Search Results

Search found 4655 results on 187 pages for 'multi homed'.

Page 33/187 | < Previous Page | 29 30 31 32 33 34 35 36 37 38 39 40  | Next Page >

  • Get the current array key in a multi dimensional array

    - by johlton
    Hi *, I have a session array *$_SESSION['cart']* with some items in it. The structure ist like this (via *print_r*): Array ( [2-1] => Array ( [color] => 7 [articlenumber] => WRG70 10 [quantity] => 1 [price] => 17.50 ) [3-8] => Array ( [color] => 2 [articlenumber] => QRG50 02 [quantity] => 1 [price] => 13.50 ) ) Looping over the values for display is fine ... foreach($_SESSION['cart'] as $item_array) { foreach($item_array as $item => $value) { echo $value . ' | '; } } ... since it results in something like this: 7 | WRG70 10 | 1 | 17.50 | 2 | QRG50 02 | 1 | 13.50 | But Now: How can I output the matching key (e.g. '2-1') as well? I tried some array functions like key() & current but couldn't get it to work (one of these days). Any quick hint on this? Thanks a lot and best from Berlin Fabian

    Read the article

  • Scroll up/down while selecting multi-page block of text in Vim through putty

    - by shan23
    I'm using vim 6.3.81 on a xterm through putty. When I use set mouse=a , I can scroll through the file , but not select text. When mouse is disabled using set mouse=, I can select text with mouse (automatically copied to a buffer) , but if the text block is over one page long , I can't scroll up/down (and have to select text one screen at a time) !! Is there any way to have the best of both - while selecting text with mouse , if i reach the end of page , it automatically scrolls up/down depending on whether i've reached top/bottom of page and allows me to select text from the next page as well ? I know vim is powerful/flexible enough to do what I want - I just don't know how !! Can anyone help ?

    Read the article

  • Multi-part template issue with Jinja2

    - by Alan Harris-Reid
    Hi, When creating templates I typically have 3 separate parts (header, body, footer) which I combine to pass a singe string to the web-server (CherryPy in this case). My first approach is as follows... from jinja2 import Environment, FileSystemLoader env = Environment(loader=FileSystemLoader('')) tmpl = env.get_template('Body.html') page_body = tmpl.render() tmpl = env.get_template('Header.html') page_header = tmpl.render() tmpl = env.get_template('Footer.html') page_footer = tmpl.render() page_code = page_header + page_body + page_footer but this contains repetitious code, so my next approach is... def render_template(html_file): from jinja2 import Environment, FileSystemLoader env = Environment(loader=FileSystemLoader('')) tmpl = env.get_template(html_file) return tmpl.render() page_header = render_template('Header.html') page_body = render_template('Body.html') page_footer = render_template('Footer.html) However, this means that each part is created in its own environment - can that be a problem? Are there any other downsides to this approach? I have chosen the 3-part approach over the child-template approach because I think it may be more flexible (and easier to follow), but I might be wrong. Anyone like to convince me that using header, body and footer blocks might be better? Any advice would be appreciated. Alan

    Read the article

  • Doxygen: grouping documentation by folder in a multi-project codebase

    - by John
    In one project, some pages were added - it was a new project in which doxygen was being tested properly - adding comments & pages - rather than simply auto-generating docs from our existing code-base. The problem is when doxygen is run on the main code-base, that project's pages show up at a top level. e.g they have a main-page and some sub-pages. But what we'd want is all those pages pushed down one level so you have main-project-main-project-pages. One question is what happens if multiple projects have a main-page? Do they get combined, or throw errors? Another question is if you can tell doxygen to use paths and containing folders to auto-generate groups or sections or page-hierarchies in some way? Going through all out projects to properly assign classes to groups is a mammoth task, so ideally everything in a directory would get put in a group of that name, as a way to make the documentation of non-doxygen codebases better. Sorry my question is a bit vague, the problem is even after reading the docs the terminology isn't totally clear yet. Hopefully the kind of question I'm asking is clear, if not I'll try to cobble an example file-structure together.

    Read the article

  • .htaccess whit multi domains problem

    - by NeoNmaN
    i have this .htaccess rule RewriteCond %{HTTP_HOST} ^domain1.com RewriteRule (.*) http://www.domain2.com/$1 [R=301,L] RewriteCond %{HTTP_HOST} www.domain1.com RewriteRule (.*) http://www.domain2.com/$1 [R=301,L] and right now i hav this code RewriteCond %{HTTP_HOST} ^www\.(.*) RewriteRule (.*) http://%1/$1 [R=301,L] the problem is right now if i type www.domain1.com its return to domain1.com and not www.domain2.com i will have it to if i type www.domain1.com, domain1.com or domain3.com its will return to this domain ( www.domain2.com ) and if i type ( domain2.com ) its return to www.domain2.com i hob i can be help here tanks a lot.

    Read the article

  • Multi Process Configuration

    - by user200937
    Hi, I have a product built out of multiple processes. Each process uses internally commons configuration. Does anyone have an idea how to manage the config? I.e. we do not want to duplicate variables so each process will be able to read them. Additionally, DB solution is no good, as we do not want to be dependent on DB for something like configuration. Thanks Yair

    Read the article

  • Multi-site WCF Service

    - by vault
    I have implemented a WCF service that will be used at Site A with 5 computers in the LAN accessing the service. What I would like to do is have an elevated user/administrator be able to use one of the 5 machines and query an identical service at Site B (also with 5 computers) that they will need to connect to by bypassing the LAN firewall.Accessing data from Site A to Site B (and vica-versa) need only be read-only Is there a standardised way to acheive this using WCF?

    Read the article

  • what pattern to use for multi-argument method?

    - by Omid S
    I have a method with the following signature: foo (Sample sample, Aliquot aliquot) "foo" needs to alter a Sample object, either the first argument or from the second argument it can extract its Sample. For example: foo (Sample sample, Aliquot aliquot) { Sample out = null; if (sample != null) out = sample else out = aliquot.getSample() return out; } But that is so un-elegant, other than reading the API a developer does not know the reference of the first argument overrides the Sample of the second argument. Now, I could change "foo" to foo (SomeMagicalObject bar) where SomeMagicalObject is a tuple for Sample and Aliquot and holds some logic ... etc. But I am wondering, are there some patterns for this question?

    Read the article

  • Multi-dimensional array edge/border conditions

    - by kirbuchi
    Hi, I'm iterating over a 3 dimensional array (which is an image with 3 values for each pixel) to apply a 3x3 filter to each pixel as follows: //For each value on the image for (i=0;i<3*width*height;i++){ //For each filter value for (j=0;j<9;j++){ if (notOutsideEdgesCondition){ *(**(outArray)+i)+= *(**(pixelArray)+i-1+(j%3)) * (*(filter+j)); } } } I'm using pointer arithmetic because if I used array notation I'd have 4 loops and I'm trying to have the least possible number of loops. My problem is my notOutsideEdgesCondition is getting quite out of hands because I have to consider 8 border cases. I have the following handled conditions Left Column: ((i%width)==0) && (j%3==0) Right Column: ((i-1)%width ==0) && (i>1) && (j%3==2) Upper Row: (i<width) && (j<2) Lower Row: (i>(width*height-width)) && (j>5) and still have to consider the 4 corner cases which will have longer expressions. At this point I've stopped and asked myself if this is the best way to go because If I have a 5 line long conditional evaluation it'll not only be truly painful to debug but will slow the inner loop. That's why I come to you to ask if there's a known algorithm to handle this cases or if there's a better approach for my problem. Thanks a lot.

    Read the article

  • Give me a practical use-case of Multi-set

    - by Calm Storm
    I would like to know a few practical use-cases (if they are not related/tied to any programming language it will be better).I can associate Sets, Lists and Maps to practical use cases. For example if you wanted a glossary of a book where terms that you want are listed alphabetically and a location/page number is the value, you would use the collection TreeMap(OrderedMap which is a Map) Somehow, I can't associate MultiSets with any "practical" usecase. Does someone know of any uses? http://en.wikipedia.org/wiki/Multiset does not tell me enough :) PS: If you guys think this should be community-wiki'ed it is okay. The only reason I did not do it was "There is a clear objective way to answer this question".

    Read the article

  • multi-threaded proxy checker having problems

    - by Paul
    hello everyone, I am trying to create a proxy checker. This is my first attempt at multithreading and it's not going so well, the threads seem to be waiting for one to complete before initializing the next. Imports System.Net Imports System.IO Imports System.Threading Public Class Form1 Public sFileName As String Public srFileReader As System.IO.StreamReader Public sInputLine As String Public Class WebCall Public proxy As String Public htmlout As String Public Sub New(ByVal proxy As String) Me.proxy = proxy End Sub Public Event ThreadComplete(ByVal htmlout As String) Public Sub send() Dim myWebRequest As HttpWebRequest = CType(WebRequest.Create("http://www.myserver.com/ip.php"), HttpWebRequest) myWebRequest.Proxy = New WebProxy(proxy, False) Try Dim myWebResponse As HttpWebResponse = CType(myWebRequest.GetResponse(), HttpWebResponse) Dim loResponseStream As StreamReader = New StreamReader(myWebResponse.GetResponseStream()) htmlout = loResponseStream.ReadToEnd() Debug.WriteLine("Finished - " & htmlout) RaiseEvent ThreadComplete(htmlout) Catch ex As WebException If (ex.Status = WebExceptionStatus.ConnectFailure) Then End If Debug.WriteLine("Failed - " & proxy) End Try End Sub End Class Private Sub Button1_Click(ByVal sender As System.Object, ByVal e As System.EventArgs) Handles Button1.Click Dim proxy As String Dim webArray As New ArrayList() Dim n As Integer For n = 0 To 2 proxy = srFileReader.ReadLine() webArray.Add(New WebCall(proxy)) Next Dim w As WebCall For Each w In webArray Threading.ThreadPool.QueueUserWorkItem(New WaitCallback(AddressOf w.send), w) Next w End Sub Private Sub Form1_Load(ByVal sender As System.Object, ByVal e As System.EventArgs) Handles MyBase.Load srFileReader = System.IO.File.OpenText("proxies.txt") End Sub End Class

    Read the article

  • PHP multi dimensional array manipulation

    - by atif089
    Hi, This is my array Array ( [0] => Array ( [sample_id] => 3 [time] => 2010-05-30 21:11:47 ) [1] => Array ( [sample_id] => 2 [time] => 2010-05-30 21:11:47 ) [2] => Array ( [sample_id] => 1 [time] => 2010-05-30 21:11:47 ) ) And I want to get all the sample_ids in one array. can someone please help ? Can this be done without for loops (because arrays are very large).

    Read the article

  • Naming multi-instance performance counters in .NET

    - by Roger Lipscombe
    Most multiple instance performance counters in Windows seem to automatically(?) have a #n on the end if there's more than one instance with the same name. For example: if, in Perfmon, you look under the Process category, you'll see: ... dwm explorer explorer#1 ... I have two explorer.exe processes, so the second counter has #1 appended to its name. When I attempt to do this in a .NET application: I can create the category, and register the instance (using the PerformanceCounterCategory.Create that takes a CounterCreationDataCollection). I can open the counter for write and write to it. When I open the counter a second time, it opens the same counter. This means that I have two applications fighting over the counters. The documentation for PerformanceCounter.InstanceName states that # is not allowed in the name. So: how do I have multiple-instance performance counters that are actually multiple instance? And where the second (and subsequent) instances get #n appended to the name? That is: I know that I can put the process ID (e.g.) on the instance name. This works, but has the unfortunate side effect that restarting the process results in a new PID, and Perfmon continues monitoring the old counter.

    Read the article

  • multi-line pattern matching in pyhon

    - by Horace Ho
    A periodic computer generated message (simplified): Hello user123, - (604)7080900 - 152 - minutes Regards Using python, how can I extract "(604)7080900", "152", "minutes" (i.e. any text following a leading "- " pattern) between the two empty lines (empty line is the \n\n after "Hello user123" and the \n\n before "Regards"). Even better if the result string list are stored in an array. Thanks!

    Read the article

  • Multi-thread conversation in django (like Reddit)

    - by dotty
    Hay, i have an app which essentially a conversation system (much like reddits) Where a post can have multiple replies, and a reply and have multiplies, and a reply to a reply can have multiple replies (etc) I've made the model like this class Discussion(models.Model): message = models.TextField() replies = models.ManyToManyField('self') and the view discussions = Discussions.objects.all() and the template looks like this {% for discussion in discussions %} {{ discussion.message }} {% endfor %} How would i go about making a system where i can output all replies like this discussion reply reply reply reply reply reply Which would go down as far as it needs to to ensure all replies are listed. Thanks

    Read the article

  • Fixing a multi-threaded pycurl crash.

    - by Rook
    If I run pycurl in a single thread everything works great. If I run pycurl in 2 threads python will access violate. The first thing I did was report the problem to pycurl, but the project died about 3 years ago so I'm not holding my breath. My (hackish) solution is to build a 2nd version of pycurl called "pycurl_thread" which will only be used by the 2nd thread. I downloaded the pycurl module from sourceforge and I made a total of 4 line changes. But python is still crashing. My guess is that even though this is a module with a different name (import pycurl_thread), its still sharing memory with the original module (import pycurl). How should I solve this problem? Changes in pycurl.c: initpycurl(void) to initpycurl_thread(void) and m = Py_InitModule3("pycurl", curl_methods, module_doc); to m = Py_InitModule3("pycurl_thread", curl_methods, module_doc); Changes in setup.py: PACKAGE = "pycurl" PY_PACKAGE = "curl" to PACKAGE = "pycurl_thread" PY_PACKAGE = "curl_thread" Here is the seg fault i'm getting. This is happening within the C function do_curl_perform(). *** longjmp causes uninitialized stack frame ***: python2.7 terminated ======= Backtrace: ========= /lib/libc.so.6(__fortify_fail+0x37)[0x7f209421b537] /lib/libc.so.6(+0xff4c9)[0x7f209421b4c9] /lib/libc.so.6(__longjmp_chk+0x33)[0x7f209421b433] /usr/lib/libcurl.so.4(+0xe3a5)[0x7f20931da3a5] /lib/libpthread.so.0(+0xfb40)[0x7f209532eb40] /lib/libc.so.6(__poll+0x53)[0x7f20941f6203] /usr/lib/libcurl.so.4(Curl_socket_ready+0x116)[0x7f2093208876] /usr/lib/libcurl.so.4(+0x2faec)[0x7f20931fbaec] /usr/local/lib/python2.7/dist-packages/pycurl.so(+0x892b)[0x7f209342c92b] python2.7(PyEval_EvalFrameEx+0x58a1)[0x4adf81] python2.7(PyEval_EvalCodeEx+0x891)[0x4af7c1] python2.7(PyEval_EvalFrameEx+0x538b)[0x4ada6b] python2.7(PyEval_EvalFrameEx+0x65f9)[0x4aecd9]

    Read the article

  • multi word query in mysql

    - by salmane
    Hi there , in order to make things easier for users i want to add multiple keyword search to my site. so that in the input the user would do something like : " keyword1 keyword 2" ( similar to google for example. would i need to write a code that would parse that string and do queries based on that or is there something built in mysql that could do it?

    Read the article

  • Multi-platform development from one computer

    - by iama
    I am planning to build a new development computer for both Windows & Linux platforms. On Windows, my development would be primarily in .NET/C#/IIS/MSSQL Server. On Linux—preferably Ubuntu—my development would be in Ruby and Python. I am thinking of buying a laptop with Windows 7 pre-installed with 4GB RAM, Intel Core 2 Duo, and 320 GB HD; running 2 VMs for both Windows and Linux development with the host OS as my work station. Of course, I would be running DBs and web servers on the respective platforms. Is this a typical setup? My only concern is running two VMs side by side. Not sure if this configuration would be optimal. Alternative would be to do my Windows development on the host Windows 7 OS. What are your thoughts?

    Read the article

  • Why isn't multi-touch working for my imagebuttons?

    - by Droidy
    I'm using imagebuttons that play sounds using SoundPool. Here is example code of one of the imagebuttons: ImageButton Button1 = (ImageButton)findViewById(R.id.sound); Button1.setOnTouchListener(new OnTouchListener() { public boolean onTouch(View v, MotionEvent event) { if (event.getAction() == MotionEvent.ACTION_DOWN ) { mSoundManager.playSound(1); return false; } return false; } }); For some reason, it doesn't allow you to click multiple buttons at the same time. Is it because I'm building for 1.6 SDK? Thanks!

    Read the article

  • ASP.NET spellchecker - Multi language

    - by Srikanth
    I am looking at various options to implement a spellchecker in our ASP.NET 3.5 application. This applicaition is to be used worldwide so a wide language support is important. Can you guys suggest the best possible controls for this purpose? Currently, we are evaluating Obout and Telerik suites. Are there better alternatives, given that spellcheck is our primary goal? thanks

    Read the article

  • Java multi Generic collection parameters complie error

    - by Geln Yang
    Hi, So strange!Please have a look the code first: public class A { } public class B extends A { } public class C extends A { } public class TestMain { public <T extends A> void test(T a, T b) { } public <T extends A> void test(List<T> a, List<T> b) { } public static void main(String[] args) { new TestMain().test(new B(), new C()); new TestMain().test(new ArrayList<B>(), new ArrayList<C>()); } } The statement "new TestMain().test(new ArrayList(), new ArrayList())" get a "Bound mismatch" compile error, while "new TestMain().test(new B(), new C())" is compiled ok. Bound mismatch: The generic method test(T, T) of type TestMain is not applicable for the arguments (ArrayList, ArrayList). The inferred type ArrayList is not a valid substitute for the bounded parameter It seems the type of the second generic List parameter is limited by the Type of the first.Is it a feature or a bug of the compile program? ps, jdk:1.6,IDE:Eclipse 3.5.1

    Read the article

  • Clustered index - multi-part vs single-part index and effects of inserts/deletes

    - by Anssssss
    This question is about what happens with the reorganizing of data in a clustered index when an insert is done. I assume that it should be more expensive to do inserts on a table which has a clustered index than one that does not because reorganizing the data in a clustered index involves changing the physical layout of the data on the disk. I'm not sure how to phrase my question except through an example I came across at work. Assume there is a table (Junk) and there are two queries that are done on the table, the first query searches by Name and the second query searches by Name and Something. As I'm working on the database I discovered that the table has been created with two indexes, one to support each query, like so: --drop table Junk1 CREATE TABLE Junk1 ( Name char(5), Something char(5), WhoCares int ) CREATE CLUSTERED INDEX IX_Name ON Junk1 ( Name ) CREATE NONCLUSTERED INDEX IX_Name_Something ON Junk1 ( Name, Something ) Now when I looked at the two indexes, it seems that IX_Name is redundant since IX_Name_Something can be used by any query that desires to search by Name. So I would eliminate IX_Name and make IX_Name_Something the clustered index instead: --drop table Junk2 CREATE TABLE Junk2 ( Name char(5), Something char(5), WhoCares int ) CREATE CLUSTERED INDEX IX_Name_Something ON Junk2 ( Name, Something ) Someone suggested that the first indexing scheme should be kept since it would result in more efficient inserts/deletes (assume that there is no need to worry about updates for Name and Something). Would that make sense? I think the second indexing method would be better since it means one less index needs to be maintained. I would appreciate any insight into this specific example or directing me to more info on maintenance of clustered indexes.

    Read the article

  • Multi-threaded .NET application blocks during file I/O when protected by Themida

    - by Erik Jensen
    As the title says I have a .NET application that is the GUI which uses multiple threads to perform separate file I/O and notice that the threads occasionally block when the application is protected by Themida. One thread is devoted to reading from serial COM port and another thread is devoted to copying files. What I experience is occasionally when the file copy thread encounters a network delay, it will block the other thread that is reading from the serial port. In addition to slow network (which can be transient), I can cause the problem to happen more frequently by making a PathFileExists call to a bad path e.g. PathFileExists("\\\\BadPath\\file.txt"); The COM port reading function will block during the call to ReadFile. This only happens when the application is protected by Themida. I have tried under WinXP, Win7, and Server 2012. In a streamlined test project, if I replace the .NET application with a MFC unmanaged application and still utilize the same threads I see no issue even when protected with Themida. I have contacted Oreans support and here is their response: The way that a .NET application is protected is very different from a native application. To protect a .NET application, we need to hook most of the file access APIs in order to "cheat" the .NET Framework that the application is protected. I guess that those special hooks (on CreateFile, ReadFile...) are delaying a bit the execution in your application and the problem appears. We did a test making those hooks as light as possible (with minimum code on them) but the problem still appeared in your application. The rest of software protectors that we tried (like Enigma, Molebox...) also use a similar hooking approach as it's the only way to make the .NET packed file to work. If those hooks are not present, the .NET Framework will abort execution as it will see that the original file was tampered (due to all Microsoft checks on .NET files) Those hooks are not present in a native application, that's why it should be working fine on your native application. Oreans support tried other software protectors such as Enigma Protector, Engima VirtualBox, and Molebox and all exhibit the exact same problem. What I have found as a work around is to separate out the file copy logic (where the file exists call is being made) to be performed in a completely separate process. I have experimented with converting the thread functions from unmanaged C++ to VB.NET equivalents (PathFileExists - System.IO.File.Exists and CreateFile/ReadFile - System.IO.Ports.SerialPort.Open/Read) and still see the same serial port read blocked when the file check or copy call is delayed. I have also tried setting the ReadFile to work asynchronously but that had no effect. I believe I am dealing with some low-level windows layer that no matter the language it exhibits a block on a shared resource -- and only when the application is executing under a single .NET process protected by Themida which evidently installs some hooks to allow .NET execution. At this time converting the entire application away from .NET is not an option. Nor is separating out the file copy logic to a separate task. I am wondering if anyone else has more knowledge of how a file operation can block another thread reading from a system port. I have included here example applications that show the problem: https://db.tt/cNMYfEIg - VB.NET https://db.tt/Y2lnTqw7 - MFC They are Visual Studio 2010 solutions. When running the themida protected exe, you can see when the FileThread counter pauses (executing the File.Exists call) while the ReadThread counter also pauses. When running non-protected visual studio output exe, the ReadThread counter does not pause which is how we expect it to function. Thanks!

    Read the article

< Previous Page | 29 30 31 32 33 34 35 36 37 38 39 40  | Next Page >