Search Results

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

Page 39/1699 | < Previous Page | 35 36 37 38 39 40 41 42 43 44 45 46  | Next Page >

  • Newb Question: scanf() in C

    - by riemannliness
    So I started learning C today, and as an exercise i was told to write a program that asks the user for numbers until they type a 0, then adds the even ones and the odd ones together. Here is is (don't laugh at my bad style): #include <stdio.h>; int main() { int esum = 0, osum = 0; int n, mod; puts("Please enter some numbers, 0 to terminate:"); scanf("%d", &n); while (n != 0) { mod = n % 2; switch(mod) { case 0: esum += n; break; case 1: osum += n; } scanf("%d", &n); } printf("The sum of evens:%d,\t The sum of odds:%d", esum, osum); return 0; } My question concerns the mechanics of the scanf() function. It seems that when you enter several numbers at once separated by spaces (eg. 1 22 34 2 8), the scanf() function somehow remembers each distinct numbers in the line, and steps through the while loop for each one respectively. Why/how does this happen? Example interaction within command prompt: - Please enter some numbers, 0 to terminate: 42 8 77 23 11 (enter) 0 (enter) - The sum of evens:50, The sum of odds:111 I'm running the program through the command prompt, it's compiled for win32 platforms with visual studio.

    Read the article

  • Glob() filesearch, question

    - by Peter
    Hi, a little question. I have this code, which is works perfect for files, but If am trying search on a directory name, the result is blank. How I can fix that? <?php function listdirs($dir,$search) { static $alldirs = array(); $dirs = glob($dir."*"); foreach ($dirs as $d){ if(is_file($d)){ $filename = pathinfo($d); if(eregi($search,$filename['filename'])){ print "<a href=http://someurl.com/" . $d .">". $d . "</a><br/>"; } }else{ listdirs($d."/",$search); } } } $path = "somedir/"; $search= "test"; listdirs($path,$search); ?> somedir/test/ result: blank (I want: /somedir/test/) somedir/test/test.txt result: OK I want to search also in the directory names, how I can do that?

    Read the article

  • Question regarding ideal database implementation for iPhone app

    - by Jeff
    So I have a question about the ideal setup for an app I am getting ready to build. The app is basically going to be a memorization tool and I already have an sqlite database full of content that I will be using for the app. The user will navigate through the contents of the database(using the uipickerview), and select something for memorization. If that row or cell of data is selected, it is put into a pool or a uitableview that is dedicated to showing which items you have in your "need to memorize" pool. When you go to that tableview, you can select the row, and the actual data would be populated. All information in the tableview would be deletable, in the event that they don't want it there anymore... Thats it. I know that with database interfacing, there are a few different options out there, in this particular setup, is core data the easiest approach? Is there any other way that would be better? I am just kind of looking for a point in the right direction, any help is greatly appreciated!!

    Read the article

  • A Newbie question regarding Software Development

    - by Sharif
    Hi, I'm going to complete my B.pharm (Hons.) degree and, you know, I don't have much knowledge about programing. I was wondering to build a software on my own. Could you guys tell me what to learn first for that? Is it too hard for a student of other discipline to build a software? Let me know please. The software I want to make is like a dictionary (or more specifically like "Physician's Desk Reference"). It should find the generic name, company name, indication, price etc. of a drug when I enter the brand name and vice versa. To build a software like that what programing language could help me most and what (and how many) language should I learn first? In my country, there is no practice of Community pharmacy (most of the pharmacy stores are run by unskilled people), that's why this type of thing could help them sell drugs. Would you please tell me what I'm to do and how tough it is? I'm very keen to learn programming. Thanks in advance NB: I started this post in ASKREDDIT section but it seems that was not the right place for poll type question, so I post it again in this section

    Read the article

  • Design Philosophy Question - When to create new functions

    - by Eclyps19
    This is a general design question not relating to any language. I'm a bit torn between going for minimum code or optimum organization. I'll use my current project as an example. I have a bunch of tabs on a form that perform different functions. Lets say Tab 1 reads in a file with a specific layout, tab 2 exports a file to a specific location, etc. The problem I'm running into now is that I need these tabs to do something slightly different based on the contents of a variable. If it contains a 1 I may need to use Layout A and perform some extra concatenation, if it contains a 2 I may need to use Layout B and do no concatenation but add two integer fields, etc. There could be 10+ codes that I will be looking at. Is it more preferable to create an individual path for each code early on, or attempt to create a single path that branches out only when absolutely required. Creating an individual path for each code would allow my code to be extremely easy to follow at a glance, which in turn will help me out later on down the road when debugging or making changes. The downside to this is that I will increase the amount of code written by calling some of the same functions in multiple places (for example, steps 3, 5, and 9 for every single code may be exactly the same. Creating a single path that would branch out only when required will be a bit messier and more difficult to follow at a glance, but I would create less code by placing conditionals only at steps that are unique. I realize that this may be a case-by-case decision, but in general, if you were handed a previously built program to work on, which would you prefer?

    Read the article

  • Mysql Performance Question - Essentially about normalizing efficiency

    - by freqmode
    Hi there. Just a quick question about database performance. I'll outline my site purpose below as background. I'm creating a dictionary site that saves the words users define to a database. What I'm wondering is whether or not to create a words table for each user or to keep one massive words table. This site will be used for entire schools so the single words table would be massive! The database structure is as follows: A user table with: User_ID PRIMARY KEY Username First Last Password Email Country Research Standings SendInfo Donated JoinedOn LastLogin Logins Correct Attempts Admin Active And one word table with: User_ID PRIMARY KEY Word Vocab Spell Defined DefinedAttempted Spelled SpelledAttempted Sentenced SentencedAttempted So what I'm asking is , performance-wise, should I create a new table for each user when they join the site - each user could have hundreds or thousands of words over time? Or is it better to have one massive table with thousands and thousands of records and filter by User_ID. I don't think I'll perform many table joins. My gut feeling is to create a new table for each user, but I thought I'd ask for expert advice! Thanks in advance.

    Read the article

  • smlnj rephrased question for listdir(filename, directoryname)

    - by czy1985
    i am a newbie learning sml and the question i am thrown with involves IO functions that i have no idea how it works even after reading it. Here is the 2 questions that i really need help with to get me started, please provide me with codings and some explaination, i will be able to trial and error with the code given for the other questions. Q1) listdir(filename,directoryname), which given the name of a directory, list its contents in a text file. The listing is in a form that makes it easy to seperate filenames, dates and sizes from each other. (similar to what msdos does with "dir" but instead of just listing it out, it places all the files and details into a text file. Q2) readlist(filename) which reads a list of filenames (each of which were produced by listdir in (Q1) and combines them into one large list. (reads from the text file in Q1 and then assigning the contents into 1 big list containing all the information) Thing is, i only learned from the lecturer in school on the introduction section, there isnt even a system input or output example shown, not even the "use file" function is taught. if anyone that knows sml sees this, please help. Thanks to anyone who took the effort helping me. Thanks for the reply, current I am using SMLNJ to try and do this. Basically, Q1 requires me to list the directory's files of the "directoryname" provided into a text file in "filename". The Q2 requires me to read from the "filename" text file and then place the contents into one large list. Duplicate of: smlnj listdir

    Read the article

  • A C# Refactoring Question...

    - by james lewis
    I came accross the following code today and I didn't like it. It's fairly obvious what it's doing but I'll add a little explanation here anyway: Basically it reads all the settings for an app from the DB and the iterates through all of them looking for the DB Version and the APP Version then sets some variables to the values in the DB (to be used later). I looked at it and thought it was a bit ugly - I don't like switch statements and I hate things that carry on iterating through a list once they're finished. So I decided to refactor it. My question to all of you is how would you refactor it? Or do you think it even needs refactoring at all? Here's the code: using (var sqlConnection = new SqlConnection(Lfepa.Itrs.Framework.Configuration.ConnectionString)) { sqlConnection.Open(); var dataTable = new DataTable("Settings"); var selectCommand = new SqlCommand(Lfepa.Itrs.Data.Database.Commands.dbo.SettingsSelAll, sqlConnection); var reader = selectCommand.ExecuteReader(); while (reader.Read()) { switch (reader[SettingKeyColumnName].ToString().ToUpper()) { case DatabaseVersionKey: DatabaseVersion = new Version(reader[SettingValueColumneName].ToString()); break; case ApplicationVersionKey: ApplicationVersion = new Version(reader[SettingValueColumneName].ToString()); break; default: break; } } if (DatabaseVersion == null) throw new ApplicationException("Colud not load Database Version Setting from the database."); if (ApplicationVersion == null) throw new ApplicationException("Colud not load Application Version Setting from the database."); }

    Read the article

  • DDSteps date question.

    - by Srini
    DDStep Date Question: Currently trying to pass just the date from excel. But getting the below error while doing it. Failed to convert property value of type [java.lang.String] to required type [java.util.Date] for property ...no matching editors or conversion strategy found spring for date conversion I even tried to add customEditorConfigurer in the ddsteps-context file. Still getting error. But in their pet store example looks like it works fine. Any help is appreciated. <entry key="java.util.Date"> <bean class="org.springframework.beans.propertyeditors.CustomDateEditor"> <constructor-arg> <bean class="java.text.SimpleDateFormat"> <constructor-arg value="yyyy-MM-dd" /> </bean> </constructor-arg> <constructor-arg value="false" /> </bean> </entry>

    Read the article

  • tricky interview question for C++

    - by Alexander
    Given the code below. How would you create/implement SR.h so that it produces the correct output WITHOUT any asterix in your solution? I got bummed by this question #include <cstdio> #include "SR.h" int main() { int j = 5; int a[] = {10, 15}; { SR x(j), y(a[0]), z(a[1]); j = a[0]; a[0] = a[1]; a[1] = j; printf("j = %d, a = {%d, %d}\n", j, a[0], a[1]); } printf("j = %d, a = {%d, %d}\n", j, a[0], a[1]); } //output j = 10, a = {15, 10} j = 5, a = {10, 15} #include <cstdio> #include "SR.h" int main() { int sum = 0; for (int i = 1; i < 100; i++) { SR ii(i); while (i--) sum += i; } printf("sum = %d\n", sum); } //The output is "sum = 161700".

    Read the article

  • question about counting sort

    - by davit-datuashvili
    hi i have write following code which prints elements in sorted order only one big problem is that it use two additional array here is my code public class occurance{ public static final int n=5; public static void main(String[]args){ // n is maximum possible value what it should be in array suppose n=5 then array may be int a[]=new int[]{3,4,4,2,1,3,5};// as u see all elements are less or equal to n //create array a.length*n int b[]=new int[a.length*n]; int c[]=new int[b.length]; for (int i=0;i<b.length;i++){ b[i]=0; c[i]=0; } for (int i=0;i<a.length;i++){ if (b[a[i]]==1){ c[a[i]]=1; } else{ b[a[i]]=1; } } for (int i=0;i<b.length;i++){ if (b[i]==1) { System.out.println(i); } if (c[i]==1){ System.out.println(i); } } } } // 1 2 3 3 4 4 5 1.i have two question what is complexity of this algorithm?i mean running time 2. how put this elements into other array with sorted order? thanks

    Read the article

  • PHP question about global variables and form requests

    - by user220201
    Hi, This is probably a stupid question but will ask anyway sine I have no idea. I have written basic php code which serve forms. Say I have a login page and I serve it using the login.php page and it will be called in the login.html page like this - <form action="login.php" method="post"> By this it is also implied that every POST needs its own php file, doesn't it? This kind of feels weird. Is there a way to have a single file, say code.php, and just have each of the forms as functions instead? EDIT: Specifically, say I have 5 forms that are used one after the other in my application. Say after login the user does A, B, C and D tasks each of which are sent to the server as a POST request. So instead of having A.php, B.php, C.php and D.php I would like to have a single code.php and have A(), B(), C() and D() as functions. Is there a way to do this? Also on the same note, how do I deal with say a global array (e.g. an array of currently logged in users) across multiple forms? I want to do this without writing to a DB. I know its probably better to write to a DB and query but is it even possible to do it with a global array? The reason I was thinking about having all the form functions in one file is to use a global array. Thanks, - Pav

    Read the article

  • Char C question about encoding signed/unsigned.

    - by drigoSkalWalker
    Hi guys. I read that C not define if a char is signed or unsigned, and in GCC page this says that it can be signed on x86 and unsigned in PowerPPC and ARM. Okey, I'm writing a program with GLIB that define char as gchar (not more than it, only a way for standardization). My question is, what about UTF-8? It use more than an block of memory? Say that I have a variable unsigned char *string = "My string with UTF8 enconding ~ çã"; See, if I declare my variable as unsigned I will have only 127 values (so my program will to store more blocks of mem) or the UTF-8 change to negative too? Sorry if I can't explain it correctly, but I think that i is a bit complex. NOTE: Thanks for all answer I don't understand how it is interpreted normally. I think that like ascii, if I have a signed and unsigned char on my program, the strings have diferently values, and it leads to confuse, imagine it in utf8 so.

    Read the article

  • The question about the basics of LINQ to SQL working

    - by Alex
    I just started learning LINQ to SQL, and so far I'm impressed with the easy of use and good performance. I used to think that when doing LINQ queries like from Customer in DB.Customers where Customer.Age > 30 select Customer Get all customers from the database ("SELECT * FROM Customers"), move them to the Customers array and then make a search in that Array using .NET methods. This is very inefficient, what if there are hundreds of thousands of customers in the database? Making such big SELECT queries would kill the web application. Now after experiencing how actually fast LINQ to SQL is, I start to suspect that when doing that query I just wrote, LINQ somehow converts it to a SQL Query string SELECT * FROM Customers WHERE Age > 30 And only when necessary it will run the query. So my question is: am I right? And when is the query actually run? The reason why I'm asking is not only because I want to understand how it works in order to build good optimized applications, but because I came across the following problem. I have 2 tables, one of them is Books, the other has information on how many books were sold on certain days. My goal is to select books that had at least 50 sales/day in past 10 days. It's done with this simple query: from Book in DB.Books where (from Sale in DB.Sales where Sale.SalesAmount >= 50 and Sale.DateOfSale >= DateTime.Now.AddDays(-10) select Sale.BookID).Contains(Book.ID) select Book The point is, I have to use the checking part in several queries and I decided to create an array with IDs of all popular books: var popularBooksIDs = from Sale in DB.Sales where Sale.SalesAmount >= 50 and Sale.DateOfSale >= DateTime.Now.AddDays(-10) select Sale.BookID; BUT when I try to do the query now: from Book in DB.Books where popularBooksIDs.Contains(Book.ID) select Book It doesn't work! That's why I think that we can't use thins kinds of shortcuts in LINQ to SQL queries, like we can't use them in real SQL. We have to create straightforward queries, am I right?

    Read the article

  • Basic question in XSL regarding preceding text

    - by Rachel
    I am new to XSL and i have a basic question on the context of using preceding text. My template match is on the text node. I am iterating over an xml file and within my for loop i am trying to take the preceding text of the text node. Unfortunately preceding::text() is not working if i use it within a for loop. I want to use it within the for loop but how can do it? <xsl:template match="text()"> <xsl:variable name="this" as="text()" select="."/> <xsl:for-each select="$input[@id = generate-id(current())]"> <xsl:variable name="preText" as="xsd:integer" select="sum(preceding::text()[. >> //*[@id=@name]]/string-length(.))"/> ... ... </xsl:for-each> </xsl:template>

    Read the article

  • Elegant code question: How to avoid creating unneeded object

    - by SeaDrive
    The root of my question is that the C# compiler is too smart. It detects a path via which an object could be undefined, so demands that I fill it. In the code, I look at the tables in a DataSet to see if there is one that I want. If not, I create a new one. I know that dtOut will always be assigned a value, but the the compiler is not happy unless it's assigned a value when declared. This is inelegant. How do I rewrite this in a more elegant way? System.Data.DataTable dtOut = new System.Data.DataTable(); . . // find table with tablename = grp // if none, create new table bool bTableFound = false; foreach (System.Data.DataTable d1 in dsOut.Tables) { string d1_name = d1.TableName; if (d1_name.Equals(grp)) { dtOut = d1; bTableFound = true; break; } } if (!bTableFound) dtOut = RptTable(grp);

    Read the article

  • C/C++ I18N mbstowcs question

    - by bogertron
    I am working on internationalizing the input for a C/C++ application. I have currently hit an issue with converting from a multi-byte string to wide character string. The code needs to be cross platform compatible, so I am using mbstowcs and wcstombs as much as possible. I am currently working on a WIN32 machine and I have set the locale to a non-english locale (Japanese). When I attempt to convert a multibyte character string, I seem to be having some conversion issues. Here is an example of the code: int main(int argc, char** argv) { wchar_t *wcsVal = NULL; char *mbsVal = NULL; /* Get the current code page, in my case 932, runs only on windows */ TCHAR szCodePage[10]; int cch= GetLocaleInfo( GetSystemDefaultLCID(), LOCALE_IDEFAULTANSICODEPAGE, szCodePage, sizeof(szCodePage)); /* verify locale is set */ if (setlocale(LC_CTYPE, "") == 0) { fprintf(stderr, "Failed to set locale\n"); return 1; } mbsVal = argv[1]; /* validate multibyte string and convert to wide character */ int size = mbstowcs(NULL, mbsVal, 0); if (size == -1) { printf("Invalid multibyte\n"); return 1; } wcsVal = (wchar_t*) malloc(sizeof(wchar_t) * (size + 1)); if (wcsVal == NULL) { printf("memory issue \n"); return 1; } mbstowcs(wcsVal, szVal, size + 1); wprintf(L"%ls \n", wcsVal); return 0; } At the end of execution, the wide character string does not contain the converted data. I believe that there is an issue with the code page settings, because when i use MultiByteToWideChar and have the current code page sent in EX: MultiByteToWideChar( CP_ACP, 0, mbsVal, -1, wcsVal, size + 1 ); in place of the mbstowcs calls, the conversion succeeds. My question is, how do I use the generic mbstowcs call instead of teh MuliByteToWideChar call?

    Read the article

  • Threadpool design question

    - by ZeroVector
    I have a design question. I want some feedback to know if a ThreadPool is appropriate for the client program I am writing. I am having a client running as a service processing database records. Each of these records contains connection information to external FTP sites [basically it is a queue of files to transfer]. A lot of them are to the same host, just moving different files. Therefore, I am grouping them together by host. I want to be able to create a new thread per host. I really don't care when the transfers finish, they just need to do all the work (or try to do) they were assigned, and then terminate once they are finished, cleaning up all resources they used in the process. I anticipate no more than 10-25 connections to be established. Once the transfer queue is empty, the program will simply wait until there are records in the queue again. Is the ThreadPool a good candidate for this or should I use a different approach? Edit: For the most part, this is the only significant custom application running on the server.

    Read the article

  • c programming malloc question

    - by user535256
    Hello guys, Just got query regarding c malloc() function. I am read()ing x number of bytes from a file to get lenght of filename, like ' read(file, &namelen, sizeof(unsigned char)); ' . The variable namelen is a type unsigned char and was written into file as that type (1 byte). Now namelen has the lenght of filename ie namelen=8 if file name was 'data.txt', plus extra /0 at end, that working fine. Now I have a structure recording file info, ie filename, filelenght, content size etc. struct fileinfo { char *name; ...... other variable like size etc }; struct fileinfo *files; Question: I want to make that files.name variable the size of namelen ie 8 so I can successfully write the filename into it, like ' files[i].name = malloc(namelen) ' However, I dont want it to be malloc(sizeof(namelen)) as that would make it file.name[1] as the size of its type unsigned char. I want it to be the value thats stored inside variable &namelen ie 8 so file.name[8] so data.txt can be read() from file as 8 bytes and written straight into file.name[8? Is there a way to do this my current code is this and returns 4 not 8 files[i].name = malloc(namelen); //strlen(files[i].name) - returns 4 //perhaps something like malloc(sizeof(&namelen)) but does not work Thanks for any suggestions Have tried suggested suggestions guys, but I now get a segmentation fault error using: printf("\nsizeofnamelen=%x\n",namelen); //gives 8 for data.txt files[i].name = malloc(namelen + 1); read(file, &files[i].name, namelen); int len=strlen(files[i].name); printf("\nnamelen=%d",len); printf("\nname=%s\n",files[i].name); When I try to open() file with that files[i].name variable it wont open so the data does not appear to be getting written inside the read() &files[i].name and strlen() causes segemntation error as well as trying to print the filename

    Read the article

  • .net printing multiple reports in one document (architecture question)

    - by LawsonM
    I understand how to print a single document via the PrintDocument class. However, I want to print multiple reports in one document. Each "report" will consist of charts, tables, etc. I want to have two reports per page. I've studied the few examples I can find on how to combine multiple documents into one; however, they always seem to work by creating a collection of objects (e.g. customer or order) that are then iterated over and drawn in the OnPrintPage method. My problem and hence the "architecture" question is that I don't want to cache the objects required to produce the report since they are very large and memory intensive. I'd simply like the resulting "report". One thought I had was to print the report to a metafile, cache that instead in a "MultiplePrintDocument" class and then position those images appropriately two to a page in the OnPrintPage method. I think this would be a lot more efficient and scalable in my case. But I'm not a professional programmer and can't figure out if I'm barking up the wrong tree here. I think the Graphics.BeginContainer and Graphics.Save methods might be relevant, but can't figure out how to implement or if there is a better way. Any pointers would be greatly appreciated.

    Read the article

  • java - question about thread abortion and deadlock - volatile keyword

    - by Tiyoal
    Hello all, I am having some troubles to understand how I have to stop a running thread. I'll try to explain it by example. Assume the following class: public class MyThread extends Thread { protected volatile boolean running = true; public void run() { while (running) { synchronized (someObject) { while (someObject.someCondition() == false && running) { try { someObject.wait(); } catch (InterruptedException e) { e.printStackTrace(); } } // do something useful with someObject } } } public void halt() { running = false; interrupt(); } } Assume the thread is running and the following statement is evaluated to true: while (someObject.someCondition() == false && running) Then, another thread calls MyThread.halt(). Eventhough this function sets 'running' to false (which is a volatile boolean) and interrupts the thread, the following statement is still executed: someObject.wait(); We have a deadlock. The thread will never be halted. Then I came up with this, but I am not sure if it is correct: public class MyThread extends Thread { protected volatile boolean running = true; public void run() { while (running) { synchronized (someObject) { while (someObject.someCondition() == false && running) { try { someObject.wait(); } catch (InterruptedException e) { e.printStackTrace(); } } // do something useful with someObject } } } public void halt() { running = false; synchronized(someObject) { interrupt(); } } } Is this correct? Is this the most common way to do this? This seems like an obvious question, but I fail to come up with a solution. Thanks a lot for your help.

    Read the article

  • PHP Object Access Syntax Question with the $

    - by ImperialLion
    I've been having trouble searching for this answer because I am not quite sure how to phrase it. I am new to PHP and still getting my feet on the ground. I was writing a page with a class in it that had the property name. When I originally wrote the page there was no class so I just had a variable called $name. When I went to encapsulate it in a class I accidental changed it to be $myClass->$name. It tool me a while to realize that the syntax I needed was $myClass->name. The reason it took so long was the error I kept getting was "Attempt to access a null property" or something along those lines. The error lead me to believe it was a data population error. My question is does $myClass->$name have a valid meaning? In other words is there a time you would use this and a reason why it doesn't create a syntax error? If so what is the semantic meaning of that code? When would I use it if it is valid? If its not valid, is there a reason that it doesn't create a syntax error?

    Read the article

  • C++ design question, container of instances and pointers

    - by Tom
    Hi all, Im wondering something. I have class Polygon, which composes a vector of Line (another class here) class Polygon { std::vector<Line> lines; public: const_iterator begin() const; const_iterator end() const; } On the other hand, I have a function, that calculates a vector of pointers to lines, and based on those lines, should return a pointer to a Polygon. Polygon* foo(Polygon& p){ std::vector<Line> lines = bar (p.begin(),p.end()); return new Polygon(lines); } Here's the question: I can always add a Polygon (vector Is there a better way that dereferencing each element of the vector and assigning it to the existing vector container? //for line in vector<Line*> v //vcopy is an instance of vector<Line> vcopy.push_back(*(v.at(i)) I think not, but I dont really like that approach. Hopefully, I will be able to convince the author of the class to change it, but I cant base my coding right now to that fact (and i'm scared of a performance hit). Thanks in advance.

    Read the article

  • C++ design question, container of instances and pointers

    - by Tom
    Hi all, Im wondering something. I have class Polygon, which composes a vector of Line (another class here) class Polygon { std::vector<Line> lines; public: const_iterator begin() const; const_iterator end() const; } On the other hand, I have a function, that calculates a vector of pointers to lines, and based on those lines, should return a pointer to a Polygon. Polygon* foo(Polygon& p){ std::vector<Line> lines = bar (p.begin(),p.end()); return new Polygon(lines); } Here's the question: I can always add a Polygon (vector Is there a better way that dereferencing each element of the vector and assigning it to the existing vector container? //for line in vector<Line*> v //vcopy is an instance of vector<Line> vcopy.push_back(*(v.at(i)) I think not, but I dont really like that approach. Hopefully, I will be able to convince the author of the class to change it, but I cant base my coding right now to that fact (and i'm scared of a performance hit). Thanks in advance.

    Read the article

  • C++ design question, container of instances and pointers

    - by Tom
    Hi all, Im wondering something. I have class Polygon, which composes a vector of Line (another class here) class Polygon { std::vector<Line> lines; public: const_iterator begin() const; const_iterator end() const; } On the other hand, I have a function, that calculates a vector of pointers to lines, and based on those lines, should return a pointer to a Polygon. Polygon* foo(Polygon& p){ std::vector<Line> lines = bar (p.begin(),p.end()); return new Polygon(lines); } Here's the question: I can always add a Polygon (vector Is there a better way that dereferencing each element of the vector and assigning it to the existing vector container? //for line in vector<Line*> v //vcopy is an instance of vector<Line> vcopy.push_back(*(v.at(i)) I think not, but I dont really like that approach. Hopefully, I will be able to convince the author of the class to change it, but I cant base my coding right now to that fact (and i'm scared of a performance hit). Thanks in advance.

    Read the article

< Previous Page | 35 36 37 38 39 40 41 42 43 44 45 46  | Next Page >