that adds the canvas and toolbar elements etc. needs to be run somehow, but I can't determine how to call this from the extension. Note that once injected I don't need this script to interact with the extension (as the extension just takes a screenshot of the entire document (and removes the added page elements) when the user presses the save button in the extension chrome).

  Read the article

• In App Purchase Unique Identifying Data

  - by dageshi

  O.K so I'm writing a iPhone travel guide, you purchase a subscription to a travel guide for 3 months, it downloads a fairly hefty database and for 3 months that database gets updated weekly with new stuff. Now what I'd like to do is make the user enter their email address as a one off action before they purchase their first guide, for China say. The purpose for doing this is 1) To allow me to contact the user by email when they add a note/tip for a particular place (the app will allow them to send notes & information to me) 2) To Uniquely identify who has purchased the subscription so that if they wipe their device and reinstall the app they can plug the email address in and pickup their subscriptions again. Or so they can use the same subscription on another device they own. My concerns are 1) Will Apple allow the email method of restoring functionality to a second or restored device? 2) As long as I tell the user what I'm using their email address for (aka I won't sell it to anyone else and use it for X purposes) will it be o.k to ask for said email address? And as a side note, can I tack the devices unique id onto my server comms to track devices or is apple going to through a hissy fit about that as well?

  Read the article

• How do i make divs go into another row when full?

  - by acidzombie24

  My code is something like the below. When theres 3 images everything is fine once theres 4 it gets full and moves the entire div.top into another row. How do i make the div inside top just start a new row instead? I tried writing .top width=500px but once it hits or passes it instead the images inside are squeeze together instead of each being 150x150. I tried max-width on top instead and in opera and chrome i see the border of top as 500width but the images continue to render pass it. (i have a firefox problem with my div so the width looks fixed to something else). So how do i make these divs go into another row? and not try to squeeze together <div class="top"> <div><a href><img/></a></div> <div><a href><img/></a></div> <div><a href><img/></a></div> </div>

  Read the article

• Looping through a file in VB.NET

  - by Ousman

  I am writing a VB.NET program and I'm trying to accomplish the following: Read and loop through a text file line by line Show the event of the loop on a textbox or label until a button is pressed The loop will then stop on any number that happened to be at the loop event and When a button is pressed again the loop will continue. Code Imports System.IO Public Class Form1 'Dim nFileNum As Integer = FreeFile() ' Get a free file number Dim strFileName As String = "C:\scb.txt" Dim objFilename As FileStream = New FileStream(strFileName, _ FileMode.Open, FileAccess.Read, FileShare.Read) Dim objFileRead As StreamReader = New StreamReader(objFilename) 'Dim lLineCount As Long 'Dim sNextLine As String Private Sub btStart_Click(ByVal sender As System.Object, _ ByVal e As System.EventArgs) _ Handles btStart.Click Try If objFileRead.ReadLine = Nothing Then MsgBox("No Accounts Available to show!", _ MsgBoxStyle.Information, _ MsgBoxStyle.DefaultButton2 = MsgBoxStyle.OkOnly) Return Else Do While (objFileRead.Peek() > -1) Loop lblAccounts.Text = objFileRead.ReadLine() 'objFileRead.Close() 'objFilename.Close() End If Catch ex As Exception MessageBox.Show(ex.Message) Finally 'objFileRead.Close() 'objFilename.Close() End Try End Sub Private Sub Form1_Load(ByVal sender As System.Object, _ ByVal e As System.EventArgs) _ Handles MyBase.Load End Sub End Class Problem I'm able to read the text file but my label will only loop if I hit the start button. It goes to the next line, but I want it to continue to loop through the entire file until I hit a button telling it to stop.

  Read the article

• How to merge objects in php ?

  - by The Devil

  Hey everybody, I'm currently re-writing a class which handles xml files. Depending on the xml file and it's structure I sometimes need to merge objects. Lets say once I have this: <page name="a title"/> And another time I have this: <page name="a title"> <permission>administrator</permission> </page> Before, I needed only the attributes from the "page" element. That's why a lot of my code expects an object containing only the attributes ($loadedXml-attributes()). Now there are xml files in which the <permission> element is required. I did manage to merge the objects (though not as I wanted) but I can't get to access one of them (most probably it's something I'm missing). To merge my objects I used this code: (object) array_merge( (array) $loadedXml->attributes(), (array) $loadedXml->children() ); This is what I get from print_r(): stdClass Object ( [@attributes] => Array ( [name] => a title ) [permission] => Array ( [0] => administrator ) ) So now my question is how to access the @attributes method ? Thanks in advance, The Devil

  Read the article

• Controlling shell command line wildcard expansion in C or C++

  - by Adrian McCarthy

  I'm writing a program, foo, in C++. It's typically invoked on the command line like this: foo *.txt My main() receives the arguments in the normal way. On many systems, argv[1] is literally *.txt, and I have to call system routines to do the wildcard expansion. On Unix systems, however, the shell expands the wildcard before invoking my program, and all of the matching filenames will be in argv. Suppose I wanted to add a switch to foo that causes it to recurse into subdirectories. foo -a *.txt would process all text files in the current directory and all of its subdirectories. I don't see how this is done, since, by the time my program gets a chance to see the -a, then shell has already done the expansion and the user's *.txt input is lost. Yet there are common Unix programs that work this way. How do they do it? In Unix land, how can I control the wildcard expansion? (Recursing through subdirectories is just one example. Ideally, I'm trying to understand the general solution to controlling the wildcard expansion.)

  Read the article

• ncurses - resizing glitch

  - by ryyst

  I'm writing an ncurses program and am trying to make it respond correctly to terminal resizing. While I can read the terminal dimensions correctly in my program, ncurses doesn't seem to deal with new dimensions correctly. Here's a (somewhat lengthy) sample program: #include <ncurses.h> #include <string.h> #include <signal.h> #include <sys/ioctl.h> void handle_winch(int sig){ struct winsize w; ioctl(0, TIOCGWINSZ, &w); COLS = w.ws_col; LINES = w.ws_row; wresize(stdscr, LINES, COLS); clear(); mvprintw(0, 0, "COLS = %d, LINES = %d", COLS, LINES); for (int i = 0; i < COLS; i++) mvaddch(1, i, '*'); refresh(); } int main(int argc, char *argv[]){ initscr(); struct sigaction sa; memset(&sa, 0, sizeof(struct sigaction)); sa.sa_handler = handle_winch; sigaction(SIGWINCH, &sa, NULL); while(getch() != 27) {} endwin(); return 0; } If you run it, you can see that the terminal dimensions are correctly retrieved. But the second line, which is supposed to draw *-characters across the screen, doesn't work. Try resizing the window horizontally to make it larger, the line of *s will not get larger. What's the problem here? I'm aware that one can temporarily leave curses mode, but I'd prefer a cleaner solution. Thanks!

  Read the article

< Previous Page | 380 381 382 383 384 385 386 387 388 389 390 391  | Next Page >