Search Results

Search found 5672 results on 227 pages for 'progamming tips'.

Page 221/227 | < Previous Page | 217 218 219 220 221 222 223 224 225 226 227  | Next Page >

  • Need Google Map InfoWindow Hyperlink to Open Content in Overlay (Fusion Table Usage)

    - by McKev
    I have the following code established to render the map in my site. When the map is clicked, the info window pops up with a bunch of content including a hyperlink to open up a website with a form in it. I would like to utilize a function like fancybox to open up this link "form" in an overlay. I have read that fancybox doesn't support calling the function from within an iframe, and was wondering if there was a way to pass the link data to the DOM and trigger the fancybox (or another overlay option) in another way? Maybe a callback trick - any tips would be much appreciated! <style> #map-canvas { width:850px; height:600px; } </style> <script type="text/javascript" src="http://maps.google.com/maps/api/js?sensor=true"></script> <script src="http://gmaps-utility-gis.googlecode.com/svn/trunk/fusiontips/src/fusiontips.js" type="text/javascript"></script> <script type="text/javascript"> var map; var tableid = "1nDFsxuYxr54viD_fuH7fGm1QRZRdcxFKbSwwRjk"; var layer; var initialLocation; var browserSupportFlag = new Boolean(); var uscenter = new google.maps.LatLng(37.6970, -91.8096); function initialize() { map = new google.maps.Map(document.getElementById('map-canvas'), { zoom: 4, mapTypeId: google.maps.MapTypeId.ROADMAP }); layer = new google.maps.FusionTablesLayer({ query: { select: "'Geometry'", from: tableid }, map: map }); //http://gmaps-utility-gis.googlecode.com/svn/trunk/fusiontips/docs/reference.html layer.enableMapTips({ select: "'Contact Name','Contact Title','Contact Location','Contact Phone'", from: tableid, geometryColumn: 'Geometry', suppressMapTips: false, delay: 500, tolerance: 8 }); ; // Try W3C Geolocation (Preferred) if(navigator.geolocation) { browserSupportFlag = true; navigator.geolocation.getCurrentPosition(function(position) { initialLocation = new google.maps.LatLng(position.coords.latitude,position.coords.longitude); map.setCenter(initialLocation); //Custom Marker var pinColor = "A83C0A"; var pinImage = new google.maps.MarkerImage("http://chart.apis.google.com/chart?chst=d_map_pin_letter&chld=%E2%80%A2|" + pinColor, new google.maps.Size(21, 34), new google.maps.Point(0,0), new google.maps.Point(10, 34)); var pinShadow = new google.maps.MarkerImage("http://chart.apis.google.com/chart?chst=d_map_pin_shadow", new google.maps.Size(40, 37), new google.maps.Point(0, 0), new google.maps.Point(12, 35)); new google.maps.Marker({ position: initialLocation, map: map, icon: pinImage, shadow: pinShadow }); }, function() { handleNoGeolocation(browserSupportFlag); }); } // Browser doesn't support Geolocation else { browserSupportFlag = false; handleNoGeolocation(browserSupportFlag); } function handleNoGeolocation(errorFlag) { if (errorFlag == true) { //Geolocation service failed initialLocation = uscenter; } else { //Browser doesn't support geolocation initialLocation = uscenter; } map.setCenter(initialLocation); } } google.maps.event.addDomListener(window, 'load', initialize); </script>

    Read the article

  • Draw rectangle-like objects on a bitmap

    - by _simon_
    I am performing OCR (optical character recognition) on a bunch of images. Images are grouped into different projects (tickets, credit cards, insurance cards etc). Each image represents an actual product (for instance, if we have images of credit cards, picture1.jpg is image of my credit card, picture2.jpg is image of your credit card,... you get it). I have a settings.xml file, which contains regions of the image, where OCR should be performed. Example: <Project Name="Ticket1" TemplateImage="...somePath/templateTicket1.jpg"> <Region Name="Prefix" NumericOnly="false" Rotate="0"> <x>470</x> <y>395</y> <width>31</width> <height>36</height> </Region> <Region Name="Num1" NumericOnly="true" Rotate="0"> <x>555</x> <y>402</y> <width>123</width> <height>35</height> </Region> </Project> </Project Name="CreditCard" TemplateImage="...somePath/templateCreditCard1.jpg"> <Region Name="SerialNumber" NumericOnly="false" Rotate="90"> <x>332</x> <y>12</y> <width>20</width> <height>98</height> </Project> I would like to set these parameters through GUI (now I just write them into xml file). So, first I load a template image for a project (an empty credit card). Then I would like to draw a rectangle around a text, where OCR should be performed. I guess this isn't hard, but it would be great if I could also move and resize this rectangle object in the picture. I have to display all regions (rectangles) on the picture also. Also - there will probably be a list of regions in a listview, so when you click a region in this listview, it should mark it on the picture in a green color for example. Do you know for a library, which I could use? Or a link with some tips how to create such objects?

    Read the article

  • Using hidden values with jQuery (and ASP.NET MVC) -- not working?

    - by SlackerCoder
    Im using a couple of JSON calls to render data, etc etc. In order to keep the proper key value, I am storing it in a tag. I have this in several places in my code, none of which cause an issue like this one is. Here is the jQuery: The call that "sets" the value: $("a[id^='planSetupAddNewPlan']").live('click', function() { var x = $(this).attr('id'); x = x.substring(19); $("#hidPlanSetupCurrentGroupKey").val(x); $.getJSON("/GroupSetup/PlanSetupAddNewList", { GroupKey: x }, function(data) { $("#planSetupAddNew").html('' + data.TableResult + ''); alert('First Inside 2 ' + x); $.blockUI({ message: $("#planSetupAddNew") }); }); }); The call that "gets" the value: $("#ddlPlanSetupAddNewProduct").live('change', function() { var a = $("#hidPlanSetupCurrentGroupKey").val(); var prod = $(this).val(); alert(a); $.getJSON("/GroupSetup/PlanSetupChangePlanList", { GroupKey: a, Product: prod }, function(data) { if (data.Message == "Success") { $("#planSetupAddNewPlan").html('' + data.TableResult + ''); } else if (data.Message == "Error") { //Do something } }); }); Here is the html in question: <div id="planSetupAddNew" style="display:none; cursor: default;"> <input type="hidden" id="hidPlanSetupCurrentGroupKey" /> <div id="planSetupAddNewData"> </div> </div> In the first section, the alert ('First Inside 2 ' + x) returns what I expect (where x = the key value), and if I add a line to display the contents of the hidden field, that works as well: ie. var key = $("#hidPlanSetupCurrentGroupKey").val(); alert(key); In the "alert(a);" call, I am getting "undefined". I have looked at the other code in the same view and it is the same and it works. I must be missing something, or have some sort of mistype that I havent caught. Just an overview of the controller events: The first call (/GroupSetup/PlanSetupAddNewList) will return an html string building a "form" for users to enter information into. The second call (/GroupSetup/PlanSetupChangePlanList) just changes a second dropdown based on the first dropdown selection (overwriting the html in the div). If you need more info, let me know! Any thoughts/tips/pointers/suggestions?!?! Thanks for all your help :)

    Read the article

  • Web Services, Memory Leaks and CRM

    - by Neil
    Hi, I have a website that allows users to upload a csv file. This calls a service that reads the information from the csv, puts it into DynamicEntity objects and calls the CRM service to Create/Update entities in CRM. When this service creates/updates an entity this kicks off other plugins to apply certain business rules. These rules can also Create or Update entites in CRM. The issue here is that the handle count of the w3wp.exe process that the website is calling increases every time the an entity is created or updated and it never comes back down. I tried putting Garbage Collection code in the business rules and this reduces the handle count of the CRM w3wp process (run by the Network Service), but not the other w3wp process. Should I have Dispose methods on the Web Service that calls the CRM service? I hope that makes sense. I'm not overly familiar with memory management issues so any help is appreciated. Can anybody give me some tips on how to stop this from occurring? Thanks, Neil -- EDIT Okay well the handle count goes up when I call the Service.Create(DynamicEntity) method. I don't think placing any code here would be beneficial. When I exit the method/class/service that contains this call the handle count stays as it is. What I need to know is whether this is something I should be managing or is it something CRM takes care of (or doesn't take care of but I can't do anything about it) -- Another Edit Right this is how it works. 1) We have CRM and its related services 2) We have another service independent of CRM that uses the CRM services (number 1 above) to create entities based on csv info passed into it 3) We have a website that allows a user to upload a csv, and calls service no 2 above to Create/Update entities in CRM 4) We have plugins fired by CRM which use Service 1 above to create/update entities So the user uploads a csv to the website (3), this fires a service(2). When service 2 creates an entity using service 1, Service 4 fires. Service 4 calls also uses service 1 to Create entities, and when these services are called (using the Service.Create() method) the handle count of the process increases. When the method/class/services finish the handle count remains the same, and so when the whole process occurs again the handle count will increased again.

    Read the article

  • A program where user enters a string and the program counts the instances of the letters

    - by user1865183
    This is the first C++ program I have ever written and I'm having trouble understanding the order in which operands must be put in. This is for a class, but it looks like I'm not supposed to use the homework tag. Sorry if I'm doing this wrong. This is my input // Get DNA string string st; cout << "Enter the DNA sequence to be analysed: "; cin >> st; This seems to work ok, but I thought I would include it incase this is what I'm doing wrong. This is what I have so far to check that the input is exclusively C,T,A, or G. It runs through the program and simply prints "Please enter a valid sequnce1, please enter a valid sequence2, ... ect. I'm sure I'm doing something very stupid, I just can't figure it out. // Check that the sequence is all C, T, A, G while (i <= st.size()){ if (st[i] != 'c' && st[i] != 'C' && st[i] != 'g' && st[i] != 'G' && st[i] != 't' && st[i] != 'T' && st[i] != 'a' && st[i] != 'A'); cout << "Please enter a valid sequence" << i++; else if (st[i] == c,C,G,t,T,a,A) i++; The second half of my program is to count the number of Cs and Gs in the sequence for (i < st.size() ; i++ ;); for (loop <= st.size() ; loop++;) if (st[loop] == 'c') { count_c++; } else if (st[loop] == C) { count_c++; } else if (st[loop] == g) { count_g++; } else if (st[loop] == G); { count_g++; } cout << "Number of instances of C = " << count_c; cout << "Number of instances of G = " << count_g; It seems like it's not looping, it will count 1 of one of the letters. How do I make it loop? I can't seem to put in endl; anywhere without getting an error back, although I know I'll need it somewhere. Any help or tips to point me in the right direction would be greatly appreciated - I've been working on this code for two days (this is embarrassing to admit).

    Read the article

  • How does Page.IsValid work?

    - by Lijo
    I have following code with a RequiredFieldValidator. The EnableClientScript property is set as "false" in the validation control. Also I have disabled script in browser. I am NOT using Page.IsValid in code behind. Still, when I submit without any value in textbox I will get error message. From comments of @Dai, I came to know that this can be an issue, if there is any code in Page_Load that is executed in a postback. There will be no validation errors thrown. (However, for button click handler, there is no need to check Page.IsValid) if (Page.IsPostBack) { string value = txtEmpName.Text; txtEmpName.Text = value + "Appended"; } QUESTION Why the server side validation does not happen before Page_Load? Why it works fine when I use Page.IsValid? UPDATE It seems like, we need to add If(Page.IsValid) in button click also if we are using a Custom Validator with server side validation. Refer CustomValidator not working well. Note: Client side validation question is present here: Whether to use Page_IsValid or Page_ClientValidate() (for Client Side Events) MARKUP <html xmlns="http://www.w3.org/1999/xhtml"> <head runat="server"> <title></title> <script type="text/javascript"> alert('haiii'); </script> </head> <body> <form id="form1" runat="server"> <div> <asp:ValidationSummary runat="server" ID="vsumAll" DisplayMode="BulletList" CssClass="validationsummary" ValidationGroup="ButtonClick" /> <asp:TextBox ID="txtEmpName" runat="server"></asp:TextBox> <asp:RequiredFieldValidator ID="valEmpName" runat="server" ControlToValidate="txtEmpName" EnableClientScript="false" ErrorMessage="RequiredFieldValidator" Text="*" Display="Dynamic" ValidationGroup="ButtonClick"></asp:RequiredFieldValidator> <asp:Button ID="Button1" runat="server" onclick="Button1_Click" Text="Button" ValidationGroup="ButtonClick" /> </div> </form> </body> </html> CODE BEHIND protected void Button1_Click(object sender, EventArgs e) { string value = txtEmpName.Text; SubmitEmployee(value); } References: Should I always call Page.IsValid? ASP.NET Validation Controls – Important Points, Tips and Tricks CustomValidator not working well

    Read the article

  • Calling managed code from unmanaged win32 assembly dll - crash

    - by JustGreg
    I'm developing a serial port dll in win32 assembly (MASM32). It has its own thread checking multiple events and at a specified buffer treshold it'd notify the managed main application by calling a callback function. It just a call with no arguments/return value. At startup the main application stores the callback function's address by calling a function in the dll: pCallBackFunction dd 0 SetCallBackPointer proc pcb:DWORD mov eax, pcb mov pCallBackFunction, eax call DWORD ptr pCallBackFunction ; verify it immediately ret SetCallBackPointer endp The upper function immediately calls back the managed application callback routine for verification purposes. It is working fine. However, when I place the call instruction to other functions in the dll it crashes the application. It doesn't matter if the call is in a simple function or in the threadproc of the dll. For example: OpenPort proc pn:byte,br:dword, inputbuffersize: dword, outputbuffersize:dword, tresholdsize: dword LOCAL dcb: DCB LOCAL SerialTimeOuts: COMMTIMEOUTS call DWORD ptr pCallBackFunction xor eax, eax mov al, pn mov [com_port+3],al etc. etc. will crash at call DWORD ptr pCallBackFunction always. Since I call SetCallBackPointer first to store a valid address in pCallBackFunction, it should have a valid address. My managed app is written in C# and the relevant part is: public partial class Form1 : Form { public delegate void CallBackDelegate(); public static CallBackDelegate mydelegate; [DllImport("serialport.dll")] private static extern void SetCallBackPointer(CallBackDelegate Delegate); [DllImport("serialport.dll")] public static extern int OpenPort(byte com, uint br, uint inbufsize, uint outbufsize, uint treshsize); public Form1() { InitializeComponent(); mydelegate =new CallBackDelegate(CallbackFunction); SetCallBackPointer(mydelegate); unsafe { int sysstat; int hResult; hResult = OpenPort(Convert.ToByte('5'), 9600, 306, 4, 4); } } public static void CallbackFunction() { MessageBox.Show( "CallBack Function Called by Windows DLL"); } The VS debugger reported that the dll had tried to read/write from/to a protected memory address. But when calling SetCallBackPointer there is no such problem. What am I doing wrong here? Any tips would be great!

    Read the article

  • Making a grid on iPhone using Opengl

    - by TheGambler
    I'm trying to make a grid similar to what you would see in these geo games(geoDefense/geometry wars). I'm wanting to apply separate transformation matrixes to each to create different effects. So, it makes since to me that I need to draw each square separately to that I can apply a different transformation to each one. The problem I'm having is mapping these grids out or connecting the squares. The coordinate systems is still confusing to me. I know it would be easy just to create a huge triangle strip but I'm not able to figure out a way to apply separate transformations to each square( quad ) if I use triangle strips. So first question: Can you apply different transformations to quads if you use a triangle strip to draw a huge grid? If so, any tips suggestions on how to do so? If not, how does one usually connect textures without using triangle strips? Here is my coord setup: const GLfloat zNear = 0.01, zFar = 1000.0, fieldOfView = 45.0; GLfloat size; glEnable(GL_DEPTH_TEST); glMatrixMode(GL_PROJECTION); size = zNear * tanf(DEGREES_TO_RADIANS(fieldOfView) / 2.0); CGRect rect = view.bounds; glFrustumf(-size, size, -size / (rect.size.width / rect.size.height), size / (rect.size.width / rect.size.height), zNear, zFar); glViewport(0, 0, rect.size.width, rect.size.height); Here is my draw: - (void)drawView:(GLView*)view; { int loop; //glColor4f(1.0, 1.0, 1.0, 0.5); glClear(GL_COLOR_BUFFER_BIT | GL_DEPTH_BUFFER_BIT); //Grid Loop for( loop = 0; loop < kNumberSectionsInGrid; loop++ ) { //glLoadIdentity(); static Matrix3D shearMatrix; Matrix3DSetShear(shearMatrix, 0.2, 0.2); static Matrix3D finalMatrix; //Matrix3DMultiply(temp2Matrix, shearMatrix, finalMatrix); glLoadMatrixf(shearMatrix); glTranslatef(-1.0f,(float)loop,-3.0f); glScalef(0.1, 0.1, 0.0); Vertex3D vertices[] = { {-1.0, 1.0, 0.5}, { 1.0, 1.0, 0.5}, { -1.0, -1.0, 0.5}, { 1.0, -1.0, 0.5} }; static const Vector3D normals[] = { {0.0, 0.0, 1.0}, {0.0, 0.0, 1.0}, {0.0, 0.0, 1.0}, {0.0, 0.0, 1.0}, }; GLfloat texCoords[] = { 0.0, 1.0, 1.0, 1.0, 0.0, 0.0, 1.0, 0.0 }; glEnableClientState(GL_VERTEX_ARRAY); glEnableClientState(GL_NORMAL_ARRAY); glEnableClientState(GL_TEXTURE_COORD_ARRAY); glBindTexture(GL_TEXTURE_2D, texture[0]); glVertexPointer(3, GL_FLOAT, 0, vertices); glNormalPointer(GL_FLOAT, 0, normals); glTexCoordPointer(2, GL_FLOAT, 0, texCoords); glDrawArrays(GL_TRIANGLE_STRIP, 0, 4); glDisableClientState(GL_VERTEX_ARRAY); glDisableClientState(GL_NORMAL_ARRAY); glDisableClientState(GL_TEXTURE_COORD_ARRAY); } } glMatrixMode(GL_MODELVIEW);

    Read the article

  • c++ recursion with arrays

    - by Sam
    I have this project that I'm working on for a class, and I'm not looking for the answer, but maybe just a few tips since I don't feel like I'm using true recursion. The problem involves solving the game of Hi Q or "peg solitaire" where you want the end result to be one peg remaining (it's played with a triangular board and one space is open at the start.) I've represented the board with a simple array, each index being a spot and having a value of 1 if it has a peg, and 0 if it doesn't and also the set of valid moves with a 2 dimensional array that is 36, 3 (each move set contains 3 numbers; the peg you're moving, the peg it hops over, and the destination index.) So my problem is that in my recursive function, I'm using a lot of iteration to determine things like which space is open (or has a value of 0) and which move to use based on which space is open by looping through the arrays. Secondly, I don't understand how you would then backtrack with recursion, in the event that an incorrect move was made and you wanted to take that move back in order to choose a different one. This is what I have so far; the "a" array represents the board, the "b" array represents the moves, and the "c" array was my idea of a reminder as to which moves I used already. The functions used are helper functions that basically just loop through the arrays to find an empty space and corresponding move. : void solveGame(int a[15], int b[36][3], int c[15][3]){ int empSpace; int moveIndex; int count = 0; if(pegCount(a) < 2){ return; } else{ empSpace = findEmpty(a); moveIndex = chooseMove( a, b, empSpace); a[b[moveIndex][0]] = 0; a[b[moveIndex][1]] = 0; a[b[moveIndex][2]] = 1; c[count][0] = b[moveIndex][0]; c[count][1] = b[moveIndex][1]; c[count][2] = b[moveIndex][2]; solveGame(a,b,c); } }

    Read the article

  • avoiding enums as interface identifiers c++ OOP

    - by AlasdairC
    Hi I'm working on a plugin framework using dynamic loaded shared libraries which is based on Eclipse's (and probally other's) extension-point model. All plugins share similar properties (name, id, version etc) and each plugin could in theory satisfy any extension-point. The actual plugin (ie Dll) handling is managed by another library, all I am doing really is managing collections of interfaces for the application. I started by using an enum PluginType to distinguish the different interfaces, but I have quickly realised that using template functions made the code far cleaner and would leave the grunt work up to the compiler, rather than forcing me to use lots of switch {...} statements. The only issue is where I need to specify like functionality for class members - most obvious example is the default plugin which provides a particular interface. A Settings class handles all settings, including the default plugin for an interface. ie Skin newSkin = settings.GetDefault<ISkin>(); How do I store the default ISkin in a container without resorting to some other means of identifying the interface? As I mentioned above, I currently use a std::map<PluginType, IPlugin> Settings::defaults member to achieve this (where IPlugin is an abstract base class which all plugins derive from. I can then dynamic_cast to the desired interface when required, but this really smells of bad design to me and introduces more harm than good I think. would welcome any tips edit: here's an example of the current use of default plugins typedef boost::shared_ptr<ISkin> Skin; typedef boost::shared_ptr<IPlugin> Plugin; enum PluginType { skin, ..., ... } class Settings { public: void SetDefault(const PluginType type, boost::shared_ptr<IPlugin> plugin) { m_default[type] = plugin; } boost::shared_ptr<IPlugin> GetDefault(const PluginType type) { return m_default[type]; } private: std::map<PluginType, boost::shared_ptr<IPlugin> m_default; }; SkinManager::Initialize() { Plugin thedefault = g_settings.GetDefault(skinplugin); Skin defaultskin = boost::dynamic_pointer_cast<ISkin>(theskin); defaultskin->Initialize(); } I would much rather call the getdefault as the following, with automatic casting to the derived class. However I need to specialize for every class type. template<> Skin Settings::GetDefault<ISkin>() { return boost::dynamic_pointer_cast<ISkin>(m_default(skin)); }

    Read the article

  • Efficiently fetching and storing tweets from a few hundred twitter profiles?

    - by MSpreij
    The site I'm working on needs to fetch the tweets from 150-300 people, store them locally, and then list them on the front page. The profiles sit in groups. The pages will be showing the last 20 tweets (or 21-40, etc) by date, group of profiles, single profile, search, or "subject" (which is sort of a different group.. I think..) a live, context-aware tag cloud (based on the last 300 tweets of the current search, group of profiles, or single profile shown) various statistics (group stuffs, most active, etc) which depend on the type of page shown. We're expecting a fair bit of traffic. The last, similar site peaked at nearly 40K visits per day, and ran intro trouble before I started caching pages as static files, and disabling some features (some, accidently..). This was caused mostly by the fact that a page load would also fetch the last x tweets from the 3-6 profiles which had not been updated the longest.. With this new site I can fortunately use cron to fetch tweets, so that helps. I'll also be denormalizing the db a little so it needs less joins, optimize it for faster selects instead of size. Now, main question: how do I figure out which profiles to check for new tweets in an efficient manner? Some people will be tweeting more often than others, some will tweet in bursts (this happens a lot). I want to keep the front page of the site as "current" as possible. If it comes to, say, 300 profiles, and I check 5 every minute, some tweets will only appear an hour after the fact. I can check more often (up to 20K) but want to optimize this as much as possible, both to not hit the rate limit and to not run out of resources on the local server (it hit mysql's connection limit with that other site). Question 2: since cron only "runs" once a minute, I figure I have to check multiple profiles each minute - as stated, at least 5, possibly more. To try and spread it out over that minute I could have it sleep a few seconds between batches or even single profiles. But then if it takes longer than 60 seconds altogether, the script will run into itself. Is this a problem? If so, how can I avoid that? Question 3: any other tips? Readmes? URLs?

    Read the article

  • Please Describe Your Struggles with Minimizing Use of Global Variables

    - by MetaHyperBolic
    Most of the programs I write are relatively flowchartable processes, with a defined start and hoped-for end. The problems themselves can be complex but do not readily lean towards central use of objects and event-driven programming. Often, I am simply churning through great varied batches of text data to produce different text data. Only occasionally do I need to create a class: As an example, to track warnings, errors, and debugging message, I created a class (Problems) with one instantiation (myErr), which I believe to be an example of the Singleton design pattern. As a further factor, my colleagues are more old school (procedural) than I and are unacquainted with object-oriented programming, so I am loath to create things they could not puzzle through. And yet I hear, again and again, how even the Singleton design pattern is really an anti-pattern and ought to be avoided because Global Variables Are Bad. Minor functions need few arguments passed to them and have no need to know of configuration (unchanging) or program state (changing) -- I agree. However, the functions in the middle of the chain, which primarily control program flow, have a need for a large number of configuration variables and some program state variables. I believe passing a dozen or more arguments along to a function is a "solution," but hardly an attractive one. I could, of course, cram variables into a single hash/dict/associative array, but that seems like cheating. For instance, connecting to the Active Directory to make a new account, I need such configuration variables as an administrative username, password, a target OU, some default groups, a domain, etc. I would have to pass those arguments down through a variety of functions which would not even use them, merely shuffle them off down through a chain which would eventually lead to the function that actually needs them. I would at least declare the configuration variables to be constant, to protect them, but my language of choice these days (Python) provides no simple manner to do this, though recipes do exist as workarounds. Numerous Stack Overflow questions have hit on the why? of the badness and the requisite shunning, but do not often mention tips on living with this quasi-religious restriction. How have you resolved, or at least made peace with, the issue of global variables and program state? Where have you made compromises? What have your tricks been, aside from shoving around flocks of arguments to functions?

    Read the article

  • css displaying divs inside a Tr fine in FireFox, but IE increases the width of the row...

    - by Ronedog
    I'm having some difficulty figuring out what is going on and how to fix it. I have some divs that contain some icons that are causing a width problem when displaying tabular data. If I remove the div's that contain the icons, the row width ends up the way I want it (See the firefox example below). Here's the view in Firefox which is what I want (notice the positioning of the icons circled in red, which is aligned on the same y coordinate, or thereabouts): Here's the view in IE7 (Notice what happens to the icons and the width of the grey line, which is the table row): Here's the HTML: <table> <tbody> <tr> <td> <span>stuff 1</span> <span>stuff 2</span> <div class="prop_edit"><img class="img_height14" src="edit.jpg"></div> <div class="prop_archive"><img class="img_height14" src="archive.jpg"></div> <div class="prop_delete"><img class="img_height14" src="delete.jpg"></div> <div style="display:none;"> <div>Links Here</div> </div> </td> </tr> </tbody> Heres the CSS: .prop_edit{ float:right; position: relative; top: 0px; right:50px; } .prop_archive{ float:right; position: relative; top: 0px; right:10px; } .prop_delete{ float:right; position: relative; top: 0px; right: -30px; } .img_height14{ height:14px; vertical-align:top; position:relative; } I've tried a bunch of different css things, but really am just hacking away hoping to figure something out. Anyone got some tips that could help me? Thanks in advance.

    Read the article

  • setting value to a parameter - always saying that it is null REALLY NEED INPUT

    - by Amina
    So I have this schedule visit page with two groups. Group 1 contains a list of: Cases. Group 2 contains a list of: Parties. each group has a checkboxes next to its item for the user to select. My Issue when i select a case and/or a party and save -- then go to the edit page of the visit i just saved and only my selected case is checked and the party i selected is not checked. After debugging i realized that the partyId is not being saved properly during the create page and thus not showing as selected or saved on the edit page. I really need help on how to properly save the party selected and setting for it a value from the parameter. Here is my code of what i have for saving a case and would like to know how to properly save with party. Controller [HttpPost] [ValidateInput(false)] public ActionResult Create(VisitViewModel viewModel, Guid[] associatedCasesSelected, Guid[] selectedParties) { if (!ModelState.IsValid) { viewModel.Time = _timeEntryHelper.Value; AddLookupsToViewModel(viewModel); return View(viewModel); } var visitEntry = Mapper.Map<VisitViewModel, VisitEntry>(viewModel); ... viewModel.CasePartyIds = selectedParties; try { _visitEntryService.Create(visitEntry, associatedCasesSelected); this.FlashInfo(string.Format(Message.ConfirmationMessageCreate, Resources.Entities.Visit.EntityName)); } catch (RulesException ex) { ex.CopyTo(ModelState); } if (ModelState.IsValid) return RedirectToAction("Edit", "Case", new { caseId = viewModel.CaseId }); AddLookupsToViewModel(viewModel); return View(viewModel); } VisitEntryService public void Create(VisitEntry visitEntry,IList<Guid>caseIds) { EnsureValid(visitEntry); _visitEntryRepository.Save(visitEntry); caseIds = AddCurrentCaseToCases(visitEntry.CaseId, caseIds); foreach (var caseId in caseIds.Distinct()) { var visit = new Visit {CaseId = caseId, VisitEntryId = visitEntry.VisitEntryId}; _visitService.Create(visit); } } AddCurrentCaseToCases private static IList<Guid>AddCurrentCaseToCases(Guid caseId, IEnumerable<Guid>caseIds) { var cases = new List<Guid>(); if (caseIds != null) { cases.AddRange(caseIds); if(!caseIds.Contains(caseId)) cases.Add(caseId); } else cases.Add(caseId); return cases; } VisitService public Visit Get(Guid visitId) { return DataContext.Visits.SingleOrDefault(v => v.VisitId == visitId); } public void Save(Visit visit) { if(visit.VisitId == Guid.Empty) { visit.VisitId = Guid.NewGuid(); DataContext.Visits.InsertOnSubmit(visit); } else { var currentVisit = Get(visit.VisitId); if (currentVisit == null) throw RepositoryExceptionFactory.Create("Visit", "VisitId"); } DataContext.SubmitChanges(); } Any TIPS or IDEAS is greatly appreciated at this time :) The entitiy for the parties will be VisitEntryParty

    Read the article

  • How can I find a URL called [link] inside a block of HTML containing other URLs?

    - by DrTwox
    I'm writing a script to rewrite Reddit's RSS feeds. The script needs to find a URL named [link] inside a block of HTML that contains other URLs. The HTML is contained in an XML element called <description>. Here are two examples of the <description> element from I need to parse and the [link] I would need to get. First example: <description>submitted by &lt;a href=&#34;http://www.reddit.com/user/wildlyinaccurate&#34;&gt; wildlyinaccurate &lt;/a&gt; &lt;br/&gt; &lt;a href=&#34;http://wildlyinaccurate.com/a-hackers-guide-to-git&#34;&gt;[link]&lt;/a&gt; &lt;a href="http://www.reddit.com/r/programming/comments/26jvl7/a_hackers_guide_to_git/"&gt;[66 comments]&lt;/a&gt;</description> The [link] is: http://wildlyinaccurate.com/a-hackers-guide-to-git Second example: <description>&lt;!-- SC_OFF --&gt;&lt;div class=&#34;md&#34;&gt;&lt;p&gt;I work a support role at a company where I primarily fix issues our customers our experiencing with our software, which is a browser based application written primarily in javascript. I&amp;#39;ve been doing this for 2 years, but I want to take it to the next level (with the long term goal being that I become proficient enough to call myself a developer). I&amp;#39;ve been reading &amp;quot;Javascript The Definitive Guide&amp;quot; by O&amp;#39;Reilly but I was wondering if any of you more experienced users out there had some tips on taking it to the next level. Should I start incorporating some PHP and Jquery into my learning? Side projects on my spare time? Any good online resources? Etc. &lt;/p&gt; &lt;p&gt;Thanks! &lt;/p&gt; &lt;/div&gt;&lt;!-- SC_ON --&gt; submitted by &lt;a href=&#34;http://www.reddit.com/user/56killa&#34;&gt; 56killa &lt;/a&gt; &lt;br/&gt; &lt;a href=&#34;http://www.reddit.com/r/javascript/comments/26nduc/i_want_to_become_more_experienced_with_javascript/&#34;&gt;[link]&lt;/a&gt; &lt;a href="http://www.reddit.com/r/javascript/comments/26nduc/i_want_to_become_more_experienced_with_javascript/"&gt;[4 comments]&lt;/a&gt;</description> The [link] is: http://www.reddit.com/r/javascript/comments/26nduc/i_want_to_become_more_experienced_with_javascript/

    Read the article

  • Is it thread safe to read a form controls value (but not change it) without using Invoke/BeginInvoke from another thread

    - by goku_da_master
    I know you can read a gui control from a worker thread without using Invoke/BeginInvoke because my app is doing it now. The cross thread exception error is not being thrown and my System.Timers.Timer thread is able to read gui control values just fine (unlike this guy: can a worker thread read a control in the GUI?) Question 1: Given the cardinal rule of threads, should I be using Invoke/BeginInvoke to read form control values? And does this make it more thread-safe? The background to this question stems from a problem my app is having. It seems to randomly corrupt form controls another thread is referencing. (see question 2) Question 2: I have a second thread that needs to update form control values so I Invoke/BeginInvoke to update those values. Well this same thread needs a reference to those controls so it can update them. It holds a list of these controls (say DataGridViewRow objects). Sometimes (not always), the DataGridViewRow reference gets "corrupt". What I mean by corrupt, is the reference is still valid, but some of the DataGridViewRow properties are null (ex: row.Cells). Is this caused by question 1 or can you give me any tips on why this might be happening? Here's some code (the last line has the problem): public partial class MyForm : Form { void Timer_Elapsed(object sender) { // we're on a new thread (this function gets called every few seconds) UpdateUiHelper updateUiHelper = new UpdateUiHelper(this); foreach (DataGridViewRow row in dataGridView1.Rows) { object[] values = GetValuesFromDb(); updateUiHelper.UpdateRowValues(row, values[0]); } // .. do other work here updateUiHelper.UpdateUi(); } } public class UpdateUiHelper { private readonly Form _form; private Dictionary<DataGridViewRow, object> _rows; private delegate void RowDelegate(DataGridViewRow row); private readonly object _lockObject = new object(); public UpdateUiHelper(Form form) { _form = form; _rows = new Dictionary<DataGridViewRow, object>(); } public void UpdateRowValues(DataGridViewRow row, object value) { if (_rows.ContainsKey(row)) _rows[row] = value; else { lock (_lockObject) { _rows.Add(row, value); } } } public void UpdateUi() { foreach (DataGridViewRow row in _rows.Keys) { SetRowValueThreadSafe(row); } } private void SetRowValueThreadSafe(DataGridViewRow row) { if (_form.InvokeRequired) { _form.Invoke(new RowDelegate(SetRowValueThreadSafe), new object[] { row }); return; } // now we're on the UI thread object newValue = _rows[row]; row.Cells[0].Value = newValue; // randomly errors here with NullReferenceException, but row is never null! }

    Read the article

  • Line Graph CGPoints from NSMutableArray

    - by Mattog1456
    I have been trying to adapt the code from the accelerometer example such that when the user depresses a uibutton a point is added to a line graph. Working on the converting two floats, which are the result of calculate as below into a CGPoint and converting the CGPoint into an NSValue and then adding this to a NSMutableArray with the following -(IBAction)calculate:(id)sender { self.points = [NSMutableArray array]; CGPoint pt = CGPointMake(d, c); [self.points addObject:[NSValue valueWithCGPoint:pt]]; NSLog(@"%@", NSStringFromCGPoint(pt)); NSLog(@"%@", [NSString stringWithFormat:@"%d points", self.points.count ]); } But for some reason I am only getting one object stored in the array, it seems everytime push the calculate button the object pt gets overwritten, on the plus side it has the correct x,y coords. Any ideas on this one? UPDATE Removed self.points = [NSMutableArray array]; and placed it in view did load, also set the first points to 0,0. so that is working ok. Now the next problem is that I have a Graph subclass where the CG Drawing is taking place. I am trying to figure out a simple way to be able to access the above NSMutableArray which is in a ViewController class from the graph class. Am so close to the end but am really stuck, any help would be great. Still trying to draw a line graph on a UIView which is on a UIScrollview. The draw rect method is in the UIView Subclass and everything is working there, I have gridlines and labels on the axis and I can draw manually onto it. But the problem I have is that I cannot read the NSMutableArray of the CGPoints, which are the x and y coords. The ViewController performs a calculation and the results are written to the NSMutable array and this is all working fine as well, I can see the CGpoints and their values being written with NSLogs in the ViewController. I have tried various ways to set the NSMutableArray up as a global but to no avail, everything runs but while I can see the points being written in the ViewController they are just not visible to the UIView Subclass. I have also tried to use the addObserver and observeValueForKeyPath methods and once again while everything runs the subclass cannot see the array. Any ideas, suggestions, tips or thoughts would be great

    Read the article

  • Michael Crump&rsquo;s notes for 70-563 PRO &ndash; Designing and Developing Windows Applications usi

    - by mbcrump
    TIME TO GO PRO! This is my notes for 70-563 PRO – Designing and Developing Windows Applications using .NET Framework 3.5 I created it using several resources (various certification web sites, msdn, official ms 70-548 book). The reason that I created this review is because a) I am taking the exam. b) MS did not create a book for this exam. Use the(MS 70-548)book. c) To make sure I am familiar with each before the exam. I hope that it provides a good start for your own notes. I hope that someone finds this useful. At least, it will give you a starting point of what to expect to know on the PRO exam. Also, for those wondering, the PRO exam does contains very little code. It is basically all theory. 1. Validation Controls – How to prevent users from entering invalid data on forms. (MaskedTextBox control and RegEx) 2. ServiceController – used to start and control the behavior of existing services. 3. User Feedback (know winforms Status Bar, Tool Tips, Color, Error Provider, Context-Sensitive and Accessibility) 4. Specific (derived) exceptions must be handled before general (base class) exceptions. By moving the exception handling for the base type Exception to after exception handling of ArgumentNullException, all ArgumentNullException thrown by the Helper method will be caught and logged correctly. 5. A heartbeat method is a method exposed by a Web service that allows external applications to check on the status of the service. 6. New users must master key tasks quickly. Giving these tasks context and appropriate detail will help. However, advanced users will demand quicker paths. Shortcuts, accelerators, or toolbar buttons will speed things along for the advanced user. 7. MSBuild uses project files to instruct the build engine what to build and how to build it. MSBuild project files are XML files that adhere to the MSBuild XML schema. The MSBuild project files contain complete file, build action, and dependency information for each individual projects. 8. Evaluating whether or not to fix a bug involves a triage process. You must identify the bug's impact, set the priority, categorize it, and assign a developer. Many times the person doing the triage work will assign the bug to a developer for further investigation. In fact, the workflow for the bug work item inside of Team System supports this step. Developers are often asked to assess the impact of a given bug. This assessment helps the person doing the triage make a decision on how to proceed. When assessing the impact of a bug, you should consider time and resources to fix it, bug risk, and impacts of the bug. 9. In large projects it is generally impossible and unfeasible to fix all bugs because of the impact on schedule and budget. 10. Code reviews should be conducted by a technical lead or a technical peer. 11. Testing Applications 12. WCF Services – application state 13. SQL Server 2005 / 2008 Express Edition – reliable storage of data / Microsoft SQL Server 3.5 Compact Database– used for client computers to retrieve and save data from a shared location. 14. SQL Server 2008 Compact Edition – used for minimum possible memory and can synchronize data with a corporate SQL Server 2008 Database. Supports offline user and minimum dependency on external components. 15. MDI and SDI Forms (specifically IsMDIContainer) 16. GUID – in the case of data warehousing, it is important to define unique keys. 17. Encrypting / Security Data 18. Understanding of Isolated Storage/Proper location to store items 19. LINQ to SQL 20. Multithreaded access 21. ADO.NET Entity Framework model 22. Marshal.ReleaseComObject 23. Common User Interface Layout (ComboBox, ListBox, Listview, MaskedTextBox, TextBox, RichTextBox, SplitContainer, TableLayoutPanel, TabControl) 24. DataSets Class - http://msdn.microsoft.com/en-us/library/system.data.dataset%28VS.71%29.aspx 25. SQL Server 2008 Reporting Services (SSRS) 26. SystemIcons.Shield (Vista UAC) 27. Leverging stored procedures to perform data manipulation for a database schema that can change. 28. DataContext 29. Microsoft Windows Installer Packages, ClickOnce(bootstrapping features), XCopy. 30. Client Application Services – will authenticate users by using the same data source as a ASP.NET web application. 31. SQL Server 2008 Caching 32. StringBuilder 33. Accessibility Guidelines for Windows Applications http://msdn.microsoft.com/en-us/library/ms228004.aspx 34. Logging erros 35. Testing performance related issues. 36. Role Based Security, GenericIdentity and GenericPrincipal 37. System.Net.CookieContainer will store session data for webapps (see isolated storage for winforms) 38. .NET CLR Profiler tool will identify objects that cause performance issues. 39. ADO.NET Synchronization (SyncGroup) 40. Globalization - CultureInfo 41. IDisposable Interface- reports on several questions relating to this. 42. Adding timestamps to determine whether data has changed or not. 43. Converting applications to .NET Framework 3.5 44. MicrosoftReportViewer 45. Composite Controls 46. Windows Vista KNOWN folders. 47. Microsoft Sync Framework 48. TypeConverter -Provides a unified way of converting types of values to other types, as well as for accessing standard values and sub properties. http://msdn.microsoft.com/en-us/library/system.componentmodel.typeconverter.aspx 49. Concurrency control mechanisms The main categories of concurrency control mechanisms are: Optimistic - Delay the checking of whether a transaction meets the isolation rules (e.g., serializability and recoverability) until its end, without blocking any of its (read, write) operations, and then abort a transaction, if the desired rules are violated. Pessimistic - Block operations of a transaction, if they may cause violation of the rules. Semi-optimistic - Block operations in some situations, and do not block in other situations, while delaying rules checking to transaction's end, as done with optimistic. 50. AutoResetEvent 51. Microsoft Messaging Queue (MSMQ) 4.0 52. Bulk imports 53. KeyDown event of controls 54. WPF UI components 55. UI process layer 56. GAC (installing, removing and queuing) 57. Use a local database cache to reduce the network bandwidth used by applications. 58. Sound can easily be annoying and distracting to users, so use it judiciously. Always give users the option to turn sound off. Because a user might have sound off, never convey important information through sound alone.

    Read the article

  • ASP.NET and WIF: Showing custom profile username as User.Identity.Name

    - by DigiMortal
    I am building ASP.NET MVC application that uses external services to authenticate users. For ASP.NET users are fully authenticated when they are redirected back from external service. In system they are logically authenticated when they have created user profiles. In this posting I will show you how to force ASP.NET MVC controller actions to demand existence of custom user profiles. Using external authentication sources with AppFabric Suppose you want to be user-friendly and you don’t force users to keep in mind another username/password when they visit your site. You can accept logins from different popular sites like Windows Live, Facebook, Yahoo, Google and many more. If user has account in some of these services then he or she can use his or her account to log in to your site. If you have community site then you usually have support for user profiles too. Some of these providers give you some information about users and other don’t. So only thing in common you get from all those providers is some unique ID that identifies user in service uniquely. Image above shows you how new user joins your site. Existing users who already have profile are directed to users homepage after they are authenticated. You can read more about how to solve semi-authorized users problem from my blog posting ASP.NET MVC: Using ProfileRequiredAttribute to restrict access to pages. The other problem is related to usernames that we don’t get from all identity providers. Why is IIdentity.Name sometimes empty? The problem is described more specifically in my blog posting Identifying AppFabric Access Control Service users uniquely. Shortly the problem is that not all providers have claim called http://schemas.xmlsoap.org/ws/2005/05/identity/claims/name. The following diagram illustrates what happens when user got token from AppFabric ACS and was redirected to your site. Now, when user was authenticated using Windows Live ID then we don’t have name claim in token and that’s why User.Identity.Name is empty. Okay, we can force nameidentifier to be used as name (we can do it in web.config file) but we have user profiles and we want username from profile to be shown when username is asked. Modifying name claim Now let’s force IClaimsIdentity to use username from our user profiles. You can read more about my profiles topic from my blog posting ASP.NET MVC: Using ProfileRequiredAttribute to restrict access to pages and you can find some useful extension methods for claims identity from my blog posting Identifying AppFabric Access Control Service users uniquely. Here is what we do to set User.Identity.Name: we will check if user has profile, if user has profile we will check if User.Identity.Name matches the name given by profile, if names does not match then probably identity provider returned some name for user, we will remove name claim and recreate it with correct username, we will add new name claim to claims collection. All this stuff happens in Application_AuthorizeRequest event of our web application. The code is here. protected void Application_AuthorizeRequest() {     if (string.IsNullOrEmpty(User.Identity.Name))     {         var identity = User.Identity;         var profile = identity.GetProfile();         if (profile != null)         {             if (profile.UserName != identity.Name)             {                 identity.RemoveName();                   var claim = new Claim("http://schemas.xmlsoap.org/ws/2005/05/identity/claims/name", profile.UserName);                 var claimsIdentity = (IClaimsIdentity)identity;                 claimsIdentity.Claims.Add(claim);             }         }     } } RemoveName extension method is simple – it looks for name claims of IClaimsIdentity claims collection and removes them. public static void RemoveName(this IIdentity identity) {     if (identity == null)         return;       var claimsIndentity = identity as ClaimsIdentity;     if (claimsIndentity == null)         return;       for (var i = claimsIndentity.Claims.Count - 1; i >= 0; i--)     {         var claim = claimsIndentity.Claims[i];         if (claim.ClaimType == "http://schemas.xmlsoap.org/ws/2005/05/identity/claims/name")             claimsIndentity.Claims.RemoveAt(i);     } } And we are done. Now User.Identity.Name returns the username from user profile and you can use it to show username of current user everywhere in your site. Conclusion Mixing AppFabric Access Control Service and Windows Identity Foundation with custom authorization logic is not impossible but a little bit tricky. This posting finishes my little series about AppFabric ACS and WIF for this time and hopefully you found some useful tricks, tips, hacks and code pieces you can use in your own applications.

    Read the article

  • Win a place at a SQL Server Masterclass with Kimberly Tripp and Paul Randal

    - by Testas
    The top things YOU need to know about managing SQL Server - in one place, on one day - presented by two of the best SQL Server industry trainers!And you could be there courtesy of UK SQL Server User Group and SQL Server Magazine! This week the UK SQL Server User Group will provide you with details of how to win a place at this must see seminar   You can also register for the seminar yourself at:www.regonline.co.uk/kimtrippsql More information about the seminar   Where: Radisson Edwardian Heathrow Hotel, London When: Thursday 17th June 2010 This one-day MasterClass will focus on many of the top issues companies face when implementing and maintaining a SQL Server-based solution. In the case where a company has no dedicated DBA, IT managers sometimes struggle to keep the data tier performing well and the data available. This can be especially troublesome when the development team is unfamiliar with the affect application design choices have on database performance. The Microsoft SQL Server MasterClass 2010 is presented by Paul S. Randal and Kimberly L. Tripp, two of the most experienced and respected people in the SQL Server world. Together they have over 30 years combined experience working with SQL Server in the field, and on the SQL Server product team itself. This is a unique opportunity to hear them present at a UK event which will:·         Debunk many of the ingrained misconceptions around SQL Server's behaviour   ·         Show you disaster recovery techniques critical to preserving your company's life-blood - the data   ·         Explain how a common application design pattern can wreak havoc in the database ·         Walk through the top-10 points to follow around operations and maintenance for a well-performing and available data tier! Please Note: Agenda may be subject to changeSessions AbstractsKEYNOTE: Bridging the Gap Between Development and Production  Applications are commonly developed with little regard for how design choices will affect performance in production. This is often because developers don't realize the implications of their design on how SQL Server will be able to handle a high workload (e.g. blocking, fragmentation) and/or because there's no full-time trained DBA that can recognize production problems and help educate developers. The keynote sets the stage for the rest of the day. Discussing some of the issues that can arise, explaining how some can be avoided and highlighting some of the features in SQL 2008 that can help developers and DBAs make better use of SQL Server, and troubleshoot when things go wrong.  SESSION ONE: SQL Server MythbustersIt's amazing how many myths and misconceptions have sprung up and persisted over the years about SQL Server - after many years helping people out on forums, newsgroups, and customer engagements, Paul and Kimberly have heard it all. Are there really non-logged operations? Can interrupting shrinks or rebuilds cause corruption? Can you override the server's MAXDOP setting? Will the server always do a table-scan to get a row count? Many myths lead to poor design choices and inappropriate maintenance practices so these are just a few of many, many myths that Paul and Kimberly will debunk in this fast-paced session on how SQL Server operates and should be managed and maintained. SESSION TWO: Database Recovery Techniques Demo-Fest Even if a company has a disaster recovery strategy in place, they need to practice to make sure that the plan will work when a disaster does strike. In this fast-paced demo session Paul and Kimberly will repeatedly do nasty things to databases and then show how they are recovered - demonstrating many techniques that can be used in production for disaster recovery. Not for the faint-hearted! SESSION THREE: GUIDs: Use, Abuse, and How To Move Forward Since the addition of the GUID (Microsoft’s implementation of the UUID), my life as a consultant and "tuner" has been busy. I’ve seen databases designed with GUID keys run fairly well with small workloads but completely fall over and fail because they just cannot scale. And, I know why GUIDs are chosen - it simplifies the handling of parent/child rows in your batches so you can reduce round-trips or avoid dealing with identity values. And, yes, sometimes it's even for distributed databases and/or security that GUIDs are chosen. I'm not entirely against ever using a GUID but overusing and abusing GUIDs just has to be stopped! Please, please, please let me give you better solutions and explanations on how to deal with your parent/child rows, round-trips and clustering keys! SESSION 4: Essential Database MaintenanceIn this session, Paul and Kimberly will run you through their top-ten database maintenance recommendations, with a lot of tips and tricks along the way. These are distilled from almost 30 years combined experience working with SQL Server customers and are geared towards making your databases more performant, more available, and more easily managed (to save you time!). Everything in this session will be practical and applicable to a wide variety of databases. Topics covered include: backups, shrinks, fragmentation, statistics, and much more! Focus will be on 2005 but we'll explain some of the key differences for 2000 and 2008 as well.    Speaker Biographies     Paul S.Randal  Kimberley L. Tripp Paul and Kimberly are a husband-and-wife team who own and run SQLskills.com, a world-renowned SQL Server consulting and training company. They are both SQL Server MVPs and Microsoft Regional Directors, with over 30 years of combined experience on SQL Server. Paul worked on the SQL Server team for nine years in development and management roles, writing many of the DBCC commands, and ultimately with responsibility for core Storage Engine for SQL Server 2008. Paul writes extensively on his blog (SQLskills.com/blogs/Paul) and for TechNet Magazine, for which he is also a Contributing Editor. Kimberly worked on the SQL Server team in the early 1990s as a tester and writer before leaving to found SQLskills and embrace her passion for teaching and consulting. Kimberly has been a staple at worldwide conferences since she first presented at TechEd in 1996, and she blogs at SQLskills.com/blogs/Kimberly. They have written Microsoft whitepapers and books for SQL Server 2000, 2005 and 2008, and are regular, top-rated presenters worldwide on database maintenance, high availability, disaster recovery, performance tuning, and SQL Server internals. Together they teach the SQL MCM certification and throughout Microsoft.In their spare time, they like to find frogfish in remote corners of the world.  

    Read the article

  • 202 blog articles

    - by mprove
    All my blog articles under blogs.oracle.com since August 2005: 202 blog articles Apr 2012 blogs.oracle.com design patch Mar 2012 Interaction 12 - Critique Mar 2012 Typing. Clicking. Dancing. Feb 2012 Desktop Mobility in Hospitals with Oracle VDI /video Feb 2012 Interaction 12 in Dublin - Highlights of Day 3 Feb 2012 Interaction 12 in Dublin - Highlights of Day 2 Feb 2012 Interaction 12 in Dublin - Highlights of Day 1 Feb 2012 Shit Interaction Designers Say Feb 2012 Tips'n'Tricks for WebCenter #3: How to display custom page titles in Spaces Jan 2012 Tips'n'Tricks for WebCenter #2: How to create an Admin menu in Spaces and save a lot of time Jan 2012 Tips'n'Tricks for WebCenter #1: How to apply custom resources in Spaces Jan 2012 Merry XMas and a Happy 2012! Dec 2011 One Year Oracle SocialChat - The Movie Nov 2011 Frank Ludolph's Last Working Day Nov 2011 Hans Rosling at TED Oct 2011 200 Countries x 200 Years Oct 2011 Blog Aggregation for Desktop Virtualization Oct 2011 Oracle VDI at OOW 2011 Sep 2011 Design for Conversations & Conversations for Design Sep 2011 All Oracle UX Blogs Aug 2011 Farewell Loriot Aug 2011 Oracle VDI 3.3 Overview Aug 2011 Sutherland's Closing Remarks at HyperKult Aug 2011 Surface and Subface Aug 2011 Back to Childhood in UI Design Jul 2011 The Art of Engineering and The Engineering of Art Jul 2011 Oracle VDI Seminar - June-30 Jun 2011 SGD White Paper May 2011 TEDxHamburg Live Feed May 2011 Oracle VDI in 3 Minutes May 2011 Space Ship Earth 2011 May 2011 blog moving times Apr 2011 Frozen tag cloud Apr 2011 Oracle: Hardware Software Complete in 1953 Apr 2011 Interaction Design with Wireframes Apr 2011 A guide to closing down a project Feb 2011 Oracle VDI 3.2.2 Jan 2011 free VDI charts Jan 2011 Sun Founders Panel 2006 Dec 2010 Sutherland on Leadership Dec 2010 SocialChat: Efficiency of E20 Dec 2010 ALWAYS ON Desktop Virtualization Nov 2010 12,000 Desktops at JavaOne Nov 2010 SocialChat on Sharing Best Practices Oct 2010 Globe of Visitors Oct 2010 SocialChat about the Next Big Thing Oct 2010 Oracle VDI UX Story - Wireframes Oct 2010 What's a PC anyway? Oct 2010 SocialChat on Getting Things Done Oct 2010 SocialChat on Infoglut Oct 2010 IT Twenty Twenty Oct 2010 Desktop Virtualization Webcasts from OOW Oct 2010 Oracle VDI 3.2 Overview Sep 2010 Blog Usability Top 7 Sep 2010 100 and counting Aug 2010 Oracle'izing the VDI Blogs Aug 2010 SocialChat on Apple Aug 2010 SocialChat on Video Conferencing Aug 2010 Oracle VDI 3.2 - Features and Screenshots Aug 2010 SocialChat: Don't stop making waves Aug 2010 SocialChat: Giving Back to the Community Aug 2010 SocialChat on Learning in Meetings Aug 2010 iPAD's Natural User Interface Jul 2010 Last day for Sun Microsystems GmbH Jun 2010 SirValUse Celebration Snippets Jun 2010 10 years SirValUse - Happy Birthday! Jun 2010 Wim on Virtualization May 2010 New Home for Oracle VDI Apr 2010 Renaissance Slide Sorter Comments Apr 2010 Unboxing Sun Ray 3 Plus Apr 2010 Desktop Virtualisierung mit Sun VDI 3.1 Apr 2010 Blog Relaunch Mar 2010 Social Messaging Slides from CeBIT Mar 2010 Social Messaging Talk at CeBIT Feb 2010 Welcome Oracle Jan 2010 My last presentation at Sun Jan 2010 Ivan Sutherland on Leadership Jan 2010 Learning French with Sun VDI Jan 2010 Learning Danish with Sun Ray Jan 2010 VDI workshop in Nieuwegein Jan 2010 Happy New Year 2010 Jan 2010 On Creating Slides Dec 2009 Best VDI Ever Nov 2009 How to store the Big Bang Nov 2009 Social Enterprise Tools. Beipiel Sun. Nov 2009 Nov-19 Nov 2009 PDF and ODF links on your blog Nov 2009 Q&A on VDI and MySQL Cluster Nov 2009 Zürich next week: Swiss Intranet Summit 09 Nov 2009 Designing for a Sustainable World - World Usabiltiy Day, Nov-12 Nov 2009 How to export a desktop from VDI 3 Nov 2009 Virtualisation Roadshow in the UK Nov 2009 Project Wonderland at EDUCAUSE 09 Nov 2009 VDI Roadshow in Dublin, Nov-26, 2009 Nov 2009 Sun VDI at EDUCAUSE 09 Nov 2009 Sun VDI 3.1 Architecture and New Features Oct 2009 VDI 3.1 is Early-Access Sep 2009 Virtualization for MySQL on VMware Sep 2009 Silpion & 13. Stock Sommerparty Sep 2009 Sun Ray and VMware View 3.1.1 2009-08-31 New Set of Sun Ray Status Icons 2009-08-25 Virtualizing the VDI Core? 2009-08-23 World Usability Day Hamburg 2009 - CfP 2009-07-16 Rising Sun 2009-07-15 featuring twittermeme 2009-06-19 ISC09 Student Party on June-20 /Hamburg 2009-06-18 Before and behind the curtain of JavaOne 2009-06-09 20k desktops at JavaOne 2009-06-01 sweet microblogging 2009-05-25 VDI 3 - Why you need 3 VDI hosts and what you can do about that? 2009-05-21 IA Konferenz 2009 2009-05-20 Sun VDI 3 UX Story - Power of the Web 2009-05-06 Planet of Sun and Oracle User Experience Design 2009-04-22 Sun VDI 3 UX Story - User Research 2009-04-08 Sun VDI 3 UX Story - Concept Workshops 2009-04-06 Localized documentation for Sun Ray Connector for VMware View Manager 1.1 2009-04-03 Sun VDI 3 Press Release 2009-03-25 Sun VDI 3 launches today! 2009-03-25 Sun Ray Connector for VMware View Manager 1.1 Update 2009-03-11 desktop virtualization wiki relaunch 2009-03-06 VDI 3 at CeBIT hall 6, booth E36 2009-03-02 Keyboard layout problems with Sun Ray Connector for VMware VDM 2009-02-23 wikis.sun.com tips & tricks 2009-02-23 Sun VDI 3 is in Early Access 2009-02-09 VirtualCenter unable to decrypt passwords 2009-02-02 Sun & VMware Desktop Training 2009-01-30 VDI at next09? 2009-01-16 Sun VDI: How to use virtual machines with multiple network adapters 2009-01-07 Sun Ray and VMware View 2009-01-07 Hamburg World Usability Day 2008 - Webcasts 2009-01-06 Sun Ray Connector for VMware VDM slides 2008-12-15 mother of all demos 2008-12-08 Build your own Thumper 2008-12-03 Troubleshooting Sun Ray Connector for VMware VDM 2008-12-02 My Roller Tag Cloud 2008-11-28 Sun Ray Connector: SSL connection to VDM 2008-11-25 Setting up SSL and Sun Ray Connector for VMware VDM 2008-11-13 Inspiration for Today and Tomorrow 2008-10-23 Sun Ray Connector for VMware VDM released 2008-10-14 From Sketchpad to ILoveSketch 2008-10-09 Desktop Virtualization on Xing 2008-10-06 User Experience Forum on Xing 2008-10-06 Sun Ray Connector for VMware VDM certified 2008-09-17 Virtual Clouds over Las Vegas 2008-09-14 Bill Verplank sketches metaphors 2008-09-04 End of Early Access - Sun Ray Connector for VMware 2008-08-27 Early Access: Sun Ray Connector for VMware Virtual Desktop Manager 2008-08-12 Sun Virtual Desktop Connector - Insides on Recycling Part 2 2008-07-20 Sun Virtual Desktop Connector - Insides on Recycling Part 3 2008-07-20 Sun Virtual Desktop Connector - Insides on Recycling 2008-07-20 lost in wiki space 2008-07-07 Evolution of the Desktop 2008-06-17 Virtual Desktop Webcast 2008-06-16 Woodstock 2008-06-16 What's a Desktop PC anyway? 2008-06-09 Virtual-T-Box 2008-06-05 Virtualization Glossary 2008-05-06 Five User Experience Principles 2008-04-25 Virtualization News Feed 2008-04-21 Acetylcholinesterase - Second Season 2008-04-18 Acetylcholinesterase - End of Signal 2007-12-31 Produkt-Management ist... 2007-10-22 Usability Verbände, Verteiler und Netzwerke. 2007-10-02 The Meaning is the Message 2007-09-28 Visualization Methods 2007-09-10 Inhouse und Open Source Projekte – Usability verankern und Synergien nutzen 2007-09-03 Der Schwabe Darth Vader entdeckt das Virale Marketing 2007-08-29 Dick Hardt 3.0 on Identity 2.0 2007-08-27 quality of written text depends on the tool 2007-07-27 podcasts for reboot9 2007-06-04 It is the user's itch that need to be scratched 2007-05-25 A duel at reboot9 2007-05-14 Taxonomien und Folksonomien - Tagging als neues HCI-Element 2007-05-10 Dueling Interaction Models of Personal-Computing and Web-Computing 2007-03-01 22.März: Weizenbaum. Rebel at Work. /Filmpremiere Hamburg 2007-02-25 Bruce Sterling at UbiComp 2006 /webcast 2006-11-12 FSOSS 2006 /webcasts 2006-11-10 Highway 101 2006-11-09 User Experience Roundtable Hamburg: EuroGEL 2006 2006-11-08 Douglas Adams' Hyperland (BBC 1990) 2006-10-08 Taxonomien und Folksonomien – Tagging als neues HCI-Element 2006-09-13 Usability im Unternehmen 2006-09-13 Doug does HyperScope 2006-08-26 TED Talks and TechTalks 2006-08-21 Kai Krause über seine Freundschaft zu Douglas Adams 2006-07-20 Rebel At Work: Film Portrait on Weizenbaum 2006-07-04 Gabriele Fischer, mp3 2006-06-07 Dick Hardt at ETech 06 2006-06-05 Weinberger: From Control to Conversation 2006-04-16 Eye Tracking at User Experience Roundtable Hamburg 2006-04-14 dropping knowledge 2006-04-09 GEL 2005 2006-03-13 slide photos of reboot7 2006-03-04 Dick Hardt on Identity 2.0 2006-02-28 User Experience Newsletter #13: Versioning 2006-02-03 Ester Dyson on Choice and Happyness 2006-02-02 Requirements-Engineering im Spannungsfeld von Individual- und Produktsoftware 2006-01-15 User Experience Newsletter #12: Intuition Quiz 2005-11-30 User Experience und Requirements-Engineering für Software-Projekte 2005-10-31 Ivan Sutherland on "Research and Fun" 2005-10-18 Ars Electronica / Mensch und Computer 2005 2005-09-14 60 Jahre nach Memex: Über die Unvereinbarkeit von Desktop- und Web-Paradigma 2005-08-31 reboot 7 2005-06-30

    Read the article

  • Interview with Lenz Grimmer about MySQL Connect

    - by Keith Larson
    Keith Larson: Thank you for allowing me to do this interview with you.  I have been talking with a few different Oracle ACEs   about the MySQL Connect Conference. I figured the MySQL community might be missing you as well. You have been very busy with Oracle Linux but I know you still have an eye on the MySQL Community. How have things been?Lenz Grimmer: Thanks for including me in this series of interviews, I feel honored! I've read the other interviews, and really liked them. I still try to follow what's going on over in the MySQL community and it's good to see that many of the familiar faces are still around. Over the course of the 9 years that I was involved with MySQL, many colleagues and contacts turned into good friends and we still maintain close relationships.It's been almost 1.5 years ago that I moved into my new role here in the Linux team at Oracle, and I really enjoy working on a Linux distribution again (I worked for SUSE before I joined MySQL AB in 2002). I'm still learning a lot - Linux in the data center has greatly evolved in so many ways and there are a lot of new and exciting technologies to explore. Keith Larson: What were your thoughts when you heard that Oracle was going to deliver the MySQL Connect conference to the MySQL Community?Lenz Grimmer: I think it's testament to the fact that Oracle deeply cares about MySQL, despite what many skeptics may say. What started as "MySQL Sunday" two years ago has now evolved into a full-blown sub-conference, with 80 sessions at one of the largest corporate IT events in the world. I find this quite telling, not many products at Oracle enjoy this level of exposure! So it certainly makes me feel proud to see how far MySQL has come. Keith Larson: Have you had a chance to look over the sessions? What are your thoughts on them?Lenz Grimmer: I did indeed look at the final schedule.The content committee did a great job with selecting these sessions. I'm glad to see that the content selection was influenced by involving well-known and respected members of the MySQL community. The sessions cover a broad range of topics and technologies, both covering established topics as well as recent developments. Keith Larson: When you get a chance, what sessions do you plan on attending?Lenz Grimmer: I will actually be manning the Oracle booth in the exhibition area on one of these days, so I'm not sure if I'll have a lot of time attending sessions. But if I do, I'd love to see the keynotes and catch some of the sessions that talk about recent developments and new features in MySQL, High Availability and Clustering . Quite a lot has happened and it's hard to keep up with this constant flow of new MySQL releases.In particular, the following sessions caught my attention: MySQL Connect Keynote: The State of the Dolphin Evaluating MySQL High-Availability Alternatives CERN’s MySQL “as a Service” Deployment with Oracle VM: Empowering Users MySQL 5.6 Replication: Taking Scalability and High Availability to the Next Level What’s New in MySQL Server 5.6? MySQL Security: Past and Present MySQL at Twitter: Development and Deployment MySQL Community BOF MySQL Connect Keynote: MySQL Perspectives Keith Larson: So I will ask you just like I have asked the others I have interviewed, any tips that you would give to people for handling the long hours at conferences?Lenz Grimmer: Wear comfortable shoes and make sure to drink a lot! Also prepare a plan of the sessions you would like to attend beforehand and familiarize yourself with the venue, so you can get to the next talk in time without scrambling to find the location. The good thing about piggybacking on such a large conference like Oracle OpenWorld is that you benefit from the whole infrastructure. For example, there is a nice schedule builder that helps you to keep track of your sessions of interest. Other than that, bring enough business cards and talk to people, build up your network among your peers and other MySQL professionals! Keith Larson: What features of the MySQL 5.6 release do you look forward to the most ?Lenz Grimmer: There has been solid progress in so many areas like the InnoDB Storage Engine, the Optimizer, Replication or Performance Schema, it's hard for me to really highlight anything in particular. All in all, MySQL 5.6 sounds like a very promising release. I'm confident it will follow the tradition that Oracle already established with MySQL 5.5, which received a lot of praise even from very critical members of the MySQL community. If I had to name a single feature, I'm particularly and personally happy that the precise GIS functions have finally made it into a GA release - that was long overdue. Keith Larson:  In your opinion what is the best reason for someone to attend this event?Lenz Grimmer: This conference is an excellent opportunity to get in touch with the key people in the MySQL community and ecosystem and to get facts and information from the domain experts and developers that work on MySQL. The broad range of topics should attract people from a variety of roles and relations to MySQL, beginning with Developers and DBAs, to CIOs considering MySQL as a viable solution for their requirements. Keith Larson: You will be attending MySQL Connect and have some Oracle Linux Demos, do you see a growing demand for MySQL on Oracle Linux ?Lenz Grimmer: Yes! Oracle Linux is our recommended Linux distribution and we have a good relationship to the MySQL engineering group. They use Oracle Linux as a base Linux platform for development and QA, so we make sure that MySQL and Oracle Linux are well tested together. Setting up a MySQL server on Oracle Linux can be done very quickly, and many customers recognize the benefits of using them both in combination.Because Oracle Linux is available for free (including free bug fixes and errata), it's an ideal choice for running MySQL in your data center. You can run the same Linux distribution on both your development/staging systems as well as on the production machines, you decide which of these should be covered by a support subscription and at which level of support. This gives you flexibility and provides some really attractive cost-saving opportunities. Keith Larson: Since I am a Linux user and fan, what is on the horizon for  Oracle Linux?Lenz Grimmer: We're working hard on broadening the ecosystem around Oracle Linux, building up partnerships with ISVs and IHVs to certify Oracle Linux as a fully supported platform for their products. We also continue to collaborate closely with the Linux kernel community on various projects, to make sure that Linux scales and performs well on large systems and meets the demands of today's data centers. These improvements and enhancements will then rolled into the Unbreakable Enterprise Kernel, which is the key ingredient that sets Oracle Linux apart from other distributions. We also have a number of ongoing projects which are making good progress, and I'm sure you'll hear more about this at the upcoming OpenWorld conference :) Keith Larson: What is something that more people should be aware of when it comes to Oracle Linux and MySQL ?Lenz Grimmer: Many people assume that Oracle Linux is just tuned for Oracle products, such as the Oracle Database or our Engineered Systems. While it's of course true that we do a lot of testing and optimization for these workloads, Oracle Linux is and will remain a general-purpose Linux distribution that is a very good foundation for setting up a LAMP-Stack, for example. We also provide MySQL RPM packages for Oracle Linux, so you can easily stay up to date if you need something newer than what's included in the stock distribution.One more thing that is really unique to Oracle Linux is Ksplice, which allows you to apply security patches to the running Linux kernel, without having to reboot. This ensures that your MySQL database server keeps up and running and is not affected by any downtime. Keith Larson: What else would you like to add ?Lenz Grimmer: Thanks again for getting in touch with me, I appreciated the opportunity. I'm looking forward to MySQL Connect and Oracle OpenWorld and to meet you and many other people from the MySQL community that I haven't seen for quite some time! Keith Larson:  Thank you Lenz!

    Read the article

  • Finding a person in the forest

    - by PointsToShare
    © 2011 By: Dov Trietsch. All rights reserved finding a person in the forest or Limiting the AD result in SharePoint People Picker There are times when we need to limit the SharePoint audience of certain farms or servers or site collections to a particular audience. One of my experiences involved limiting access to US citizens, another to a particular location. Now, most of us – your humble servant included – are not Active Directory experts – but we must be able to handle the “audience restrictions” as required. So here is how it’s done in a nutshell. Important note. Not all could be done in PowerShell (at least not yet)! There are no Windows PowerShell commands to configure People Picker. The stsadm command is: stsadm -o setproperty -pn peoplepicker-searchadcustomquery -pv ADQuery –url http://somethingOrOther Note the long-hyphenated property name. Now to filling the ADQuery.   LDAP Query in a nutshell Syntax LDAP is no older than SQL and an LDAP query is actually a query against the LDAP Database. LDAP attributes are the equivalent of Database columns, so why do we have to learn a new query language? Beats me! But we must, so here it is. The syntax of an LDAP query string is made of individual statements with relational operators including: = Equal <= Lower than or equal >= Greater than or equal… and memberOf – a group membership. ! Not * Wildcard Equal and memberOf are the most commonly used. Checking for absence uses the ! – not and the * - wildcard Example: (SN=Grant) All whose last name – SurName – is Grant Example: (!(SN=Grant)) All except Grant Example: (!(SN=*)) all where there is no SurName i.e SurName is absent (probably Rappers). Example: (CN=MyGroup) Common Name is MyGroup.  Example: (GN=J*) all the Given Names that start with J (JJ, Jane, Jon, John, etc.) The cryptic SN, CN, GN, etc. are attributes and more about them later All the queries are enclosed in parentheses (Query). Complex queries are comprised of sets that are in AND or OR conditions. AND is denoted by the ampersand (&) and the OR is denoted by the vertical pipe (|). The general syntax is that of the Prefix polish notation where the operand precedes the variables. E.g +ab is the sum of a and b. In an LDAP query (&(A)(B)) will garner the objects for which both A and B are true. In an LDAP query (&(A)(B)(C)) will garner the objects for which A, B and C are true. There’s no limit to the number of conditions. In an LDAP query (|(A)(B)) will garner the objects for which either A or B are true. In an LDAP query (|(A)(B)(C)) will garner the objects for which at least one of A, B and C is true. There’s no limit to the number of conditions. More complex queries have both types of conditions and the parentheses determine the order of operations. Attributes Now let’s get into the SN, CN, GN, and other attributes of the query SN – is the SurName (last name) GN – is the Given Name (first name) CN – is the Common Name, usually GN followed by SN OU – is an Organization Unit such as division, department etc. DC – is a Domain Content in the AD forest l – lower case ‘L’ stands for location. Jerusalem anybody? Or Katmandu. UPN – User Principal Name, is usually the first part of an email address. By nature it is unique in the forest. Most systems set the UPN to be the first initial followed by the SN of the person involved. Some limit the total to 8 characters. If we have many ‘jsmith’ we have to somehow distinguish them from each other. DN – is the distinguished name – a name unique to AD forest in which it lives. Usually it’s a CN with some domain or group distinguishers. DN is important in conjunction with the memberOf relation. Groups have stricter requirement. Each group has to have a unique name - its CN and it has to be unique regardless of its place. See more below. All of the attributes are case insensitive. CN, cn, Cn, and cN are identical. objectCategory is an element that requires special consideration. AD contains many different object like computers, printers, and of course people and groups. In the queries below, we’re limiting our search to people (person). Putting it altogether Let’s get a list of all the Johns in the SPAdmin group of the Jerusalem that local domain. (&(objectCategory=person)(memberOf=cn=SPAdmin,ou=Jerusalem,dc=local)) The memberOf=cn=SPAdmin uses the cn (Common Name) of the SPAdmin group. This is how the memberOf relation is used. ‘SPAdmin’ is actually the DN of the group. Also the memberOf relation does not allow wild cards (*) in the group name. Also, you are limited to at most one ‘OU’ entry. Let’s add Marvin Minsky to the search above. |(&(objectCategory=person)(memberOf=cn=SPAdmin,ou=Jerusalem,dc=local))(CN=Marvin Minsky) Here I added the or pipeline at the beginning of the query and put the CN requirement for Minsky at the end. Note that if Marvin was already in the prior result, he’s not going to be listed twice. One last note: You may see a dryer but more complete list of attributes rules and examples in: http://www.tek-tips.com/faqs.cfm?fid=5667 And finally (thus negating the claim that my previous note was last), to the best of my knowledge there are 3 more ways to limit the audience. One is to use the peoplepicker-searchadcustomfilter property using the same ADQuery. This works only in SP1 and above. The second is to limit the search to users within this particular site collection – the property name is peoplepicker-onlysearchwithinsitecollection and the value is yes (-pv yes) And the third is –pn peoplepicker-serviceaccountdirectorypaths –pv “OU=ou1,DC=dc1…..” Again you are limited to at most one ‘OU’ phrase – no OU=ou1,OU=ou2… And now the real end. The main property discussed in this sprawling and seemingly endless monogram – peoplepicker-searchadcustomquery - is the most general way of getting the job done. Here are a few examples of command lines that worked and some that didn’t. Can you see why? C:\Program Files\Common Files\Microsoft Shared\Web Server Extensions\12\BIN>stsa dm -o setproperty -url http://somethingOrOther -pn peoplepicker-searchadcustomfi lter -pv (Title=David) Operation completed successfully. C:\Program Files\Common Files\Microsoft Shared\Web Server Extensions\12\BIN>stsa dm -o setproperty -url http://somethingOrOther -pn peoplepicker-searchadcustomfi lter -pv (!Title=David) Operation completed successfully. C:\Program Files\Common Files\Microsoft Shared\Web Server Extensions\12\BIN>stsa dm -o setproperty -url http://somethingOrOther -pn peoplepicker-searchadcustomfi lter -pv (OU=OURealName,OU=OUMid,OU=OUTop,DC=TopDC,DC=MidDC,DC=BottomDC) Command line error. Too many OUs C:\Program Files\Common Files\Microsoft Shared\Web Server Extensions\12\BIN>stsa dm -o setproperty -url http://somethingOrOther -pn peoplepicker-searchadcustomfi lter -pv (OU=OURealName) Operation completed successfully. C:\Program Files\Common Files\Microsoft Shared\Web Server Extensions\12\BIN>stsa dm -o setproperty -url http://somethingOrOther -pn peoplepicker-searchadcustomfi lter -pv (DC=TopDC,DC=MidDC,DC=BottomDC) Operation completed successfully. C:\Program Files\Common Files\Microsoft Shared\Web Server Extensions\12\BIN>stsa dm -o setproperty -url http://somethingOrOther -pn peoplepicker-searchadcustomfi lter -pv (OU=OURealName,DC=TopDC,DC=MidDC,DC=BottomDC) Operation completed successfully.   That’s all folks!

    Read the article

  • Using SQL Source Control with Fortress or Vault &ndash; Part 2

    - by AjarnMark
    In Part 1, I started talking about using Red-Gate’s newest version of SQL Source Control and how I really like it as a viable method to source control your database development.  It looks like this is going to turn into a little series where I will explain how we have done things in the past, and how life is different with SQL Source Control.  I will also explain some of my philosophy and methodology around deployment with these tools.  But for now, let’s talk about some of the good and the bad of the tool itself. More Kudos and Features I mentioned previously how impressed I was with the responsiveness of Red-Gate’s team.  I have been having an ongoing email conversation with Gyorgy Pocsi, and as I have run into problems or requested things behave a little differently, it has not been more than a day or two before a new Build is ready for me to download and test.  Quite impressive! I’m sure much of the requests I put in were already in the plans, so I can’t really take credit for them, but throughout this conversation, Red-Gate has implemented several features that were not in the first Early Access version.  Those include: Honoring the Fortress configuration option to require Work Item (Bug) IDs on check-ins. Adding the check-in comment text as a comment to the Work Item. Adding the list of checked-in files, along with the Fortress links for automatic History and DIFF view Updating the status of a Work Item on check-in (e.g. setting the item to Complete or, in our case “Dev-Complete”) Support for the Fortress 2.0 API, and not just the Vault Pro 5.1 API.  (See later notes regarding support for Fortress 2.0). These were all features that I felt we really needed to have in-place before I could honestly consider converting my team to using SQL Source Control on a regular basis.  Now that I have those, my only excuse is not wanting to switch boats on the team mid-stream.  So when we wrap up our current release in a few weeks, we will make the jump.  In the meantime, I will continue to bang on it to make sure it is stable.  It passed one test for stability when I did a test load of one of our larger database schemas into Fortress with SQL Source Control.  That database has about 150 tables, 200 User-Defined Functions and nearly 900 Stored Procedures.  The initial load to source control went smoothly and took just a brief amount of time. Warnings Remember that this IS still in pre-release stage and while I have not had any problems after that first hiccup I wrote about last time, you still need to treat it with a healthy respect.  As I understand it, the RTM is targeted for February.  There are a couple more features that I hope make it into the final release version, but if not, they’ll probably be coming soon thereafter.  Those are: A Browse feature to let me lookup the Work Item ID instead of having to remember it or look back in my Item details.  This is just a matter of convenience. I normally have my Work Item list open anyway, so I can easily look it up, but hey, why not make it even easier. A multi-line comment area.  The current space for writing check-in comments is a single-line text box.  I would like to have a multi-line space as I sometimes write lengthy commentary.  But I recognize that it is a struggle to get most developers to put in more than the word “fixed” as their comment, so this meets the need of the majority as-is, and it’s not a show-stopper for us. Merge.  SQL Source Control currently does not have a Merge feature.  If two or more people make changes to the same database object, you will get a warning of the conflict and have to choose which one wins (and then manually edit to include the others’ changes).  I think it unlikely you will run into actual conflicts in Stored Procedures and Functions, but you might with Views or Tables.  This will be nice to have, but I’m not losing any sleep over it.  And I have multiple tools at my disposal to do merges manually, so really not a show-stopper for us. Automation has its limits.  As cool as this automation is, it has its limits and there are some changes that you will be better off scripting yourself.  For example, if you are refactoring table definitions, and want to change a column name, you can write that as a quick sp_rename command and preserve the data within that column.  But because this tool is looking just at a before and after picture, it cannot tell that you just renamed a column.  To the tool, it looks like you dropped one column and added another.  This is not a knock against Red-Gate.  All automated scripting tools have this issue, unless the are actively monitoring your every step to know exactly what you are doing.  This means that when you go to Deploy your changes, SQL Compare will script the change as a column drop and add, or will attempt to rebuild the entire table.  Unfortunately, neither of these approaches will preserve the existing data in that column the way an sp_rename will, and so you are better off scripting that change yourself.  Thankfully, SQL Compare will produce warnings about the potential loss of data before it does the actual synchronization and give you a chance to intercept the script and do it yourself. Also, please note that the current official word is that SQL Source Control supports Vault Professional 5.1 and later.  Vault Professional is the new name for what was previously known as Fortress.  (You can read about the name change on SourceGear’s site.)  The last version of Fortress was 2.x, and the API for Fortress 2.x is different from the API for Vault Pro.  At my company, we are currently running Fortress 2.0, with plans to upgrade to Vault Pro early next year.  Gyorgy was able to come up with a work-around for me to be able to use SQL Source Control with Fortress 2.0, even though it is not officially supported.  If you are using Fortress 2.0 and want to use SQL Source Control, be aware that this is not officially supported, but it is working for us, and you can probably get the work-around instructions from Red-Gate if you’re really, really nice to them. Upcoming Topics Some of the other topics I will likely cover in this series over the next few weeks are: How we used to do source control back in the old days (a few weeks ago) before SQL Source Control was available to Vault users What happens when you restore a database that is linked to source control Handling multiple development branches of source code Concurrent Development practices and handling Conflicts Deployment Tips and Best Practices A recap after using the tool for a while

    Read the article

  • When OneTug Just Isn&rsquo;t Enough&hellip;

    - by onefloridacoder
    I stole that from the back of a T-shirt I saw at the Orlando Code Camp 2010.  This was my first code camp and my first time volunteering for an event like this as well.  It was an awesome day.  I cannot begin to count the “aaahh”, “I did-not-know I could do that”, in the crowds and for myself.  I think it was a great day of learning for everyone at all levels.  All of the presenters were different and provided great insights into the topics they were presenting.  Here’s a list of the ones that I attended. KodeFuGuru, “Pirates vs. Ninjas” He touched on many good topics to relax some of the ways we think when we are writing out code, and still looks good, readable, etc.  As he pointed out in all of his examples, we might not always realize everything that’s going on under the covers.  He exposed a bug in his own code, and verbalized the mental gymnastics he went through when he knew there was something wrong with one of his IEnumerable implementations.  For me, it was great to hear that someone else labors over these gut reactions to code quickly snapped together, to the point that we rush to the refactor stage to fix what’s bothering us – and learn.  He has some content on extension methods that was very interesting.  My “that is so cool” moment was when he swapped out AddEntity method on an entity class and used a With extension method instead.  Some of the LINQ scales fell off my eyes at that moment, and I realized my own code could be a lot more powerful (and readable) if incorporate a few of these examples at the appropriate times.  And he cautioned as well… “don’t go crazy with this stuff”, there’s a place and time for everything.  One of his examples demo’d toward the end of the talk is on his sight where he’s chaining methods together, cool stuff. Quotes I liked: “Extension Methods - Extension methods to put features back on the model type, without impacting the type.” “Favor Declarative Code” – Check out the ? and ?? operators if you’re not already using them. “Favor Fluent Code” “Avoid Pirate Ninja Zombies!  If you see one run!” I’m definitely going to be looking at “Extract Projection” when I get into VS2010. BDD 101 – Sean Chambers http://github.com/schambers This guy had a whole host of gremlins against him, final score Sean 5, Gremlins 1.  He ran the code samples from his github repo  in the code github code viewer since the PC they school gave him to use didn’t have VS installed. He did a great job of converting the grammar between BDD and TDD, and how this style of development can be used in integration tests as well as the different types of gated builds on a CI box – he didn’t go into a discussion around CI, but we could infer that it could work. Like when we use WSSF, it does cause a class explosion to happen however the amount of code per class it limit to just covering the concern at hand – no more, no less.  As in “When I as a <Role>, expect {something} to happen, because {}”  This keeps us (the developer) from gold plating our solutions and creating less waste.  He basically keeps the code that prove out the requirement to two lines of code.  Nice. He uses SpecUnit to merge this grammar into his .NET projects and gave an overview on how this ties into writing his own BDD tests.  Some folks were familiar with Given / When / Then as story acceptance criteria and here’s how he mapped it: “Given <Context>  When <Something Happens> Then <I expect...>”  There are a few base classes and overrides in the SpecUnit framework that help with setting up the context for each test which looked very handy. Successfully Running Your Own Coding Business The speaker ran through a list of items that sounded like common sense stuff LLC, banking, separating expenses, etc.  Then moved into role playing with business owners and an ISV.  That was pretty good stuff, it pays to be a good listener all of the time even if your client is sitting on the other side of the phone tearing you head off for you – but that’s all it is, and get used to it its par for the course.  Oh, yeah always answer the phone was one simple thing that you can do to move  your business forward.  But like Cory Foy tweeted this week, “If you owe me a lot of money, don’t have a message that says your away for five weeks skiing in Colorado.”  Lots of food for thought that’s on my list of “todo’s and to-don’ts”. Speaker Idol Next, I had the pleasure of helping Russ Fustino tape this part of Code Camp as my primary volunteer opportunity that day.  You remember Russ, “know the code” from the awesome Russ’ Tool Shed series.  He did a great job orchestrating and capturing the Speaker Idol finals.   So I didn’t actually miss any sessions, but was able to see three back to back in one setting.  The idol finalists gave a 10 minute talk and very deep subjects, but different styles of talks.  No one walked away empty handed for jobs very well done.  Russ has details on his site.  The pictures and  video captured is supposed to be published on Channel 9 at a later date.  It was also a valuable experience to see what makes technical speakers effective in their talks.  I picked up quite a few speaking tips from what I heard from the judges and contestants. Design For Developers – Diane Leeper If you are a great developer, you’re probably a lousy designer.  Diane didn’t come to poke holes in what we think we can do with UI layout and design, but she provided some tools we can use to figure out metaphors for visualizing data.  If you need help with that check out Silverlight Pivot – that’s what she was getting at.  I was first introduced to her at one of John Papa’s talks last year at a Lakeland User Group meeting and she’s very passionate about design.  She was able to discuss different elements of Pivot, while to a developer is just looked cool. I believe she was providing the deck from her talk to folks after her talk, so send her an email if you’re interested.   She says she can talk about design for hours and hours – we all left that session believing her.   Rinse and Repeat Orlando Code Camp 2010 was awesome, and would totally do it again.  There were lots of folks from my shop there, and some that have left my shop to go elsewhere.  So it was a reunion of sorts and a great celebration for the simple fact that its great to be a developer and there’s a community that supports and recognizes it as well.  The sponsors were generous and the organizers were very tired, namely Esteban Garcia and Will Strohl who were responsible for making a lot of this magic happen.  And if you don’t believe me, check out the chatter on Twitter.

    Read the article

< Previous Page | 217 218 219 220 221 222 223 224 225 226 227  | Next Page >