Search Results

Search found 42453 results on 1699 pages for 'question'.

Page 37/1699 | < Previous Page | 33 34 35 36 37 38 39 40 41 42 43 44  | Next Page >

  • php/mysql question, retrieving information

    - by chance
    Hey SO, I have a question, pretty new at PHP and MySql but was curious is it possible to display only the most recent column to a table, from a database? I know it is possible to display a list of all information, or if you have just one column of information in your table to display that using the Here is my code for my basic script (just loads 1 column of information) I would like it to only print out the recent column i added to the table $con = mysql_connect ("localhost","username","password"); if (!$con) { die ('Could not connect:' . mysql_error()0; } mysql_select_db("db_name", $con); $result = mysql_query(Select column FROM table"); while($row = mysql_fetch_array($result)) { echo $row['column']; } mysql_close($con); * keep in mind, im not looking for someone to shoot me my code with it completed, though it would be handy, however just giving me the options of what I have to do, or what the script is called would be plentiful enough for my research, thanks!

    Read the article

  • question about Tetration

    - by davit-datuashvili
    i have question how write program which calculates following procedures http://en.wikipedia.org/wiki/Tetration i have exponential program which returns x^n here is code public class Exp{ public static long exp(long x,long n){ long t=0; if (n==0){ t= 1; } else{ if (n %2==0){ t= exp(x,n/2)* exp(x,n/2); } else{ t= x*exp(x,n-1); } } return t; } public static void main(String[]args){ long x=5L; long n=4L; System.out.println(exp(x,n)); } } but how use it in Tetration program?please help

    Read the article

  • Shedule tasks question

    - by gabac
    Hi everyone I have a though question :) Let's say you get some data which is due at a certain point int the future. You get A saying my task is due in 1h and other which says his task is due in 1.5h. The information is collected on your server. In which programming language or even how would solve that? So task a will be executed in 1h, task B in 1.5h. I read something about java scheduler but I'm not yet sure if this is the right way. What are ur ideas? Cheers

    Read the article

  • What will or won't cause a thread to block (a question from a test)

    - by fingerprint211b
    I've had a test, and there was a question I lost some points on, because I wasn't able to answer it : Which of the following is NOT a condition which can cause a thread to block : Calling an objects's wait() method Waiting for an I/O operation Calling sleep() Calling yield() Calling join() As far as I know, all of these are blocking calls : wait() returns when an something calls notify(), blocks until then If the thread is WAITING for an I/O operation then it's obviously blocked sleep(), obviously, blocks until the time runs out, or something wakes up the thread yield() "cancels the rest of the thread's timeslice" (lacking a better term), and returns only when the thread is active again join() blocks until the thread it's waiting for terminates. Am I missing something here?

    Read the article

  • Question regarding the iPhone In App Purchase capabilities

    - by Wesley
    Hi everyone. I am working on releasing an app that will greatly benefit from using the in app purchase model. The app is a sort of book viewer, and the content I would like to make available for purchase will be more books in various languages. Each book is stored in sqlite format, in separate .db files. Now, the way that my developer has set it all up is that he added a line onto the bottom of the info.plist file called databases. Inside that database section, I can type in 'es' as the key, and for the value the name of the spanish database, and it populates in a uitableview in spanish, using NSLocale. So it is very easy to setup and implement different databases, which is great, but now I am confused about how I can implement my in app purchase model. Sorry about the long winded intro, here is my question: Is it possible to have an in app purchase add a line to my info.plist file? Or if not, is it possible to have the purchase reveal a new and updated plist file altogether that I would have already setup correctly? Any help here would be appreciated. Thanks

    Read the article

  • General Web Programming/designing Question: ?

    - by Prasad
    hi, I have been in web programming for 2 years (Self taught - a biology researcher by profession). I designed a small wiki with needed functionalities and a scientific RTE - ofcourse lot is expected. I used mootools framework and AJAX extensively. I was always curious when ever I saw the query strings passed from URL. Long encrypted query string directly getting passed to the server. Especially Google's design is such. I think this is the start of providing a Web Service to a client - I guess. Now, my question is : is this a special, highly professional, efficient / advanced web design technique to communicate queries via the URL ? I always felt that direct URL based communication is faster. I tried my bit and could send a query through the URL directly. here is the link: http://sgwiki.sdsc.edu/getSGMPage.php?8 By this , the client can directly link to the desired page instead of searching and / or can automate. There are many possibilities. The next request: Can I be pointed to such technique of web programming? oops: I am sorry, If I have not been able to convey my request clearly. Prasad.

    Read the article

  • LINQ to DataSet Dataclass assignment question

    - by Overhed
    Hi all, I'm working on a Silverlight project trying to access a database using LINQ To DataSet and then sending data over to Silverlight via .ASMX web service. I've defined my DataSet using the Server Explorer tool (dragging and dropping all the different tables that I'm interested in). The DataSet is able to access the server and database with no issues. Below is code from one of my Web Methods: public List<ClassSpecification> getSpecifications() { DataSet2TableAdapters.SpecificationTableAdapter Sta = new DataSet2TableAdapters.SpecificationTableAdapter(); return (from Spec in Sta.GetData().AsEnumerable() select new ClassSpecification() { Specification = Spec.Field<String>("Specification"), SpecificationType = Spec.Field<string>("SpecificationType"), StatusChange = Spec.Field<DateTime>("StatusChange"), Spec = Spec.Field<int>("Spec") }).ToList<ClassSpecification>(); } I created a "ClassSpecification" data class which is going to contain my data and it has all the table fields as properties. My question is, is there a quicker way of doing the assignment than what is shown here? There are actually about 10 more fields, and I would imagine that since my DataSet knows my table definition, that I would have a quicker way of doing the assignment than going field by field. I tried just "select new ClassSpecification()).ToList Any help would be greatly appreciated.

    Read the article

  • Starter question of declarative style SQLAlchemy relation()

    - by jfding
    I am quite new to SQLAlchemy, or even database programming, maybe my question is too simple. Now I have two class/table: class User(Base): __tablename__ = 'users' id = Column(Integer, primary_key=True) name = Column(String(40)) ... class Computer(Base): __tablename__ = 'comps' id = Column(Integer, primary_key=True) buyer_id = Column(None, ForeignKey('users.id')) user_id = Column(None, ForeignKey('users.id')) buyer = relation(User, backref=backref('buys', order_by=id)) user = relation(User, backref=backref('usings', order_by=id)) Of course, it cannot run. This is the backtrace: File "/Library/Python/2.6/site-packages/SQLAlchemy-0.5.8-py2.6.egg/sqlalchemy/orm/state.py", line 71, in initialize_instance fn(self, instance, args, kwargs) File "/Library/Python/2.6/site-packages/SQLAlchemy-0.5.8-py2.6.egg/sqlalchemy/orm/mapper.py", line 1829, in _event_on_init instrumenting_mapper.compile() File "/Library/Python/2.6/site-packages/SQLAlchemy-0.5.8-py2.6.egg/sqlalchemy/orm/mapper.py", line 687, in compile mapper._post_configure_properties() File "/Library/Python/2.6/site-packages/SQLAlchemy-0.5.8-py2.6.egg/sqlalchemy/orm/mapper.py", line 716, in _post_configure_properties prop.init() File "/Library/Python/2.6/site-packages/SQLAlchemy-0.5.8-py2.6.egg/sqlalchemy/orm/interfaces.py", line 408, in init self.do_init() File "/Library/Python/2.6/site-packages/SQLAlchemy-0.5.8-py2.6.egg/sqlalchemy/orm/properties.py", line 716, in do_init self._determine_joins() File "/Library/Python/2.6/site-packages/SQLAlchemy-0.5.8-py2.6.egg/sqlalchemy/orm/properties.py", line 806, in _determine_joins "many-to-many relation, 'secondaryjoin' is needed as well." % (self)) sqlalchemy.exc.ArgumentError: Could not determine join condition between parent/child tables on relation Package.maintainer. Specify a 'primaryjoin' expression. If this is a many-to-many relation, 'secondaryjoin' is needed as well. There's two foreign keys in class Computer, so the relation() callings cannot determine which one should be used. I think I must use extra arguments to specify it, right? And howto? Thanks

    Read the article

  • Rails has_and_belongs_to_many relationship question

    - by Kevin Whitaker
    Hello all, I'm sure that this question has been asked somewhere before, as the habtm relationship seems to be very confusing. I have two models, users and promotions. The idea is that a promotion can have many users, and a user can have many promotions. class User < ActiveRecord::Base has_and_belongs_to_many :promotions end class Promotion < ActiveRecord::Base has_and_belongs_to_many :users end I also have a promotions_users table/model, with no id of its own. It references user_id and promotions_id class PromotionsUsers < ActiveRecord::Base end So, how do I add a user to a promotion? I've tried something like this: user = User.find(params[:id]) promotion = Promotion.find(params[:promo_id]) promo = user.promotions.new(promo) This results in the following error: NoMethodError: undefined method `stringify_keys!' for #<Promotion:0x10514d420> If I try this line instead: promo= user.promotions.new(promo.id) I get this error: TypeError: can't dup Fixnum I'm sure that there is a very easy solution to my problem, and I'm just not searching for the solution the right way. Thank you for your time, and any help you can provide.

    Read the article

  • UX question: is better to have "serious delete" or have "trash"

    - by ftrotter
    I am developing an application that allows for a user to manage some individual data points. One of the things that my users will want to do is "delete" but what should that mean? For a web application is it better to present a user with the option to have serious delete or to use a "trash" system? Under "serious delete" (would love to know if there is a better name for this...) you click "delete" and then the user is warned "this is final and tragic action. Once you do this you will not be able to get -insert data point name here- back, even if you are crying..." Then if they click delete... well it truly is gone forever. Under the "trash" model, you never trust that the user really wants to delete... instead you remove the data point from the "main display" and put into a bucket called "the trash". This gets it out of the users way, which is what they usually want, but they can get it back if they make a mistake. Obviously this is the way most operating systems have gone. The advantages of "serious delete" are: Easy to implement Easy to explain to users The disadvantages of "serious delete" are: it can be tragically final sometimes, cats walk on keyboards The advantages of the "trash" system are: user is safe from themselves bulk methods like "delete a bunch at once" make more sense saves support headaches The disadvantages of the "trash" system are": For sensitive data, you create an illusion of destruction users think something is gone, but it is not. Lots of subtle distinctions make implementation more difficult Do you "eventually" delete the contents of the trash? My question is which one is the right design pattern for modern web applications? With enough discussion to justify your answer... Would love to be pointed towards some relevant research. -FT

    Read the article

  • Land of Lisp example question

    - by cwallenpoole
    I've read a lot of good things about Land of Lisp so I thought that I might go through it to see what there was to see. (defun tweak-text (lst caps lit) (when lst (let ((item (car lst)) (rest (cdr lst))) (cond ; If item = space, then call recursively starting with ret ; Then, prepend the space on to the result. ((eq item #\space) (cons item (tweak-text rest caps lit))) ; if the item is an exclamation point. Make sure that the ; next non-space is capitalized. ((member item '(#\! #\? #\.)) (cons item (tweak-text rest t lit))) ; if item = " then toggle whether we are in literal mode ((eq item #\") (tweak-text rest caps (not lit))) ; if literal mode, just add the item as is and continue (lit (cons item (tweak-text rest nil lit))) ; if either caps or literal mode = true capitalize it? ((or caps lit) (cons (char-upcase item) (tweak-text rest nil lit))) ; otherwise lower-case it. (t (cons (char-downcase item) (tweak-text rest nil nil))))))) (the comments are mine) (FYI -- the method signature is (list-of-symbols bool-whether-to-caps bool-whether-to-treat-literally) but the author shortened these to (lst caps lit).) But anyway, here's the question: This has (cond... (lit ...) ((or caps lit) ...)) in it. My understanding is that this would translate to if(lit){ ... } else if(caps || lit){...} in a C style syntax. Isn't the or statement redundant then? Is there ever a condition where the (or caps lit) condition will be called if caps is nil?

    Read the article

  • General question about DirectShow.NET, DirectShow and Windows Media Format

    - by Paul Andrews
    I searched and googled for an answer but couldn't find one. Basically I'm developing a webcam/audio streaming application which should capture audio and video from a pc (usb webcam/microphone) and send them to a receiving server. What the server will do with that it's another story and phase two (which I'm skipping for now) I wrote some code using DirectShow and Windows Media Format and it worked great for capture audio/video and sending them to another client, but there's a major problem: latency. Everywhere in the internet everyone gave me the same answer: "sorry dude but media format isn't for video conferencing, their codecs have too high latency". I thought I could skip the .wmv problems but seems like it's not possible to do... this road ends here then. So I saw a few examples with DirectShow.NET which were faster for both audio and video.. my question is: how come that DirectShow.NET is faster and better for video/audio conferencing? Shouldn't it be just a .NET porting of C++'s DirectShow? Am I missing something? I'm a bit confused at this point

    Read the article

  • MVC implementation/best-practices question

    - by Vivin Paliath
    I have to work with some code that isn't truly MVC (i.e., it doesn't use an explicit framework among other things). Right now we make do with servlets that pass data to services. Here is my problem. I am receiving a post to a servlet that contains a whole bunch of address data that I have to save to the database. The data is (obviously) in the HttpRequest object. My question is, how do I pass this data into a service? I am reluctant to do it like this: AddressService.saveAddress(request); Because I don't think the service should have a dependency on the request. My other option is to do something like this: String addressLine = request.getParameter("addressLine"); .. .. about 7 other parameters .. String zip = request.getParameter("zip"); AddressService.saveAddress(addressLine, ... 7 other parameters ..., zip); But I don't like having a function with a huge number of parameters either. I was thinking of making an intermediate object called AddressData that would hold data from the request, and then passing that into the service. Is that an acceptable way of doing things?

    Read the article

  • Question about SQL Server HierarchyID depth-first performance

    - by AndalusianCat
    I am trying to implement hierarchyID in a table (dbo.[Message]) containing roughly 50,000 rows (will grow substantially in the future). However it takes 30-40 seconds to retrieve about 25 results. The root node is a filler in order to provide uniqueness, therefor every subsequent row is a child of that dummy row. I need to be able to traverse the table depth-first and have made the hierarchyID column (dbo.[Message].MessageID) the clustering primary key, have also added a computed smallint (dbo.[Message].Hierarchy) which stores the level of the node. Usage: A .Net application passes through a hierarchyID value into the database and I want to be able to retrieve all (if any) children AND parents of that node (besides the root, as it is filler). A simplified version of the query I am using: @MessageID hierarchyID /* passed in from application */ SELECT m.MessageID, m.MessageComment FROM dbo.[Message] as m WHERE m.Messageid.IsDescendantOf(@MessageID.GetAncestor((@MessageID.GetLevel()-1))) = 1 ORDER BY m.MessageID From what I understand, the index should be detected automatically without a hint. From searching forums I have seen people utilizing index hints, at least in the case of breadth-first indexes, as apparently CLR calls may be opaque to the query optimizer. I have spent the past few days trying to find a solution for this issue, but to no avail. I would greatly appreciate any assistance, and as this is my first post, I apologize in advance if this would be considered a 'noobish' question, I have read the MS documentation and searched countless forums, but have not came across a succinct description of the specific issue.

    Read the article

  • Compare new Integer Objects in ArrayList Question

    - by thechiman
    I am storing Integer objects representing an index of objects I want to track. Later in my code I want to check to see if a particular object's index corresponds to one of those Integers I stored earlier. I am doing this by creating an ArrayList and creating a new Integer from the index of a for loop: ArrayList<Integer> courseselectItems = new ArrayList(); //Find the course elements that are within a courseselect element and add their indicies to the ArrayList for(int i=0; i<numberElementsInNodeList; i++) { if (nodeList.item(i).getParentNode().getNodeName().equals("courseselect")) { courseselectItems.add(new Integer(i)); } } I then want to check later if the ArrayList contains a particular index: //Cycle through the namedNodeMap array to find each of the course codes for(int i=0; i<numberElementsInNodeList; i++) { if(!courseselectItems.contains(new Integer(i))) { //Do Stuff } } My question is, when I create a new Integer by using new Integer(i) will I be able to compare integers using ArrayList.contains()? That is to say, when I create a new object using new Integer(i), will that be the same as the previously created Integer object if the int value used to create them are the same? I hope I didn't make this too unclear. Thanks for the help!

    Read the article

  • pthread windows event equivalent question

    - by ScaryAardvark
    I have the following code which replicates the windows manual and auto reset events. class event { public: event( bool signalled = false, bool ar = true ) : _auto( ar ), _signalled( signalled ) { pthread_mutex_init( &_mutex, NULL ); pthread_cond_init( &_cond, NULL ); } ~event() { pthread_cond_destroy( &_cond ); pthread_mutex_destroy( &_mutex ); } void set() { pthread_mutex_lock( &_mutex ); // only set and signal if we are unset if ( _signalled == false ) { _signalled = true; pthread_cond_signal( &_cond ); } pthread_mutex_unlock( &_mutex ); } void wait() { pthread_mutex_lock( &_mutex ); while ( _signalled == false ) { pthread_cond_wait( &_cond, &_mutex ); } // if we're an autoreset event, auto reset if ( _auto ) { _signalled = false; } pthread_mutex_unlock( &_mutex ); } void reset() { pthread_mutex_lock( &_mutex ); _signalled = false; pthread_mutex_unlock( &_mutex ); } private: pthread_mutex_t _mutex; pthread_cond_t _cond; bool _signalled; bool _auto; }; My question surrounds the "optimisation" I've put in place in the set() method where I only call pthread_cond_signal() if the event was unsignalled. Is this a valid optimisation or have I introduced some subtle flaw by doing so.

    Read the article

  • Ordering view by highest group count question - Ruby on Rails

    - by bgadoci
    I've read the couple of questions about this on stack overflow but can't seem to find the answer. I am trying to display the tags in my blog by the ones with the highest count in the tags table. Thanks to KandadaBoggu for helping me get the tags feature of the blog I am designing working. Here is the basics and my question. Tag belongs_to :post and Post has_many :tags. The tags table is simple really, consisting of the normal scaffolded fields plus post_id and tag_name (I actually called the column 'tag_name' instead of just 'name'). in my /views/posts/index.html/erb file I correctly am displaying the tags by group and the amount of times they are being used (appearing in the tags table). I just want to know how to order them by the highest count. Here is the code, and I currently have it set to updated_at: PostsController def index @tag_counts = Tag.count(:group => :tag_name, :order => 'updated_at DESC', :limit => 10) conditions, joins = {}, nil unless(params[:tag_name] || "").empty? conditions = ["tags.tag_name = ? ", params[:tag_name]] joins = :tags end @posts=Post.all(:joins => joins, :conditions=> conditions, :order => 'created_at DESC').paginate :page => params[:page], :per_page => 5 respond_to do |format| format.html # index.html.erb format.xml { render :xml => @posts } format.json { render :json => @posts } format.atom end end

    Read the article

  • Newbie question about controllers in ASP.Net MVC.

    - by Sergio Tapia
    I'm following a tutorial on creating the NerdDinner using ASP.Net MVC. However, I'm using Visual Studio 2010 Ultimate edition and there was only MVC2 to choose from. So I've following the tutorial so far, and everything is really clicking and being really well explained, until this little hitch. The guide is asking me to create new methods on a Controller file like so: using System; using System.Collections.Generic; using System.Linq; using System.Web; using System.Web.Mvc; namespace NerdDinner.Controllers { public class DinnersController : Controller { public void Index(){ Response.Write("<h1>Coming Soon: Dinners</h1>"); } public void Details(int id) { Response.Write("<h1>Details DinnerID: " + id + "</h1>"); } } } However, when I created the Controllers file, Visual Studio created an Index method already, but it looks vastly different to what the tutorial shows. Maybe this is the new way to do things using MVC2? using System; using System.Collections.Generic; using System.Linq; using System.Web; using System.Web.Mvc; namespace NerdDinner.Controllers { public class DinnersController : Controller { // // GET: /Dinners/ public ActionResult Index() { return View(); } } } My question is, how can I reproduce the Details and Index method (they're in MVC) to the MVC2 way? Is this even relevant? Thank you!

    Read the article

  • question about combining jquery tabs and a toggling horizontal slide panel

    - by Clay
    Hi, I'm attempting to combine a horizontally sliding panel and the ui tabs plugin. The idea is that you press a button, a panel slides open and inside is a series of tabs that expand content below. In terms of getting this to function in the most basic sense, so far so good: http://www.claylerner.com/bio_alt/ And my source looks like this: $(document).ready(function(){ $(".btn-slide").click(function(){ $("#panel").animate({width: 'toggle'}); }); }); $(function() { $("#tabs").tabs({ selected: -1 }); }); However, the problem I'm running into occurs when you have a tab exapnded and then you collapse the horizontal panel--as you can see the animation that hides the currently expanded tab looks pretty clunky. I get that this is happening because there's not a separate function hiding the expanded tab. So my question is, how can I make it so that when the button to collapse the panel (".btn-slide") is toggled it first hides any currently expanded tabs and then hides the open panel. Thanks so much!

    Read the article

  • SQL Server: A Grouping question that's annoying me

    - by user366729
    I've been working with SQL Server for the better part of a decade, and this grouping (or partitioning, or ranking...I'm not sure what the answer is!) one has me stumped. Feels like it should be an easy one, too. I'll generalize my problem: Let's say I have 3 employees (don't worry about them quitting or anything...there's always 3), and I keep up with how I distribute their salaries on a monthly basis. Month Employee PercentOfTotal -------------------------------- 1 Alice 25% 1 Barbara 65% 1 Claire 10% 2 Alice 25% 2 Barbara 50% 2 Claire 25% 3 Alice 25% 3 Barbara 65% 3 Claire 10% As you can see, I've paid them the same percent in Months 1 and 3, but in Month 2, I've given Alice the same 25%, but Barbara got 50% and Claire got 25%. What I want to know is all the distinct distributions I've ever given. In this case there would be two -- one for months 1 and 3, and one for month 2. I'd expect the results to look something like this (NOTE: the ID, or sequencer, or whatever, doesn't matter) ID Employee PercentOfTotal -------------------------------- X Alice 25% X Barbara 65% X Claire 10% Y Alice 25% Y Barbara 50% Y Claire 25% Seems easy, right? I'm stumped! Anyone have an elegant solution? I just put together this solution while writing this question, which seems to work, but I'm wondering if there's a better way. Or maybe a different way from which I'll learn something. WITH temp_ids (Month) AS ( SELECT DISTINCT MIN(Month) FROM employees_paid GROUP BY PercentOfTotal ) SELECT EMP.Month, EMP.Employee, EMP.PercentOfTotal FROM employees_paid EMP JOIN temp_ids IDS ON EMP.Month = IDS.Month GROUP BY EMP.Month, EMP.Employee, EMP.PercentOfTotal Thanks y'all! -Ricky

    Read the article

  • std::vector iterator or index access speed question

    - by Simone Margaritelli
    Just a stupid question . I have a std::vector<SomeClass *> v; in my code and i need to access its elements very often in the program, looping them forward and backward . Which is the fastest access type between those two ? Iterator access std::vector<SomeClass *> v; std::vector<SomeClass *>::iterator i; std::vector<SomeClass *>::reverse_iterator j; // i loops forward, j loops backward for( i = v.begin(), j = v.rbegin(); i != v.end() && j != v.rend(); i++, j++ ){ // some operations on v items } Subscript access (by index) std::vector<SomeClass *> v; unsigned int i, j, size = v.size(); // i loops forward, j loops backward for( i = 0, j = size - 1; i < size && j >= 0; i++, j-- ){ // some operations on v items } And, does const_iterator offer a faster way to access vector elements in case i do not have to modify them? Thank you in advantage.

    Read the article

  • Beginner SQL question: querying gold and silver tag badges in Stack Exchange Data Explorer

    - by polygenelubricants
    I'm using the Stack Exchange Data Explorer to learn SQL, but I think the fundamentals of the question is applicable to other databases. I'm trying to query the Badges table, which according to Stexdex (that's what I'm going to call it from now on) has the following schema: Badges Id UserId Name Date This works well for badges like [Epic] and [Legendary] which have unique names, but the silver and gold tag-specific badges seems to be mixed in together by having the same exact name. Here's an example query I wrote for [mysql] tag: SELECT UserId as [User Link], Date FROM Badges Where Name = 'mysql' Order By Date ASC The (slightly annotated) output is: as seen on stexdex: User Link Date --------------- ------------------- // all for silver except where noted Bill Karwin 2009-02-20 11:00:25 Quassnoi 2009-06-01 10:00:16 Greg 2009-10-22 10:00:25 Quassnoi 2009-10-31 10:00:24 // for gold Bill Karwin 2009-11-23 11:00:30 // for gold cletus 2010-01-01 11:00:23 OMG Ponies 2010-01-03 11:00:48 Pascal MARTIN 2010-02-17 11:00:29 Mark Byers 2010-04-07 10:00:35 Daniel Vassallo 2010-05-14 10:00:38 This is consistent with the current list of silver and gold earners at the moment of this writing, but to speak in more timeless terms, as of the end of May 2010 only 2 users have earned the gold [mysql] tag: Quassnoi and Bill Karwin, as evidenced in the above result by their names being the only ones that appear twice. So this is the way I understand it: The first time an Id appears (in chronological order) is for the silver badge The second time is for the gold Now, the above result mixes the silver and gold entries together. My questions are: Is this a typical design, or are there much friendlier schema/normalization/whatever you call it? In the current design, how would you query the silver and gold badges separately? GROUP BY Id and picking the min/max or first/second by the Date somehow? How can you write a query that lists all the silver badges first then all the gold badges next? Imagine also that the "real" query may be more complicated, i.e. not just listing by date. How would you write it so that it doesn't have too many repetition between the silver and gold subqueries? Is it perhaps more typical to do two totally separate queries instead? What is this idiom called? A row "partitioning" query to put them into "buckets" or something?

    Read the article

  • Codechef practice question help needed - find trailing zeros in a factorial

    - by manugupt1
    I have been working on this for 24 hours now, trying to optimize it. The question is how to find the number of trailing zeroes in factorial of a number in range of 10000000 and 10 million test cases in about 8 secs. The code is as follows: #include<iostream> using namespace std; int count5(int a){ int b=0; for(int i=a;i>0;i=i/5){ if(i%15625==0){ b=b+6; i=i/15625; } if(i%3125==0){ b=b+5; i=i/3125; } if(i%625==0){ b=b+4; i=i/625; } if(i%125==0){ b=b+3; i=i/125; } if(i%25==0){ b=b+2; i=i/25; } if(i%5==0){ b++; } else break; } return b; } int main(){ int l; int n=0; cin>>l; //no of test cases taken as input int *T = new int[l]; for(int i=0;i<l;i++) cin>>T[i]; //nos taken as input for the same no of test cases for(int i=0;i<l;i++){ n=0; for(int j=5;j<=T[i];j=j+5){ n+=count5(j); //no of trailing zeroes calculted } cout<<n<<endl; //no for each trialing zero printed } delete []T; } Please help me by suggesting a new approach, or suggesting some modifications to this one.

    Read the article

  • JSF Managed Property question

    - by kidvid
    I have a search page that I'll called "Parent." The search page references a country lookup page that I'll call "Child." When the user selects a country on Child's page and clicks on OK, I set the country back into the parent page. I do this by calling a method on the Parent page called "UpdateCountryCodeWithLookupValue(Child child)" When the user clicks on OK on the Child page, that method gets called in the parent, wherein it'll get the selected country code out of the Child page and set it into a text entry field. My question has to do with the proper way to set up this relationship in the faces config file. The way I have it now is that the child has a managed property for the parent. I.e., in my Child page I defined a method called "SetParent(Parent parent)". Is there any drawback to doing it this way? Would it be preferable to set the managed property so that the Child page class is a property of the parent instead of vice-versa? Let's say that I could have two Parent (search) pages open at the same time, and each of these was able to open the Child page (country code lookup). What would be the ramification for that circumstance in terms of the managed property in the faces config file? Thanks, Adrian

    Read the article

  • Rhino Mocks - Fluent Mocking - Expect.Call question

    - by Ben Cawley
    Hi, I'm trying to use the fluent mocking style of Rhino.Mocks and have the following code that works on a mock IDictionary object called 'factories': With.Mocks(_Repository).Expecting(() => { Expect.Call(() => factories.ContainsKey(Arg<String>.Is.Anything)); LastCall.Return(false); Expect.Call(() => factories.Add(Arg<String>.Is.Anything, Arg<Object>.Is.Anything)); }).Verify(() => { _Service = new ObjectRequestService(factories); _Service.RegisterObjectFactory(Valid_Factory_Key, factory); }); Now, the only way I have been able to set the return value of the ContainsKey call is to use LastCall.Return(true) on the following line. I'm sure I'm mixing styles here as Expect.Call() has a .Return(Expect.Action) method but I can't figure out how I am suppose to use it correctly to return a boolean value? Can anyone help out? Hope the question is clear enough - let me know if anyone needs more info! Cheers, Ben

    Read the article

< Previous Page | 33 34 35 36 37 38 39 40 41 42 43 44  | Next Page >